The more things change: economic institutions and maritime trade in the Archaic Western Mediterranean

PDF

Download

Open document

Transcript (if available)

Content The More Things Change: Economic Institutions and Maritime Trade In the Archaic Western Mediterranean Paul W. Salay, Jr. Faculty of the USC Graduate School The University of Southern California A dissertation submitted for the degree of: Doctor of Philosophy (Classics) May 2018 ii Parentibus meis: Et terra sit levis vobis iii Table of Contents Acknowledgements ...................................................................................................................... vi Notes on the Text ........................................................................................................................ viii List of Figures ................................................................................................................................ x Introduction ................................................................................................................................ xiii Section One .................................................................................................................................... 1 Chapter One: Towards a New Theoretic Approach .............................................................................1 Where We Are Now ...............................................................................................................................1 Ancient Economies? ............................................................................................................................14 Chapter Two: Ancient Economics? ......................................................................................................16 How Did We Get Here? .......................................................................................................................17 Conclusions ..........................................................................................................................................34 Chapter Three: The Role of Institutions .............................................................................................47 Section Two: The Literary Evidence ......................................................................................... 85 Introduction ............................................................................................................................................85 The Beginnings of the Archaic Period .................................................................................................85 Homeric Questions .............................................................................................................................101 Chapter Four: The Homeric Epics .....................................................................................................104 Introduction ........................................................................................................................................104 The World of Homer ..........................................................................................................................106 Economic Activity in Homer .............................................................................................................119 Conclusions ........................................................................................................................................159 Chapter Five: Hesiod ...........................................................................................................................164 Introduction ........................................................................................................................................164 iv The World of Hesiod .........................................................................................................................170 Economic Activity in Hesiod .............................................................................................................185 Conclusions ........................................................................................................................................201 Chapter Six: The Theognidea .............................................................................................................212 Introduction ........................................................................................................................................212 Theognis and Institutions ...................................................................................................................232 What Does Theognis Say? .................................................................................................................235 What Does It Mean? ..........................................................................................................................238 Conclusions ........................................................................................................................................241 Section Three: The Archaeological Evidence ......................................................................... 246 Introduction ..........................................................................................................................................246 Chapter Seven: Shipwrecks as a Proxy for Maritime Trade Growth in the 6 th c. BCE ...............250 Introduction ........................................................................................................................................250 Ancient Greek Naval Architecture .....................................................................................................251 Greek Ship Construction of the 6 th c. BCE ........................................................................................263 Cargo Evidence ..................................................................................................................................277 Conclusions ........................................................................................................................................283 Chapter Eight: Pottery Distribution as an Indicator of Developing Trade Networks ..................291 Introduction ........................................................................................................................................291 The Project Outline ............................................................................................................................294 The Preliminary Picture Visible in the Ceramics ..............................................................................305 The Data .............................................................................................................................................307 Conclusions ........................................................................................................................................328 Conclusions ...........................................................................................................................................333 Introduction ........................................................................................................................................333 v The Analysis ......................................................................................................................................339 Final Thoughts ...................................................................................................................................366 Figures ........................................................................................................................................ 373 Bibliography .............................................................................................................................. 399 vi Acknowledgements I suspect that with any project of this size and scope, the final product is the end result of a collaborative effort much more so than an individual achievement. At the very least, I can say with all certainty that this has very much been the case for me. It may be my name on the title page, but the list of those who deserve to be recognized, and who have my undying gratitude and appreciation, is long indeed. At the risk of being self-indulgent, I would like to take this opportunity to thank those who have been instrumental in offering their support and encouragement at so many points along the way. Of course, it is a virtual guarantee that I will leave some deserving individuals out, so I apologize in advance for any omissions. First and foremost, I must give thanks to my parents, both of whom sadly passed away during the writing process. Their support has been worth more to me than I could ever express, and it saddens me that I could not share this milestone with them. I am certain that they would have actually enjoyed it more than I. Thanks also to the support of my family, especially my brother Chris, sister-in-law Stephanie, and nephew Cary. Likewise to my dear friend Roy Sahachaisere, who encouraged me and supported me in more ways that I can count and that he even realizes. Of course, I owe a great debt to the members of my committee who, without any compensation or complaint, carefully read the manuscript and offered many thoughtful suggestions and insightful comments: Lynn Dodd, Vincent Farenga, and especially Alain Bresson, who has been a great source of inspiration for me professionally and has always been kind and generous with his time. Most of all, though, credit must be given to Greg Thalmann, my dissertation supervisor, who has spent innumerable hours reviewing chapters, offering suggestions and edits, and providing counsel. The dissertation would have been finished much sooner, and the process gone much more smoothly, had I simply swallowed my ego and followed his advice at every turn. In addition, I wish to recognize a pair of mentors who passed away recently: first, David Hood who was a constant source of support, and second, the irrepressible Gene Cooper, whom I shall miss as much for his sharp wit and humor as for his ceaseless encouragement vii and tireless intellect. My thanks also to David Tandy, whom I admired for his scholarship long before I was lucky enough to meet him in person and now call friend, and Archer Martin, who has forgotten more about pottery than I will ever know. I also count myself lucky to have been able to study at the University of Southern California. The Department of Classics has provided an exceptionally supportive environment in every way. The atmosphere has always been collegial in the best sense, and all the faculty have done their utmost to guide their students intellectually, personally, and professionally. In addition, the university has been more than generous in providing all the resources that one could possibly expect in order to ensure the greatest chance of success. Finally, there is a laundry list of institutions that have provided various resources, from financial support like the Wenner-Grenn Foundation and the USC Del Amo Fund, to research facilities and access to materials. In particular, the staff at the Centre d’Arqueologia Subacuàtica de Catalunya, including their director Gustau Vivar and librarian Josep Llorens, have gone above and beyond in allowing me access to their materials and facilities. Also, the library at the American Academy in Rome has been a place of refuge and a great resource, as have been the libraries at the Museu d’Arqueologia de Catalunya in Barcelona and the Biblioteca de Catalunya. For anyone I have omitted, my sincere apologies, and for everyone who helped make this dissertation a reality, I hope that I have done you proud. viii Notes on the Text Before venturing further, the reader should be aware of several features of the text that will (I hope) make it somewhat easier to navigate. First of all, as those who are familiar with ancient Greek are no doubt aware, there are certain letters of the Greek alphabet that have no direct analogue in the Latin alphabet used in English. Over time, it has become more popular to attempt to transliterate Greek words as closely as possible into English, so that for example the Greek word “πρόµαχος” that means “one who fights in the front ranks” is often written in English as “promakhos.” All things being equal, I prefer this approach, and for the most part, when transliterating generic Greek vocabulary this is the method which I have chosen to employ. However, there is also a long-standing practice of using the system by which ancient Roman writers transliterated Greek into Latin, so that “πρόµαχος” is written as “promachus,” etc. At the same time, many people are much more familiar with the Latin-ized version of Greek names in the Iliad and the Odyssey, so that they more readily recognize “Achilles” instead of “Akhilleus,” for instance. As a result, I have chosen to retain the more traditional spellings of Greek personal and place names, so that “Εὔµαιος,” the loyal swine-herd of Odysseus, is written as “Eumaeus” rather than “Eumaios.” In a similar way, I have preferred “Ajax” to “Aias,” “Ithaca” to “Ithaka,” “Telemachus” to “Telemakhos,” and so on. Also, I have chosen to cite passages from the Iliad and the Odyssey in a slightly indiosyncratic way. The traditional method for citation of the Homeric epics is by book and line number. So, Book 13, line 360 of the Odyssey is written as “Odyssey, 13.360.” Meanwhile, the same line from the Iliad is written as “Iliad 13.360.” This is all well and good, but when citing dozens of lines of texts from both sources, I have sought a simpler way to distinguish one from the other. The simplest solution, or so it seemed to me, was to cite passages from the Iliad using Arabic numerals for the book number and passages from the Odyssey using Roman numerals. Thus, Book 13, line 360 of the Iliad is cited as 13.360, while the same combination from the Odyssey is cited as XIII.360. To be perfectly honest, this practice was adopted more for my own convenience than anything else, and I hope that it does not cause any confusion. For the other ix texts, including the Theognidea, the Works and Days, and the Theogony, I have followed traditional practice. Finally, as for the translations of the Greek passages themselves, I wish to make clear that the English translations are my own. I take full responsibility for any errors, omissions, misinterpretations, etc. x List of Figures Figure 1: Reconstruction of a pegged mortise-and-tenon system utilizing a "false" or "loose" tenon; from Casson (1981). ...................................................................................................................................373 Figure 2: Hull remains of Uluburun wreck (late 14 th c. BCE), demonstrating the use of pegged mortise- and-tenon joinery; from Pulak (2003) ................................................................................................374 Figure 3: Reconstruction of tetrahedral notches and tension pegs typical of Greek sewn-plank ship construction; this feature can be seen in the Jules-Verne 9 and the Giglio, Bon Porté, Pabuç Burnu, and Cala Sant Vicenç wrecks; from Polzer (2009). ...........................................................................375 Figure 4: "Crow's foot" stitching pattern, which in tandem with the distinctive tetrahedral notches, works to inhibit lateral displacement; from Polzer (2009). ..........................................................................375 Figure 5: Illustration of construction of the Cala Sant Vicenç wreck, showing the typical stitching pattern, trapezoidal crenelated frames, hook scarf between frame and futtock, and coaks in between strakes; from Nieto & Santos (2008b). ...........................................................................................................376 Figure 6: Illustration of various architectural elements and construction aspects of wooden vessels. Clockwise from top left: keel, stem, and sternpost; keel in profile joined to strakes with pegged mortises and tenons, and hull joined to floor-timber/frame with treenails and metal nails; hook scarf; detail of strakes joined with pegged mortises and tenons; from Steffy (1994). ................................376 Figure 7: Architectural elements of a typical ancient wooden vessel, including strakes/planking, garboard, half-frames, floor timbers, and futtocks; from Steffy (1994). ...........................................................377 Figure 8: Sites of shipwrecks from 6th-4th c. BCE discussed in the text, indicating relative dates and method of construction; sizes of icons are proportional to the estimated length of the vessel; map from the author. ..................................................................................................................................377 Figure 9: Illustration of presumed interlocking trade routes connecting the Aegean to the western Mediterranean in the 6th c. BCE, based on cargoes from contemporary shipwrecks, and indicating the excavated sites (Salento, Tarquinia, Emporion) which provided the pottery analyzed in the text; from the author. ..................................................................................................................................378 Figure 10: Basic developmental timeline of the transition of Greek wooden ship-building from sewn- plank construction to pegged mortise-and-tenon joinery, 6th-4th c. BCE; from the author. ............378 Figure 11: Distribution chart of pottery from Oria (all periods) ................................................................379 Figure 12: Distribution chart of pottery from all areas of Salento except Oria (all periods) .....................379 Figure 13: Distribution chart of pottery from all areas of Salento, including Oria (all periods) ...............380 Figure 14: Time-series distribution of Salento pottery by pottery class, 650-500 BCE ............................381 Figure 15: Salento pottery classes as a share of the whole, 650-600 BCE ................................................382 Figure 16: Salento pottery classes as a share of the whole, 600-550 BCE ................................................382 xi Figure 17: Salento pottery classes as a share of the whole, 550-550 BCE ................................................383 Figure 18: Time-series distribution of Salento pottery by provenience, 650-550 BCE ............................384 Figure 19: Total numbers of Black Gloss Pottery found at Salento, by sherd count, 650-500 BCE .........385 Figure 20: Attic Pottery found at Salento by pottery class, 650-500 BCE ................................................385 Figure 21: Breakdown of the distribution of specific forms among the different pottery classes found at Salento, 650-500 BCE .......................................................................................................................386 Figure 22: Ancient sites at Salento where pottery forming the database was excavated; map by author, data from Semeraro (1997) ................................................................................................................387 Figure 23: Distribution of Tarquinia Pottery by Functional Group ...........................................................388 Figure 24: Time-series of Tarquinia pottery by functional group, 700-500 BCE .....................................388 Figure 25: Tarquinia fine wares by class, 700-500 BCE ...........................................................................389 Figure 26: Tarquinia Fine Wares by Class, 700-650 BCE ........................................................................389 Figure 27: Tarquinia Fine Wares by Class, 650-600 BCE ........................................................................390 Figure 28: Tarquinia Fine Wares by Class, 600-550 BCE ........................................................................390 Figure 29: Tarquinia Fine Wares by Class, 550-500 BCE ........................................................................391 Figure 30: Time-series data of Tarquinia fine-ware pottery by class, 700-500 BCE ................................392 Figure 31: Time-series data of the distribution of pottery from Emporion by functional group, 650-500 BCE ....................................................................................................................................................393 Figure 32: Distribution of pottery from Emporion by functional group, 650-625 BCE (Phase IIA) ........394 Figure 33: Distribution of pottery from Emporion by functional group, 625-580 BCE (Phase IIB) ........394 Figure 34: Distribution of pottery from Emporion by functional group, 580-560 BCE (Phase IIIA) .......395 Figure 35: Distribution of pottery from Emporion by functional group, 560-540 BCE (Phase IIIB) .......395 Figure 36: Pottery distribution from Emporion by functional group, 540-520 BCE (Phase IIIC) ............396 Figure 37: Distribution of pottery from Emporion by functional group, 520-500 BCE (SU 5056) ..........396 Figure 38: Distribution of pottery from Emporion by provenience, 650-580 BCE (Phase II) ..................397 Figure 39: Distribution of pottery from Emporion by provenience, 580-560 BCE (Phase IIIA) ..............397 Figure 40: Distribution of Pottery from Emporion by provenience, 560-520 BCE (Phase IIIB and IIIC) ............................................................................................................................................................398 xii Figure 41: Distribution of pottery from Emporion, 520-500 BCE (SU 5056) ..........................................398 xiii Introduction In recent years, interest in the phase of Greek history known as the Archaic period has gained newfound interest and attention. 1 Long regarded simply as a transitional phase, its considered importance was largely based on the formation of nascent social and political structures that emerged fully during the Classical period. More recently, though, a greater appreciation of the Archaic period as a dynamic phase of Greek history, important for its own sake and unique in its particulars, has emerged. Yet even while so much scholarly attention has been devoted in recent years both to the historical significance of the period as a whole as well as to specific developments and currents, one aspect of the Archaic period remains underappreciated: the expansion of maritime trade from the time of the Homeric epics in the late 8 th c. BCE to the end of the 6 th c. BCE. This gap in the academic literature is particularly ironic in light of the resurgent interest in questions of ancient economic activity. 2 At a time when ancient historians are dedicating so much energy to the Archaic period in general and various aspects of archaic society and politics in particular, all while there is a concurrent renewal of interest in ancient economic behavior, new scholarly treatments of archaic commerce have been conspicuous in their absence. Several possible explanations exist for this situation. For instance, in the wake of Finley’s Ancient Economy (1973), debate over pre-modern economic behavior within the context of the ancient world was largely stalled. His larger thesis was that modern 1 This can be seen, for instance, in the appearance of several edited volumes such as Fisher & van Wees (1998), Shapiro (2007), and Raaflaub & van Wees (2009), as well as numerous works on various particular aspects of the period, including the development of the polis (Mitchell & Rhodes, 1997), trade (Möller, 2000), law (Papakonstantinou, 2008), colonization, (Tsetskhladze, 2008), networks (Malkin, 2011), class (Rose, 2012), and gender (Eaverly, 2013), among others. 2 The most notable illustration of this renewed interest is the recent publication of the Cambridge Economic History of the Greco-Roman World (2007), along with the edited volumes by Manning & Morris (2005) and Scheidel & von Reden (2002); on the Greek side there are also various published works by Bresson (2000, 2007, 2008, 2015), Moreno (2007), and Engen (2010), among others, while on the Roman side there has been an explosion of scholarship, visible in part by the growth of the Oxford Roman Economy Project with an excellent collection of bibliographies on various aspects of Roman economic activity on their website (http://oxrep.classics.ox.ac.uk/bibliographies/). xiv commerce was of a fundamentally different nature than pre-modern exchange (which finds no analogue in the ancient world), and therefore could not be understood by or interpreted through the lens of modern economic theory that was formulated specifically to analyze the historically unique forms of modern economic activity. This approach was so influential that it had the effect of curtailing economic questions about the ancient world for several decades. However, the repressive influence of Finley does not explain the continued relative lack of interest in the subject of archaic trade when, as we have already noted, there has been a surge of interest both in the period and in the topic of ancient economic behavior. I believe that this gap in the scholarship is due to the fact that the two most influential historians who have written about archaic Greek economic activity, Walter Donlan and David Tandy, have largely followed Finley’s lead in denying the significance of market exchange in the period (Donlan in particular), and together they have greatly shaped the debate. Still, merely the existence of a gap in the scholarship is not in and of itself sufficient reason to promote an intensive study. It is a necessary, but not sufficient cause. Instead, the research question should be able to provide an additional level of insight into the degree and nature of our understanding of the period. Herein lies the crux of my investigation, and where I depart from the arguments of Tandy and Donlan. For Donlan (1999), the Archaic Period produced sweeping changes in the Greek world, including demographic, political, and ideological changes in Greek society. For him, the level and even the nature of trade also evolved, but this was a function of the broader transformation of Greek society, a symptom rather than a formative element. For Tandy (1997), the nature of commerce was a powerful influence in the course of Greek socio-political development, but trade was carefully controlled by archaic elites who were eager to reap the profits from rising levels of market exchange while obscuring the reality of its existence. For these elites, the “disembedded” nature of market exchange was contrary to prevailing norms of social obligation, and therefore must be kept hidden and tightly controlled. xv In broader terms, I agree strongly with the insights of Tandy, with one critical exception: rather than constituting a source of income predominantly for archaic elites, I believe that market exchange, and particularly the expansion of maritime trade, provided an unprecedented opportunity for non-elites to amass personal wealth. Whereas Tandy believes that archaic elites were motivated to disguise their own profits generated through trade, I believe that they were desperate to repress market exchange because it constituted a means of wealth generation that allowed non-elites to enrich themselves at the expense of traditional elites, which in turn served as a foundation which they could leverage in order to effect political and social changes. In order to make this broader argument, I have divided the project into three main sections: first, a introductory section in which I outline a theoretic approach that I believe is most effective for understanding the changing dynamics of economic behavior in antiquity more broadly, and during the Archaic Period more specifically. To make this case, I will first explain what I see as the major failing of most economic analysis of the ancient world to this point, namely the erroneous assumption that orthodox, or neo-classical, economic theory is synonymous with modern economic theory more generally. Consistent with this premise, I will briefly discuss the major tenets of orthodox theory, pointing out both why this body of thought has experienced such popularity, as well as why its underlying assumptions are inconsistent with the goals of ancient economic history. In order to understand this dynamic, as well as to get beyond the sterilizing influence of the primitivist-modernist dyad, I suggest utilizing an approach that takes advantage of the New Institutional Economics (NIE) pioneered most prominently by Douglass North. What NIE emphasizes is the influence of institutions, or “rules of the game,” that shape collective activity. For North and other neo-institutionalists, a major failing of modern neo-classical economic theory is that it either ignores or takes for granted an enormous range of institutions and enforcement arrangements that profoundly shape the ways in which trade takes place. This body of thought is critical for our investigation, because it provides a framework for analyzing xvi economic activity without lapsing into the binary choice between whether or not ancient forms of exchange were either wholly like or wholly unlike modern exchange. In the second section, I analyze the main literary sources for the period, which are the Homeric epics of the Iliad and the Odyssey, the poetry of Hesiod, especially the Works and Days, and finally the corpus of elegiac poetry attributed to Theognis of Megara. In these sources, we can detect an ongoing struggle for social standing and legitimacy which is frequently played out in economic terms. More specifically, I suggest that different social groups during the period engaged in an ideological struggle over the establishment of norms of economic behavior that were conducive to their respective interests, with archaic elites advocating symbolic forms of gift-exchange that both validated their elevated social standing and limited access to others. At the same time, non-elite free landowners favored producing goods specifically for export, distribution via maritime trade routes, and forms of trade that were explicitly aimed at generating and maximizing profits. In the terminology of NIE, this is a struggle over the establishment of prevailing rules, or “informal” institutions. As North (2000) points out, institutions are not ideologically neutral, but rather they are products of human agency, and therefore they frequently take forms that are amenable to those who have the power to establish and maintain them. In the third section, I analyze the archaeological evidence for the period, specifically shipwrecks and pottery. The shipwrecks admittedly represent a small sample size, and this is further complicated by the fact that there are no wrecks in the western Mediterranean securely assigned to dates earlier than the 6 th c. BCE. Nevertheless, they constitute an important source of information for our understanding of maritime trade, both through an examination of the naval architecture of the period and their cargoes. On the basis of their cargoes, we can see evidence for different forms, or levels, of maritime trade: small-scale, localized distribution known as “cabotage”; a middle-range of distribution utilizing nodal distribution points that Nieto (2008) has labeled “emporic”; and a third level of point-to-point distribution consisting of larger, much more homogeneous cargoes that by their very nature must have relied on extended distribution networks. This picture is contrary to the traditional view that ancient seaborne trade, xvii especially at such an early date, was limited exclusively to coastal-hugging cabotage. Also, it is clear that the 6 th c. BCE was a period of rapid change in the construction techniques of Greek wooden ships, which I believe can only be adequately explained as a result of demands to produce larger vessels with greater displacement, capable of traversing longer distances of open ocean and carrying larger cargoes. Indeed, the 6 th c. BCE witnessed a profound and dramatic change in Mediterranean wooden shipbuilding that would see few, if any, parallels for nearly two thousand years, 3 and I suggest that this shift must have been motivated by the intensifying pressures resulting from rapidly increasing levels of maritime commercial activity. Both the increased levels of trade and developed distribution networks that are suggested by the shipwrecks can be seen in the pottery evidence. By conducting a statistical analysis of entire ceramic assemblages from three different sites in the western Mediterranean (Salento, Tarquinia, Ampùries), I find clear evidence for substantial increases in both the volume and variety of imported pottery at all three sites. Also, there are sufficient differences in the composition of the assemblages to indicate the influence of regional consumer preferences, contrary to the supply-side depiction of imported Greek pottery. In spite of various difficulties resulting from the challenges of collating data from different published sources, rudimentary statistical measurements, and finding sufficient equivalencies for cross-comparison, I believe that the data clearly show the effects of sophisticated, integrated networks capable of distributing high volumes of trade goods across large geographic areas, accommodating divergent consumer preferences, and involving frequent contact and repeated exchanges over time, all of which are consistent with the conclusions drawn from the shipwreck evidence. 3 The only other change in ancient wooden shipbuilding in the Mediterranean that could conceivably rival the shift in Greek shipbuilding from sewn-plank to pegged mortise-and-tenon construction might be the transition to frame- first from shell-first construction that probably began primarily during the 5 th c. CE, but was almost certainly completely underway by the 6 th c. CE; see McGrail (2013), Pomey, Kahanov & Rieth (2012). xviii By combining these various sources into our analysis, we can see clearly that commercial activity during the 7 th and 6 th centuries BCE was highly contested, with elites engaged in a protracted contest against non-elite free landowners to legitimize opposing forms of exchange that benefited their respective advocates. The traditional landed nobility sought to valorize a ritualized form of gift-giving that protected their status while limiting opportunities for non-elites, while free farmers opportunistically sanctioned a form of profit-driven trade that gave them the financial footing from which to challenge the existing social hierarchy. The material evidence stands as mute witness to the rising levels of maritime commerce and the networks that facilitated it, whereas the literary sources demonstrate the degree to which both elites and non-elites clearly understood the connections between symbolic, cultural, and financial capital, as well as the potential for the introduction and redistribution of wealth to destabilize the traditional social order. Within this context, we must now recognize that the emergence of seaborne trade was not a side effect of larger political, demographic, and social changes, but rather an important potential source of power, and different groups within archaic Greek society consciously and deliberately sought to manipulate the institutions that regulated it for their own interests. Ultimately, archaic elites were unable to repress the growth of maritime trade, while non-elites seized on the opportunities that it presented to amass personal wealth that they were able to leverage in a struggle for social and political change. Thus, the growth of market exchange during this period was a powerful tool for the transformation of Greek society, and no history of the Archaic Period would be complete without a fuller appreciation of its importance. Section One Chapter One: Towards a New Theoretic Approach Where We Are Now It is not an overstatement to claim that the debate over the size and nature of “the economy” in the ancient world has been one of the more vexing issues and contentious subjects in the field of classical studies for at least the last several decades. Nearly thirty years ago, Keith Hopkins colorfully and provocatively described the ancient economy as “an academic battleground,” in which “contestants campaign under various colours – apologists, Marxists, modernizers, primitivists…these categories are neither mutually exclusive, nor internally united…But no new weapon is lethal; and none of the battles is finally decisive…” (Hopkins, 1983). Unfortunately, one of the by-products of this debate has been to limit discussion to a choice between two opposing poles. In the early stages of the discussion, lines were drawn between primitivists on the one hand, and modernists on the other. Broadly speaking, primitivism denies that the pre-modern economy ever reached a stage of development such that it could be considered to resemble the modern market economy, and consequently was not a subject of fruitful study using the tools of modern economic theory and analysis, whereas modernism (at least for our intents and purposes) argues that at various points in human history, the fundamental elements of a so-called modern market economy were in place, such as banking systems, mechanisms of lending and credit, and most importantly, price-making markets, and that the only salient difference between these earlier economies and that of the modern world was one of degree. Since then, these terms have been refined and changed, with many now characterizing the debate as one between formalists and substantivists (Morris, 1999b), or more recently, continuationists and transformationists (Denemark, Freidman, Gills, & Modelski, 2000), but the starkly binary nature of the terms of the debate remains constant, restricting the discussion over the nature of the ancient economy to an almost exclusively dyadic choice. In the words of Douglass North, this polarizing dyadic interpretive framework “has led to one of the more sterile debates in history…” (1981). 2 I would suggest that it is important to understand the numerous factors that have influenced the course of the discussion, including the historical background of the debate, the fundamental assumptions that lie at the heart of the respective positions, and some of the different ways that the economics of the ancient world have been studied from different perspectives and disciplinary approaches. As is so often the case, the basic assumptions that support each position have become, over time, submerged beneath layer after layer of successive debate, with the result that they continue to exert tremendous influence, although (or because) they are relied upon so tacitly. 4 This of course is precisely one of the points made by Polanyi in describing the position of the advocates of economic liberalism, who (according to his argument) have formulated their position in such a way that there is no possible outcome that would be regarded as sufficient evidence to overturn the fundamental assumptions that support the argument (Polanyi, 1957a [1944]). The persistence of this unshakeable faith in some of the core principles of orthodox economic theory continues to be well chronicled by critics of the neo-classical economic tradition (Arnsperger & Varoufakis, 2006). Yet, others have demonstrated that even under optimal conditions, some of the most central assumptions of mainstream economics that are fundamental to the ostensible predictability of economic decision making are not supported by the available evidence (Romer, 2006). Second, I would argue that another factor contributing to the problematic nature of the debate concerns the frequent lack of rigor in delineating precisely what constitutes the realm of economics. The very term “economic” is all too often thrown about carelessly under the apparent assumption that the definition of the term is self-evident. However, as Belshaw (1965) for one notes, there is widespread disagreement 4 Despite the informational value of such an understanding, this is not the place for a thorough review of the development of economic thought from its 18 th /19 th c. foundations with Smith, Mill, Ricardo, and others, to the various debates of the 20 th c., including the early institutional economics of Veblen, Keynesian thought, the post-war quantitative shift, and the growth of monetary policy and the response of the New Keynesians (which still continues to the present). It will have to be enough for our present purposes to accept the basic premise that for the most part (with early institutional economics standing as an exception), modern economic theory can trace its intellectual heritage along a fairly continuous developmental progression, and that it still rests firmly on a set of entrenched assumptions that follow a direct line of descent from Smith and other early thinkers. 3 over just what phenomena belong under the heading of “economic.” 5 It is precisely this terminological difficulty that Polanyi noted in the writings of Menger, among others (1971), and which prompted him to formulate his distinction between the “formal” and “substantive” types of economic activity. Furthermore, I would suggest that our notions of economics are strongly culturally constituted, and that ethnocentric conceptions of the economic sphere and the nature of economic behavior are often applied to cases that would seem to demand a more culturally specific approach. However, because these concepts are so familiar to us, the degree to which they are products of a particular cultural environment is underappreciated, and they are therefore imbued with an air of universal applicability that is inappropriate given the growing body of evidence. Third, it is a central feature of mainstream economic theory that it seeks to eliminate the effects of society and culture from its analysis. This is of course a major point of contention among substantivist critics, who observe that this position depends upon a universalist view of human nature as hedonic, motivated by self-interest and dedicated to the application of rational calculation in order to maximize individual utility. 6 Admittedly, this concept is an abstraction which is designed to facilitate reliable economic modeling, and it is certainly true that mainstream economic theory has shown itself to be the most effective predictive tool for economists. However, when the concept of the “economic man” is tested in different cultural contexts, the validity of this model appears much more limited and culturally specific than its supporters would claim (Henrich, et al., 2001, 2005; Sahlins, 1999). 5 For example, some like Marshall employ the term very expansively (1890 [1920]), whereas others like Pigou argue for a use of the term that applies to those activities that can be measured and related to money, or Boulding who considers the economic sphere as encompassing the system of production, distribution and consumption (Boulding, 1948). 6 This view of human nature, often referred to as homo economicus, has its roots in the earliest classical economic writings, including J.S. Mill (1848) and Adam Smith, who famously asserted, “The propensity to truck, barter and exchange one thing for another is common to all...” (The Wealth of Nations, 2003 [1776]). This conception of the rational, self-interested agent is one that continues to be a central (and largely unquestioned) tenet of orthodox economic theory to the present. 4 Still, the mainstream economic approach continues to thrive. As a result, I wish to make several points about the nature of orthodox economic thought, for several reasons – first, to understand more completely precisely what are some of these assumptions; second, how these are problematic, with a particular view from the perspective of ancient history; and third, why neo-classical thought has continued to occupy a place of such dominance within the discipline of economics. To do so, I wish to emphasize several important points: The term “economic” has too often been carelessly applied, in a myriad of contexts, and with diverse meanings. Seemingly from the beginning, there has been a struggle to find a satisfactory definition of the economic sphere, with mixed results. In the case of modern economics, the classical theorists like Smith, Ricardo and Mill adopted a more inclusive sense of the term that accounted for the effects of social and political institutions on individual behavior. Since at least the end of the 19 th century with the work of Alfred Marshall (1890) and the work of the marginalists like Jevons, Walras, and Menger, the discipline has been characterized by a more scientific approach than the work of the classical economists whose approach was grounded in moral philosophy. This emphasis on the scientific quality of economic analysis tended to marginalize or even ignore completely the effects of culture on individual agents, who instead were presumed to operate according to universal motives of maximization, and led ultimately to the quantitative, mathematical-deductive formalism of the 1950s and beyond. In each case, the topic of study has been “economics,” but how narrowly or broadly that can be conceived has often changed. For instance, most recently, economists have begun to shift their focus away from the impersonal and mechanical processes of production, consumption and distribution towards an approach that favors behavioral considerations of decision making and the framing of alternative choices At the risk of burying the lead, this is a crucial point for the sake of my larger argument. All too often, economic 5 analysis and economic history take as a given that the mechanical processes of economic activity are unconnected to rules that govern collective behavior.. For this reason, I believe that it is imperative from the outset to establish a clear and explicit definition of economics, first and foremost, and at the same time to recognize that when the term is used by others, it is often informed by very different assumptions. Consequently, I have chosen to use Lionel Robbins’ definition of economics as the field that studies “human behavior as a relationship between ends and scarce means which have alternative uses…” (Robbins, 2007 [1932]), primarily because Robbins’ definition is more inclusive than some others, and therefore provides ample scope to employ a comparative approach. There is nothing in Robbins’ definition that in any way requires the kind of maximizing, economic rationality with which it is so often associated. Certainly, it is not incompatible with the hedonic homo economicus of Mill, and in fact would seem to imply a certain degree of maximizing behavior, but there is nothing that makes such a connection intrinsic. Robbins’ focus is on choice, and therefore is just as applicable to a model of human nature as constrained by institutions as to a model of man as a rational maximizer. Mainstream economic theory, often referred to as neoclassical or orthodox economics, maintains a strong connection to the classical economic precepts of Smith and Ricardo, among others. For obvious reasons, those who propose innovation are careful to be as explicit as possible regarding the ways in which they seek to break with earlier traditions. However, they are often silent when it comes to the ways that they accept traditional dogma. This phenomenon frequently occurs at an almost unconscious level, since many assumptions become accepted as unquestioned truth over time, but it does not negate the fact that these assumptions continue to require empirical testing in the face of new available evidence, and often what is taken as conventional wisdom is shown to be false in the face of such tests. I would argue that the various “revolutions” in economic thought, including the marginalism of Walras, Menger, and Jevons, the insights of Keynes and the later neoclassical synthesis, and the 6 formalism of the 1950s and beyond have produced significant changes in economic analysis, but have not fundamentally altered the core principles established by the classical economists. This is in spite of the fact that many innovators such as the early pioneers of marginalism saw themselves as constituting a distinct break from the classical tradition (Jaffé, 1976), and would almost certainly have taken issue with the neoclassical label coined by Veblen in 1900. It is also made problematic by the changes that took place following the Second World War, changes that produced economic analyses that would very likely have been unrecognizable to the orthodox economics of prior generations (Blaug, 1997). Of course, this is precisely the point, namely that over the course of time, with each succeeding generation and revolution, the underlying precepts have become submerged beneath successive layers of theory but continue to wield a powerful and pervasive influence on mainstream economic thinking (Blaug, 1996). Identifying these foundational tenets, however, is not nearly so straightforward. Weintraub introduces three principles of neoclassical economics: rational preferences, utility maximization, and perfect information (Weintraub, 2002), but as Arnsperger and Varoufakis point out, in many cases these do not constitute necessary conditions of mainstream economic analysis (Arnsperger & Varoufakis, 2006). Imperfect competition models, for instance, directly address the implications of asymmetric information, uncertainty, and other conditions directly contrary to the assumption of perfect information. Similarly, behavioral finance seeks to explore the effects of preferences and bounded rationality on economic decision making, 7 which at the very least place limits on the traditional notion of utility maximization. Still, there are others who point to entirely different assumptions that are made in mainstream economics. In particular, Douglass North points out a number of presumptions that are almost universally made in mainstream economic analysis. Unlike the schemas of Weintraub and Arnsperger & Varoufakis that focus on some of the more positivist tenets of orthodox theory, North instead highlights the phenomena that are 7 The foundational work on cognitive biases and economic decision making is that of Kahneman and Tversky (1973, 1975, 1979, 1984, 1985). 7 typically excluded from conventional economic models. In particular, he contends that modern approaches almost invariably make a number of assumptions: perfectly competitive markets as a way of ignoring the potential complications of market frictions; perfectly specified and costlessly enforced property rights as a means of eliding information and enforcement costs; neutral government; and unchanging tastes (North, 1981). In short, it is his view that orthodox theory presupposes the tendency towards equilibrium produced by competition, and to do so ignores the various factors that would inhibit the efficiency of the price mechanism. All told, I would argue that neoclassical economic theory is defined by a set of shared premises, of which it is sometimes possible to exclude or modify one or several, but never all of them. In addition to those identified by North above, these also include: • Atomism – In neoclassical analysis, it is almost unanimously presumed that the fundamental category of analysis is the individual agent, an axiom that Arnsperger and Varoufakis refer to as methodological individualism (Arnsperger & Varoufakis, 2006). Reliance on this assumption has been criticized, but remains a central feature of mainstream economic analysis (Kirman, 1989). • Equilibrium –Of course, the roots of this theory clearly go back to Smith and his concept of the invisible hand of the market, and the assumption of general equilibrium has been a standard feature of neoclassical analysis since the late 19 th century and continues to be a staple of mainstream economics. It is often criticized by detractors as a means of preserving the status quo, and general equilibrium theorists are frequently guilty of creating models that presuppose equilibrium as an ideal state as a way of examining forces of disequilibrium, a process that is dangerously tautological. Even prominent supporters of general equilibrium theory admit that the assumption of states that give no reason to suppose that they will ever actually occur “is easily convertible into an apologia for existing economic arrangements and it is frequently so converted…” (Hahn, 1984). 8 • Hedonic human nature/utility maximization – Since Smith, economists have based their models on the assumption that human beings can be relied upon to act in predictably self-interested ways. When this supposition has been challenged, it is often defended by expanding the notion of utility, with the end result that essentially any choice, including altruism or other self-sacrificing behavior, can be explained as a manifestation of the individual satisfaction of selfish motives, and as a result can frequently assume a form that is not subject to falsification. However, when tested even under the most optimal conditions, outcomes diverge from what would be expected from the maximization ideal (Romer, 2006). • Competition – Beginning with Smith and certainly reinforced by Malthus, economists have relied on the assumption that competition among economic agents is the engine that drives the economy towards equilibrium. However, the concept of unlimited wants that is necessary for the general principle to apply has often been criticized as a culturally constructed value. In addition, imperfect competition models have often produced examples where this assumption is in question. • Rational agency – Related to the atomistic tendency of neoclassical economics, forecasting models almost invariably depend to some extent on the predictability of economic agents, which must by necessity rely to some degree on rational expectation. This idea is well established by Smith, and found its fullest expression with J.S. Mill, who admitted that the model of homo economicus was an abstraction, but a valuable and defensible one that was true in general terms. Behavioral finance has been instrumental in introducing the notion of bounded rationality into a broader economic discourse, but even this only attempts to explain economic behavior in terms of constraints or limits on rational agency and does not fundamentally question the basic assumption. Ultimately, although neoclassical economists are more willing to admit the limits of rational agency, the fundamental presumption is still that economic agents are sufficiently rational to allow for the construction of predictive models. 9 The success of mainstream economic theory does not lie in its explanatory ability or the truth value of its premises, but rather in its predictive value. Mainstream economics, broadly conceived, strives to create predictive models that are by nature abstractions, and therefore must necessarily elide or ignore important elements of economic behavior in favor of efficiency and tractability. It is this aspect of neoclassical analysis that most likely explains the persistence of the assumptions set forth above. To be sure, it is not my goal to discount the achievements of mainstream economics. In fact, orthodox analysis has proven, time and again, to be an extraordinarily effective heuristic device, and far and away the most successful predictive approach. However, part of the explanation for this success may very well lie in the fact that the presuppositions of orthodox theory are much more historically and culturally contingent than is often believed. As North for one argues, “the legal structure of American society that had emerged by mid-nineteenth century was one that explicitly reflected the same efficiency criteria as that of neoclassical theory…It is not surprising that neoclassical economic theory emerged in this era…” (North, 1978). 8 Certainly, mainstream economic analysis has not remained perfectly insulated from contemporary historical events. It seems highly unlikely, for instance, that Keynesian theory would have developed along the lines that it did, or that it would have found such acceptance, were it not for the Great Depression. Nevertheless, modern mainstream analysis continues to rely on many of the fundamental premises that are part of its intellectual heritage from classical economic theory. As a result, it seems safe to say that it is at least possible, if not even probable, that neoclassical theory has increasingly diverged from actual real-world conditions over time, or perhaps more accurately, that real-world conditions have moved away from the theory. In fact, if one accepts the arguments put forth by Blaug (2003) and Lawson (2005), the practice of mainstream economics has moved significantly towards the abstract, giving ever 8 If we accept this basic premise, then mainstream economics simultaneously ignores institutions and relies upon their effects to produce incentive structures that foster behavior consistent with their models. 10 greater priority to the model itself, and less and less to the actual conditions that the models are ostensibly designed to represent. Indeed, this disconnect has become so great that no less a figure than Nobel prize winner Ronald Coase was moved to observe, “Mainstream economics…is in fact little concerned with what happens in the real world…” (Coase, 1998). For many neoclassical economists, this accusation causes surprisingly little concern. Although there are still many who, like Oliver Williamson, maintain that analytic models should strive to approximate as closely as possible empirically observable conditions (Williamson, 1989), there are others who reject this standard of measurement. For example, no less a figure than Milton Friedman has argued that the ability to predict outcomes should serve as the only standard by which economic theory should be judged. According to Friedman, “theory is to be judged by its predictive power for the class of phenomena which it is intended to explain…the only relevant test of the validity of a hypothesis is comparison of its predictions with experience…” (Friedman, 1953). Even setting aside the fact that by his own standard of success, the monetary policy approach advocated by Friedman failed abysmally to achieve the goal just described as witnessed by the global economic meltdown of recent years, it is quite clear that the emphasis of mainstream economics is vastly different from that of the historian. For Friedman, the construction of a predictive model is of paramount importance, and the explanation of actual events a distant second, and even then only significant in terms of how such explanation can refine and improve the efficacy of the predictive model. Consequently, it seems clear that the tools of mainstream economists have been fashioned for a much different purpose from the practice of history. It may very well have once been the case that the aims of orthodox economists and contemporary historians were closely aligned, but it seems clear that this is far from the case today, and as North points out, the divide between economic historians and professional economists continues to widen (North, 1976). In short, the priorities of the two groups are very different, and as a result, the respective analytic approaches have come to reflect that difference. Therefore, two things are clear: first, that neoclassical economic theory has consistently moved further from a reflection 11 of contemporary economic conditions, and the conditions themselves have substantially changed in ways that are increasingly inconsistent with the central propositions of neoclassical theory; second, that mainstream economists are less and less concerned with those analytic approaches that do not explicitly further their aims of predictive accuracy and creating more effective policy making tools. Mainstream economic theory is often treated as static and homogeneous, with universal agreement among economists regarding the theoretic foundations that form the basis of what is often simply referred to as “modern” economics. As we have seen, for much of its history, classical as well as neoclassical economics have embraced a significant degree of diversity. However, one of the distinguishing features of economic thought since the Second World War has been a growing trend towards homogenization, accompanied by a decreasing tolerance for divergent approaches. Already by 1955, Paul Samuelson could confidently assert that ninety percent of economists had adopted what was a largely unified theoretic approach (Samuelson, 1955). Since that time, the field has exhibited an “extraordinary uniformity in the global economics profession that we now characterize as neoclassical economics…” (Blaug, 2003). Also, this appearance of uniformity is often no mere accident, but rather a deliberate construct. For example, the formation of voluntary secondary school standards for education in economics that were formulated in the 1990s was conducted with an explicitly restrictive neoclassical agenda that self-consciously eliminated minority views (Cohn, 2003). Members of the writing committee claimed that the standards constituted “the fundamental propositions of economics…” and did so expressly by using “the majority paradigm…” (Siegfried & Meszaros, 1997). Granted, the standards were voluntary, and not designed for college-level instruction. Nevertheless, the project serves as an illustration of the view that the entire field of economics can (and in certain cases even should) be understood as a unified body of theory. Thus, it should not be surprising that among those outside the discipline, the field would be treated as internally consistent. However, this creates genuine obstacles to mutual intelligibility when comparing the 12 work of various writers who are separated by time. For instance, the so-called “Bücher-Meyer controversy” of the 1890s involved two diametrically opposing views, those of the primitivist Bücher and the modernist Meyer, with Bücher insisting that the economic behavior of the ancient world was drastically different from that of the modern world of the late 19 th century, and Meyer arguing the opposite. Of course, the “modern” economics of Bücher and Meyer, as we know, was not the same as those of Karl Polanyi in 1944, which in turn differed from Finley’s in 1973. Not only were the actual contemporary economic realities different in many ways for the various authors; the theoretic approaches were likewise highly diverse. Also, drawing a stark contrast between “modern” and “pre-modern” economic behavior is inherently reductive, failing to portray accurately both the diversity of the ancient world and that of the modern. Finally, this distinction often rests on a fundamental misconception, namely that economic theory is always an accurate facsimile of actual behavior. This has been demonstrated to be false, particularly in recent decades, but the outward uniformity of economics as a field of inquiry has had the tendency to inhibit the appreciation of heterodox approaches. At the same time, while numerous mainstream economists in recent years have decried the perceived growing disconnect between theory and real world behavior, many have explicitly rejected the notion that such accuracy is even a relevant gauge of the success of economic models, and therefore does not constitute a primary objective of the discipline. As a consequence of the homogeneity of the field, however, it has been common to conflate modern theory with modern behavior, under the belief that the economic models are reliable analogues for economic activity, whereas in fact many of these models are reflective of neither ancient nor modern practices. Many economic thinkers, particularly since the middle of the 20th century, have considered economics as a science. As has already been discussed, the so-called formalist revolution of the 1950s and beyond has produced an environment within economics that is overwhelmingly predisposed towards technical, formal models 13 that are frequently mathematical. This of course is far from a new development. In fact, many of the marginalist pioneers of the late 19 th century, Walras in particular, emphasized explicitly scientific approaches that relied heavily on quantitative analysis (Vaggi & Groenewegen, 2003). Also working at the end of the 19 th century, Alfred Marshall likewise sought to establish the scientific credentials of economics (Backhouse, 2006). While this has contributed to the sense of differentiation between the economic activities of the modern and pre-modern worlds described above, it has also contributed to the impression that the analytic tools of modern economics are inappropriate to study economic activity in other historical contexts. For instance, Finley rejects the application of modern economic theory to ancient practices at least in part on the grounds that the ancient world does not offer the kind of statistical data favored by modern economists (Finley, 1999 [1973]). By rejecting modern economic analysis wholesale because its statistical approach is ill-suited to pre-modern evidence, Finley both misses the opportunity to exploit some of the insights available from various heterodox approaches, and at the same time helps to perpetuate the false dichotomy between ancient and modern economic behavior. In sum, one of the obstacles to an analysis of pre-modern economic behavior has been a persistent misappropriation of neoclassical economic theory, in its various guises. So far, I have attempted to eliminate some of the more trenchant misconceptions regarding mainstream economics. For example, the disconnect between the abstractions of the orthodox economic approach and real world, empirically observable practices has consistently elided many examples of the diversity with which economic agents perceive choices, frame decisions, and form preferences. 9 This has had the effect of fostering a false 9 This notion of diversity is an important one for my argument, and therefore deserves to be emphasized. First of all, it is a major departure from traditional microeconomic theory, which relies on an assumption that human nature is essentially consistent, and therefore behavior is predictable. However, this assumption ignores the effects of institutions, which allow culturally fluent actors to develop a wide array of strategies in response to them. This range of options includes conformity, avoidance, challenge, alteration, and more. In addition, the incentive structure produced by the prevailing institutions frequently incentivizes different behavior for different social groups. Most importantly, the potential of institutions to influence behavior creates the incentive to alter them in distinct ways that are favorable to specific individuals or groups, which can be inconsistent with considerations of efficiency. 14 dichotomy between ancient and modern economic activity, and has therefore been a significant contributing factor to the frustratingly contentious and unproductive debate over the nature of the ancient economy that we described at the beginning of this chapter. Second, treatments that represent orthodox economic theory as synchronically and diachronically uniform have inhibited the application of potentially valuable insights to the ancient evidence. Many powerful observations have been made by heterodox economists at various points in time, but these contributions to our understanding of economic behavior have been limited by the degree to which neoclassical economic theory has dominated the discourse. In particular, I will propose the incorporation of the neo- institutionalist approach of North, Williamson, and others as one that holds particular promise for the practice of history, and I will present a fuller discussion of the principles of this approach in the following pages. Finally, it has also been my goal to illustrate the degree to which neoclassical economic theory is only a limited reflection of the vast range of potential strategies which economic agents are capable of adopting. In this regard, the field of anthropology offers a host of ethnographic studies that suggest some of the alternative paths that have been taken by various societies in the past, as well as many of the ways that different cultures are incorporating different economic institutions into their own culture. As a result, I will attempt to move past the ethnocentric emphasis on the neoclassical prioritization of resource allocation through the markets. Ancient Economies? After a discussion of the flaws inherent in orthodox economic theory, we might reasonably ask what relevance an analysis of mainstream economic thought holds for the discipline of ancient history. The simple answer to this question is that, as we have already observed, neo-classical theory has become the dominant mode of economic inquiry. So much so, in fact, that it can often appear to represent the entire body of modern economic thought, especially to those like ancient historians who are understandably not 15 versed in the nuances and developments in the history of economic theory. While this is perfectly understandable, the result has often been to contribute to a reductive binary nature of the debate over pre- modern economic behavior. I have already stated my aim of examining the economic dimension of the Archaic Period through the lens of the New Institutional Economics. In order to understand what analytic tools NIE offers beyond those of neo-classical analysis, I have tried to present the intellectual context of mainstream economic thought. Next, in order to better grasp the potential benefits of this approach for ancient historians, I believe it is important to offer a similar context as well, with the same goal of demonstrating the ways in which NIE can provide nuance to the discussion. 16 Chapter Two: Ancient Economics? We have already noted that the discussion over the nature of pre-modern economic behavior and institutions has been largely moribund for more than a century (Reibig, 2001), inhibited by the terms of a binary debate that have affected a range of disciplines, including economic anthropology, ancient history, economic history, and classical studies. This limiting and reductive debate for Wilk and Cliggett (2007) that “was once the centerpiece of economic anthropology, has now become an obstacle instead of an inspiration.” For economic historians, the controversy has produced “one of the most sterile debates in history…” (North, 1990), and among ancient historians and classicists, “Ancient economic history has been the arena for what must be the most voluminous debate in ancient history over the past thirty years…” (Saller, 2002). There is one point on which nearly all can agree, namely that the level of academic discourse on topics related to pre-modern economic activity has remained frustratingly suppressed by this trenchant and constraining ideological schema. I should say it is not my intent to denigrate any one school of thought. Instead, I believe that both sides of the formalist-substantivist debate have highlighted important truths about transactional behavior and the way that we study it, and that each side has also quite rightly pointed out many of the misapprehensions of the other, but that both approaches are ultimately founded on a series of false notions, often held in common, and that in the final analysis, these false notions make both positions untenable. In addition, both the formalist and substantivist schools implicitly treat neo-classical theory as an analogue for modern economic thought, full stop. It is this which I believe is ultimately the fatal flaw of the controversy, since both positions frame their arguments based on the assumption that the tenets of orthodox theory hold true for modern economic behavior (and for the formalists, for ancient behavior as well). In part, this has led to the “New Orthodoxy” of Finley. 17 How Did We Get Here? The issue of how the debate over the nature of the ancient economy has developed (or perhaps more accurately, has not developed) over time has been extensively discussed elsewhere, 10 and there is no need to cover extensively the territory that has already been covered so thoroughly. However, several points are particularly notable. For one, the work of Karl Polanyi has been of singular importance and remains extremely influential in numerous disciplines, including ancient history and anthropology. In the field of ancient history, one of his former colleagues, Moses Finley, is generally accepted as the single most important figure, and his influence has persisted for several decades since the 1973 publication of his Ancient Economy. 11 Of course, just as their work has been extremely influential on subsequent thinkers, they were in turn heavily influenced by their predecessors. For instance, Max Weber was a singularly important influence, most notably on Finley, but also on Polanyi. Second, one of the most salient characteristics of the debate has been its pronounced binary character. This has been an immanent feature from the very outset, 12 and clearly one of the factors that has contributed most to the polarizing nature of the debate, which in turn has inhibited productive discourse on the subject (Saller, 2002). In its broad 10 The following is just a partial list of some of the authors, in no particular order, who have chronicled the background of the debate, either in full or in part, from the point of view of various disciplines, including economic anthropology, economic history, and ancient history: North (1977); Humphries (1983); Morris (1994); Silver (1995); Tandy (1997); Cartledge (1983, 2002); Bresson (2000, 2007); Engen (2004, 2010); Wilk & Cliggett (2007). Of especial note are Bresson (2007), for the attention he pays to framing the debate within contemporary currents of thought in economics, and Engen (2010), for an especially concise and thorough synthesis of the major theorists and their respective contributions and legacies. 11 See Morris (1999) for a favorable assessment of Finley’s work and its continued impact on the field; Bresson (2006) and Engen (2010) offer more critical perspectives. 12 Although Morris (1999) claims that economic questions entered the field of ancient history rather late, this position is inconsistent with the widely-held belief that the origins of the debate over the ancient economy began at least as early as the late 1800s with the so-called “Bücher-Meyer controversy” (see Finley [1980]; Morris [1999]; Bresson [2006]; Engen [2010]); also, Rostovtzeff’s Social and Economic History of the Roman Empire (1926) stands as a major early work defending the modernist position, and Hasebroeck’s Staat und Handel im Alten Griechenland (1928) takes up the primitivist argument. Although his influence has been more profound in other fields such as sociology and anthropology, Max Weber’s Wirtschaft und Gesellschaft (1922 [2013]) has also been influential, inspiring the likes of Karl Polanyi and Moses Finley. 18 outlines, the debate has followed a similar line of development within the discipline of anthropology as well. 13 The Work of Karl Polanyi Without doubt, Polanyi’s influence on a broad range of disciplines is widely acknowledged (Mark, Ramlogan, & Randles; Mendell and Salée 1991). In ancient history, for example, Bresson (2007) maintains that Polayni “a exercé la plus grande influence sur la conceptualisation de l’économie des societies d’avant le capitalisme, et plus spécialement du monde antique.” He is a central figure in both ancient history and anthropology, and because of his influence on the course of subsequent scholarship, his work is an important nexus where economic anthropology and ancient history intersect. His greatest impact has doubtless been the result of several core concepts: the major modes of exchange, the notion of embeddedness, and the distinction between the substantive and the formal concept of economy (Humphreys 1984; Tandy and Neale 1994). Polanyi’s Core Concepts Modes of Exchange For Polanyi, it was critically important to recognize that not all exchanges are equivalent by nature. They can occur in a widely divergent range of social and institutional contexts, governed by radically disparate rules, and can stem from vastly different motives. In the modern market economy, the overwhelming majority of material exchange takes place through the institutions of the market system, but this was not the case for the greater portion of human history (Polanyi, 1957a [1944]). Therefore, in order to 13 One of the main foci of the early fieldwork by Boas (1888) was the exchange systems, especially the potlatch, of the Pacific Northwest Coast peoples. This interest in gift-giving was prompted to a great extent by the fact that it appeared to violate the neo-classical assumption of rational self-interest, and was a prominent feature of Malinowski’s groundbreaking study of the Trobriand Islanders (1984 [1922]), as well as a central topic for Mauss’ seminal Essai sur le don (2000 [1923-1924]). Another similarity between the early works of ancient historians and anthropologists is the presence of a distinct normative bias, in which the European economic system of the 19 th century, wherein resources are allocated primarily through the institutions of the market, is considered the standard of comparison. 19 understand pre-modern economic activity, one must be able to recognize and identify the most common modes of exchange – reciprocity, redistribution, and market exchange. Reciprocity tends to occur most frequently in simple societies, in which face-to-face transactions are most frequent. This mode assumes symmetrically arranged groupings as a background (Polanyi, 1957), and does not foster hierarchical social structures. Redistribution, by contrast, depends on some measure of centricity within the group, and holds when the allocation of goods is collected by a single, central authority; it can take place by virtue of custom, law or ad hoc central decision (Polanyi, 1957). It should be noted that according to Polanyi, redistribution is not specific to a particular social structure, and that it can take place for many reasons, and at all levels of civilization. In addition, it can apply to different subgroups, such as the household or manor, and can therefore exist within a society, regardless of the mode of exchange by which the society as a whole is integrated. Finally, market exchange, or what Polanyi sometimes refers to simply as “exchange,” is the mode by which resources are allocated through a system of price-making markets. Polanyi recognized that markets for various goods existed at different points in human history, but believed that these were historically rare, and were generally “peripheral” to the operation of pre-modern economies (Polanyi, 1963). It was only in Europe in the nineteenth century that market exchange became the dominant mode, and produced a market economy in which the entire economic system consisted of a complex of integrated price-making markets where fluctuations in one market would manifest themselves in other parts of the system (Polanyi, 1947). The appearance of a system of interdependent markets was, for Polanyi, a historically unprecedented phenomenon. It was more than the result of an aggregate collection of discrete (but separate) individual markets, but rather an integrative, institutionalized structure that spread the effects of prices on ancillary markets (Polanyi, 1957), an “economy directed by market prices and nothing but market prices…” (Polanyi, 1957a [1944]). What was unique about this development was that, for the first time in history (according to Polanyi’s argument), society was organized according to the dictates of the economic system, rather than the other way around: “Previously to our time no economy has ever existed that, even in principle, was controlled 20 by markets…Though the institution of the market was fairly common since the later Stone Age, its role was never more than incidental to economic life…” (Polanyi, 1957a [1944]). The Embedded Economy Consistent with his belief that the market system was an historically unprecedented development, Polanyi argued that what made this particular economic system so radically different from any other previous economy was that for the first time, transactional processes were conducted first and foremost for the sake of the items being exchanged, as opposed to the entire expanse of previous human experience, where the relationships of the actors involved in a transaction took precedence over the material considerations. With the shift to a market economy, the relationship of the participants to each other was secondary to the transaction itself, and in many cases their relationship was limited solely to the actual process of exchange. As a result, an economic sphere came into existence that was sharply delimited from other institutions in society (Polanyi, 1947). As Tandy (1997) points out, in The Great Transformation Polanyi uses the term “submerged” to describe this concept, and the term “embedded” that has become commonly used to describe this concept did not appear in his work until later. However, this terminological distinction does not reflect any underlying conceptual development. The core notion remains the same, namely that prior to the appearance of the market system in Europe in the 19 th century, it was the case for pre-modern societies that economic activity was submerged in social relationships, and that the economic system was conducted on the basis of “non-economic” motives. As a result, the notion of rational self-interest, so central to classical and then neo-classical economic theory, fails to describe accurately the motives of individuals in non-market economies, where social pressures in tribal societies prompt the individual to ignore self-interest in the classical economic sense, making him or her unable “even to comprehend the implications of his own actions in terms of such an interest…” (Polanyi, 1957a [1944]). Furthermore, not only did the market economy present the historically unique conditions whereby the economy was removed from its social context, but in fact the relationship of the economy to the society was reversed, so that social relations 21 came to be a function of the market: “The market mechanism became determinative for the life of the body social…” (Polanyi, 1947). According to this model, both reciprocal and redistributive exchanges have been dictated for the most part by more dominant social and cultural considerations. In Polanyi’s view, prior to the modern era, the material fact of the goods being transferred has been secondary to the relationships that the process of exchange validates and reifies, and the distribution of goods has traditionally been subordinate to social and cultural dictates. However, a distinguishing feature of the modern economy has been an inversion of the relationship between society and exchange. In the modern economy, society has been organized according to the dictates of the market rather than the reverse, and the mechanisms of reciprocal and redistributive exchange that had previously served to integrate human societies and promote social cohesion were sublimated to the priorities of what Polanyi called “market exchange,” and thus could no longer function effectively. With the emergence of the market economy, the constituent elements of society were inevitably dislocated and thrown into disarray. Formal and Substantive Economic Activity Without a doubt, any discussion about economics should proceed from a clear, unequivocal understanding of just what is meant by the terms “economy” and “economic.” However, this has often proven difficult, with some commentators regarding certain types of behavior as clearly economic in nature, while others would classify them as political, or social. For instance, is economic activity strictly limited to the kind of maximizing, profit-seeking, materialist behavior described by Smith, Mill, or Ricardo, or can we consider the elaborate, highly ritualized and ceremonial exchange of gifts in the Trobriand kula as economic as well, albeit conducted according to a very different type of logic? Perhaps Polanyi’s greatest contribution, and certainly one that has had the most staying power, has been his attempt to resolve this dilemma, represented by his distinction between two forms of economic activity, the formal and the substantive. 22 Expressed most fully in “The Economy as Instituted Process,” his contribution to the collected volume Trade and Market in the Early Empires, Polanyi (Polanyi, 1957) points out that one of the most persistent issues that has inhibited discussion on pre-modern economics has been the tendency uncritically to use the term “economic” in two very different ways. The first meaning of the term, that he calls the “substantive,” refers to the interaction with the natural and social environment in terms of the satisfaction of material wants. This type of activity operates according to the laws of nature, and “implies neither choice nor insufficiency of means; man’s livelihood may or may not involve the necessity of choice and…need not be induced by the limiting effect of a ‘scarcity’ of the means…” (Polanyi, 1957). By contrast, the second, “formal” meaning derives from the logical character of the means-end relationship, and refers to a definite situation of choice between the different uses of means induced by insufficiency. Rather than being dictated by the material facts of existence, the formal definition of “economic” refers to ideological considerations, and “implies a set of rules referring to choice between the alternative uses of insufficient means…” (Polanyi, 1957). According to Polanyi, the indiscriminate conflation of these two fundamentally different forms of activity has hindered our understanding of economic behavior, and led us to a number of false assumptions, with perhaps none more significant than the erroneous belief that the analytic tools of modern economic theory are applicable to behavior other than that which occurs in a market system (Polanyi, 1971). Consistent with his belief that the modern market economy was historically unprecedented, Polanyi argued that formal economic behavior was a particular type of activity that was uniquely consistent with the socially disembedded, self-interested market form of exchange. Extending this argument even further, he posited that modern economic theory, formed as it was to explain the phenomena of the market system, was appropriate solely to analysis of the particular formal exchange that it was created to describe. As a result, it was not possible to employ the tools of formal economic analysis (i.e., the neo- classical approach) in order to explain pre-modern economic activity. In the modern world, the economic system had been structured according to the principles of market exchange, and therefore modern society 23 satisfied its material needs through the mechanism of the market system. In other words, in the modern market economy, the substantive economic operation of the society was essentially coterminous with the formal processes of exchange. It is this distinction that ultimately forms the basis of Polanyi’s ideology, justifying and supporting his central claim for the singularity of the modern market economy in human history. In addition, the influence of this particular aspect of his thought on subsequent analysis has been almost incalculable. Since Polanyi, the terms of the debate over the nature of pre-modern economic behavior have been indelibly drawn along the lines of the formalist-substantivist distinction, with nearly every subsequent theorist adopting a position on one side or the other of this division. For Polanyi and his substantivist followers, the economy is an instituted process, indivisible from its social and cultural context for nearly the entirety of human history, until the appearance of the market system created the conditions by which exchange became disembedded from the society. For the dissenting formalists, it is a simple truism that human beings virtually without exception pursue what they perceive to be in their individual self- interests, and that conditions of scarcity will create inevitable competition, the end result being that those actors who do not adopt effective strategies will be unable to compete. These principles, they argue, so familiar to us from the classical economists onwards, are timeless and universal, and can be applied to the economic activity of any society, in any place, and in any time period. Of course, as has already been noted, it is this binary choice between formalism and substantivism that has dominated the field, in one form or another, ever since, and even though there are hints that the field is gradually divesting itself of this model that seems to have lost its potency, the precise shape of what will replace it remains hazy (Engen, 2010). Perhaps Polanyi’s greatest contribution, though, has come in the form of his ability to constitute a point of departure from mainstream economic theory. As we have already seen, modern economic theory has been dominated by a distinct orthodoxy in the neo-classical tradition. There have long been dissenters, both within and without the discipline of economics, of whom Marx is probably the most notable and 24 influential. However, Marx’ political and social agenda was clear, and his lasting appeal has been far greater in debates over contemporary policy and political economy rather than history, especially outside of continental Europe. Although Polanyi also had an explicitly reformist agenda, his work has been far more instrumental in the social sciences and economic history, while having a negligible effect among economists (North, 1990). This is perhaps due to his position within the academic community, or because of his work with the Interdisciplinary Project at Columbia that was dedicated to increasing the understanding of the origin of economic institutions (leading to the publication of Trade and Markets in the Early Empires in 1957). In any case, his critiques of mainstream economic theory, or “formal” theory to use his own term, have done much to illustrate the limitations of economic orthodoxy, especially in terms of its relevance to pre-market societies. More than anything, Polanyi argued persuasively for the necessity of understanding pre-market transactional processes within their social, political, cultural, and religious contexts. Mainstream economic theory, which deliberately ignores these considerations for the sake of model-building, is of limited value at best for pre- or non-market economic systems, and his success in demonstrating the nature of pre-market economies as, in his words, “instituted processes,” is arguably his most important innovation. Despite his tremendous contributions, there are a number of problematic aspects of Polanyi’s thought. Admittedly, it is far easier to criticize his work in the light of nearly seventy years of hindsight, and the concomitant decades of scholarship. In addition, no criticism could elide the positive impact that he has had on the study of economic behavior in past societies. From 1989 to 2008, the Karl Polanyi Institute for Political Economy conducted a web-based research program attempting to track references to Polanyi, and identified approximately 25,000 during that period alone (Mendell, 2008). There is no questioning his lasting impact. However, in part because of this influence, the more problematic features of his work have been commensurately magnified. First of all, despite claims to the contrary, Polanyi’s model is fundamentally, albeit implicitly, teleological. As Wilk and Cliggett (2007) argue, Polanyi’s model incorporates a strong evolutionist 25 foundation, tracing the path from the simple (or “primitive”) to the complex (or “modern”), and depicting modern society as a distinct break from the past. At times, he seems to be aware of this tendency, even going so far in his discussion of the three modes of exchange that they “do not represent ‘stages’ of development. No sequence in time is implied…” (Polanyi, 1957). However, his denials do not seem to be confirmed by the structure of his arguments. For one, in his analysis of Ports of Trade, Polanyi opposes the idea that they represent an example of market activity in the ancient world (1963). Instead, he claims that the Port of Trade represents a form of pre-market development, and in so doing, places the Port of Trade at a particular stage along a developmental continuum. A fuller range of this linear developmental model can be glimpsed in his treatment of Menger’s contrast between the modern market economy and previous economic structures: “The corresponding economies can be contrasted as ‘advanced’ and ‘backward,’ specifiable as market economies and pre-industrial non-market economies…” (Polanyi, 1971). Finally, it is well known that Polanyi relied on the work of several prominent ethnographers, in particular Thurnwald on the Bánaro and Malinowski on the Trobriand Islanders (Tandy & Neale, 1994). In so doing, he incorporates their work by citing the Bánaro and Trobriand Islanders as examples of “primitive” or “savage” societies (Polanyi 1944). Without question, this tendency rests on a number of tacit assumptions about the nature of social and economic development that necessarily affected his interpretations. This evolutionist bent is certainly not enough in and of itself to invalidate his views, of course, but it remains the case that many of his assertions are highly problematic. According to North (1977), it is easy to find fault with Polanyi’s analytical framework, on the grounds that numerous parts of his analysis “show a failure to grasp elementary economic principles...” and “the evidence he makes use of from the ancient world is highly selective…He clearly understates [the use of markets] throughout history.” As Bresson (2007) points out, Polanyi observed many unexpected features in the ancient world, including a market system familiar with price fluctuations, transporting numerous products for mass consumption and embracing large geographic zones. At the same time, he denied the existence of price- making markets in the ancient world. Such discrepancies, whether they stem from imprecisions in expression, or from any other reason, are difficult to reconcile. 26 Also, Polanyi’s criticism of what he calls “formal” economic theory, or more particularly the application of this theory to the ancient world, rests on a rather hazy grasp of so-called modern economic thought. For one, Polanyi consistently represents the body of economic thought as monolithic and static (Polanyi 1944 [1957], 1947, 1957, 1963). However, as we have already seen, this is far from accurate. Economic analysis is a highly dynamic field, and any criticism that fails to recognize diversity and change within the field at the very least generates the sense that it is arbitrary. Without a doubt, mainstream economic theory maintains a strong connection to the work of the classical economists, and the continued reliance on and tacit acceptance of many of their underlying principles is a notable feature of modern orthodox economics, to the extent that the label “neo-classical” still seems quite applicable. However, not only has neo-classical economics changed and developed over time, but also such a reductive portrayal of the entirety of modern economic thought completely ignores the contributions of any number of heterodox schools. In addition, it fails entirely to recognize the tremendous changes that took place in actual economic conditions from the time of the “Great Transformation” in the 19 th century that Polanyi describes to the latter half of the 20 th century (North, 1978). However, perhaps the most enduring aspect of Polanyi’s broader argument is that it continues to reinforce the polarizing, binary nature of the earlier primitivist-modernist debate. Placing all pre-modern, non- Western economies on one side of an interpretive division is more a way of evaluating modern economic policy than it is an analysis of pre-modern behavior (Figueira, Karl Polanyi and Ancient Greek Trade: The Port of Trade, 1984). It is one thing to make the argument that the modern economic system is unique, and quite another to do so by constructing an interpretive framework that systematically ignores or elides the distinctive features of particular economic systems at various points in the past. More than anything, the result of this approach has been to validate a new binary division, albeit in different terms. Just as Bücher and Meyer ushered in the primitivist-modernist debate, the work of Polanyi has given rise to the formalist-substantivist debate. Unfortunately, this aspect of his legacy may be the one that defines him, and ultimately overshadows his contributions to the field. The implications of this polarizing outlook 27 can be seen in the way he clearly underestimates the role of markets in early societies, but also in many ways overestimates their role in the modern world. As North (1977) explains, Polanyi has a specific and highly restrictive definition of a price-making market that demands that several elements be present, such as a large number of buyers and sellers, a variety of goods offered for sale, an agreed upon medium of exchange, and an enforced set of property rights. Under these conditions, most exchanges, whether ancient or modern, do not take place as market exchange. Therefore, it is not surprising that Polanyi would find that markets played a minimal role in the ancient world. Even so, the evidence suggests that even in spite of these restrictive conditions, the “prerequisite functions demanded by Polanyi – the allocation of consumer goods, land, and labor through the supply-demand-price mechanism; risk organized as a market function; and credit markets – can be seen plainly in the record from the earliest times…” (Silver, 1983). Some have argued that this tendency to minimize market activity in the pre- modern world stems from an idealization of the primitive, a utopian vision that minimizes conflict; as a result, Polanyi attributes altruistic and cooperative tendencies to actors in non-market conditions, and de- emphasizes self-seeking, aggressiveness, and competition (Cook 1966, 1969). Without question, Polanyi regards pre-modern society as essentially static and changeless in the absence of outside influences: “As long as social organization runs in its ruts [italics mine], no individual economic motives need come into play…In such a community the idea of profit is barred; higgling and haggling is decried; giving freely is acclaimed as a virtue; the supposed propensity to barter, truck and exchange does not appear…” (Polanyi, 1957a [1944]). This is of course a major shortcoming of the model, which represents reciprocal and redistributive systems as essentially changeless, and nothing in his framework explains changes in the mix of the system over time (North, 1977). Whatever the reason, it seems clear that Polanyi is pre- disposed to overlook the importance of ancient markets, and in many cases to ignore their existence altogether, and as a result his model presents a distorted picture. With the benefit of a half-century of reflection, it is apparent that many of Polanyi’s predictions have not been borne out historically. As Clark (2008) observes, “The great puzzle of Polanyi's book is thus its 28 enduring allure, given the disconnect between his predictions and modern realities. The fans of Polanyi seem to be responding to his general belief that markets corrupt societies, and his assertion that free market economies are a shocking recent departure from a socially harmonious past.” Polanyi’s work has inspired numerous followers across a range of disciplines. In anthropology, those who initially were strong supporters of Polanyi’s theses included George Dalton, who worked on markets and primitive money, Paul Bohannon, who worked among the Tiv of Nigeria and was instrumental in developing the concept of exchange spheres, and Marshall Sahlins, who further developed and refined Polanyi’s discussions on reciprocal exchange. Also often included in this list are Manning Nash, Cyril Belshaw, Maurice Godelier, Neil Smelser, and Walter Neale (Humphries, 2004 [1978]). His influence has also been prominent in the study of the ancient world, with his work influencing the likes of Harry Pearson, David Tandy, Walter Neale, and most significantly, Moses Finley. The Work of Moses Finley For classical studies and ancient history, and especially for Homeric studies, the influence of Moses Finley has been incalculable. For Andreau (2002), Finley’s 1973 The Ancient Economy “marked without a doubt a decisive turning point in the field of Greek and Roman economic history. The ‘Finley Method’ triumphed.” Saller (2002) too, in describing the sheer volume of debate on the topic of ancient economic history, places Finley squarely at the center of that debate, and going even further, Morris (1999a) contends that “no book in this century has had such a great influence on the study of Greek and Roman history…The Ancient Economy’s humane vision of the Greek and Roman past will remain at the center of our arguments for the foreseeable future.” First and foremost, Finley (1999 [1973]) was emphatic about the fact that it was not possible to apply a “market-centered analysis” to the ancient world. There was a fundamental incomparability between economies prior to the 18 th century and the “modern European industrializing economies of the late- eighteenth, nineteenth, and twentieth centuries analyzed by Adam Smith, Ricardo, Marx, and the neo- classical economists…” (Saller, 2002). Crediting Weber for first introducing this concept, and then “his 29 most important disciple” Johannes Hasebroeck as well as Polanyi with developing the notion, Finley argues that for several reasons, such an approach is untenable for the ancient world, and that new concepts and models must be found. For one thing, he claims that the volume of data available is “below the minimum required for standard statistical procedures”; there was no time series available in antiquity, and “without a time series there can be no reasoning by figures, no statistics…” (Finley, 1999 [1973]). Of course, on this point at least, Finley seems to imply that quantitative analysis is a sine qua non of modern economic theory. It is certainly true that at the time of the publication of The Ancient Economy in 1973, quantitative analysis had become a prominent feature of economic analysis. 14 However, the lack of sufficiently robust data sets to produce reliable statistical models is not in and of itself an insurmountable barrier to the application of the theory that he rejects. Elaborating on this position even further, Finley denies the existence of an ancient economy altogether. According to this argument, if one uses Smith’s political economy or Marshall’s principles of economics as a point of comparison, the ancients “lacked the concept of an ‘economy,’ and, a fortiori, that they lacked the conceptual elements which together constitute what we call ‘the economy.’” Of course, numerous difficulties are present in this position. First of all, and perhaps most importantly, such a statement is not supported by the evidence. In Xenophon’s Poroi or Aristotle’s Politics it is clear that there is precisely the kind of explicit deliberation about balance of trade, profit, and calculated policy at the state level that seems impossible to describe by anything other than economic policy (Bresson 1987; Meikle 1997; Engen 2010). Secondly, even if one could discount such examples as notable exceptions and accept this as true in a general sense, it is not clear how this implies the absence of an economy altogether. The principles of Smith and Marshall are formulated to describe particular forms of activity and behavior, and even if they were never expressed, the activity still can exist. By the same token, whether or not ancient authors described economic behavior in terms that we recognize from the model of 14 Blaug (2003), Lawson (2005). 30 Smith does not indicate whether the behavior itself existed. In addition, this argument seems to rest in part on an absence of evidence, itself a dangerous proposition given the fragmentary nature of the literature that has survived from antiquity. For such a blanket statement to be possible, it was necessary for Finley to treat the ancient economy as a totality. In fact, Finley argues for a single economy, of the same basic nature and possessing the same fundamental characteristics, that predominated from the end of the Bronze Age to the fall of the Roman Empire (Andreau 2002; Engen 2010). In so doing, he elides the tremendous social, political, and institutional differences across both time and space. Finley defends this position, at least in part, by claiming that the counter-examples that would suggest his picture is over-simplified are isolated instances, and not sufficient to invalidate the general truth of his position. Morris (1999a) also defends the general truth of the Finley model, along very similar grounds. For Morris, it is an immanent feature of generalizing models that they must necessarily ignore particularities in order to maintain the integrity of the broader picture. This is of course true, but defending a universalizing model simply because it is universalizing is tautological, and fails to confront the fact that the literary evidence, augmented by a continuously growing body of archaeological evidence, suggests that the economic conditions in antiquity were far too diverse and complex to support Finley’s totalizing vision (Engen, 2010). In keeping with this tendency to generalize and oversimplify ancient economic conditions, Finley also denies the existence of economic growth in the ancient world. As he says, “inelastic markets and traditional methods of technology and agricultural organization blocked any significant growth in productivity, in what we should call the gross national product…” (Finley, 1999 [1973]). Once again, this statement is at odds with the evidence. Of course, it is certainly true that the pace of technological development in the industrialized world has had a tremendous impact on levels of productivity, and therefore economic growth. Consequently, it would clearly be a mistake to expect comparable levels of growth in antiquity. However, this does not mean that growth could not, or did not, occur. In fact, it seems quite possible to document economic growth in the ancient world, albeit on a much different scale 31 (Morris 2004; Morris, Saller and Scheidel 2007). Ultimately, for Finley, status serves to explain in large part the economic conditions of the ancient world. For him, the primacy of status concerns had five consequences for the ancient economy (Morris, 1999a): 1. Status considerations inhibited the development of markets in land, labor, and capital 2. The development of citizen status posed challenges for the rich to employ wage labor, which led at the same time to the growth of freedom as an element of citizenship, and of slavery; the ancient economy was in fact the world’s first true slave economy 3. Status concerns pushed even profitable activities to the margins of society; finance and trade, for example, were dominated by foreigners and metics 4. Greeks and Romans pursued wealth through legal and political channels instead of economic activities 5. War and imperialism were major sources of wealth in Greco-Roman society Not surprisingly, Finley has had his critics. As both Morris (1999a) and Engen (2010) observe, these criticisms fall into three main categories: empiricist, oversocialized, and undersocialized. The empiricist criticisms are based on Finley’s oversimplification of the ancient economy. Such critics generally attempt to illustrate the complexity and dynamism of ancient economic behavior and to point out instances of disembedded economic activity, concluding that Finley’s model cannot account for the details and/or particular parts of the Greco-Roman world between 1000 BCE and 500 CE. In defending Finley, Morris (1999a) adopts Weber’s argument that there is an inverse relationship between the general applicability of a model and its ability to accommodate reality. This defense, however, is self-reinforcing, in that it a priori rejects evidence that is contrary to the model on no other basis than that it is contradictory. As Morris himself admits, Finley consistently underestimated the scale of ancient trade, banking, industry, and other non-agricultural production, and at times his position strongly leans toward a more traditionally primitivist model. 32 The criticism of oversocialization borrows a term from Mark Granovetter (1985), who argues that models of pre-modern economic behavior that emphasize the impact of social considerations are often overly deterministic, failing to account for individual agency and the fact that social structures are a product of collective activity, and therefore malleable. Criticisms on this basis also point out that the literary sources favor elite viewpoints and mask more mundane, quotidian practices. According to this view, we cannot take the ancient writers at face value, but should instead be looking for places where “economic realities peep through the masks of culture…” (Morris, 1999a). Such an argument is that proposed by Edward Cohen (Cohen, 1992), who in drawing on modern analyses of “underground” or “shadow” economies, argues that there is evidence for a fully developed banking system in classical Athens, in which many transactions were deliberately underplayed, and certain types of wealth were consciously and systematically made aphanes (“invisible”). In addition, although Finley explicitly marginalized the potential of material evidence to contribute to our understanding of ancient economics, there is a growing body of archaeological evidence for growth in the ancient world. Finally, the criticism of undersocialization is drawn from the growing popularity of the so-called New Cultural History of Lynn Hunt (1989), Roger Chartier (1988), and others, according to which the focus is on the creation of social categories and how they were formed and contested in the discursive practices of knowledgeable actors. By this view, Finley’s analysis fails to account for the representational nature of the evidence, and fails to acknowledge that the written sources are not objective sources of real conditions, only evidence of how individuals wanted to represent it to further their own perceived interests. This is a point emphasized by Sitta von Reden (1995), who criticizes Finley for ignoring ideological negotiations about the limits of the institutions that he analyzed. Ultimately, there is concern that Finley failed to recognize sufficiently the fact that “the men who wrote our texts did not simply distort realities; they constructed alternative, context-dependent realities…” (Morris, 1999a). As Engen (2010) argues, all of these criticisms are generally valid. Finley makes very broad generalizations and discounts contrary evidence as exceptional. He also tends to take the literary evidence 33 at face value, assuming that elite voices speak for a broader spectrum of society, and also overemphasizes the power of society to control individual economic behavior. Finally, in his desire to prove that the ancient economy is qualitatively different from the modern economy, even going so far as to claim that the ancients did not have an economy as such, his argument is in many ways more about what the ancient economy was not (Foxhall, 1990). In addition to these criticisms, there are several other points that deserve mention. For one, Finley frequently vacillates between an argument that is distinctly grounded in the substantivist position that the difference between the ancient and modern economies is qualitative, and the primitivist position that the difference is simply quantitative. As we have already noted above, Finley’s propensity for underestimating the scale of various non-agricultural economic activities leans towards a primitivist approach. Also, in his discussion of the nature of the ancient city, Finley makes it a point of emphasis to contrast the size of ancient cities with modern ones, an observation that is more relevant to a primitivist argument than a substantivist one. However, Finley’s reliance on status as the central point of difference between ancient and modern economic behavior is much more consistent with a substantivist argument. Second, Finley’s definition of “economy” is based on that of Erich Roll, who regarded the “economic system as an enormous conglomeration of interdependent markets, [and] the central problem of economic inquiry becomes the explanation of the formation of price…” (Finley, 1999 [1973]). This definition can only be described as extremely narrow and restrictive. As a result, it is not surprising that he finds little in the ancient world that corresponds to this concept. In fact, much of what would be considered economic activity in the modern world would fail to meet these criteria as well. Even further, in his denial of even the existence of an ancient economy, Finley stretches the definition almost to the breaking point, describing the economy as “conceptual elements which together constitute what we call ‘the economy’” (1999 [1973]), and completely removing economic activity from its material context. Third, it cannot be ignored that Finley’s model reflects a limited grasp of contemporary economic literature. His characterizations of the modern economy rely heavily on the classical economists and early marginalists, particularly Marshall, but seem to ignore completely the range of contributions that were made since the end of the 19 th century. As a result, Finley fails to take advantage of the 34 dynamism and diversity of 20 th century economic theory, and represents the entire body of modern economic thought as static and monolithic. In fact, his view of modern economic thought is limited to orthodox theory, and is a reductive and simplistic view at that. Conclusions Although both have been supremely important for their contributions to their respective fields, Polanyi and Finley have left problematic intellectual legacies that have had the overall impact of suppressing debate in economic history and anthropology. In large part as a result of their continued influence, discussion of pre-modern economic behavior has split along formalist-substantivist lines, with little or no room for compromise. Most importantly in my view, both formalists and substantivists have consistently operated under and shared a number of fundamental misconceptions that have distorted the discussion and obstructed constructive debate and progress. 1) Price-making markets are a feature of the modern market system. For substantivists, the absence of these markets constitutes a fundamental difference between modern and pre-modern economic systems, while formalists see evidence for them throughout human history. We have already noted the often careless use of terms, including what is meant by “economic” and “market,” among others. In fact, one of Polanyi’s most important contributions was to point out the polyvalent meanings of “economic” that all too often were used uncritically and without distinction. Similarly, the term “market” can be used to describe a physical location, a type of exchange, or a set of conditions. By “price-making markets,” it is clear that what is meant is the orthodox economic definition of a market as a set of conditions, where both buyers and sellers are present (either physically or virtually), and price is solely a function of impersonal factors of supply and demand. The problem with this concept is that it is an abstraction designed to explain the formation of price, but it does not represent an accurate reflection of real-world conditions in any economic system, modern, capitalist, or otherwise. Price formation has been a central concern of modern economists ever since Smith tried to resolve the so- 35 called called “Paradox of Water and Diamonds” (Smith A. , 2003 [1776]). In order to explain price when resources are allocated primarily through the market mechanism, the orthodox economic approach makes a number of assumptions, including perfect competition, efficient markets, and frictionless transactions. In addition, in its focus solely on the allocative mechanism of the market, such a view ignores the underlying cultural and social components of supply and demand factors. For instance, the demand for diamonds is driven in large part by culturally constituted aesthetic and symbolic values. Also, the supply is affected by the symbolic connection of diamonds to notions of permanence, so that they are often bought in order to validate romantic relationships by virtue of their symbolic representation of love and commitment. As a result, they are usually retained and only rarely re-enter the market, fostering conditions of scarcity. Therefore, it is only in an artificially narrow sense that price can be considered a function solely of the impersonal factors of supply and demand. So-called “price-making markets” are not an intrinsic feature of modern economics, and therefore cannot serve as a point of contrast between modern and pre-modern economic systems. By the same token, it is clear from the evidence that conditions of supply and demand were capable of affecting price in the ancient world. However, modern economic systems are frequently institutionalized in such a way as to encourage more frictionless transactions and therefore enhance the operation of price formation in response to supply and demand factors. Merely observing the existence in the ancient world of price fluctuations in response to conditions of scarcity or abundance does not automatically represent conditions identical to modern economic systems. 2) Economic rationality is an underlying component of behavior in modern market systems. For formalists, it is a function of the fundamental human “propensity to truck, barter, and exchange one thing for another.” For substantivists, this is a function of the historically specific conditions of the market system that fosters and even demands such behavior, but is not present in reciprocal or redistributive systems. 36 Once again, part of the difficulty stems from terminological imprecision. In this particular case, economic rationality is in fact narrowly construed to mean a particular form of means-ends calculus, whereby the goal is the maximization of utility, often referred to as “economizing.” According to classical and neo- classical theory, the innate self-interest of economic actors, when allowed to operate under conditions of competition, will encourage rational agency through differential levels of success. Commonly misunderstood to imply that human beings are rational, this is in fact not the point. Instead, the presumption is that in a competitive atmosphere, rational strategies will be those that end up being more successful, and those that are less rational will simply be out-competed. Many, Polanyi (1957a [1944]) included, have rejected the notion that human beings are inherently any more rational in the modern world, or less rational in pre-modern societies. However, the presence of rational, strategic thinking in the ancient world is often erroneously interpreted as evidence for a consistent, universal form of behavior that is connected with economizing. For example, James Redfield (1983) sees in Odysseus a figure who is constantly challenged with situations in which he must choose from a range of options, and it is a marker of his success that he does so by careful calculation that is based on achieving goals that are to his greatest advantage, and he interprets this as an ancient analogue for maximizing behavior. However, the problematic issue of economic rationality lies not in the application of rational, strategic thinking, but rather in the actual ends to which that behavior is directed. Goals that are considered “reasonable” are typically conceptually linked to economic rationality, whereas those that are not, are not. For example, the first ethnographic analyses of the potlatch of the Pacific Northwest Coast peoples were initially greeted by many in the anthropological community as providing evidence for the antithesis of economic rationality, and the potlatch was often characterized as anti-economic, irrational, and even contrary to human nature. However, ethnographic studies of similar forms of ceremonial gift exchange among the Siuai of New Guinea frequently indicate elaborate, highly developed, rational strategies with the goal of accumulating the greatest amount of resources in order to distribute them (Oliver, 1955). Such strategies are regarded as somehow “irrational” not because they are not carefully formulated according to 37 reasonable expectations, but rather because the ultimate goal is not considered rational. However, as we have already seen, demand is a strongly culturally contingent phenomenon, and distinguishing between one goal and another because it is more rational to prefer one over another is a fundamentally flawed position. As Bresson (2007) points out, individuals within a given society are perfectly capable of acting equally rationally, and the goals which are fostered by social and cultural values must be judged by the internal coherence of the society, and are not subject to an external standard of value. 3) The definition of “economic” and “economy” continues to remain elusive, and frequently those on either side of the formal-substantive divide construct definitions that implicitly support a particular model. Despite the efforts of Polanyi to clarify this issue and to point out some of the pitfalls that are the result of less than careful use of terms, terminological imprecision continues to be a persistent problem. Frequently, proponents of a particular view will adopt a definition that supports their broader argument. For instance, Finley (1999 [1973]) employs a very narrow and unique definition of the economy, and in so doing argues for a fundamental distinction between modern and ancient economics, even going so far as to suggest the absence of an ancient economy altogether. It is in part his restrictive definition of the term that allows him to make such claims (Lowry, 2003). In fact, Polanyi himself is guilty of a similar error in his denial of markets in pre-modern societies. As North (1977) explains, Polanyi has a specific and very restrictive definition of a price-making market that demands that several elements be present (large number of buyers and sellers, variety of goods, an agreed medium of exchange, and an enforced set of property rights). By such a definition, most exchanges do not take place in markets, either in the modern world or in any other. Complicating this issue further is the frequent conflation of the capitalist system with the modern economic system in general. However, capitalism is based on specific elements, including private ownership of the means of production and the creation of goods or services for profit or income by 38 individuals or corporations (Heilbroner, 2008). As we have seen, the vast majority of arguments in favor of or against the existence of economizing behavior in the ancient world, for instance, are clearly dependent on such a system. In fact, it may be fairly said that the “market system” that forms such a central component of Polanyi’s analysis is indistinguishable from a specifically capitalist system. All too often, the specificity of this system is completely ignored, and this has powerful implications. For instance, according to such a definition, no communist or socialist system would be considered economic. In large part, the distinction being made by substantivism is not the qualitative difference between the so- called “modern economy” and pre-modern systems, but the relative lack of specifically capitalist forms. 4) Orthodox/mainstream/neo-classical economic theory, often referred to as “formal,” is assumed to be an accurate and comprehensive representation of the totality of modern economic behavior. For the substantivists, evidence for economic systems that do not conform to the expectations and assumptions of neo-classical theory is an indication of the inapplicability of this body of thought to the entire pre-modern world. For formalists, instances of behavior that conforms to neo- classical assumptions is regarded as evidence for the universal applicability of the theory. As I have already argued from the very beginning, mainstream economic theory is inherently neo- classical, by virtue of the continued reliance on a number of fundamental precepts and assumptions that can be traced to the classical economists. These assumptions include atomism, equilibrium, rational agency, and others. However, the orthodox economic model explicitly and deliberately produces an abstraction, not for the sake of historical explanation, but with the express goal of prediction, and many prominent economists have even overtly rejected any obligation to tailor their approach to conform to real-world conditions. As a consequence of this predictive emphasis, mainstream economists continue to rely on assumptions that may not hold true under even optimal conditions. For instance, David Romer (2006) has examined the play-calling tendencies of football teams in order to test the premise that firms maximize. He observes that even under the optimal conditions of intense competition, clear and unequivocal indicators of success orfailure, ready access to accurate and relevant information, and high 39 rewards for success, NFL coaches regularly adopt strategies that demonstratively and negatively affect their teams’ odds of success. Similarly, even the most ardent proponents of general equilibrium theory readily admit that they rely on an assumption that does not and might never exist in the real world (Lawson, 2005). In a certain sense, Polanyi was absolutely correct in his assertion, made first in The Great Transformation, that formal economic theory did not adequately explain behavior in non-market societies. His mistake lay in believing orthodox economic theory to be an accurate reflection of modern society. More accurately, his error stemmed from his belief that modern society was a market society. To be sure, the political and economic institutions of Europe and the U.S. for the greater part of the 19 th century were structured to adhere extremely closely to contemporary economic precepts, but this was a historically specific phenomenon, and even then was not a perfect facsimile of a free market ideal (North, 1978). Indeed, as Polanyi himself occasionally admitted, all societies have elements of reciprocity, redistribution, and market exchange to the extent that it is often difficult to distinguish where one begins and the other ends (Polanyi, 1957). This is just as true for the modern, so-called market economy as it is for any other. As North (1977) explains, “substitutes for markets not only have dominated exchange in past societies, but do so today as well…Price-making markets have never completely dominated economic decision making throughout history, including the nineteenth century….Reciprocity and redistribution are everywhere characteristic today as in the past in resource allocation within households, voluntary organizations, and in government.” On a similar note, but from the anthropological perspective, Wilk and Cliggett (2007) argue that the “Western economy is therefore just as deeply embedded in social (gendered) relations as the Trobriand and Tlingit economies are.” Thus, Kaplan (1968) in his challenge to formalism was quite right not only to question the relevance of formal economic theory to non-market societies, but in a broader sense to suggest its limitations for modern economic activity as well. To an extent, mainstream economics has become self-reinforcing in the degree to which it selectively approaches certain types of resource allocation, and then denies the 40 importance of reciprocal and redistributive exchange because they do not constitute a substantial component of the activity under analysis. However, this has begun to change. As Cohen (1992) for one points out, in recent years economists have become increasingly aware of the importance and sheer volume of “underground” or “shadow” economies, often referred to as black markets. Economists have largely viewed these parallel markets as nefarious challenges to officially sanctioned authority, but this prevailing opinion has changed dramatically, and analysts now regard seriously the role of underground or “informal” economies as often crucial elements of greater economic efficiency. However, because of the tendency to focus on “official” activity, mainstream economics has failed to grasp the significance of these kinds of transactions. In any case, both formalism and substantivism have operated under the assumption that orthodox economics is more or less an accurate reflection of modern economic behavior, and this underlying presumption has distorted the debate. 5) Both formalism and substantivism have grounded their arguments, either explicitly or implicitly, in reference to modern economic systems, and this has produced a tendency towards economic solipsism and “historical parallax.” Both sides of the debate have consistently maintained an approach that uses modern economic systems as their frame of reference. For substantivism, of course, the model seeks to distance the supposedly unique modern market system from all prior periods in human history. This has a tendency to produce what I have called historical parallax. As we know, parallax is a difference in the apparent placement of objects when viewed along different lines of sight. Objects that are nearby have a larger parallax than those far away, so that as the viewpoint moves laterally the more distant objects appear to move more slowly. From a historical perspective, there is a similar phenomenon for those events that are farther removed in time. As the historical viewpoint changes, the more temporally distant events seem to change less rapidly. In the case of economic history, this has fostered a sense that the economic conditions of the present or recent past are much more dynamic and fluid, while those of the more distant past are more static. For example, as we have seen, Polanyi (1957a [1944]) regards the imposition of the market system in the 19 th 41 century as a radical, violent break from previous human history, and his focus is prompted to a great extent by his genuine concern over the social disruptions that he attributed to this development. However, he seems to marginalize the economic dynamism of past societies. This perception has an analogue in anthropology, which Sahlins (1999) calls the “historyless character of indigenous cultures.” This anthropological vision sees indigenous peoples prior to contact with the West as remote and unknown, and therefore “pristine and aboriginal…as if they had no historical relations with other societies, were never forced to adapt their existence.” Both the historical and anthropological tendencies are produced by a sense of distance and removal, either in time or in space, and have contributed to the view that the modern, industrialized economy is entirely unique. Of course, there is no doubt that the modern economic system is transformative and dynamic, and to a degree that does not have a historical parallel. However, this is not sufficient to justify the view of Finley, who sees one single ancient economy, from the end of the Bronze Age to Late Antiquity, or better yet Polanyi, who frames the entirety of all past economic activity against the backdrop of the last century and a half. As Figueira (1984) notes, putting all pre-modern, non-Western economies on one side of the ledger is more a way of talking about modern economic policy than about primitive economies, and yet this model has been a central component of the discussion of pre-modern economic systems. Formalism, on the other hand, argues that the tools of modern mainstream economic theory are applicable not only to the modern economic system, but to all economic behavior in general. To a certain extent, it might be said that this totalizing vision is the antithesis of the kind of normative view that characterizes the substantivist position. However, the arguments in favor of formalism nevertheless betray a similar tendency to use modern conditions as the lens through which the societies of the past should be judged. There is certainly some momentum for this view in the legacy of the earlier primitivist-modernist debate, where the standard of comparison was explicitly that of the modern economic system. This predilection is a sort of ethnocentrism, what Polanyi calls “economic solipsism,” in which the underlying meaning and 42 rationale for observable behavior are attributed to the same or similar motives as those which direct modern economic agents. As Andreau (2002) argues: To contrast, term by term, everything pre-industrial with everything modern, and endlessly to scour antiquity for all possible and imaginable signs of archaism, results in a very reductionist view of history. Besides, whether deliberate or not, such an approach has the effect of providing present-day institutions and situations with an intellectual justification which they do not always merit, and of strengthening our reassuring (but illusory) impression that they are eternal, or at least immortal, since we have now entered modernity. In any case, whether a historical parallax produced by temporal distancing, viewing indigenous cultures as “historyless,” or attributing modern motives to pre-modern behavior, the result has been a failure to appreciate the internal coherence of other societies (Bresson, 2007). 6) In part because of the institutional structure of the academy, ancient history and anthropology have tended to engage largely within their own disciplinary boundaries, and interaction with other disciplines, especially economics, has been limited. One of the features of the debate over the nature of pre-modern economic behavior has been a degree of parochialism, in which members of each discipline interact in an internal dialogue with their colleagues. In fact, this potential pitfall was recognized by Polanyi, and served as the impetus for his Interdisciplinary Research Project while at Columbia. However, despite his efforts, interdisciplinarity has been largely absent from the larger discussion. In particular, this has been manifest in ancient history and anthropology by a lack of contribution from formally trained economists. Considering the amount of ink that has been spilled over the subject of the relevance of modern economic theory to non-market or pre-modern societies, it seems an obvious and natural step to solicit opinions from those who are professionally engaged in the discipline in question. However, this has all too rarely been the case. Accompanying a resurgence in interest on the topic in ancient history, a number of collected volumes have appeared in recent years, including Scheidel and von Reden (2002), Manning and Morris (2005), Bang, Ikeguchi and Ziche (2006), and the recent Cambridge Economic History of the Greco-Roman World (2007), but there is little if any discernible effort to invite the participation of trained economists. As a result, many of the 43 misconceptions and misapprehensions that we have described here continue to go unchallenged and are perpetuated. 7) So-called “self-regulating” markets are often thought to be a distinctive feature of a market economy. However, markets and market systems are institutionalized and administered. Even in the most economically liberal historical conditions, self-regulating markets have never existed. For many substantivists, the self-regulating market is a phenomenon that constitutes a point of difference between market and non-market societies (Mayhew, Tandy and Neale 1985; Tandy and Neale 1994). According to this argument, the market economy is characterized in part by the presence of “a supply- demand feedback mechanism determining the pattern of output and its distribution and the use of resources…” (Mayhew, Tandy, & Neale, 1985). In other words, resources are allocated through the market which is permitted to dictate production strategies, and of course determine price. However, this notion implies that markets exert an influence that is not moderated by any institutional constraints. In fact, nothing could be further from the truth. As the transaction cost economic analysis of North (1977, 1987, 1990), Williamson (1981, 1989, 1998) and others has demonstrated, the independent functioning of markets is anything but independent. In fact, the modern economy can arguably be said to be characterized by a greater degree of regulation, oversight and administration than at any other point in time. In part, this institutionalized structure is such a familiar element of the market system that it often goes unrecognized. Also, I suspect that few ancient historians or anthropologists are familiar with the extensive, elaborate regulatory environment of various modern markets, nor with the different institutions that are charged with monitoring, oversight, and enforcement. These institutions tend to enhance the security of property rights, reduce transaction costs, and augment the enforceability of contractual arrangements, thereby avoiding common impediments to efficient economic performance. When functioning properly, they give the impression that the markets operate more or less independently and foster the illusion of a self-regulating market. However, the simple fact is that such a perception is 44 erroneous, and that self-regulating markets have never been a feature of any economic system, modern or otherwise. 8) In general, orthodox economic theory is intended to serve a predictive function, whereas economic history and economic anthropology have an explanatory role. The former predicts future outcomes, and the latter explains past results. As a result, it is often impossible to apply mainstream economic models in an explanatory capacity, but this is not an indication of a lack of heuristic value. As we have noted numerous times already, one of the central issues of contention in the formalist- substantivist debate concerns the applicability (or lack thereof) of formal, neo-classical economic models to pre-modern phenomena. On a certain level, this is an impossible task, due in large part to the intended function of these models. Although it is certainly true that mainstream economists attempt to account for past performance, their motivation is not explanation, but almost invariably the refinement of their models in order to enhance the reliability of future predictions. Their concerns are simply not those of anthropologists or historians, who are interested in a fuller understanding of the particular social and cultural context that gives rise to individual economic systems, or explaining the course of events that have already taken place. This may seem like a rather esoteric distinction, but the implications are profound. Because of the emphasis on producing reliable predictions, mainstream economics is not concerned, and cannot afford to concern itself, with factors that do not lend themselves readily to quantification or accurate forecasting, or with conditions that are socially specific or historically unique. On the other hand, these are often precisely the issues that anthropologists and historians are most keen to examine. Economists typically ignore these considerations, not because they are uninteresting, but simply because they are irrelevant to, or even distractions from, their project. This is not an indictment of the heuristic value of orthodox approaches. In fact, as Hodgson (1998) points out, mainstream economic theory has proven itself to be by far the most effective at prediction. However, 45 historians often fail to recognize that when they defend or reject the use of these tools, they are debating the merits of their use for purposes for which they were never designed. In fact, it may even be the case that the underlying assumptions of neo-classical analysis would hold as true for ancient markets as for modern ones, but this is an untestable hypothesis given the absence of necessary and relevant data. As a result, a large part of the formalist-substantivist debate may simply be the result of an attempt to hammer a conceptual square peg into a round hole. * * * In conclusion, both sides of the formal-substantive divide have achieved some genuine insights. For instance, substantivists have quite rightly pointed out the often profound effects of social and behavioral norms on economic activity, whereas formalists have been able to demonstrate the ways in which maximizing behaviors frequently take place in any number of pre-modern societies. At the same time, they both are operating under a series of flawed assumptions. For example, substantivists frequently underestimate the existence of market exchange in pre-modern societies while failing to recognize the impact of social obligations in modern economies, while formalists often ignore the existence and impact of rules on economic behavior in both modern and pre-modern contexts. Perhaps most importantly, they both seem to operate on the tacit assumption that mainstream economic theory is an analogue for modern economic theory, thereby failing to appreciate or take advantage of other contributions to modern economic thought. These assumptions have prevented or at least inhibited their ability to move beyond what has become an unbridgeable divide. The terms of the debate are such that the inevitable result is a polarizing, binary choice between mutually exclusive options. There are indications that this situation is changing, and that attempts are being made to move beyond the confines of the current paradigm and towards a more productive approach. As Engen (2010) argues, this shift is still in its early stages, and the outlines of the emerging paradigm have yet to emerge. A number of different conceptual frameworks offer promise, from the New Institutional Economics of North, Coase, Hodgson and others, to the 46 distinction between various forms of capital proposed by Bourdieu, and many more (Wilk & Cliggett, 2007). 47 Chapter Three: The Role of Institutions Having identified the main obstacles to our understanding of ancient economic behavior, we must obviously now find a solution to these problems. Given the way that research into questions of pre- modern economic behavior has been dominated by the question of whether or not it is either wholly unlike or entirely analogous to modern economic activity, it is imperative to establish an interpretive approach that can provide an additional level of nuance, and thereby bridge this trenchant divide. To accomplish this goal, I believe that there are several available resources that have the ability to offer substantial contributions. First of these is a more developed and nuanced notion of class. Because the concept is so strongly associated with the work of Marx, there has been a general reluctance among ancient historians to engage with it. 15 However, given that my main argument centers on what I perceive as a protracted struggle for legitimacy between archaic Greek elites and non-elite landowners, particularly in the arena of maritime trade, I believe that a more sophisticated approach is needed in which it is possible to understand the complex ways in which class, status, and political power are interwoven. The second is the body of ethnographic evidence related to economic behavior. Admittedly, this is not a novel idea, especially in the field of Homeric studies, as Finley (1954) introduced anthropological models to elucidate the social structure depicted in the Homeric epics more than sixty years ago. Also, Tandy (1997) makes use of ethnographic evidence in his analysis of the Greek archaic economy, and there have been notable examples in classical studies that have sought to take advantage of the contributions of the social sciences, especially anthropology 16 . Still, one of the features of ethnographic data is that it can continue to be mined for additional analysis, as the work of Malinowski on the Trobriand Islanders continues to provide insights a century later. 15 With some notable exceptions, esp. de Ste. Croix (1981) 16 Ex., Humphreys (2004 [1978]); Redfield (1991); Cohen (1992). 48 Most importantly, anthropological fieldwork can reveal the variety of ways that various societies have accommodated the introduction of integrated markets within the framework of traditional practices and social norms. As noted, the neo-classical approach takes as a given a specific view of human nature as inclined to engage in commerce and trade. This basic premise has had a number of implications for our interpretation of economic activity in pre-modern societies. First, it has contributed to a teleological vision of economic development, according to which the most advanced or developed form of exchange occurs in a “free” market system where impersonal conditions of supply and demand are the sole determinants of price. Of course, this is one of the significant flaws of Polanyi’s typology of the different modes of exchange, although it has arguably been exaggerated further by his adherents. Also, this underlying premise has fostered the assumption that when traditional societies are exposed to integrated markets, they quickly abandon traditional forms of exchange. However, this is not the picture indicated by ethnographic analysis. For instance, as Tandy (1997) points out, when the Tiv of Nigeria were subjugated by European colonists, they forcibly resisted the imposition of market exchange, seeing it as a foreign threat to the social fabric. However, this is an extreme example, and one that Tandy pushes too far in his desire to draw an indelible line between modern and pre-modern exchange. A more nuanced picture is suggested by Wilk (1997) in his study of the Kekchi Maya in Belize, who have adopted production for trade in larger markets within their traditional family structures. The main point here is that a host of ethnographic studies demonstrate the diversity of economic practices in a wide range of circumstances (introduced externally, imposed from above, or developed internally), and an ability for traditional rules and norms to persist and coexist with all modes of exchange. 17 17 Another work dealing with these issues that is likely more familiar to classicists and ancient historians is Money and the Morality of Exchange (Parry & Block, 1989), in which the authors examine the diverse ways in which money is often incorporated into different cultural contexts. In particular, they cast doubt on many long-held assumptions about money, such as that it erodes social cohesion and depersonalizes relationships. Instead, they argue that the majority of societies make specific arrangements to accommodate impersonal and maximizing behavior without compromising traditional social and cultural norms. 49 Finally, and most importantly, I wish to apply the insights of modern heterodox economics, particularly the New Institutional Economics (NIE) of North, Coase, and others. I believe that this approach is uniquely applicable both to the present case and to ancient economics more generally, especially in the way that it erodes the problematic and persistent distinction between modern and pre-modern economic behavior. What NIE emphasizes is the impact of institutions, often referred to as the “rules of the game,” on the ways in which economic activity is organized and carried out. In particular, the work of the neo- institutionalists has argued forcibly against the atomistic view of modern economic behavior where agents act independently in pursuit of their own perceived self-interest. Maintaining that orthodox economic theory either ignores or takes for granted the existence of these rules when analyzing economic decision- making, North (1991, 1998) points out the various complex ways that systems of rules condition such decisions. As we shall see, the arguments of NIE offer an important interpretive framework for understanding both modern and pre-modern economic behavior in a way that avoids the necessity of viewing them either as qualitatively different, or formally identical. Primarily, my goal is to establish an approach that bridges the modern/pre-modern divide. According to Wilk & Cliggett (2007), the debate is part of a much larger, and often unacknowledged, debate about fundamental human nature. According to this argument, various opposing dyads are introduced into and become part of the very fabric of the debate, such as free will v. determinism, rationalism v. romanticism, and selfishness v. altruism. In essence, the two sides continue to foster an ongoing ideological contest of wills with the formalists adopting a Hobbes-ian approach, and the substantivists standing firmly in the Rousseau-ian camp. Even more problematic for the sake of achieving some sort of resolution, the opposing positions begin from their respective assumptions about human nature without ever testing them. This is, I would suggest, only partially true. The pronounced dualism of the discussion over the nature of non-market behavior makes more sense when considered as part and parcel of the seemingly eternal debate in the social sciences over the primacy of structure versus agency, which is itself a refinement of 50 the “nature versus nurture” question. However, instead of the firm determinism of the latter, which attributes the majority of human behavior to influences (either genetic or cultural) that are beyond their control, the structure-agency dualism considers the degree to which human activity is a function of individual choice or whether or not it is more accurate to analyze behavior in terms of larger structures. Of course, the debate over this issue has been a part of the social sciences seemingly from the very beginning, and the amount of time and space necessary for an even partial treatment would be prohibitive. Instead, I will focus on the more influential and recent efforts to provide some resolution to this long- standing divide. Two of the more influential thinkers on this topic who have strongly shaped my approach are Anthony Giddens and Pierre Bourdieu. According to Giddens (1984), the social system is not simply an inert product of human activity, nor is it an external complex of rules and resources that constrains choice. Instead, what occurs is a sort of feedback loop, whereby the rules and resources drawn upon in the production and reproduction of social action are at the same time the means of system reproduction. As a consequence, the traditional view of structure and agency as a dualism (two independent entities) is erroneous. Instead, they exist as a duality, inseparable from one another, and mutually interdependent. The social system is thus continuously recreated by activities to which it in turn gives meaning, and these are not discrete, individual actions, but rather part of an ongoing, self-referential process: “Human social activities…are recursive. That is to say, they are not brought into being by social actors but continually recreated by them via the very means whereby they express themselves as actors. In and through their activities agents reproduce the conditions that make these activities possible…” (Giddens, 1984). Above all, agency exists in such a way that actions take place with reference to the social system but are not dictated by structure. Every social system both limits some choices and enables others for agents who are reflexive and conscious both of the system and of their own past actions. As Giddens (1984) explains, “All competent members of society are vastly skilled in the practical accomplishments of social activities and are expert ‘sociologists.’ The knowledge they possess is not incidental to the persistent patterning of 51 social life but is integral to it…Structure is not to be equated with constraint but is always both constraining and enabling.” Bourdieu likewise dedicated his efforts to creating an interpretive model for human behavior that accounted for both the influence of systems of rules and individual choice. Bourdieu (1977 [1972]) saw daily activity as a form of improvisation in which people work within systems of rules and norms in a strategic, purposive manner. Agents are culturally informed and leverage their knowledge to pursue their interests, and rules are not deterministic; individuals know what they should do but break and manipulate rules all the time. Nevertheless, they are not completely unfettered, but influenced by cultural experience that produces a set of assumptions about the way the world works. Also important for this project is the portion of Bourdieu’s thought that is specifically relevant to economic behavior, broadly conceived, specifically his notion of the different forms of capital. Bourdieu (1986) postulated that there were three fundamental forms of capital (conceived as “accumulated labor” in material form): economic, cultural, and social. Economic capital is immediately and directly convertible into money and may be institutionalized in the forms of property rights, whereas cultural capital refers to non-financial assets that promote social mobility and is convertible, on certain conditions, into economic capital and may be institutionalized in the forms of educational qualifications. Finally, social capital consists of social obligations ('connections') and is convertible, in certain conditions, into economic capital and may be institutionalized in the forms of a title of nobility. Those who possessed a greater share of one form of capital were at a comparative advantage, and could and often did leverage this advantage in order to acquire a greater share of other forms of capital. He called this process “conversion,” observing that, “different types of capital can be derived from economic capital, but only at the cost of a more or less great effort of transformation, which is needed to produce the type of power effective in the field in question…" (Bourdieu, 1986). However, this process of conversion needs to be disguised by those in positions of power in order to obfuscate the otherwise arbitrariness of their authority, resting as it does in part simply on the possession of greater economic resources. As he explains, “every reproduction 52 strategy is at the same time a legitimation strategy aimed at consecrating both an exclusive appropriation and its reproduction…" (Bourdieu, 1986). The anthropological perspective has also made contributions in a very similar vein, but even more directly related to the formal-substantive debate. In particular, Mark Granovetter (1985) confronts the issue of “embeddedness” directly in his attempt to resolve the conflict between the model of the rational, self- interested individual posited by classical and neo-classical economics on the one hand, and the argument of embeddedness that sees individual behavior as so constrained by social relations that it is not possible to analyze or understand it independently. For him, both extremes are unsustainable, in particular because “the level of embeddedness in economic behavior is lower in nonmarket societies than is claimed by substantivists and development theorists, and it has changed less with 'modernization' than they believe; but I argue that this level has always been and continues to be more substantial than is allowed for by formalists and economists…" (Granovetter, 1985). Granovetter finds fault first with the “over-socialized” concept of man in modern sociology that sees people as overwhelmingly sensitive to the opinions of others, and therefore so obedient to the dictates of consensually developed systems of norms and values that obedience is not perceived as an imposition, and second with the “under-socialized” concept of man in the neo-classical tradition that disallows “any impact of social structure and social relations on production, distribution, or consumption…” (Granovetter, 1985). Adopting a middle position, Granovetter argues that networks of social relations penetrate irregularly and in different degrees in different sectors of economic life, and that distrust, opportunism, and disorder must be acknowledged. What is needed is an approach that can account for individual choices within the context of the structured social environment, whereby choices are framed within the structural system without being wholly determined by it (Engen, 2010; Hodgson, 2006; Granovetter, 1985). It is of course this middle position that is both intuitively plausible, and also supported by the empirical evidence. In addition, such approaches have found support from the social sciences (Bourdieu, 1977, 1986; Giddens, 1979, 1984; Granovetter, 1985; Sahlins, 2004) as well as from economics (Williamson, 1975, 1981, 1998; North, 53 1991, 1998, 2005; Hodgson, 1997, 1998, 2006). Also, despite the frequent tendency to erroneously conflate neo-classical approaches with the entire body of modern economic thought, these arguments have long been a part of economic theory. From Veblen (1973 [1899]) on, institutional approaches to economic behavior have long stressed the importance of social and cultural considerations in any analysis of economic activity. It was Veblen’s emphasis on institutions as more than simple constraints on individual action but rather embodiments of generally accepted values and behavioral norms that was to constitute the foundation of the institutionalist approach: “At the most basic level, the most important element in the institutionalist approach is the conception of the economic system as a set of evolving social institutions…” (Rutherford M. , 2003). 18 In recent years, however, many economists have become increasingly frustrated with the perceived shortcomings of the neo-classical approach and its tendency towards greater abstraction at the expense of consistency with empirically observable phenomena. That has coincided with a growing interest in developing models that are more consistent with real-world conditions. One of the more promising models is that of the New Institutional Economics made popular by Coase (1937, 1960, 1998), Williamson (1975, 1981, 1998), and North (1980, 1981, 1987, 1991, 2005), among others. Working self- consciously within the tradition of the “old” institutional economics (OIE) of Veblen, Commons, Mitchell, and others, New Institutional Economics (NIE) as a term was coined by Oliver Williamson in order to distinguish it from the older institutionalist approach (Dugger 1995). However, as the name implies, NIE shares many of the same premises of OIE. In particular, both schools stress the role of institutions, broadly conceived as systems of rules, on economic behavior, and that in all economic systems, market economies included, the institutional framework of the society produces the incentive structure and determines economic performance. In addition, both approaches recognize that institutions 18 The history of the development of institutional economics is obviously beyond the scope of the present discussion. For a fuller treatment of the subject, there are excellent reviews in Rutherford (2000, 2001, 2003), Blaug (1997), and Williamson (1998). 54 change over time and often respond to economic factors (Rutherford M. , 1995). While the older institutionalism has diminished in importance, it has not disappeared entirely (Furobotn & Richter, 2005), and the term “institutional economics” has been applied to an increasing variety of economic approaches (Rutherford M. , 2001), and even within the narrower category of New Institutional Economics, a lack of unanimity can make it difficult to assess the core principles involved (Hodgson, 1998). Nevertheless, in light of its potential to offer an interpretive approach that resolves the persistent formal-substantive controversy, it is to this approach that we now turn. What is the New Institutional Economics? Although there is nowhere near enough space in this current project to trace the full development of institutional economic thought from its origins to the present, in order to apprehend the nature of the New Institutional Economics, it is necessary to spend a modicum of effort on a brief survey on the origins of this approach, if for no other reason than to separate out the central tenets from those which are more tangential. Also, many criticisms of NIE rest on fundamental misapprehensions of the approach, which necessitates a basic understanding of the concepts involved in order to properly address them. In addition, as we have already seen, the New Institutional Economics of North and Williamson shares much in common with the older institutionalism of Veblen and Commons. As a result, we need a firmer grasp on the essential differences between them if we are to understand just what it is about the New Institutional Economics that makes it “new.” As noted previously, NIE is linked conceptually to the institutionalism of the first half of the twentieth century. However, the inspirational spark for NIE is not the work of Veblen, Commons, Mitchell, or any of the early institutionalists. Instead, it is almost universally acknowledged that NIE begins with the transaction cost approach of North and Williamson, which was itself inspired by the work of Ronald Coase (Williamson 1981; Eggertsson 1990; Dugger 1995). Coase (1998) himself admits that his work ultimately gave rise to NIE, although he also points out that what has developed is far beyond what he could have possibly envisioned from its rather modest beginnings. Regardless, the fact is that transaction 55 cost analysis, and in turn NIE, can at least provisionally be attributed to Coase’s seminal 1937 article, “The Nature of the Firm.” In that article, Coase (1937) raises the simple and (in retrospect) glaringly obvious question of why firms exist at all in a market-based economy, and in so doing exposed one of the major flaws of neo-classical theory, and the implications of his initial question continue to reverberate even today. In short, Coase was struck by the apparent inability of neo-classical economic theory to explain the existence of the firm. According to the conventional view of the market as self-regulating, in which the “normal economic system works itself...by a process that is automatic, elastic and responsive…” (Salter, 1921), there should be no reason for the firm, in which several production functions were coordinated under a single governance structure, to exist. As Coase (1937) asks, “if co-ordination will be done by the price mechanism, why is such organisation necessary?” The simple and obvious answer, for Coase, was that there were costs associated with using the price mechanism, and the firm was a form of co-ordination intended to reduce these costs. It was this basic revelation that transactions are costly, and that transaction costs “that arise when individuals exchange ownership rights to economic assets and enforce their exclusive rights…” (Eggertsson, 1990) have profound implications for the allocation of resources and the structure of economic organization, that inspired the transaction cost approach of Williamson and North. For Williamson (1981), the transaction becomes the basic unit of analysis, and transaction costs are the driving force behind the observable forms of economic organization. From this point, it is easy to see how the emphasis on transaction costs and their effects on governance structures could give rise to an examination of the effects of institutions on economic activity. As more and more forms of organization were examined through the transaction cost lens, it became clear that neo- classical theory simply accepted as given an enormously complex matrix of institutions in order to make economic models tractable. As Coase (1998) would later explain, “Mainstream economics, as one sees it in the journals and the textbooks and in the courses taught in economics departments has become more and more abstract over time, and although it purports otherwise, it is in fact little concerned with what 56 happens in the real world…Economists study how supply and demand determine prices but not with the factors that determine what goods and services are traded on markets and therefore are priced. It is a view disdainful of what happens in the real world.” To be sure, under conditions of perfect efficiency/costless transacting, the neo-classical model was an accurate description. However, such a model requires numerous stringent requirements, none of which can survive critical scrutiny (North, 1990). In this context, the New Institutional Economics was formed in order to account for the enormous range and variety of structures that were highlighted by the recognition of the importance of transaction costs. As the thread was followed, it became clear that the costliness of transactions frequently gave rise to complex institutional arrangements, and that these institutions in turn had dramatic and far-reaching effects on the costs of transacting. However, these institutions are not ideologically neutral. Taking individuals as given must presume a number of rules, cultural and social norms, etc., and ultimately one is forced to the conclusion that even the ultimate “given” of mainstream economics, namely the market, "is structured by, operates within, and gives effect to the institutions which organize it…” (Samuels, 1984). In fact, the implication of institutional analysis is that the market itself is an institution, composed of a host of subsidiary institutions, through which resources are allocated, and that it is but one of many possible mechanisms for the allocation of resources. As Bresson (2007) argues, NIE is characterized by its focus on transaction costs. This is true to an extent, but is gradually becoming less accurate as NIE continues to develop. For one, the transaction cost approach still shares much in common with neo-classical theory. Most importantly, it assumes efficiency as the basis for the observed forms of economic organization. Admittedly, this can have heuristic value, especially in relatively more efficient developed economic systems, but as Williamson (1981) admits, “Economic approaches to the study of organization...generally focus on efficiency. To be sure, not every interesting organizational issue can be usefully addressed...in efficiency terms.” To the extent that transaction cost analysis is focused on an examination of the governance structures that arise in response to transaction costs in a market system, this efficiency postulate is sustainable. However, when applied to 57 an examination of economic activity in history, or even in many parts of the modern world, where inefficient institutions are abundant and persistent, the efficiency postulate is not realistic. As economists began increasingly to question neo-classical models on the basis of their shortcomings in explaining real-world economic phenomena, some went even further and began to question the behavioral assumptions of neo-classical theory. Of course, such questions had already been raised, perhaps most notably by Nobel laureate Herbert Simon (1957), who observed that the rationality postulate of classical and neo-classical economics did not account for the significant effect of social considerations in the ways that economic actors make choices. He noted that while agents do behave in predictable, rational ways in many instances, in others they respond more powerfully to emotional or non-rational motivations. In addition, human beings are simply incapable of performing the kind of complex evaluative calculus that neo-classical theory postulates. As a result, he concluded that human beings exhibit a kind of limited, “bounded” rationality in framing economic choices. Both Simon’s work and that of the transactional cost approach were conceived as ways of refining mainstream economic theory, and introducing more realism into traditional models. However, as these attempts to make mainstream economics more realistic continued to highlight its flaws, many have begun to doubt the feasibility of such a project. As a result, the development of NIE appears to be moving further and further from a set of assumptions shared with mainstream economic theory. For now, it seems that there are several questions that must be answered about NIE before it is possible to use this approach for an examination of the ancient world. First, what exactly are “institutions”? Second, how do institutions work? Third, what are the unique features and contributions of NIE? Finally, and perhaps most important, how is an institutional approach specifically relevant to an examination of pre-modern economic activity? It is to these questions that we now turn. 58 What are Institutions? As Hodgson (1998) has observed, there is less than unanimous agreement among neo-institutionalist economists about their core precepts. However, he also admits that there are shared features among the more prominent neo-institutional theorists, including North, Williamson, and Posner. Of course, the most fundamental concept that must be addressed in order to achieve an adequate grasp of this approach is a basic definition of institutions. Furobotn and Richter (2005) provide an introductory definition of the term as “a set of formal and informal rules, including their enforcement arrangements.” This of course requires further elaboration, and the following are several notable definitions: • “systems of established and prevalent social rules that structure social interactions. Language, money, law, systems of weights and measures, table manners, and firms (and other organizations) are thus all institutions…” (Hodgson, 2006) • “"Institutions are the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interaction. In consequence they structure incentives in human exchange, whether political, social, or economic…” (North, 1990) • “the humanly devised constraints that structure political, economic and social interaction. They consist of both informal constraints (sanctions, taboos, customs, traditions, and codes of conduct), and formal rules (constitutions, laws, property rights)…" (North, 1991) • "the sets of working rules that are used to determine who is eligible to make decisions in some arena, what actions are allowed or constrained, what aggregation rules will be used, what procedures must be followed, what information must or must not be provided, and what payoffs will be assigned to individuals dependent on their actions...All rules contain prescriptions that forbid, permit, or require some action or outcome. Working rules are those actually used, monitored, and enforced when individuals make choices about the actions they will take…" (Ostrom, 1986) 59 • “a set of formal or informal rules, including their enforcement arrangements (the ‘rules of the game’), whose objective it is to steer individual behavior in a particular direction. Examples: Ownership, marriage, friendship, market system, customer relations, monetary system, the state, the firm, industrial networks…” (Furobotn & Richter, 2005) These are just a few examples of many, and there are clearly different points of emphasis. However, there are several main points on which nearly all agree. First, institutions provide a structure to human interaction. This is a central issue in the dynamics of how institutions work, a topic that will be addressed directly in the next section, so I will hold off on treating it here. Second, there are several components of institutions: formal rules, informal constraints, and enforcement arrangements. According to North (1981), formal institutions are formulated and codified, and include political and judicial rules, economic rules, and contracts, whereas informal constraints are norms of behavior embodied in customs, traditions, and codes. For others, the distinction between formal and informal is much more nebulous and complex than North acknowledges. For instance, Hodgson (2006) argues that the basis for this distinction is unclear; at times, North seems to suggest that difference is between tacit and explicit rules, or perhaps legal versus non-legal, and at other times designed versus spontaneous. In fact, he contends that it is a mistake to attempt to make such a distinction at all, since there is no observable instance where formal institutions do not depend on informal rules and tacit norms in order to operate. In addition, an over- emphasis on legal rules can ignore the existence of rules and institutions that can also constrain and mold human behavior in significant ways (Hodgson, 2006). In any case, formal and informal institutions are only part of the story, and they are seriously weakened, if not rendered wholly ineffective, in the absence of some sort of enforcement arrangements. In the case of formal rules, these usually take the form of penalties under the law and are enforceable by the state. For informal constraints, penalties can include ostracism, expulsion from a group, public shame, and others. The main point here, however, is that all agree that institutions are inclusive of all three. 60 The third point of emphasis is that institutions steer behavior in a particular direction. Institutional constraints include what individuals are prohibited from doing, and also under what conditions some individuals are permitted to undertake certain activities (North, 1990). However, it is not the case that institutions function solely as constraints on human behavior. By forming a structure that creates more or less stable expectations about the behavior of others, institutions create a framework within which agents can develop strategies. In fact, the very existence of an institutional structure often creates possibilities that would not have existed in its absence. Finally, we should remember that institutions are not external to human activity. They are, in fact, the products of collective action, and as such, even if they often precede any particular individual, they remain subject to alteration. However, this process is frequently neither simple nor transparent. While formal institutions are subject to change by fiat, and can therefore be altered rapidly, informal institutions typically change much more incrementally, and generally without deliberate, concerted direction. Ultimately, institutions combine to form a framework that consists of the political structure that specifies the way political choices are developed and aggregated, the property rights structure that defines the formal economic incentives, and the social structure (i.e., norms and conventions) that defines the informal incentives in the economy. The features of neo-institutional analysis that constitute a departure from prevailing orthodox economic theory are the notion of bounded rationality, the introduction of transaction costs, and an emphasis on the vital importance of property rights as a determinant of economic performance (Ankarloo, 2002). How Do Institutions Work? Of course, beyond simply providing a definition of institutions and listing some of their central characteristics, it is vital to be able to describe the process by which they operate. To begin with, quite simply, human interaction takes place within a context of rules. This is a constant that is all too often uncritically considered as a given in mainstream economic theory, but one that any anthropologist knows full well has profound implications. For the transaction cost approach, relying as it does on the 61 assumption of efficiency, institutions can be explained as a response to the fact that transactions are costly, and that institutions contribute to the reduction of these costs (see Williamson 1973, 1981, 1989, 1998). In other words, transaction cost analysis sees institutions as efficient solutions to problems of organization in a competitive framework. This approach assumes market exchange, vertical integration, etc. as efficient solutions, but does not address the varied performance of economies in different times and places (North, 1991). In addition, such an approach is incapable of explaining the persistence of inefficient institutions that in many cases can be observed to last over extended periods of time, even under nearly optimal conditions. Clearly, the explanation for the existence, ubiquity, and pervasive impact of institutions must lie elsewhere. The answer lies in the most basic function of rules, namely the reduction of uncertainty. It is the existence of an embedded set of institutions that makes it possible for actors to avoid extensive calculations or engage in complex analyses for every transaction. For the vast majority of people, daily life is made up of routines in which the matter of choices appears to be regular (North, 1990). In fact, the evidence would suggest that greater degrees of uncertainty actually cause rule-governed behavior to become more and more predictable. As Heiner (1983) points out, greater uncertainty causes behavioral rules to become more restrictive in eliminating particular actions or response patterns, which in turn further constrains behavior to simpler, less sophisticated patterns that are more predictable. Contrary to the expectation of orthodox economics, “rule-governed behavior means that an agent must ignore actions which are actually preferred under certain conditions…[and] greater uncertainty will cause rule-governed behavior to exhibit increasingly predictable regularities, so that uncertainty becomes the basic source of predictable behavior.." (Heiner, 1983). In fact, in many cases institutions can arise that affect behavior so strongly that they appear to virtually eliminate the effects of individual choice. These are what Hodgson (2003) calls “agent insensitive” institutions, or those that are largely unaffected by changing agent preferences. Going even further along these lines, it may even be said that the institutional structure of the market system is substantially 62 responsible for reproducing behavior that is consistent with the assumptions of mainstream economic analysis. As he explains, numerous experiments have been conducted that demonstrate, for example, that behavior ruled by habits or inertia is just as capable as rational optimization of predicting the standard downward-sloping demand curve and profit-seeking activity of firms. He concludes, “Ordered market behavior can result from the existence of resource and institutional constraints and may be largely independent of the 'rationality,' or otherwise, of the agents...We thus face the possibility of a study of markets that focuses largely on institutions and structures, to a degree independent of the assumptions made about agents…" (Hodgson, 2006). If we accept the premise that the market is an institution, itself composed of a host of subsidiary institutions, and that the resulting incentive structure is such that the institutional matrix is highly insensitive to variabilities of agency, then the entire explanatory basis of classical and neo-classical economics, namely that market exchange is a product of agent preferences, is reversed. Of course, such a situation as that described above would seem to lend credence to the structure side of the structure-agency debate. However, this phenomenon is not universal. As Hodgson (1997) also explains, there are numerous counter-examples of agent sensitive institutions that are highly responsive to changing tastes, and the impact of agent preferences is often substantial. According to North (1990), “It is simply impossible to make sense out of history (or contemporary economies) without recognizing the central role that subjective preferences play in the context of formal institutional constraints.” Individual utility functions are simply more complicated than the simple assumptions so far incorporated in orthodox economic theory, and “even with a constant set of rules, detection procedures, and penalties there is immense variation in the degree to which individual behavior is constrained. Strong moral and ethical codes of a society is the cement of social stability which makes an economic system viable…" (1981). Ultimately, the institutions arise in response to conditions of uncertainty, but once in place, they are not wholly determinative, and they present a range of potential choices to the individual agent. As a result, the reduction of uncertainty is rarely completely diffuse. The deep underlying source of institutions 63 has been and continues to be the effort to structure the environment to make it more predictable, but this effort can (and often does) make for increased uncertainty for some of the players. This function of institutions as a means to reduce uncertainty is consistent with recent work in cognitive science, demonstrating pattern-based thinking as a strong tendency of human thought. This is due quite simply to the sheer impossibility of calculating and predicting possible outcomes in the tremendous complexity of the real world. The variables are such that except in extremely rare circumstances, the degree of computational difficulty involved in the process suggested by the optimization of orthodox economic theory is far too high to make such calculus feasible, or even possible: "Combinatorial explosion paralyzes even moderately domain-general systems when encountering real-world complexity. As generality is increased by adding new dimensions to a problem space or new branch points to a decision tree, the computational load increases with catastrophic rapidity…" (Cosmides & Tooby, 1994). As a result, human beings resort by necessity to rules and habits, and construct institutions that tend to limit the potential variables and reduce the complexity of the environment. However, as we have already noted, institutions are not invariably the creation of deliberate strategies. Broadly speaking, there are two basic types of institutional creation. On the one hand, there is what Furobotn and Richter (2005) call “deliberate design,” Hayek (1973) calls “made order,” and Williamson (1991) refers to as “intentional.” On the other hand are institutions that are typically termed “spontaneous.” In general, formal institutions are produced deliberately, and informal institutions arise more organically and spontaneously. As a result, it is usually much easier to analyze the ways that formal institutions change as opposed to informal ones. This of course leads us to an important point: institutions change human behavior, but are themselves a product of human activity and are also both reinforced through repetition. They can also be altered, modified, or abolished through collective action. As a result, they both shape and are shaped by human activity in an ongoing feedback loop whereby behavior is informed by institutions, and institutions are affected by behavior. This dynamic must be understood as a continuous process. Actions do not take place as a series of discrete but separate events. Actors are reflexive, and participate in a recursive, 64 developing environment in which past actions inform present ones, and shape the incentive structure so that the available choices are constantly evolving (Giddens, 1984). As a result, "Actors do not behave or decide as atoms outside a social context, nor do they adhere slavishly to a script written for them by the particular intersection of social categories that they happen to occupy. Their attempts at purposive action are instead embedded in concrete, ongoing systems of social relations…" (Granovetter, 1985). Put another way, “Because institutions simultaneously depend upon the activities of individuals and constrain and mold them, through this positive feedback they have strong self-reinforcing and self-perpetuating characteristics…” (Hodgson, 2006). This process has particularly important implications for the formal-substantive debate. Instead of requiring a binary choice between two mutually exclusive options, the feedback loop of behavior and institutions explains how the modern market economy is much more embedded in a social and cultural context than is presented by mainstream economic analysis, and at the same time that pre- modern, non-market systems present a wide range of choices for the individual actor, including the pursuit of self-interested optimization through the application of rational strategies within the particular incentive structure produced by the institutional matrix. Agents are no more autonomous in the modern economic world than in the ancient world, and they are no more bound by social and cultural constraints in the ancient world than in the modern. There is abundant evidence that demonstrates the extent to which business relations are merged with social relationships, and the blending of business and social activity, especially among business elites, is exceptionally well-documented. For example, dispute settlement frequently relies upon embedded business relationships, to the extent that there appears to be strong reluctance to use the formal institution of the legal system to enforce contracts, and they are often only a last resort, and examples abound of the extensive use of subcontracting in many industries that also provides opportunities for sustained relationships (Dumhoff, 1971; Useem, 1979). In fact, Granovetter (1985) insists that “the anonymous market of neoclassical models is virtually nonexistent in economic life and that transactions of all kinds are rife with…social connections.” 65 However, we must not ignore the powerful implications of the fact that institutions are subject to change. Since they can be altered by actors, the possibility always exists that they can be changed according to the perceived benefit of an individual, if powerful enough, or certainly a dominant minority. This is particularly true for formal institutions, since they can be changed much more readily. As a result, formal rules are often much more reflective of the prevailing power structure than informal institutions. Of course, this raises the question of just how institutions change. As we have already noted, formal institutions cannot function in the absence of informal support, and the evolution of informal rules is often seen as the force that in the end may lead to institutional change (Furobotn & Richter, 2005). One process that has the potential to cause change in the institutional framework is growth; also, technological change, more efficient markets, demographic changes, etc. can all have destabilizing effects. Of course, stagnation can also be destabilizing in a competitive environment (North, 2005). Ultimately, the reasons for institutional change are many and varied, and it may simply be impossible to describe with perfect clarity the precise workings of institutional change. Nevertheless, history demonstrates that institutions are constantly evolving, which in turn continually alters the available choices. This process of change is typically incremental rather than discontinuous, for a number of reasons. For one, since informal institutions are not subject to change by fiat, they tend in turn to make it more difficult for the rapid change of formal rules, which can be changed instantaneously as a result of political or judicial decisions. Also, the path of change is shaped by the “lock-in” that comes from the symbiotic relationship between organizations that have evolved as a consequence of the incentive structure of the institutions. This lock-in is the product of the dependence of organizations on extant institutions, and both formal and informal institutional constraints give rise to particular exchange organizations (North, 1990). Third, change is affected by the feedback process by which human beings perceive and react to changes in the opportunity set. The reality is that political and economic markets are simply not efficient, and incremental change is produced by the perceptions of actors that they could do better by altering the existing institutional framework at some margin (North, 1990). Finally, broad based 66 support for action frequently requires ideological commitment in order to overcome the free rider problem. The fact is that change is continually occurring as entrepreneurs adopt strategies to improve their competitive position. The result is alteration of the institutional matrix, and thus revisions of perceptions of reality, and then in turn new efforts by entrepreneurs in a never-ending process of change, but this process is hindered by the fact that dominant beliefs over time result in the accretion of an elaborate structure of institutions that determine economic and political performance (North, 2005). Of course, it is important to recognize that institutional structures are rarely, if ever, equally or universally beneficial. For one thing, the limited capacity of agents to assess their own needs or interests accurately can have a dramatic effect on the efficacy of proposed changes. Both the knowledge and computational power of the decision maker are severely limited (Simon H. A., 1986). The fact is that institutions often do not produce the desire result, even when they are the product of deliberate direction. As North (1998) points out, deliberate policies are at best only blunt instruments, and limited in their impact only to the formal rules. Also, it is important to avoid the functionalist trap; even when the consequences of the adoption of a habit or rule are beneficial, this does not automatically explain why individuals adopt the habit or rule (Hodgson, 1997). It is also vital to evade the evolutionist trap; growth and improvement are not inevitable. Path dependence can lead to stagnation or decline (Dugger, 1995), and institutions are not established for their own efficiency, but rather to serve the interests of those powerful enough to establish them (North, 1990, 1991, 1993). In fact, while norms can be sensible, they do not need to be. There are numerous examples of norms that are not obviously adaptive, or even are spectacularly maladaptive (Boyd & Richerson, 2001; Furobotn & Richter, 2005). The major role of institutions in a society is to reduce uncertainty by establishing a stable, but not necessarily efficient, structure to human interaction. The actual institutional framework is usually a mixed bag of institutions that promote productivity-raising activities and those that provide barriers to entry, encourage monopolistic restrictions, and impede the low-cost flow of information (North, 1990). 67 Further complicating matters is the fact that the institutional matrix consists of a mix of informal constraints, formal rules, and their enforcement arrangements, and these do not always function cohesively. One example that can serve as an illustration of the way that this dynamic can take place is through the analogy of a sporting contest. There is, of course, a codified system of formal rules, established by a governing body. There are also enforcement arrangements, in the form of umpires or referees, for these formal rules. However, as any player can tell you, there is an enormous significance assigned to the “unwritten rules.” For example, it is unquestionably a violation of the informal rules to try and deliberately hurt another player, and there are enforcement arrangements for violations of the informal rules as well. In fact, in some sports there are players whose job it is to physically punish opposing players in the event of a perceived transgression, and the presence of these players is thought to be a deterrent; such role players are frequently referred to as “enforcers.” However, there is a complex script that must be followed, and there are limits to how such an attempt at enforcement can play out. Also, it is not only the players who are bound by these conventions. For instance, in most sports there are some formal rules that are considered significant and others that are considered trivial, and umpires or referees are typically expected to refrain from enforcing the supposedly trivial rules, especially during particularly crucial points in the contest. Finally, informal rules can be extremely influential in the development of strategies (Romer, 2006). In fact, they are such a commonly accepted part of strategic planning in sport that following these unwritten conventions is often referred to as doing things “by the book” (Tango, Lichtman, & Dolphin, 2007). In the same way, informal rules (conventions, norms, customs), formal rules (laws, governments), and their enforcement arrangements are inextricably linked and integral to economic performance (North 1990). In particular, while politics and economics are conventionally considered separately, what NIE demonstrates is the myopia of such a view. As Furobotn and Richter (2005) explain, “Political and economic processes cannot be separated…People are not merely consumers and producers; they are also citizens in a variety of polities that not only regulate markets but can expropriate directly the resources 68 markets allocate. Correspondingly it is impossible to predict market outcomes without also predicting the political response that alternative outcomes engender.” Of course, such a view represents a fundamental break with mainstream economic analysis that considers economic performance independently, and regards associated political structures as given. This does raise the question of the degree of reconciliation that is possible between NIE and mainstream economic analysis. 19 To be sure, opinions range widely on this subject, ranging between two extremes. Some, like Hodgson (1997, 1998, 2006), argue that the institutional approach, with its focus on the impact of rules systems that shape individual behavior and the consequent rejection of the methodological individualism of orthodox economic theory, is incompatible with the neo-classical approach. At the other extreme are those like Eggertsson (1990) who believe that NIE is simply a means of adding a greater degree of realism to orthodox analysis by including areas of inquiry that have typically been ignored, but that the fundamental neo-classical assumptions remain intact. Not surprisingly, there has been some change as neo-institutional analysis has developed. Generally speaking, the early NIE work, including North (1976, 1978), Coase (1937, 1960), and Williamson (1973), originated from attempts to address some of the shortcomings of mainstream economics, but stopped far short of actually questioning the very foundations of neo-classical theory. Indeed, this continues to be the basic approach of the transaction cost approach (Williamson 1989, 1991, 1998) that rejects the notion of perfectly efficient markets, but still explains the existence and form of institutions as efficient responses to the costs of transacting. More generally, NIE initially developed as an attempt to apply the apparatus of neo-classical theory to explain the workings and evolution of institutional arrangements, and to expand the scope of neo-classical economics (Furobotn & Richter, 2005). 19 It is impossible in this forum to engage in a full discussion of the potential for neo-classical and neo-institutional analyses to be reconciled. However, many of the criticisms of the neo-institutional approach to economic history are largely grounded in critiques of the supposed neo-classical features of NIE. For an excellent review of this issue, see Furobotn and Richter (2005). 69 As NIE has progressed, however, increasing doubts have arisen about the possibility of maintaining even a tenuous connection to mainstream economics. Granovetter (1985) contends that the main thrust of NIE is to deflect the analysis of institutions from sociological, historical, and legal argumentation, and to show instead that they are the result of efficient solutions to economic problems. However, it is generally recognized that in the more than two centuries since Adam Smith, mainstream economic research has primarily involved the examination of a single set of idealized rules governing market exchange (Eggertsson 1990; Dugger 1995). This has produced increasing tension as NIE appears to move gradually further from an orthodox foundation. Even Eggertsson (1990), who argues in favor of maintaining the link with orthodox economics, is careful to point out that this position is not consistent with that of many other neo-institutionalists who reject fundamental neo-classical approaches such as the rational choice model. As Furobotn and Richter (2005) explain, “In general, the representatives of the NIE have not been content to merely criticize orthodox neoclassical theory. Rather, they have attempted to produce new theoretical constructs capable of explaining areas of economic life that have previously been ignored and of showing why certain neoclassical theorems cannot be validated in real life.” Going even further, they contend that “a proper understanding of institutional phenomena requires the development of a new paradigm and that significant progress cannot be made simply by seeking to extend and generalize the neoclassical model…” (Furobotn & Richter, 2005). In fact, it would seem that neo-institutional and neo- classical economics have been steadily moving in opposite directions, with mainstream economics favoring an increasingly technical path of development (North, 1976, 1978; Furobotn & Richter, 2005). At the same time, many neo-institutional theorists, in particular Douglass North, have questioned and even flatly rejected many assumptions that are fundamental to orthodox theory, and North’s work appears to be moving even further away from this foundation (Dugger, 1995; Rutherford, 1995, 2001). In fact, North (1981) has gone so far as to insist that the “economic historian who has constructed his model in neoclassical terms has built into it a fundamental contradiction since there is no way for the neoclassical model to account for a good deal of the change we observe in history.” 70 Of course, many of the supporters of orthodox economics defend it on the basis of the fact that models, by their nature, must necessarily approach a level of abstraction that precludes the possibility of accounting for every single permutation of economic activity. Interestingly, this response to empiricist criticism is much the same as that raised by Ian Morris (1999a) in defense of Moses Finley’s substantivist position regarding the ancient economy. This is of course widely recognized, and North (1991) admits that we “construct in our minds a synthetic framework designed to explain how the economy works…The model is always a very imperfect reflection of how the economy really works.” The question is not whether or not the traditional model can account for every single instance of economic behavior, or each and every empirically observable phenomenon, but rather, given the complexity and diversity of the various forms of economic organization in history, whether a universally applicable model is even feasible. Certainly, NIE has no aspirations to produce “an elegant, all-embracing, axiomatic institutional economics in the style of the Arrow-Debreu general equilibrium model…” (Furobotn & Richter, 2005). Instead, neo- institutionalist economic analysis is characterized by several common features: first, a primary concern for institutions, habits, rules, and their development through specific and historically located approaches; second, an approach that moves from general ideas about human agency, institutions, and evolution of economic processes to specific ideas and theories related to specific institutions or economic types; third, an incorporation of psychological, anthropological, and sociological research into how people behave (as opposed to standard theoretical models of rational individuals); fourth, the absence of a general theory of price in favor of a set of guideline approaches to specific problems that produces historically and institutionally specific studies with more operational value than an all-embracing general theory (Hodgson, 1998). What Does NIE Bring to the Table? To be sure, at this point it would seem appropriate to raise the question about what particular positive contributions a neo-institutional approach can add to the discussion. NIE rejects a wide range of basic tenets of formal economic theory, including the rationality assumption (Heiner 1983; Henrich et al. 2001), 71 optimality (Hodgson 1997, 1998, 2003), general equilibrium (Furobotn & Richter, 2005), economics as a science (Samuels, 1984), the self-regulating market (Coase, 1937), the lack of opportunism (Williamson 1975, 1981, 1998), efficient markets (Paarlberg, 1993), and utility maximization (North, 2005). Thus, it is clear that NIE differs from mainstream economic theory on a number of levels, but how exactly is this helpful? First of all, perhaps the single most important contribution of NIE has been its ability to demonstrate unequivocally the powerful and pervasive influence of rules (institutions) on economic behavior. This is a dynamic that orthodox economics takes as a given, and as a result is simply unable to explain a wide range of economic behavior, in the modern world as well as the ancient world. Although NIE does not produce a general model, it nevertheless offers a means of explaining and understanding the diversity of economic systems in different times and places. The implications of this development are extremely powerful. For one, NIE accommodates the process of change, and allows for the contextualization of economic change within the framework of broader processes of technological, cultural, and political change. As a result, NIE demonstrates the arbitrariness of the distinction between modern and pre-modern economic systems, first by demonstrating the diversity of economic organization in the modern world, and second by revealing that the underlying dynamics that produce this diversity can be understood diachronically. In other words, the particular effects of institutions are historically contingent, but their importance in creating historically unique structures is consistent over time. Also, NIE coincides with the larger trend within the social sciences, notably through the work of Giddens and Bourdieu, among others, which attempts to resolve the structure-agency antinomy. Giddens’ work on structuration is strongly supportive of the general precepts of neo-institutional theory, which are in turn finding greater empirical support as NIE is applied to more specific cases. In particular, NIE provides an approach that allows both for the strategic choices of individuals within a larger institutional matrix, and a recognition of the ways that the incentive structure produced by the institutional framework both 72 constrains the behavior of individual agents and offers opportunities for action. In addition, both structuration and NIE account for the fact that decision-making and rule-following often take place on a level that is not explicitly deliberative, and therefore is a more realistic approximation of human behavior that is finding increasing support in cognitive science. Finally, it should be noted that NIE explicitly rejects the evolutionist model that is implicit in development economics. Of course, evolutionism is an explicit feature of the Austrian School (North, 2005), but it is a recurrent feature of nearly all mainstream theory, which regards economic structures as the product of efficient responses in conditions of competition. From the very beginning, a tacitly evolutionist approach was a fundamental feature of classical economics, and it continues to be a mainstay of neo-classical explanatory models (Williamson, 1991). In addition, this basic assumption is a common feature in numerous other disciplines, appearing as a prominent element in theorists from Marx to Weber (Sahlins, 1999). By contrast, NIE does not presume progress toward a pre-figured goal, recognizing that institutions are capable of enhancing or reducing efficiency (North, 1990). As a result, the teleological assumption that is inherent in the contrast between “primitive” and “modern” is absent from the neo-institutional approach. In addition, many of the lessons from anthropology lend further credence to this position. For instance, Sahlins (1999) points out that one of the surprises of late capitalism is that many hunter-gatherer groups in the world still live by hunting and gathering while living in the shadow of the market system. Citing examples from the Bering Sea to New Guinea, he demonstrates that the common assumption that market exchange would come to dominate and overrun societies that are exposed to it has been clearly shown to be false. The survival of unique, culturally specific forms of economic organization is not a simple function of their isolation, since the subsistence of many contemporary hunter-gatherer societies is dependent upon modern means of production, transportation, and communication. Also, contrary to the famous arguments of Bohannon (1955, 1959) and Polanyi (1944, 1947, 1957), among others, the introduction of so-called “general purpose” or “modern” money does not 73 inevitably annihilate traditional exchange relationships. Quite the contrary, it appears that the potential for variation in the way that money is represented in different cultures is limited only by the human capacity for innovation (Parry & Bloch, 1989; Gregory, 1997). For instance, among the Enga, Mendi, Siane, and many others of New Guinea, money has enabled the reproduction of traditional exchange practices: “Rather than the antithesis of community, money has thus been the means. High- value bank notes replace pearl shells as key exchange valuables, gifts of Toyota land cruisers complement the usual pigs, and large quantities of beer function as initiatory presents…” (Sahlins, 1999). This is also echoed by Giddens (1999), citing evidence from empirical observations across the globe that contrary to traditional theory, the isolation of the individual, secularization of society, breakdown of kinship relations, and the impersonalization of social relationships in urban settings are not automatic or inevitable. The only possible conclusion that can be drawn from this evidence is that the evolutionary model of classical and neo-classical economics is not consistent with real-world conditions. Economic organization, both in the modern and ancient world, must be regarded not by some external standard of logic or economic rationality, but rather by the “rationalité interne” (Bresson, 2007) that is consistent with the incentive structure produced by the institutional matrix specific to a given society. The Relevance of NIE to the Ancient World Thus, in a general sense, it seems clear that a neo-institutional approach presents numerous advantages over the uncritical cross-cultural application of orthodox economic models. However, I would argue that it has even greater relevance to a study of the ancient world, and to this project in particular, for a number of reasons. For one, as has already been noted, one of the central concerns of NIE is change over time. Even a brief glance at the titles of Douglass North’s work reflects this focus, for example: Structure and Change in Economic History (1981), Understanding Economic Change (1998), and Understanding the Process of Institutional Change (2005). In so doing, North (and by extension, NIE) provides insights into some of the particular aspects of economic change. First, that institutional change is typically incremental, 74 due to the nature of the institutional matrix as a combination of formal and informal institutions, along with their enforcement arrangements. Although formal institutions can be changed by fiat, since they inevitably depend on informal institutional support for their authority, the potential for discontinuous change in the entire institutional framework is limited at best. For instance, the Emancipation Proclamation may have changed the formal status of slaves in one fell swoop in 1863, but for the proclamation to take effect required viable enforcement arrangements, in this case the presence of the Union army, in those states where the validity of the proclamation was rejected. And, of course, the informal ideological basis that allowed for the enslavement of other human beings and justified it on the basis of the supposedly objective (but obviously culturally contingent) category of race continues to be a feature of American society even to the present day. Also, NIE introduces the concept of path dependence to its analysis, helping to clarify this often misapplied notion. As North (2005) explains, “History matters...not just because we can learn from the past, but because the present and future are connected to the past by the continuity of a society's institutions.” However, path dependence is not inertia; it is the constraints on the choice set in the present that are derived from the historical experiences of the past. Path dependence consists of historically derived constraints that are supported both by existing institutions that oppose change, and the belief system that helped to produce those constraints (North, 1990). Institutions that have accumulated give rise to organizations whose survival depends on the perpetuation of those institutions, and hence these organizations will devote resources to preventing any alteration that threatens their survival. Given that the present project is an examination of the role of economic activity in the tremendous social, political, and cultural change of the Greek world in the Archaic Age, this concept is particularly relevant, especially due to the fact that NIE also accounts for the effects of differential power relations in the process of institutional change, thereby introducing economic change as an expression of power beyond the schematic Marxist approach. Changes that alter the relative bargaining power of a group of constituents can lead to the alteration of the rules (North, 1981). 75 Second, NIE provides a behavioral model that is more consistent with observable phenomena. Unlike classical and neo-classical approaches that take a pre-conceived concept of the individual as a given, NIE rejects this assumption and instead attempts to account for the non-rational, but not irrational, ways that human beings process information, frame choices, and make decisions under conditions of uncertainty or complexity. Often relying on Simon’s (1957) concept of “bounded rationality,” NIE recognizes that decision makers are not omniscient, and have real difficulties in processing information, and as a result can be seen as “intendedly” rational (Furobotn & Richter, 2005). To a degree, NIE echoes Veblen’s (1898) famous critique of the ahistorical, hedonistic conception of man as a “lightning calculator of pleasures and pains, who oscillates like a homogenous globule of desire of happiness under the impulse of stimuli that shift him about the area, but leave him intact. He has neither antecedent nor consequence. He is an isolated, definitive human datum.” Instead, the neo-institutional approach recognizes the insurmountable obstacles to maximization, such as the limitations of information processing on computing optima from known preference or utility information, the unreliable probability information about complex environmental contingencies, and the absence of a well-defined set of alternatives or consequences (Heiner, 1983). Such limitations force individuals to observe regularities of behavior that arise from the uncertainty in distinguishing preferred from less-preferred options. Although Hodgson (1997) prefers the terms “habit” and “rule,” the typical neo-institutional recognition of informal and formal rules is still generally observed, and the distinction is not one that holds a great deal of significance for this project. What is of significance is that NIE accounts for a broad range of human activity (altruism, self-sacrifice, dedication to ideological causes, religious commitment, etc.) that is simply dismissed by orthodox economic theory (North, 1990). Third, a neo-institutionalist approach provides an interpretive framework within which to understand economic behavior and performance within the context of political, social, and cultural constraints. By the emphasis on informal rules, NIE in particular allows for an analysis of economic activity in situations where the power of the state is limited or absent. During the Archaic Age, it is undeniable that one of the 76 major transformations of the period consisted of the codification of the polis, which was in turn accompanied by the rise of what Morris (1996) calls the “strong principle of equality.” However, the formal power of the state in the 8 th century was, by all indications, extremely limited. What NIE offers is a way of contextualizing economic activity in conditions both where the formal power of the state (and the concomitant enforcement arrangements) are well developed, as well as in conditions where the informal rules based on values and ideology prevailed in the absence of highly developed juridical systems. During this transformative period in which such dramatic changes were taking place to the entire institutional matrix, NIE is uniquely positioned to offer insights into the dynamics of this process. For instance, North (1990) points out that economic activity continues even in stateless societies, or when there are no formal rules. In the absence of efficient and effective formal institutions, there frequently exist underground economies that attempt to provide a structure for exchange. Given that during this period, the level of long-distance trade increased markedly, it is crucially important to be able to explain the kinds of institutional arrangements that are necessary for such a phenomenon to take place. This is perhaps the single most salient point for the current project, namely specifying what are the common characteristics of an economic structure that is capable of supporting long-distance trade, and offering a plausible model, based on empirical observations, whereby the conditions necessary for the growth of such trade can arise. Long-distance trade, after all, requires a sharp break in the characteristics of the economic structure, including substantial specialization in exchange, economies of scale, and geographic specialization. In addition, two distinct transaction cost problems must be overcome: the classic problem of agency, and the challenge of negotiating and enforcing contracts in alien parts of the world (North, 1991). In general, North (1990) identifies three major types of exchange: personalized exchange, impersonal exchange without third-party enforcement, and impersonal exchange with third- party enforcement. Each of these types of exchange faces particular challenges and constraints, and relatively typical sets of institutions are common to each type. In addition, the types of exchange are roughly correlative to the geographic area or range within which they are conducted, from small-scale, 77 village trade to regional trade, and then to long-distance trade. Traditionally, these categories have been framed within a developmental narrative, most notably in the German historical school (North 1991; Bresson 2007). To be sure, it is not impossible for an economic system to progress from one stage to the next, but NIE asserts that this is not an immanent feature of the structure, and the historical evidence presents numerous counter-examples to the traditional narrative. For example, many regional economies evolved from the beginning as export economies, building their development around the export sector (North, 1991). While we will examine the particular aspects of the aforementioned types of exchange and their accompanying institutional characteristics more fully in subsequent pages, three features are of particular importance and deserving of some mention here. First is the shift from personal exchange to impersonal exchange, and it is this change that is obviously of primary concern for our investigation. As NIE argues, any form of economic organization, from the simplest to most complex, must have provisions for the specification and enforcement of the terms of exchange. Underlying even the most rudimentary transactions are complex structures of law and enforcement, whereby uncertainty is reduced by an accepted structure of property rights and their enforcement arrangements (North, 1981). However, there are types of institutions that are characteristic of personal exchange, which is usually characterized by repeat dealing, cultural homogeneity (a common set of values), and lack of third party enforcement. These are vastly different from the typical arrangements that accompany impersonal exchange, which are in turn are strongly affected by the presence (or absence) of third party enforcement (North, 1990). Second, for exchange to take place, the existence of credible commitments is a fundamental pre-requisite. Two problems exist with the mainstream economic treatment of commitments. First, the ability to create complete and perfect contracts, without cost, is frequently taken as simply given in neo-classical models. However, given the computational limitations of human agents, complete contracts are simply impossible (Williamson, 1981). In addition, due to the non-ergodic nature of the world, unprecedented situations continually present themselves that are by their nature impossible to predict (North, 2005). Finally, even 78 if such completeness were theoretically possible, the information costs associated with the creation of such a contract would be prohibitive. The second problem is that credible commitments are usually associated with formal contracts. However, as we have seen, formal institutions do not exist independently. They must be supported by complementary informal institutions, as well as enforcement arrangements. As Keefer and Knack (2005) make clear, there is ample evidence for differential levels of trust and trustworthiness across cultures, strongly suggesting that they are culturally contingent. For obvious reasons, in societies wherein low trust levels are prevalent, the difficulty of making credible commitments is commensurately higher. Also, Fukuyama (1995b, 2000) argues that it is not only possible to label societies as high or low trust, but also that norms of trust and trustworthiness differ in what he calls the “radius” of trust, from narrow to wide. The trust radius can have powerful implications for the ability to make commitments outside of social sub-groups. For example, strong intra-ethnic trust in an ethnically heterogeneous society may restrict the scope of transacting, leading to segmented markets, and a reduction in gains from economies of scale. As a result, when long-distance, impersonal trade is observed, the clear implication is that there exists some solution to the challenge of credible commitment. Consequently, what is arguably the most important potential of NIE for the present project is the potential for making reasonable conjectures about the nature of the institutional structure, even when direct evidence is lacking. In every single observable instance, impersonal long-distance trade takes place within an institutional environment that supports such activity. Unless we assume that somehow the Archaic Greek world constitutes the single known exception to this rule, we can reasonably posit the existence of institutions that solve the problems of agency, contract fulfillment and enforcement, transaction costs, and credible commitment, even when the actual institutions themselves are not directly observable. In the same way that physicists can propose the existence of sub-atomic particles, and even form hypotheses about their nature, without directly observing the actual particles but rather their effects, 79 we can apply a similar line of reasoning to the institutional structure of the period in question, given the empirical evidence for the existence, volume, and range of long-distance trade. Class as a Concept Finally, we turn to the concept of class as it relates to archaic Greek society. The concept of class has been utilized in classical scholarship, 20 and it appears frequently as a category of analysis. 21 It is often not given separate treatment and rarely appears in the indices of even substantial accounts of the period, despite the fact that there are plentiful listings under “aristocrats” and occasionally “peasants” and “slaves” but rarely any account of the conceptual framework within which the relations of these classes are to be understood (Rose, 2009). In general, the focus is almost exclusively on status, and even when particular classes are identified and examined, it is often with a distinct lack of awareness of the theoretic issues involved. For example, following Finley (1999 [1973]), archaic Greek society has often been characterized as a “peasant” society (Millett P. , 1984), or at least one divided between “chiefs” and “peasants” (Morris, 2009), or even more problematically between “landowners” and “peasants” (Frier & Kehoe, 2007). It is clear that this term is regularly used in descriptions of the ancient Greek world although definitions – most of which are derived from studies of peasant societies in very different times and places – vary widely (Hall, 2007). From the standpoint of anthropology, there have been two dominant approaches to the study of peasantry: that of R. Redfield (1930) who studied peasantry as a social phenomenon, and that of Steward (1956) who considered their role to have been largely a function of adaptation to the environment (Silverman, 1979). In either case, however, both of these main schools of thought have 20 Ex., Thalmann (1988), Tandy (1997), and of course de Ste. Croix (1981), who made it a central tenet of his interpretation of Greek history. 21 For example, see van Wees’ (2009) discussion of the “leisure class,” Raaflaub (1988) on the Greek “landowning middle class,” Morris (1986) on the potential for oral poetry to serve “class interests,” etc. 80 generally agreed on certain characteristics as being common to all or nearly all peasant groups, and these are (from Hall, 2007): 22 • The peasantry occupies just one sector within a complex, stratified society • Peasants tend to possess the land they cultivate, but this is not a necessary condition; they may be freeholders but can also be tenants or lessees • Peasants typically utilize the resources of their own household although they may occasionally employ restricted slave or wage labor • Peasants tend to reside in village communities where they maintain links with other households • Peasants typically owe obligations to the state even if they do not enjoy full political or civil rights Admittedly, some of these characteristics are directly relevant to the putative experience of the majority of non-elite Greeks in the Archaic period, such as the tendency towards self-sufficiency at the household level, or residency in villages with links to other households. However, many of these features are clearly inconsistent with the nature of archaic Greek society, especially the concept of peasants as a small part of a highly stratified society. In fact, the defining feature of the peasant as having “marginal contact with the larger society (the state, the world market), but…still influenced by it in restricted, but often essential ways…” (Partapuoli & Nielsen, n.d.) is contingent upon the existence of a bureaucracy that cannot be said to exist at this time. In addition, the contrast of peasants with landowners is wholly idiosyncratic, and this is but one example of a category of class that has been carelessly misapplied. For our purposes, a class is “a group of persons in a community identified by their position in a whole system of social production, defined above all according to their relationship…to the means of production…and to other classes…" (de Ste. Croix, 1981). This definition has several important implications. First, class is inherently exploitative, which means that classes are necessarily plural and 22 This view of peasant status will become particularly important when we turn to an examination of Hesiod’s Works and Days. 81 relational. In other words, class (and therefore class conflict) cannot exist in an environment of pronounced egalitarianism such as the !Kung hunter-gatherers of the Kalahari, but is a fundamental feature of stratified societies, and becomes most powerful when economic disparities are pronounced. This antagonistic relationship of exploiters and exploited extends equally both to the means of production and to the relations of production (Rose, 2009). Second, class conflict can manifest itself in open resistance, but often is not overt. For example, Marx (1965) provides by way of example how a slave tends to do as little work as possible, or engages in individual acts of sabotage, without overtly engaging in rebellion. This is markedly similar to Scott’s (1985) observations on “sub-insurrectional” resistance strategies among peasant groups, and what he calls “weapons of the weak.” The majority of Marx’ writings are best understood as critiques of ideology (Rose, 2009). For Marx, ideology is distinct from propaganda, the conscious distortion of the truth for explicit political ends. Although ideology can be deliberately exploited, it more often serves as the frame within which exploitation and resistance take place, but the limits of this frame and even its existence are often internalized by the participants, and not consciously realized (Althusser, 1971), like Bourdieu’s doxa (1977 [1972]) or Giddens’ structures (1984). In addition, ideology can both unite and divide, and one typical function of ruling class ideology is to maintain a balance between antagonistic elements in a society under the pressure of fundamental changes (Rose, 2009). In a similar vein, a common misconception is related to the issue of class consciousness. As Marx (1965) argues, for much of history people are not conscious of the mode of production into which they are born, taking it as natural. As a result, often changes are brought about at the level of production that are not consciously realized, even by those who bring about those changes, until they are nearly complete. For Rose (2012), the very blindness of actors to these factors is an indication of the inadequacy of the emphasis on status (and marginalization of class) that is a frequent feature of classical scholarship. It is true that the surviving sources are much more explicitly engaged with self-conscious struggles over status, but they are deeply embedded within a class structure. It is here that the contributions of Bourdieu (1977, 1984, 2005) are 82 especially relevant in his observations of ways that financial capital (the basis for class) is often leveraged, exploited, and converted into social and cultural capital (the bases for status), a subject to which we will have ample opportunity to return. Finally, it has been common to view Marx’ approach as deterministic. While it is true that his immediate political agenda and especially his view of capitalism were strongly informed by a belief in the unsustainability of the system, his view of history is much more nuanced. For Marx, human beings make their own history, but they are not free to choose whatever course they desire. They are constrained by circumstances transmitted from the past that condition their range of options. Human beings are neither unintentional pawns of vast unconscious forces, nor completely free actors. Instead, they make decisions within a context of historical circumstances that are a product of past decisions (both their own and others’), and these choices in turn alter the set of circumstances and condition future decisions. If this sounds familiar, it is precisely because this concept of history is very similar to North’s (1978, 1981, 1991, 1998, 2005) description of path dependence, and the behavior of economic actors within the context of institutions. It is also strikingly similar to Giddens’ (1984) explanation of structuration, whereby structures both shape the behavior of actors, and are in turn produced, reified, and altered by individual choices. At this point, it seems appropriate to examine a related concept, that of status. To be sure, “status” and “class” are problematic terms, and are often carelessly applied. In fact, they are frequently used almost interchangeably, but for our purposes they have more refined and distinct meanings. The most influential work on this subject is probably that of Weber (2013 [1925]), who described what he envisioned as two related but separate concepts, the “status group” and “social class.” For sociologists and social psychologists working within the tradition of Weber, status and class are traditionally conceived as independent from one another. Class, consistent with Marx’ approach, is a strictly economic distinction 83 based on control over the means of production and the labor power of others (Kohn & Stomczynski, 1990), 23 whereas status is a complex blend of several different elements: “(1) prestige, (2) unequal distribution of relatively scarce social resources and unequal opportunity for acquiring them, and (3) power, i.e., the ability to induce others to fulfill one’s goals…” (Schooler, 2017). As well as being differentiated from one another along the lines of economic v. social, class and status are conventionally regarded as divergent on the basis of the fact that society is separated into a limited number of classes, whereas status exists on a continuum. In other words, “social class is…an ordered categorical variable…in which part of a class’s definition is its relationship to other classes. Social status, on the other hand, is generally a more finally graded, essentially linear variable…” (Schooler, 2017). However, despite these distinctions, I believe that there is more overlap between status and class than has traditionally been represented. First and foremost, I accept the above characterizations of status and class broadly speaking, but with one major caveat: in keeping with neo-institutional theory, I maintain that it is a mistake, albeit a common one, to view economic behavior independently from its cultural and social context. What NIE has demonstrated unequivocally is the significant impact of social and cultural factors on economic performance. Also, according to the conventional view, power is an important element of status. However, this view completely ignores the complex interrelationship between power on the one hand, and preferential access to resources as well as the control over the means of production, the labor power of others, and mechanisms of distribution, on the other hand. Economic influence is both an expression of and a potential source of power. 23 I would also suggest that this definition can be amended to include ownership and control over the means of distribution. 84 It is on this point that I believe the work of Bourdieu, particularly his analysis of the different forms of capital (social, cultural, and economic) is so helpful (Bourdieu, 1979 [1984]). Instead of representing the economic sphere as separate and distinct from social and cultural spheres of activity, he demonstrated the degree of permeability in the barriers between them. Furthermore, he highlighted the extent to which the standing of social and cultural elites is frequently supported by (and even dependent upon) their control over economic capital, as well as the frequency with which economic capital is “converted” into social and cultural capital. Thus, while status and class may be regarded as separate categories, they are anything but mutually exclusive. Quite the contrary, they are interdependent and mutually reinforcing. So, how does this pertain to archaic Greek society? For one, I argue that the society of the late 8 th c. BCE was more complex and stratified than might be immediately apparent in the descriptions of Homer. When examined more closely, the archaic literature indicates not only a more pronounced degree of class division, but suggests a degree of fluidity with strong challenges to the traditional social order, and this malleability was a source of severe anxiety for elites, and an opportunity for non-elite landowners. Our concept of class offers a way of understanding the interplay between differential levels of access to resources, especially control over the means of production, and social hierarchy in the archaic Greek world. Also, it dovetails nicely with the approach of NIE that provides a framework not only for the ways in which economic activity takes place within the context of systems of rules, but also the ways in which those rules are formed, reinforced, reproduced, challenged, enforced, and altered. Finally, these concepts can be reinforced and elucidated with relevant ethnographic evidence demonstrating specific ways in which status, class, and economic activity have intersected in other societies with varying political structures and levels of complexity. Taken all together, these three elements, with NIE as the foundation, provide a powerful set of tools to examine the nature of archaic Greek economic practices in the context of contemporary social transformation. 85 Section Two: The Literary Evidence Introduction The Beginnings of the Archaic Period It may have been controversial for Austin and Vidal-Naquet (1977) to insist nearly four decades ago that the “archaic period is perhaps the most important period in Greek history, but now it raises no eyebrows to describe the period as “caractérisé par un grand dynamisme démographique et, en partie par voie de conséquence, par une grande mobilité de personnes et des biens…” (Gras, La Méditerranée archaïque, 1995) or, even more forcefully, as characterized by “an exceptional, energetic, effervescent culture which developed and expanded with extraordinary speed and innovative assurance…[that] it would be absurd to describe as ‘old-fashioned’ or antiquated…’” (Davies, 2009). It is my view that the literary sources of the period must be understood within the context of this dynamic environment, seeing the literature not only as muddled reflections of the social background in which they were produced, but also as attempts to engage directly in the dynamic social processes of the time, existing both as products of and agents of change. As a result, it is necessary to engage in at least a brief survey of the broader historical context in which the literary sources were composed and fixed in written form. I should stress that there are no illusions that this will be a comprehensive account in any way. Instead, the goal is simply to locate the texts in their historical moment in order to allow for a more informed reading. For the present purposes, I consider the Archaic age as beginning in 776 BCE (the first Olympic Games) and ending in 480 BCE (the Battle of Salamis), 24 but I realize that these are simply convenient temporal bookends. My primary concern is to trace larger developmental trends across a broad geographical expanse, and as a result, I expect the dates in question to serve more as a general guide. Also, there is no 24 I readily acknowledge the arguments of Davies (2009), among others, who points to the “trap of terminology” resulting from labels such as “Dark Age,” “Archaic,” etc. that are necessarily products of a backwards gaze, imply comparison, and reflect decisions about periodization. At the same time, despite these pitfalls, I believe that they are reflections, albeit problematic ones, of an underlying historical reality wherein Greek society underwent substantial structural, political, demographic, and economic changes. 86 question that different approaches have produced very different narratives that often seem incompatible with one another, from the conventional, text-based approach of ancient historians and then the more inclusive style of cultural history, as well as the conventional approach of classical archaeology, focused primarily (until recent decades) on establishing absolute chronologies and the history of occupation of specific sites such as the major sanctuaries of Delos, Delphi, Olympia, etc. In addition, the boundary between “Dark Age” and “Archaic” is being challenged, and even the notion of a cohesive Greek identity that may very well be anachronistic. Finally, there is a temptation to be seduced by the Hellenocentric perspective that is a product of both the tradition of scholarship and the fact that the surviving literary sources are almost entirely Greek. I will argue for a much more cosmopolitan interpretation of this period, consistent with recent trends in scholarship (i.e., Horden and Purcell 2000) that is introducing models of cultural interaction to explain Greek expansion and settlement overseas that replace models of invasion, occupation, or domination. With this in mind, we begin with the Mediterranean of the 8 th c. BCE, for a number of reasons. First, It is in the 8 th c. BCE that the Mediterranean world in general, and the Greek world in particular, experienced the initial phases of a revival of sorts that included substantial population growth, expansion in maritime trade, and widespread movements of populations in the form of colonies, especially Greek and Phoenician, across the Mediterranean. Of course, there have been various arguments for the dating of the texts of Homer and Hesiod, particularly the highly contentious debate over the dating of the Homeric epics. For now, however, I will stipulate for the sake of argument that we can reasonably locate them in time perhaps as early as the latter half of the 8 th c. and the early 7 th c. BCE. As a result, it is during this time of major demographic changes (that must themselves have placed tremendous strains on the existing social and political structures) that the works of Homer and Hesiod were produced. One of the reasons why the changes that began in the 8 th c. are so intriguing is that they occur in the wake of what Snodgrass (1980) has famously termed a “Dark Age” that followed the disintegration of the Mycenaean social structure in the 12 th c. BCE and persisted until the revitalizing movements of the 8 th c. 87 BCE. To be sure, many aspects are controversial, from the causes of decline, the level of decline/collapse, and the degree of continuity between the late Bronze Age and the early Iron Age into the Archaic period. However, the overall picture still holds. 25 This interpretation of decline and subsequent revival is important for a study of the Archaic period because it is still an open question as to what degree of continuity can be traced from the Bronze Age through the early Iron Age, and also whether or not the early Iron Age decline was so precipitous, and the consequent Archaic revival so dramatic, as has been argued. On the other end of the spectrum from Snodgrass, there are those who have called into question the very notion of a collapse and subsequent recovery, 26 whose ideas are summed up perhaps most succinctly by Susan Langdon (1997): “although the romantic appeal of the notion will linger for some time to come, the dark age of Greece now appears to have been a less blighted, impoverished, and isolated time – that is, less ‘dark’ an age – than previously believed…” In fact, numerous issues still remain opaque. For instance, even if one accepts the premise that the 12 th c. BCE ushered in a widespread collapse, the degree to which this was a distinctly Greek experience as opposed to a broader Mediterranean-wide phenomenon is at issue. Despite these questions, the general consensus is that the archaeological evidence, while perhaps leading Snodgrass to overestimate the precipitous degree of population decline, still presents a valid picture of collapse and recovery, albeit slightly moderated from the extreme position of Snodgrass. 27 The general outlines remain clear, with abundant and widespread evidence indicating a) a fall in population that is certainly detectable and may have been devastating, b) a decline or loss of certain purely material skills, c) a decline or loss of some of the more “elevated” arts, including the art of writing, d) a fall in living 25 The most ardent proponent for the view of precipitous and calamitous decline is Snodgrass (1971, 1980), who attempts to provide a rough quantitative basis for the degree of disruption and disintegration through an examination of the number of archaeologically attested Bronze Age settlements that are abandoned during the early Iron Age, along with contemporary burials. 26 See S. Morris (1992a, 1992b); de Polignac (1995); Foxhall (1995;) Thomas and Conant (1999) 27 See Boardman (1999); Scheidel (2003); Morris (2007, 2009). 88 standards and perhaps in the sum of wealth, and e) a general disappearance of contacts, commercial and otherwise, with most peoples beyond the Aegean. 28 As a result, the subsequent effervescence of the 8 th c. BCE can indeed be regarded as a “structural revolution” that took place during the Archaic period and “established the economic basis of Greek society, as well as the main outlines of its social framework…it set up…the forms of state that were to determine Greek political history…” (Snodgrass A. M., 1980). As Morris (1999b) explains, "there are few episodes in world history before the industrial revolution when a society experienced such profound change in the course of a hundred years…its core features (demography, state formation, social conflict) must remain at the heart of any balanced discussion.” At the same time, I would argue for a modification of this position. Specifically, I find it extremely unlikely that the institutions of the Greek world in the early Archaic period would have been capable of adjusting to accommodate the swift and substantial demographic changes that are attested in the archaeological record. Instead of the “Eighth Century Revolution” championed by Morris (2009) in which state institutions were reshaped and the very notions of community were redefined, it is much more plausible that during this period the traditional social order was pushed well past the breaking point, and that the social and political structures that had been functional during the early Iron Age were simply inadequate to deal with the changing realities in the 8 th c. BCE. As we have already noted in the preceding chapter, it is a characteristic of informal institutions that they are resistant to change, and even when the do change, they typically do so incrementally over the course of time (North, 2005). 29 It would require 28 There are some notable exceptions, such as the finds at Lefkandi (Coldstream 1996; Lemos 2001, 2006) which indicate continued overseas contact, but the overall picture indicated by the archaeological evidence is one of isolation. 29 To this point, I have stressed the nature of informal institutions as resistant to rapid change, especially in comparison with formal institutions. It is precisely due to this feature of informal institutions that the Archaic Period represents such a seminal stage in Greek history. As we know, the 8 th c. BCE ushered in a period of sweeping and dramatic developments, including swift expansion across a wide geographic area, significant rise in population levels, rapid increase in the volume and complexity of long-distance trade, increased contacts with diverse peoples throughout the Mediterranean, and greater mobility of goods, people, and information. Because informal institutions typically change slowly, I believe that this resulted in a lag between the accelerated rate of structural changes (demography, economy, etc.) and the ability of traditional norms and rules to remain relevant in the light of rapidly 89 much more time for the Greek world to develop and embrace the so-called “strong principle of equality” (Morris, 1999b) that would serve as the basis for Greek political identity and social cohesion in the classical polis. 30 As we shall see, the literature of the late-8 th /early 7 th c. BCE, particularly the Homeric epics and major poetic works attributed to Hesiod, reveal a society very much in flux, struggling to formulate a conceptual framework capable of supporting social cohesion in the midst of competing visions of the good society (Kistler, 2004). In fact, it would take the next two centuries for formal and informal institutions to come into being that were consistent with the new economic and demographic realities, with the result that the Greek world of the 7 th and 6 th c. BCE differed markedly from that of the 9 th and 8 th c. BCE (Osborne, 1996a). Thus, it is more accurate to say that the 8th c. BCE was less about the formation of new notions about class, gender, and ethnicity than it was about the inability of the traditional models to survive in the face of the tremendous social pressures. The formation of new institutions to replace them would take place in the 7 th and 6 th centuries, but as we will see in the Homeric epics and particularly the poetry of Hesiod, the rough outlines of these new institutional arrangements were already beginning to take form. In other words, the society depicted in the Homeric epics and Hesiodic poetry reflects the initial phases of a struggle to deal with larger destabilizing forces that would continue to be negotiated and contested over the succeeding two centuries. Admittedly, many changes were dramatic even at the time, but for the economic historian, even more substantial changes come later during the 7 th and 6 th c. BCE (Osborne, 2007). changing conditions. Consequently, new institutions needed to be adopted that were more consistent with the emergent realities and thus more capable of performing the critical function of stabilizing expectations. This created opportunities for those groups and individuals who may have been previously disenfranchised to create a more favorable institutional framework. 30 I use the term “polis” with the realization that it carries connotations and associations that are more applicable to subsequent periods of Greek history, and I do not employ it with any intent to suggest the more fully developed and formalized political entity that it would later come to signify. I do concur with the arguments of Nagy (1999) and Raaflaub (1997a, 1997b, 2000) that the nascent beginnings of the prototypical polis structure can be glimpsed in Homer, but that is not the point of emphasis. More than anything, I employ the term because it is used in the epics themselves and there does not seem to me to be another word that more effectively replaces it, despite the potential anachronistic pitfalls. 90 Of these changes, one of the most important for an understanding of the Homeric epics is the emergence and growth of Panhellenism. The archaeological evidence is quite clear that the 8 th c. witnessed the rising importance of the major Panhellenic sanctuaries at Olympia, Delos, Delphi, etc. In addition, the widespread growth of monumental architecture utilizing standard architectural orders is evidence for a growing sense of shared cultural identity, which is reflected in numerous developments, including the Panhellenic sanctuaries, games, amphictyonies, and religious homogenization. 31 According to Morris (1986), the first large-scale gatherings at Panhellenic games and sanctuaries were the only contexts in which poems as long as the Iliad and Odyssey could be performed, 32 and Nagy (1995) goes so far as to argue that the growth of Panhellenism is critical “as a hermeneutic model to help explain the nature of Homeric poetry…” For Raaflaub (1998a, 1998b, 1998c, 2005), the Homeric epics both served as an agent in the promotion of Panhellenism, and assumed their characteristic form in response to its rise. Further, Homeric society is Panhellenic in the sense that it allows broad recognition and identification, characteristics that help explain the broad appeal of the epics, but do not render them any less historical (Nagy 1990, 1999; Donlan 1985). 33 In any case, it is not surprising to discover this rise of pan-Hellenism occurring precisely at a time when the exposure of Greeks to a larger, more multi-cultural world rose dramatically. However, it should be noted that there is ample evidence to suggest that many parts of the Mediterranean that were once thought to have been unknown to the Greeks in the Bronze Age and early Iron Age were in fact familiar to them. 31 For the evidence of sanctuaries, see Rolley (1983); Morgan (1990, 1993); on religion, Vermeule (1974); de Polignac (1995); Crielard (1995). 32 Such occasions were the only ones in which the epics could be performed in their entirety, and it would obviously have been possible, even likely, that individual episodes could have been performed in other less formal settings. 33 I want to emphasize that I am not insisting that the Homeric epics represent social or political conditions that are historically accurate in their details. Instead, I would focus on two specific points: first, the 8 th c. BCE shows evidence for a developing sense of pan-Hellenic identity based both on shared characteristics (i.e., religion, language, etc.) and in relation to the diverse cultures of the Mediterranean with whom the Greeks were in contact as a result of the westward colonial expansion; second, that this growing pan-Hellenism served as a basis for the widespread appeal of various literary texts, including the Homeric epics and Hesiodic poetry, that were culturally relevant to a broad cross-section of the Greek-speaking world. As a result, they describe general issues and anxieties that reflect the shared concerns of their audience. 91 Certainly, Greek presence across large portions of the eastern Mediterranean is well attested from at least as early as the second millennium BCE (Boardman, 1999), but Minoan Cretans and Mycenaean Greeks of the Bronze Age were familiar with the route to Sicily, and the Greeks of the Archaic period knew well that they lived in a landscape long molded by previous inhabitants with whose legacy they came to terms in various ways (Davies, 2009). Without question, the western Mediterranean was from the outset a shared space with Phoenicians and Estruscans (Antonaccio, The Western Mediterranean, 2009), but Sicily and the Italian peninsula were also characterized by a diverse mix of Italic peoples, including Veneti, Messapians, Sicels, Sicans, Ligurians, Raetians, Onotrians, and others (Thomas 2009). Given this diversity, it is difficult to make sweeping generalizations about conditions that can be applied to the entirety of the Greek experience across the Mediterranean. This is particularly true given the difference on a broad level of the Greek experience in the eastern and western Mediterranean. For various reasons, I have chosen to focus on the conditions relevant to the western Mediterranean during the 7 th and 6 th c. BCE rather than the east, or for that matter, the Mediterranean as a whole. On the one hand, I maintain that in light of their Panhellenic nature, the literary texts (especially the poetry of Hesiod and the Homeric epics) can reliably be understood as reflecting general conditions relevant to the Greek world on a broad scale (see West 1997; Boardman 1999), and on the other hand the different nature of Greek occupation in the west and the east 34 visible in the archaeological record is substantial enough to preclude treating them as equivalent, or even ignoring altogether any possibility of difference and regarding the Mediterranean as a generic whole. 34 I recognize the inherently problematic nature of terms like “west” and “east” as potentially reflective of a Hellenocentric normative bias, but even in Greek antiquity, there was a concept of "the west" or the western Mediterranean, and even notions of a near and far west (Horden and Purcell 2000). Ultimately, both in geographical and in historical/social conditions the distinctions remain valid, and as such I still employ these terms, albeit cautiously. 92 Of course, although both the geography and history of the Mediterranean to this point would seem to lend themselves to thinking about the western Mediterranean as a more or less discrete unit, that is not to say that such considerations are without challenges. For one, the western Mediterranean was neither a unified ancient political territory, nor perfectly distinct topographically (Antonaccio, 2009). Although the western Mediterranean may be a bounded inland sea, the mineral resources of the Iberian peninsula, particularly silver at Tartessos, as well as potential trading partners in the Atlantic, prompted the establishment of colonies at Gades, Huelva, and Lixus west of the Straits of Gibraltar (Ruiz Mata 2002a, 2002b). As a result, it would be a mistake to regard the regionality of the western Mediterranean as entirely fixed or strictly bounded. In addition, a focus on the distinction between east and west also runs the risk of eliding some of the potentially important distinctions between north and south (Bowersock, 2005). Finally, it has occasionally been suggested that certain parts of the central Mediterranean should not be considered separately from ‘Greece’ in the Archaic period (Antonaccio, 2009). Nevertheless, such categories are strongly supported by empirical evidence. For instance, from a purely geographical perspective, the Mediterranean is divided rather neatly into two basins with the Sicilian channel between the island and Cap Bon in modern Tunisia as the dividing line. Also, the ocean currents in each basin move independently in a clockwise direction, encouraging navigation patterns that restrict movement between the east and west basins (Thomas, 2009). Furthermore, the historical patterns of settlement, at least for Greek-speaking peoples, exhibit strong regional peculiarities. For one, the early and numerous cities of the eastern Aegean littoral anchor the Greek presence in the east to a degree that is not the case in the west (Antonaccio, 2009). On a related note, there is substantial evidence in many sites in the eastern Mediterranean to suggest continuous, uninterrupted Greek occupation from the Bronze Age through the early Iron Age and into the Archaic period (Thomas and Conant, 1999; Antonaccio 2009; Thomas 2009). Also, the eastern basin of the Mediterranean was ringed with kingdoms that could either trace a continuous existence back centuries or even millennia (such as the Egyptian kingdoms of North Africa), or at the very least were political, cultural, and linguistic heirs of much earlier states (such as the 93 Neo-Assyrian empire that had roots in the disintegration of the Akkadian empire that had reached its apex in the later centuries of the third millennium BCE, or the Phoenicians who were active in the seas of the eastern basin since at least the late Bronze Age). In the west, no analogue for the lengthy traditions of the Egyptian, Mesopotamian, or various Semitic civilizations existed (Boardman, 1999). Finally, the migration patterns of Greeks were much different from east to west. As Antonaccio (2009) points out, distinguishing features of the western colonial movement include its concentration in time, the richness and success of many of the foundations, the impact on developments in the mainland, and the prominence of the western colonies in ensuing Greek history. Snodgrass (1994) goes even further, beginning first with the observation (Morel, 1984) of the striking difference between the pattern of Greek settlement in the east and west when viewed on a map, with Greek settlements on the north coast of the Aegean giving the impression that every possible area of settlement was occupied, whereas the large empty gaps in the map of the Greek colonies in the west are notable. For Snodgrass, the sites of the western colonies exhibit careful choice and deliberation, which led him to conclude that the movement to settle in the west was calculated and planned to a high degree, and possibly a historically unprecedented phenomenon. It is precisely because of this unique character of the westward colonial movement, as well as the fact that it is contemporaneous with the Hesiodic poetry and (as I shall argue) of the first appearance of written Homeric texts, that it now deserves special mention by way of introduction to the texts themselves. Westward Ho! The Greek Colonial Expansion in the Western Mediterranean Following the arguments of Snodgrass (1994), it seems clear that we can regard the Greek colonization in the western Mediterranean as historically idiosyncratic. Although the early colonists may have persuaded themselves that they were following in the footsteps of predecessors in the heroic age, such a position is not consistent with the available evidence. The Bronze Age Greeks clearly were aware of western maritime routes and had made contact with at least the western shores of Italy, Sardinia, and Sicily, but there are no indications that they engaged in the kind of deliberate, premeditated expansion and settlement that began in the 8 th c. BCE, and the result was much more than simply extending the culture of the Greek 94 world to foreign shores. Rather, what they produced was arguably a new world altogether. Since this formative event essentially ushers in the period that is the focus of the present project, it is imperative that we examine the nature of western colonization at least briefly, and to ask why such an unprecedented phenomenon happened when it did, and in the way that that it did. In order to situate ourselves chronologically, the process of Greek colonization in the west is thought to begin with the first settlement of Pithecusae, on the island of Ischia in the Bay of Naples, at some point in the first half of the 8 th c. BCE. 35 Following shortly on the heels of the foundation at Pithecusae, numerous sites in southern Italy and Sicily were colonized by Greeks of various origin (an issue that we will explore shortly) that constituted the “first wave” of colonization, lasting until the early 7 th c. This was followed by the “second wave” of colonies that, with the single exception of Gela in 688/687 BCE, was the result of colonies established in the first wave sending out colonial expeditions of their own, ending in the first decades of the 6 th c. BCE (De Angelis, 1994, 2002; Boardman, 1999). Traditionally, the explanation for Greek expansion and migration westward has been that it was fundamentally a means of relieving population pressure in a region where good arable land is scarce, 36 and the archaeological evidence (as we have already seen) indicates substantial population growth in the 8 th c. BCE. However, there are many reasons why this narrative cannot adequately be maintained in the light of the material evidence. 37 As a result, we are forced to look elsewhere for explanations, especially 35 Thomas (2009) argues for a date of roughly 780 BCE, while Coldstream (2003) and Ridgway (1992, 1994, 2000) place its foundations in the early eighth century, and Morris (2009) places it rather more vaguely sometime before 750 BCE. However, Coldstream (1994) also insists that Euboean presence can be identified at Sardinia as much as a century prior, and Ridgway (1994) postulates that the Euboean presence at Pithecusae itself involved Euboeans, Phoenicians, and probably some others living side by side for at least a generation prior to the accepted date of the site’s foundation. 36 See Bérard (1960); Lepore (1969, 2000); Snodgrass (1977, 1980, 2000); Desborough (1972); Coldstream (2003 [1977]); O. Murray (1993 [1980]). 37 As Snodgrass himself (1994) explains, overpopulation and “land hunger” as motives for the colonizing movement have always had to confront the objections that whatever the level of population in the Greek city in the second half of the eighth century, it was so much higher in the fifth that other factors must be considered. Also, the land hunger motive was based on Snodgrass’ (1977) early analysis of archaeological data which have since been reinterpreted by 95 given the fact that many colonies, especially the earlier ones, were not established in areas with large tracts of arable land (Tandy, 1997). For instance, the settlement at Pithecusae does not have near enough arable land to support the population suggested by the associated cemeteries (Antonaccio, 2009), and Ridgway (1994) argues that the Euboean precolonial story in the west starts in Sardinia, in a context that had intimate Phoenician and metallurgical connections, and may precede the establishment of Pithecusae by as much as a century. A similar pattern also emerges at Emporion on what is now the northeast coast of Spain. Evidence indicates that it was occupied by an indigenous settlement for at least a century prior to the actual Greek occupation at the site (Shefton 1994; Aquiluè, Castanyer, Santos, & Tremolada, 1999, 2001). Furthermore, instead of a simple process of land acquisition, the material evidence indicates an intimate process of coexistence with the indigenous peoples, quite different from what we would expect from expansion due to land hunger (Dietler, 2009). 38 Instead, it seems that many colonies were established in areas that seem to have been points of contact well prior to colonization. 39 I. Morris (2006, 2007); see also Scheidel (2003). As Morris explains, the dramatic increase in burials that Snodgrass interpreted as reflective of rising population levels was more indicative of changing ritual practices, specifically a growing preference for inhumation over cremation, than it was evidence for rising population levels. In so doing, he argued that the striking population increases that Snodgrass postulated on the basis of burial data (a sevenfold population increase from 780-720 BCE at Athens and Argos) was not nearly as dramatic as Snodgrass initially believed. This is not to say that populations were not growing at this time, as Tandy (1997) has shown, but simply to say that the population growth was almost certainly not sufficient to be the primary, or even a significant factor, in the Greek colonial expansion into the western Mediterranean. 38 Also, the “land hunger” explanation fails utterly to explain the so-called “second wave” of colonial expansion, in which Greek colonies, particularly in Sicily, sent out colonial expeditions of their own, often within just a few years of their own foundation. Such efforts cannot possibly be adequately explained by a lack of arable land or other natural resources. For one, the number of settlers involved in the initial colonial settlements was typically limited to a few hundred, and the likelihood of these relatively small settlements expanding beyond the ability of their territories to support them is realistically impossible. Given the rapidity with which a number of Greek colonies in eastern Sicily sent out their own expeditions, it seems impossible that the populations of the newly founded settlements could have managed to grow so swiftly as to necessitate a removal of some of the population because they could not be supported by the local resource base, especially in light of the fact that in the new world of the colonies we are regularly dealing with larger and often better expanses of land and lower population levels, at least in the initial stages (Coldstream, 1994; DeAngelis, 1994). 39 The sites chosen in the second wave of colonial expansion, in which the recently founded Greek colonies established colonies of their own, were nearly all occupied by existing Sicel populations, and in every instance these indigenous populations appear to have been ejected (Boardman, 1999). The question of what happened to these groups is a difficult one. In looking at other Greek settlements in the western Mediterranean, there does not appear to be a consistent pattern when Greeks look to inhabit already occupied locales. At Pithecusae, for instance, the 96 Thus, we are led to conclude that one of the motives for colonial expansion in the western Mediterranean, and perhaps even the primary motive (especially for the earliest settlements), was to foster maritime commerce. Trade has long been the standard explanation for Phoenician settlements abroad (Aubet 2001; Antonaccio 2009). The sites selected for early Phoenican colonies in Iberia beginning in the late 8 th c. BCE provided strategic access to the supply of silver at Tartessos, 40 and western expansion also brought access to other raw materials such as gold, tin, and copper (Antonaccio, 2009). It is evident that the Phoenicians were manning trading posts on the coasts of southern Spain, north Africa, west Sicily, and probably Sardinia, before the end of the 8 th c. BCE. It would be an oversimplification to argue that trade was the only motivation, but if the Phoenicians had been interested only or even primarily in land, there is little reason why they should have chosen to drive out the Greeks along the important routes, and ignored nearer tracts of land like Cyrenaica (Boardman, 1999). Even by such an early date, the Phoenicians were able to recognize the potential benefits from expanded trade the in western Mediterranean, suggesting that Greek colonial expansion, which followed so closely behind the Phoenicians, might have been prompted by the same motives. Also, the fact that there is so little overlap in the region between areas of Greek settlement and those with Phoenican occupation also suggests that they were in direct competition with one another for control of those territories. If the desire to acquire more land was not a realistic motivation for such expansion, the most likely impetus was the regulation of maritime trade. Greek population appears to have co-existed along with both indigenous residents and Phoenician settlers (Ridgway, 1992, 1994), whereas Greeks at Emporion initially established an independent settlement near to an existing Iberian site, then later re-located across the harbor and established a new site in which both Greek and Iberian populations co-existed within an outer circuit wall, but the two populations were separated from one another by an interior wall (Alquilué, Santos, Buxó, & Tremoleda, 1999). Meanwhile, if the excavations at Monte Polizzo are at all representative of conditions in Sicily (Morris, Jackman, Blake, & Tusa, 2001, 2002; Morris, Jackman, Blake, Garnand, & Tusa, 2003), it would appear that many of the displaced Sicels re-located to inland sites that were safer from the encroachment of both Greeks and Phoenicians whose settlements were invariably located near the sea. However, the evidence for the displaced Sicel populations is sparse, due largely to the fact that archaeological excavations in Sicily have typically focused on sites occupied primarily by Greek and Phoenician populations. 40 See Ruiz Mata (2002a, 2002b); Aubet (2002); Fernández Jurado (2002). 97 I defer to the position of Boardman (1999), who states, “Archaeology, geography, and common sense combine to suggest that trade normally preceded the flag and that in the case of some of the earliest colonies trade rather than land was the dominant factor in choosing a site…it would be idle to pretend that considerations of trade did not provide some part of the motive in founding many colonies, and the major part in a few…” This view is supported by far too much evidence to deny. For example, the very locations for colonial settlements were often away from what at the time were open and unoccupied large tracts of arable land, and instead clearly prioritize access to maritime routes (Tandy, 1997). The earliest of the Greek colonies in the west (Pithecusae, Cumae) had been deliberately planted in positions that were particularly favorable for trade with Etruria (Boardman, 1999). In addition, the results of excavations at these sites indicate that the earliest settlers were Euboean, the same people who had opened up the east to the Greek world a century before (Tandy, 1997; Boardman, 1999). The emphasis on trade is also supported by the indirect evidence of growing unrest and clashes between Greeks and Phoenicians that increased substantially in the 6 th c. BCE, and appear to have been the culmination of growing commercial rivalry. Initially, it seems that Greeks and Phoenicians coexisted quite peacefully. As we have already pointed out, the early finds at Pithecusae indicate that Greeks and Phoenicians lived side-by-side (Ridgway, 1992, 1994). The discovery of a strong presence of Greek pottery of the later 8 th c. BCE at Carthage suggest that the Phoenicians were, at least initially, ready carriers of Greek pottery, but once the trade routes via Etruria had been secured by Greeks well before the end of the 8 th c. BCE, only the Spanish route was open, and this the Phoenicians secured by their foundations at Carthage and other areas (Boardman, 1999). Consequently, the later tensions between Phoenicians and Greeks may be regarded as the product of increased competition in trade. The importance of commercial considerations in the process of colonization can also be glimpsed in the identities of the earliest colonizers. It is well-established that the early colonial expansion westward was initially dominated by the Euboeans, who had apparently maintained connections with the east from the end of the Bronze Age through the early Iron Age, as visible in the finds at Lefkandi (Lemos, 2001, 98 2006). As Snodgrass (1994) points out, we cannot ignore the presence of the Euboeans on both sides of the balance of imported and exported objects, which cannot be attributed either to coincidence or to any superior quality in their pottery. In some way, they controlled a privileged channel of communication into and out of the Aegean, and their primacy in the movement to colonize hints strongly that they were already well-traveled themselves. The continued connection between colonization and trade is visible even as the Euboean domination waned. Only three Greek states (Chalcis, Corinth, and Megara) were responsible for the settlement of the east coast of Sicily (Coldstream, 1994). Without exception, the major players in the process of western Greek colonization were already successful maritime powers. Regardless of the actual motives, the establishment of colonies had the effect of producing a dense network of distribution nodes capable of supporting extensive maritime commercial activity. Even if we somehow deny trade as a motive for the initial expansion efforts, we cannot pretend that the result was anything other than the creation of an extensive, interconnected system that Malkin (2003) has described as a rhizome, an “endless, interconnected root system (a network with no center) giving rise to leafy plants above the surface…” More importantly, this dense network was capable of supporting a high volume of exchange within a distinctly multi-cultural sphere of exchange. In general, consumption is a feature of the ancient western Mediterranean economy that has received much less attention than either trade or production, and has often been considered as a “natural” response to the availability of goods. However, recent work has challenged the assumptions underlying this perspective, and has focused on understanding the socially situated nature of indigenous demand for foreign goods (Dietler, 1998, 1999). In Sicily among the Sicans and Sicels and in southern Italy among Iapygians, Oenotrians, and others, the archaeological picture is consistent, as Greek goods found a receptive audience (Boardman, 1999). Of course, the active participation of Etruscans operating on a 99 significant scale in southern France, 41 Phoenicians and Carthaginians along the coast of southern Spain and north Africa, 42 and Greeks in southern Italy and Sicily all create a picture characterized by a complex reconfiguration of relations that were fluid, but became more confrontational later in the period (Antonaccio, 2009). Already by the late 8th c. BCE, Corinthian pottery was to be found on most sites known to have been settled by Greeks in Italy and Sicily (Payne, 1931; Amyx, 1988; Benson, 1989; Neeft, 1997; Morris and Papadopoulos, 1998), and different sites seem to have exercised distinctly different tastes for decorative patterns, which suggests that pottery was being moved by men who knew local preferences very clearly and precisely (Osborne 1996b, 2008). In short, when considered from a broad perspective, the material evidence is compelling, and presents a convincing picture of widespread, diffuse economic activity, involving the participation of several active groups, including Etruscans, Phoenicians, and Greeks (especially Euboeans in central Italy, then Chalcidians, Megarians, and Corinthians in Sicily, and Achaeans, Spartans, and others in south Italy), in a dynamic regional exchange system that also depended on the receptivity and collaboration of a host of indigenous populations. The traditional explanation for colonial expansion as a product of overpopulation in mainland Greece is wholly unsatisfying, and cannot be supported by the available evidence. Instead, the choice of settlement locations, the subsequent expansion, and the geographic relationship of colonies to each other, all strongly suggest a calculated, deliberate pattern of development (Snodgrass, 1994). In addition, it seems likely that this expansion was prompted to a great extent by a desire to enhance the economic opportunities represented by increasing maritime trade through the introduction of entirely new groups of consumers in the western Mediterranean. Finally, the result of this process of colonization established a network of distributional nodes that was capable of supporting far higher levels of interconnectivity and exchange across greater distances than had ever before been realized in the region 41 See Bouloumié (1982a); Gras (1985); Morel (1984); Py (1985). 42 See Aubet (1985, 2001); Frankenstein (1979); Moscati (1988); Niemeyer (2006). 100 (Malkin, 2004, 2011). All of this occurred with the background of resurgent population levels in mainland Greece, and the resultant combination cannot but have exerted tremendous strain on the existing social and political structures. If the picture of Homeric society in the Iliad and the Odyssey is to be believed (with some reservations), the epics present a relatively rudimentary form of social organization, what Donlan (1997) identified as a “low-level chiefdom,” 43 according to the schema of Elman Service (1962), and in its broader outlines, this social framework finds general agreement in the Hesiodic poetry, albeit from a different perspective. This form of organization is capable of integrating larger groups numbering in the thousands or even tens of thousands (Service, 1962), and can even maintain loose confederations among groups, as in the case of Polynesia (Sahlins, 1963). However, there is no observable instance in which it supported the diversity, complexity, and geographical displacement evident in the colonial expansion of the Greek world beginning in the 8 th c. BCE. In order to maintain the levels of connectivity indicated by the archaeological evidence, the low-level chiefdom, with its dependence for cohesion on the personal relationships between chiefs, would have been structurally insufficient to accommodate the logistical demands. Thus, we find ourselves at a point in which the traditional social order found itself unable to cope with new pressures, but it was not yet clear what form the new arrangements would take in order to cope with newfound realities. It was in this context that the Homeric epics and the major Hesiodic works were composed, and it is from this perspective that I propose that they must be read. 43 At various other points, Donlan refers to Homeric society as a “Big Man” society (1979, 1989a, 1994), apparently seeing elements of social organization that to him were representative of both types. The term “Big Man” was coined by Marshall Sahlins (1963) to describe the type of leader that he saw as emblematic of flexible ranked societies described by Fried (1967), one whose status was achieved and required constant reification or be lost to potential challengers. There are some indications in the epics that rank in Homeric society was flexible, as when Telemachus speaks to his mother’s suitors and admits that while he may be entitled to inherit his father’s property, he does not inherit the right to rule (I.394-404). For the most part, however, the texts seem to indicate a more fixed type of rank consistent with chiefdoms (see note 152 below). 101 Homeric Questions Before entering into my analysis of the Homeric epics and their insights into the economic conditions of the period, there are several questions that have been hotly contested within the field of Homeric studies and which deserve to be addressed. This is not the place for an extensive account of the various debates (which would likely consume several volumes in and of itself). Instead, I simply wish to state my position on these issues from the outset for the purposes of clarification and because they are important to my interpretation of the texts. These questions revolve around the date of the written texts, the historicity of the texts (particularly of the social background depicted in the Iliad and the Odyssey), the date of Homeric society relative to that of the texts themselves, and whether or not we can consider the two epics more as a matched pair, or whether or not they are sufficiently different from one another to require separate examination. As for the date of the texts themselves, I follow the early arguments of Janko (1982) that place the epics as we have them 44 in the second half of the 8 th c. BCE. 45 Not only is such a date range consistent with the general scholarly consensus and hence validated somewhat by the “wisdom of the crowd,” it is in 44 The most prominent voice in favor of a later date for the epics continues to be Nagy (1990, 1992, 1995, 1996a, 1996b, 1999, 2001, 2003, 2004), who contends that the multiformity of early Homeric poetry indicates the lack of any written texts during what he calls a “formative period” from the mid-8 th to the mid-6 th c. BCE until they were codified during the “Peisistratean Recension” beginning in the mid-6 th c. BCE. However, the variants on which Nagy bases his arguments are minor deviations, and he is unable to identify a single major departure in the narrative structure, plot, or individual episodes. As Merkelbach (1952) shows through a comparison with other epics recorded in writing, the Iliad and Odyssey demonstrate a much greater degree of textual stability, and Finkelberg (2000) comes to similar conclusions through a comparison with the other surviving elements of the Homeric Cycle. West (1998) finds few departures in either the Ptolemaic papyri or quotations by Attic orators from the medieval vulgate, and even the variations within the Ptolemaic papyri have been discounted by some (see Dué, 2001a, 2001b, 2002). Also, consistent with the arguments of Janko (1982), Haslam (1997) notes that the degree of linguistic evolution in the Homeric poems appears to have been arrested at a very early point, prior to Hesiod. 45 This date is supported by Morris (1986), Raaflaub (1997a), and Papadopoulos (1996) on the basis of archaeological evidence, and by Kirk (1976, 1985), Edwards (1990) Haslam (1997), and most prominently Janko (1982) on the basis of linguistic evidence. However, Janko (1996, 1998) has revised his opinion on the basis of recent arguments by Ruijgh (1995), and now places the Iliad in the second quarter of the 8 th c. BCE, with the Odyssey slightly later, but this revised dating is based largely on a single alphabetic inscription and therefore remains tenuous, nor does such a date range comport with what I believe are suggestions in the text of the early westward colonization (esp. 9.116-172). 102 agreement with the body of both linguistic and archaeological evidence. 46 Concerning the historicity of the society depicted in the epics, this remains a controversial issue as well. For the most part, the consensus had shifted predominantly in favor of the position advocated initially by Finley (1954) that the epics offer a largely consistent depiction of the social practices and value system that are indicative of a cohesive society, but in recent years a number of people a number of people have moved away from this position. 47 However, my emphasis is not on establishing whether or not the specifics of the social background depicted in the epics are historically accurate reflections of actual historical conditions, either one much earlier than the date of the texts (Finley, 1954) or roughly contemporaneous. Instead, my focus is on 1) the ways in which the texts reflect prevailing concerns, tensions, and anxieties, and 2) the ideological character of the texts, specifically the norms and values that are promoted (as well as behaviors that are denigrated and discouraged), and the possible motivations for the ideology manifested in the texts. On this point, I rely on the arguments of Thalmann (1998) and others who maintain that by their nature, the ideological aspects of the Iliad and the Odyssey must be relevant to the audience for whom they were performed. Also, ethnographic evidence vividly demonstrates the protean nature of oral poetry, making it impossible to retain holdovers beyond a very short time horizon (Morris I. , 1986). Therefore, such elements belong to roughly the same date as the texts themselves, namely the late 8 th /early 7 th c. BCE. 46 On the basis of linguistic evidence, West (1998, 2010, 2011, 2012) has argued for an early 7 th c. BCE date for the Iliad, and a date for the final third of the 7 th c. BCE for the Odyssey. 47 There have been some detractors, especially Snodgrass (1974), but also Long (1970) and more recently Cartledge (2002), who insist that the epics present a disjointed, garbled mélange of descriptions of different societies, or at least a single society at different stages of development, accumulated in the epics over time as they were told and re- told, thus preserving, albeit incompletely, snapshots of Greek society at various points in time. According to this argument, none of these snapshots are sufficiently detailed, comprehensive, or cohesive to allow for the attribution of a reliable date. An excellent review of this position can be found in Morris (1986). However, the weight of scholarly opinion has come down on the side of Finley (see Finley [1954, 1978]; Morris [1986]; Herman [1987]; van Wees [1992]; Tandy [1997]; Raaflaub [1997, 1998a, 1998b, 1998d, 2005]; Donlan [1981, 1999]; Ulf [2009]). 103 Finally, it is my intention to treat the Iliad and Odyssey as a single source. Although the evidence suggests that the Iliad was written first and the Odyssey a bit later, the dates of their respective codification in writing do not appear to be separated by a significant span of time (Janko, 1982). Although there are some who insist that there are significant ideological differences reflected in the texts, 48 they are generally treated by historians as if they were, to all intents and purposes, interchangeable. 49 Ultimately, I believe that the social background depicted in the epics does not differ in any appreciable way between the texts, and therefore I will rely upon them as a single source. 50 48 Ex., Farron (1979); I find Farron’s arguments largely unconvincing, especially when he insists that the Odyssey is ideologically motivated by an anti-aristocratic sentiment, whereas the Iliad is simply a “good story.” In particular, such a view fails to recognize the power of literary performance as a discursive practice in which cultural mores are established, challenged, and mediated, regardless of whether or not this forms a conscious goal of the performer (see Dougherty and Kurke, [1993]). 49 Ex., Finley (1954); Morris (1986); Raaflaub (1997). 50 This is not to say that there are no differences. There are several aspects of contemporary Greek society that are discussed at length in one, but not the other. For instance, the Iliad contains extended musings on the nature of what is now often referred to as the “heroic code,” including the desirability of kleos and the justifications for war, especially in defense of another’s honor. Likewise, there are numerous elements in the Odyssey that are at best touched on only briefly in the Iliad, such as the proper elements of a functioning polis, the appropriate role of slaves in a functioning society, and the proper organization of an elite household. Simply due to the requirements of the narrative, the Odyssey represents aspects of the contemporary Greek world that appear only fleetingly or even not at all in the Odyssey. This does not in any way mean that the two epics are contradictory or exclusive, and in fact there is a substantial degree of overlap which serves to confirm their mutual intelligibility. 104 Chapter Four: The Homeric Epics Introduction Having now addressed the aforementioned questions, I wish to move on to my particular approach to the texts. In no way do I mean to suggest that the lengthy and involved debates over these issues are inconsequential or undeserving of attention, but it is not my goal to venture into these deep waters at the expense of losing focus. I have introduced the issues with a full awareness of their complex histories, and I have staked my positions because I believe they are integral to my understanding and interpretation of the texts. Instead, I prefer to focus on what I believe is a new approach to understanding the social conditions of the Homeric epics, and in particular the economic behavior reflected in them. Specifically, I refer to the potential of the New Institutional Economics (NIE) to provide a different and valuable perspective. In particular, I wish to focus not only on the behavioral dicta that are reflected in the poems which neo- institutionalists refer to as the “rules of the game,” but on the implications of these rules, including their enforcement arrangements, the processes by which they are established, reinforced, challenged, or altered, and the various strategies that are adopted by culturally fluent actors in response to these rules. As we have already noted, modern economists have been guilty of ignoring the powerful effect of institutions on the variety of economic strategies developed by agents in response to their existence and particular forms. By and large, it is my belief that the scholarship dedicated to questions of economic activity in Homer has fallen prey to the same temptation that has affected mainstream economic theory, namely that rational self-interest represents the “purest” expression of human nature. Thus, economies that obey a different logic are paler reflections of this reality and modern behavior that fails to conform to this assumption is somehow anomalous, a conclusion grounded in an ethnocentric assumption that this vision of modern economic activity represents the apex of sophistication. This tacit assumption has led to a failure to recognize the complexity and diversity of pre-modern economic behavior, as well as obscured the reality that all economic agents act within a complex matrix of rules, both formal and informal. 105 As already discussed, when well-informed actors engage with each other within these institutional matrices, the strategic options are myriad. In general, the temptation is to regard conformity with the prevailing institutions as the only, or at least dominant, option. For this reason, institutions are often mistakenly referred to as constraints. However, their most important function is to create stable expectations. The establishment of these expectations then gives rise to an enormous range of possible strategies, many of which would not have been possible in their absence, and they are formulated in response to a recognition that the various actors involved are well aware of the prevailing institutions. In other words, not only can I adopt a particular strategy based on my own familiarity with the rules of the game, but my actions are informed by the fact that I know that the other actors are also familiar with them, and what is more, that their own actions are formulated with the same knowledge. Therefore, my interest in the texts is both in the mechanical aspects of economic activity (production, consumption, distribution), and in the rules that inform these activities. I wish to explore not only what those rules appear to be, but how actors engage with the rules, paying particular attention to how the institutions are formed, established, and perpetuated. Also, I am interested in the variety of responses to the existence of these institutions, including of course compliance, but also avoidance, rejection, shirking, and even outright challenges. In so doing, I maintain that it is crucial to keep in mind that what the texts present is an ideologically motivated representation, and thus it is important to consider not only what the epics suggest are the prevailing norms, but whether there are valuative judgements, and if so, what the motives for such assessments would be. On a related note, I also wish to examine the apparent connections between the institutions and the social hierarchy, and in particular how changes in the institutional structure might be correlated to changing power relationships within the society. Finally, given what we know (or at least can guess) about the changes that were already taking place by this time, we should also be alert to indications of stress upon prevailing institutions in light of the expansion of the Greek world, and how different groups in the contemporary society appear to be adapting to these changing conditions, with particular emphasis on the role of economic activity. 106 The World of Homer Of course, in order to understand the ways that economic activity was affected by and affected social and political conditions in the Archaic Period, we need to establish clearly what we believe those conditions were, as reflected or refracted through the lens of the Homeric epics. Much ink has been spilled on this issue, 51 and I have little desire to re-argue a case that has already received such thorough treatment. I merely wish to make a few brief points. First, I believe that Homeric society was more complex and stratified 52 than has typically been believed. This is important not simply as a corrective to the current scholarship, but is especially salient for my argument because this level of stratification produced potential lines of fracture in archaic Greek society and offered greater possibilities for class rivalries to develop. 53 Second, control of material resources played an under-appreciated role in the formation and negotiation of contemporary power relationships. This too is of particular importance for my argument, since I shall argue that changing economic conditions were not simply a product of larger socio-political developments, but rather that they produced a seismic shift in the balance of power in archaic Greek society because of the degree to which cultural and social capital were intertwined with economic capital. Third, I wish to examine the myriad justifications in the texts for elite standing. The epics are rife with various defenses of elite authority, so much so that I believe they collectively indicate a strong undercurrent of anxiety regarding the stability of the traditional hierarchy. Rather than indicating a static 51 The work of Finley (1954, 1998) continues to exert tremendous influence in the field, while Donlan (1973,1979, 1989a, 1989b, 1989c, 1994, 1997, 1999) has arguably been the most prolific on the subject; see also Redfield (1975), Raaflaub (1993, 1997, 1998a, 1998b, 1998c, 1998d), Morris (1996), and Tandy (1997). 52 I employ the term “stratified” here in a general sense in order to indicate a relative level of complexity, although I recognize that the term often has a particular meaning when used to describe social structures, based in particular on the work of Morton Fried (1967). Fried employed the term as part of his classification system of social complexity, identifying four general categories of social ranking (egalitarian, flexible rank, fixed rank, and stratified) to describe societies that ranged in degree of social complexity and inequality from least to most. Especially when considered in tandem with the work of Elman Service (1962), stratified social structures are generally equated with state-level political systems. Clearly, that is not an accurate depiction of Homeric society, and I wish to avoid any such possible confusion. 53 The most influential view characterizing Homeric society as structurally simple is that of Finley (1954), who maintained that 1) the Homeric epics offer a faithful representation of the structure of Greek society in the Early Iron Age, and 2) that this society was defined by a “deep horizontal cleavage” separating elites from non-elites. 107 social structure, the repeated emphasis on vindicating elite power is evidence for the strength of anti- aristocratic thought. Also, these justifications take numerous forms, but systematically elide the role of wealth in maintaining existing power relationships, a phenomenon that indicates a recognition of the importance of economic capital. Social Stratification From the outset, I wish to be clear that I believe that the Homeric epics represent a strongly elite-centric point of view. While this position is far from controversial, 54 it is central to my interpretation of the texts, and so it is important to establish that it forms a basis for my analysis. For one, this bias in the texts is responsible in my view for presenting an over-simplified picture of Homeric society. 55 The general picture that emerges from the poems is of a very simple socio-political structure, strongly consistent in many ways with the flexible ranked, “Big Man” type society. 56 It is equally clear that they maintain a consistent viewpoint, according to which the basileis are glorified, and the demos is ignored often to the point of total exclusion (Morris I. , 1986). However, I suggest that this representation of Homeric society 54 To be sure, the position that the Homeric epics are pro-elite is not universally accepted. The most extreme view to the contrary is that of Farron (1979), who sees the Odyssey as an anti-aristocratic statement, but Rose (1975) also finds what he calls “class ambivalence” in the Odyssey. I am much more strongly in favor of the arguments of Donlan (1973) or Thalmann (1988) who believe that there was a current of anti-aristocratic thought that was sufficiently powerful to intrude in the epics, but this in no way should be taken as an indication that the poems were generally supportive of such a view. 55 For the present purposes, I will refer to and make use of the classificatory schemes of Elman Service (1962) and Morton Fried (1967) to describe the structural properties of archaic Greek society. Service identified four major organizational structures (band, tribe, chiefdom, and state) that he believed were applicable, at least in general terms, to all human societies. In a similar vein, Fried described four different organizational principles (egalitarianism, flexible ranking, fixed ranking, and stratification) that he saw as underlying and producing the observable structures delineated by Service. For both, the goal was to establish a means of comparing societies and elucidating the ways in which they developed and perpetuated inequalities. I have chosen these models in large part because they have proven extremely influential, particularly among those scholars who have most prominently studied archaic Greek society such as Donlan (1981, 1989a), Qviller (1981), and Tandy (1997). Admittedly, these classification systems are not without their detractors. For example, Pauketat (2007) finds fault with them on the basis of the fact that they are a) evolutionary, and b) overly reductive. Still, Fried and Service have remained popular among anthropologists and archaeologists, especially those who study chiefdoms; see Earle (1987, 1991, 1997); King (2003); Pollack (2004). 56 Using the schema of Service (1962), Donlan (1979, 1985, 1989a) refers to Homeric society alternately as a “Big Man” type and as a “low-level chiefdom” (1997). 108 is deliberately over-simplified. Given the significant structural changes that were already taking place in the Greek world by this time, I do not believe that the relatively undifferentiated social structure depicted in the epics would have been capable of maintaining political and social cohesion as the Greek world expanded. Instead, in the face of mounting pressures and an undercurrent of anti-aristocratic sentiment, I believe that the social structure represented in the epics is an attempt to valorize an idealized past, a “golden age” in which elite primacy was largely unquestioned and unchallenged. In addition, the elite focus of the Iliad and the Odyssey produces a parallax view of Homeric society, with the elites occupying a disproportionate share of the narrative, while non-elites are regarded as a more or less homogeneous group, defined more by the fact that they do not belong among the elites rather than by any immanent characteristic. There are numerous terms used to describe elites, ranging from descriptive attributes (exochos, megathymos, protos, aristos) to particular titles (agos, hēgētōr, hēgemōn, basileus, anax). 57 By contrast, non-elites are usually lumped together (laos, dēmos, plēthys). In fact, this distinction is so prevalent in the poems that Finley (Finley, 1954) regarded this “cleavage” as a defining feature of Homeric society. To be sure, I do not mean to suggest that this divide did not exist, nor that it was not a significant feature. 58 Indeed, there is ample evidence to indicate that membership in the aristocracy had 57 The precise nature of these titles has been hotly contested, and the fact that they seem to change over time has done little to simplify things. Particularly problematic are the terms basileus and anax, that are clearly etymological descendants of the Mycenaean titles pa-si-re-u and wa-na-ka attested in the Linear B tablets. In Linear B, the basileus is evidently a local potentate of limited importance, while the anax is a title indicative of much greater influence. However, they are used interchangeably in Homer, and numerous other political titles found in the tablets (lawagetas, korete, prokorete, telestai) did not survive into the 8 th c. BCE, strongly suggesting that they have lost any sense of their earlier distinctions and become simply generic terms for a leader. The translation as ‘kings’ is problematic, given the culturally specific and anachronistic connotations of the term. Some even argue that ‘kings’ ceased to exist in the Early Iron Age (Qviller, 1981; Rihll, 1992). Others have contended that ‘king’ is an appropriate translation based evidence for some degree of heritability of the position; Donlan (1997), using the terminology of Service (1962), equates the basileus to a “low-level chief,” and Raaflaub (1997) calls them “aristocracy in formation” and rejects the notion that a basileus is a ruler or monarch; Tandy (1997) admits it is possible to conceive of heritable power or leadership, but insists that whatever is being inherited is no more secure than when held by its previous holder. 58 For instance, see how the young warrior Thoas describes himself and the other Achaean leaders as the “best in the army” (15.295-299), or the markedly different way in which Odysseus addresses common soldiers as opposed to the 109 become such a profoundly important identity marker that Homeric elites felt a stronger sense of shared identity with other elites from exogenous groups than with commoners within their own groups. 59 This is precisely the point emphasized by Herman (1987), who argues that intra-aristocratic bonds that were perpetuated and reinforced by the system of xeinia often created friction within the polis, as aristoi frequently felt a greater sense of obligation to their foreign xenoi that to their own fellow citizens. However, I maintain that this view oversimplifies the social structure. By focusing on the distinction between demos and aristoi, we risk ignoring many other ways in which the epics imply the existence of a more heterogeneous, complex social structure than has typically been observed. For one, the epics are replete with indications of class and status hierarchy within the aristocracy. Indeed, one of the primary sources of tension in the Iliad is the power struggle between Agamemnon and Achilles. At the same time, we should also recognize the profound disparity between slave and free. 60 However, because of their relative lack of agency and their alienation from the means of production (a situation that did not change during the period), the role and status of slaves is not especially pertinent to my particular argument. This is not to diminish the importance of this topic to the broader history of the period, only to keep the focus on the topic at hand. 61 basileis on the basis of their rank (2.198-202), reflecting what Thalmann (1988) characterizes as an assumption of an inherent qualitative difference. 59 A powerful indication of this emphasis on shared status and how it takes priority over other forms of group membership can be seen in the episode in which Diomedes and Glaucus, soldiers on opposing sides, recognize their relationship towards one another as hereditary guest-friends/xeinoi (6.215-231). 60 As Thalmann (1998) argues, it is even appropriate to think of them as a class, given that they are united by, among other things, exclusion from the means of production and alienation from the products of their own labor. Their identity is formed largely etically, but it is an identity sustained by countless reminders and reinforcements. Whether they share a class consciousness is not clear, and in light of the fact that the only voices we have for them are mediated through an elite perspective, any conclusions about what actual slaves may or may not have thought of themselves and their condition would be tenuous at best. However, class consciousness is not necessary to the existence of class divisions (Rose, 2009, 2012), an issue that Marx (1963 [1852]) himself recognized. 61 Certainly, the topic of slavery has been hotly debated by classicists and ancient historians for some time and remains an extremely controversial topic. The field even today is a potential mine field (Scheidel, 1998) that I seek to avoid as much as possible. In addition, the subject has seen a number of important studies in recent years that provide much more thorough and careful treatment of the topic than I could possibly hope to achieve (Fisher, 1993, 110 With that in mind, I would suggest that diversity among non-elites is a feature of Homeric society that can be glimpsed at various points in the epics, and represents the potential for collective action by non-elite groups acting in their own self-interest, and therefore a means of challenging elite primacy. For instance, in both the Iliad and the Odyssey, there are mentions of unskilled labor. “Country folk” (aneres agroiōtai) are mentioned as a general category (11.549, 15.272), but more often, specific occupations appear, including woodcutters (drytomos 62 ), hunters (epaktēr, 63 thērētēr, 64 elaphēbolos 65 ), and generic herdsmen (nomeus, 66 poimēn 67 ). 68 At the same time, there are still many mentions of rural labor, particularly animal herding. In addition to the noble swineherd (sybōtēs) Eumaeus, who is accompanied by several other herdsmen, there is the goat herder (aipolos) Melanthius, also assisted by others, and the cattle herder (boukolos) Philoetius. It is important to note that we should be careful not to assume that all such laborers are servile. Although it is clear that both Melanthius and Eumaeus are slaves, we find out that Philoetius has stayed serving the household of Odysseus out of a sense of loyalty, not out of compulsion. He claims that the situation with the suitors rankles him, and were it not for his personal attachment to Odysseus, he would have left to serve another basileus long ago (XX.218-223), very possibly not an option that would have been available to a slave. 1995, 2006; Thalmann, 1998; Wiedemann, 2000, 2003; DuBois, 2003; Andreau & Descat, 2006; McKeown, 2007, 2011). 62 Ex., 1.86, 14.633, 23.315. 63 Ex., 17.135. 64 Ex., 4.51, 11.292, 15.581, 17.726, 21.574. 65 Ex., 18.319. 66 Ex., 11.697, 14.632, 18.525, 18.577, 18.583. 67 Ex., 3.11, 4.455, 5.137, 9.559, 12.451, 13.493, 15.354, 17.162, 23.835; in a more specific sense as sheep and goat herders, 5.137, 12.451, 13.493. 68 As Donlan (1999) points out, many of these, such as woodcutter and hunter, are more likely simply terms for the person who is in the act of cutting wood or hunting, and not an indication of professionalization. In many instances, this is doubtless true, although Donlan also admits that many other instances exist that suggest at least in the agricultural sphere that there was a significant degree of labor specialization. 111 There are also mentions of wage laborers. Despite the prevailing view that agricultural production was dominated by small, relatively self-sufficient farming households, 69 the Iliad offers numerous glimpses into various types of wage labor, including a wool-worker struggling to earn enough to feed her children (12.433-435) and day laborers (erithoi) working a large estate (18.550, 18.560), and thetes also appear at various points in the Odyssey, most notably in the exchange between Odysseus and the shade of Achilles in the Underworld (XI.489-491). Finally, we also find descriptions of various skilled professions, among them religious figures. There is, of course, the example of the priest (arētēr 70 ) Chryses (1.11), who came to the Greek camp to ransom his daughter and was rebuffed by Agamemnon, as well as the seer (theopropos 71 ) and bird-diviner (oiōnistēs 72 or oiōnopolos 73 ) Calchas (1.92, 13.68-71) who is responsible for revealing that the plague afflicting the Greek force was inflicted by Apollo at the behest of his priest. We also find mention of the dream-interpreter (oneiropolos 74 ) Eurydamas and his two sons (13.68-71), Laogonos (16.603-605), the priest (hiereus 75 ) of Zeus, as well as Theoclymenus (XV.224-264) the seer and son of Melampous (mantis 76 ). In addition to religious professionals there is also evidence in the Iliad for the existence of healers. Although the text does not suggest that healers (iētroi 77 ) are full-time professionals, they are obviously held in high esteem, and their knowledge and skill are seen as important assets. This high regard foreshadows the later developments in the Archaic Age, in which physicians would come to be so prestigious that they could become quite famous, and find their services in high demand throughout the Mediterranean (Wees, 2009). 69 Ex., Finley (1999 [1973]); Donlan(1989c, 1997); contra van Wees (2009). 70 Cf. 1.94, 5.78 71 Cf. 12.228. 72 Cf. 2.858, 17.218. 73 Cf. 6.76. 74 Cf. 1.63. 75 Cf. 5.10, 24.221. 76 Cf. XVII.384. 77 See also 2.732, 13.213, 16.28 112 Finally, we see the existence of specialized craft labor. In fact, craft specialists are described as belonging to a separate category: τίς γὰρ δὴ ξεῖνον καλεῖ ἄλλοθεν αὐτὸς ἐπελθὼν ἄλλον γ᾽, εἰ µὴ τῶν οἳ δηµιοεργοὶ ἔασι, µάντιν ἢ ἰητῆρα κακῶν ἢ τέκτονα δούρων, ἢ καὶ θέσπιν ἀοιδόν, ὅ κεν τέρπῃσιν ἀείδων; οὗτοι γὰρ κλητοί γε βροτῶν ἐπ᾽ ἀπείρονα γαῖαν… “For who, traveling abroad, calls another man ‘foreigner’ Unless he is one of those workers for the people, A prophet, a healer of sickness, a carpenter, Or an inspired singer who pleases with his song? These of all men are welcomed across the boundless earth…” XVII.382-386 The message here is unmistakable – there are certain people who, by virtue of the universally acknowledged value of their work, are considered to belong to a special category. 78 The distinguishing feature of these specialists, and one that the text themselves recognize, is their advanced training and skill: µήτι τοι δρυτόµος µέγ᾽ ἀµείνων ἠὲ βίηφι: µήτι δ᾽ αὖτε κυβερνήτης ἐνὶ οἴνοπι πόντῳ νῆα θοὴν ἰθύνει ἐρεχθοµένην ἀνέµοισι: µήτι δ᾽ ἡνίοχος περιγίγνεται ἡνιόχοιο. 23.315-318 It is through skill that the woodcutter becomes much better, not force; It is also through skill that the pilot keeps a ship On course as it is buffeted by the winds on the wine-dark sea; And so it is through skill that one charioteer surpasses another. Several points are to be made here. The binary distinction between elite and non-elite is not without value, but it is incomplete and fails to reflect the degree of complexity and diversity that becomes 78 As Tandy (1997) points out, the general consensus is that such workers for the people were characteristically outsiders, and probably itinerant. Weber (1925) argues that this category includes all those whose occupations served indeterminate large groups, while Murakawa (1957) insists that they were not strictly limited to craftsmen, but included all those occupations that were indispensable, and further, that dēmioergoi were especially skilled in such tasks and thus frequently invited from abroad. Along with Finley (1973), I see nothing in this passage that would require them to be itinerant, merely that their labor was so highly regarded that they would be welcomed wherever they went. Regardless, it seems evident that these non-elites have attained a different, more elevated status by virtue of their skill and the value of their work. 113 apparent on a closer reading. The picture becomes even more complicated when we introduce the example of heralds (kērykes 79 ) who seem to occupy a more preferential position. They are occasionally mentioned by name, 80 and are often mentioned in a supervisory capacity, including watching over the drawing of lots 81 or regulating the assembly. 82 To be sure, it is not possible to say with any certainty in what numbers various skilled professionals existed at this time, nor to establish with precision to what degree labor specialization was a feature of the economic system. However, their existence cannot be reasonably denied, and the existence of specific terms to indicate particular professions seems incompatible with a view that they were no more than activities associated with a general category of work, or at least not in every instance. A wood-cutter (drytomos) may simply be someone who is in the act of cutting wood, but a potter (kerameus), sailor (nautēs), or bronze-smith (chalkeus) is not to be explained so easily. 83 The existence of these multiple categories produced conditions that seem fertile ground for the formation of specific group interests, and along which fracture lines could appear in the society. Regardless of whether or not such groups developed a conscious identity, as Rose (2012) points out, the lack of class consciousness is not an automatic barrier to collective action, and the greater the degree of social complexity, the greater the potential for the existence of class rivalries. In addition, the division of labor indicates a level of economic sophistication well beyond that which has typically been associated with Homeric society. From the beginnings of modern economic theory with Adam Smith (2003 [1776]), labor specialization has been considered an analogue for systemic economic complexity and development. The presence of various types of laborers, including wage labor, unskilled workers, and highly skilled craft 79 Ex., 2.184, 4.198, 9.174, 12.342, 18.558, 23.897, 24.149, 24.178, etc. 80 4.192, 10.315. 81 7.183. 82 2.50, 9.10. 83 An additional wrinkle is represented by the case of charioteers (hêniochoi), who sometimes appear simply to be whoever holds the reins, whereas at other times seem to belong to the category of attendants (therapontes); see 5.580, 8.89, 8.119, 23.132. 114 specialists, suggests a degree of economic development sufficient to offer a range of opportunities for economic cooperation well beyond the strictly limited modes of exchange indicated in the Homeric poetry. Class and Status in Homer For Tandy, the Archaic Period was one in which the traditional social order was disrupted by a changing dynamic between class and status. Specifically, he contends that for the most part, differential access to material resources was a function of social standing early in the period, whereas by the end of the period, control over resources largely determined one’s rank. As he puts it, initially class followed status but later, status followed class (1997). However, I believe that this is a misconception born largely from the fact that Homeric elites sought to elide the role of wealth in status formation. The elite vision of the epics obscures the fact that, as Piketty (2011) illustrates, wealth is a social relation of power and control. Thus, even from the very beginning, it was an integral part of the formation and propagation of the social hierarchy. In general, status typically is either based on one’s own achievements (“achieved status”), or one’s position is inherited (“ascribed status”). 84 In the Iliad, claims to achieved status almost invariably are based on two complementary categories, the ability to inspire others through public speaking on the one hand, and prowess on the battlefield on the other hand. 85 At the same time, age is often referred to as a 84 For details regarding the distinction between ascribed and achieved status, the foundational work is Linton (1936); additional elaboration and clarification can be found in Foladare (1969). 85 For example, Peleus entrusted his young son Achilles to Phoenix in order to learn to be a “speaker of words and a doer of deeds” (9.442-443), Diomedes is esteemed among the Achaeans both for his preeminence in battle and in counsel (9.53-54), and the young Thoas is singled out both for his excellence in battle and in speaking (15.281-283). There are, however, suggestions that martial prowess carries greater prestige than skill as a public speaker as evidenced by both Achilles (15.105-106) and Hector (18.251-252), although Hektor makes his fatal mistake in the assembly in Book 18, so that he is shown to be lacking in speaking and taking counsel. The ideal seems to be a balance, as in 9.442-43. 115 marker of superiority. 86 Homeric elites assert their status on the basis of age, descent/inheritance, and ability (both in deliberation and in action), and in the case of Agamemnon, on the basis of political power. 87 This sentiment is echoed by the narrator in the Catalogue of Ships, describing the Mycenaeans: τῶν ἑκατὸν νηῶν ἦρχε κρείων Ἀγαµέµνων Ἀτρεΐδης: ἅµα τῷ γε πολὺ πλεῖστοι καὶ ἄριστοι λαοὶ ἕποντ᾽: ἐν δ᾽ αὐτὸς ἐδύσετο νώροπα χαλκὸν κυδιόων, πᾶσιν δὲ µετέπρεπεν ἡρώεσσιν οὕνεκ᾽ ἄριστος ἔην πολὺ δὲ πλείστους ἄγε λαούς. 2.576-580 Ruling over each of the hundred ships was Lord Agamemnon, Son of Atreus, and with him by far the most and bravest People followed; among them he himself stood, proudly decked out in shining bronze, Distinguished among all the heroes Since he was the best, and ruled by far the most people. If we accept the notion that power is one of the elements of status, 88 and that power is the ability to induce others to fulfill one’s own goals (Schooler, 2017), then this passage is an example of Agamemnon’s power, and thus a component of his status. Meanwhile, in the peacetime world of the Odyssey, the justification is often framed in form of a mutually beneficial relationship between the generous elite protector and the loyal, non-elite supporter. 89 So, elite standing can be justified by claims on numerous bases, including age, birth, military prowess, skill at public speaking, mutual benefit, and political power. However, the role of wealth in the formation of status in Homeric society is frequently downplayed. Nevertheless, I believe that in the epics we can still see some of the subtle yet powerful ways that wealth was manipulated to reinforce social hierarchies, while at the same time masking the material basis for 86 Odysseus, for instance, claims equal standing with Achilles, despite the latter’s superior skills in battle, on the basis of his age and experience (19.216-219). 87 In his dispute with Achilles (1.177-187), Agamemnon admits that Achilles is stronger (karteros), but lays claim to superior standing because he is more powerful (pherteros), basing his claim on political power, an argument that Nestor validates (1.278-281); see Rose (1992) on the novelty of Agamemnon’s argument. 88 See p. 95ff. above. 89 For example, Mentor refers to Odysseus as a “gentle father”/patêr êpios (II.234), and Eumaeus calls him a “kindly lord”/anax euthymos (XIV.64). 116 these distinctions. In other words, we can see some of the complex ways that class 90 overlaps and intersects with status, both achieved and ascribed. One of the more vivid examples of this process can be seen through the penalty (apoina) offered to Achilles by Agamemnon as recompense for his original transgression of snatching Briseis, the female war-captive awarded to Achilles. The offer is beyond extravagant, including gold, horses, tripods, slaves, one of his own daughters to take as a wife without bride-price, a dowry that Agamemnon promises will be unprecedented, and seven whole inhabited cities along with their associated income (9.121-157). However, he makes this offer with an important proviso: δµηθήτω: Ἀΐδης τοι ἀµείλιχος ἠδ᾽ ἀδάµαστος, τοὔνεκα καί τε βροτοῖσι θεῶν ἔχθιστος ἁπάντων: καί µοι ὑποστήτω ὅσσον βασιλεύτερός εἰµι ἠδ᾽ ὅσσον γενεῇ προγενέστερος 91 εὔχοµαι εἶναι. 9.121-161 Let him yield; Hades is unforgiving and unyielding, And therefore to mortals he is the most hateful of all the gods; And let him accept his place below me, on account of how much kinglier I am And how much I can claim to be superior by birth. In this episode, we can see some of the complex ways that status and class overlap. As we know, Agamemnon has instigated a conflict with Achilles because of a perceived slight to his status (the loss of Chryseis). Ostensibly in order to resolve the conflict, he then makes an offer of expensive gift items (that are reflective of his class), but with the proviso that in accepting the “gifts,” Achilles acknowledge Agamemnon’s superior status. Meanwhile, part of his offer includes seven poleis under his control that are rich in sheep and oxen and whose people will shower Achilles with gifts and carry out his commands. In essence, the seven poleis embody the overlap between class and status for the aristoi. Agamemnon emphasizes their value both on the basis that they will carry out Achilles’ decrees (themistai), thus 90 For our purposes, class is “defined by an individual’s access to available material resources…and perception of rank in society relative to others…” (Kraus, Piff, Mendoza-Denton, Rheinschmidt, & Keltner, 2012). 91 The word that I have translated “superior by birth” (progenesteros) can also mean “older.” However, in the context of Agamemnon’s general claims to elevated status, I believe that the former translation is more fitting. 117 enhancing his authority, as as well as making the clear implication that they will provide material benefits as well, thereby augmenting his wealth. However, both of these benefits appear to be complementary. In other words, it is not that power precedes wealth, or that wealth precedes power, but that both are mutually reinforcing. This dynamic is consistent with the Agamemnon’s larger strategy, by which he attempts to use his wealth as a vehicle for elevating his status. That this is his primary objective is beyond question, as evident by his demand that Achilles explicitly acknowledge Agamemnon’s superior standing. Ultimately, the passage illustrates the way in which Agamemnon’s political power provides him with his preferential access to resources, which he then leverages in order to validate his position. The conclusion to this recitation of material gifts is that he is “kinglier” (basileuteros) and “superior in birth” (progenesteros). Agamemnon attempts to convert his economic capital into social capital (and vice versa), and disguises this rhetorical move by clothing it in a manner more consistent with conventional terminology. From these passages, I wish to make a couple of important points. The first is to note how frequently and prominently various justifications of elite standing are featured throughout the text. 92 Again and again, the poems offer various rationalizations for the primacy of elites in Homeric society. The only reasonable explanation for such an abiding concern for validating elite authority, or so it seems to me, is an underlying anxiety that suggests palpable challenges to the status quo. As Raaflaub (1998a) argues, there are numerous social tensions reflected in the poems, including a growing recognition of the importance of 92 In reality, claims to elite superiority take two main forms in the epics. The first is intra-elite rivalry between individual aristoi (such as the dispute between Agamemnon and Achilles) when one of the elite claims to be ascendant over another. The second concerns broad claims of group preeminence for the elite more as a group, and is thus an argument in favor of elite primacy on a systemic level. The most explicit and extended justification for the hierarchical structure of Homeric society is in the dialogue between Glaucus and Sarpedon, in which Sarpedon goes to great length to explain the reason why they, and others like them, hold a place of primacy (12.308-328); see also 2.198-202, 15.295-299 where the aristoi are explicitly contrasted with the demos. 118 the rank and file soldiers in the military, the shortcomings of aristocratic leadership, and the essential problems and needs of the community. 93 Second, there appears to be an undercurrent of tension regarding the overlap between wealth and power. One of the more prominent complaints by Thersites (2.225-242), for example, is that Agamemnon has profited disproportionately, taking a greater share than what he has earned. As Thersites explains, he and the other Achaean soldiers are the ones who bring in the spoils and the ransom for captives that Agamemnon receives (2.229-231). Although he does not make the connection between power and wealth explicit, the implication of his criticism is that there is a flaw in the system that allows Agamemnon to exploit his political authority to obtain a greater share of the spoils than what he has earned through his performance on the field of battle. I believe that this episode indicates a degree of elite anxiety concerning the process of conversion of capital. What Agamemnon asserts is a status that is immutable and immanent. However, as we have seen, that status is grounded to a certain extent on access to resources, which are external to the individual. As Bourdieu (1986) argues, it is common for elites to be concerned about “bringing to light the arbitrariness of the entitlements transmitted and of their transmission…” and the episode of Thersites serves to illustrate this concern. However, I do not regard the accusation as an authentic criticism of elite privilege. Rather, it seems to be a way for the elite-centric Iliad to undercut the accusation. By giving voice to the argument through the figure of Thersites, whom the text describes as vain, quarrelsome, constantly looking to curry favor, and physically malformed, the text calls the validity of the accusation into question by discrediting the accuser. Then, he is beaten and publicly humiliated by Odysseus with the approval of 93 As stated previously (see n. 48 above), I am unconvinced by Farron’s (1979) assertion that the Odyssey is an anti- aristocratic diatribe. However, that does not diminish the possibility of a contemporary current of thought that questioned the assumptions which supported the traditional social hierarchy. Most notably, Donlan (1973) extensively chronicles numerous examples in archaic poetry that prompt him to identify what he calls a tradition of anti-aristocratic thought. 119 the assembled Greek host, who laugh and call it “the greatest thing [Odysseus] has ever done for the Argives…” (νῦν δὲ τόδε µέγ᾽ ἄριστον ἐν Ἀργείοισιν ἔρεξεν). 94 Thus, I maintain that the Thersites episode functions as a release valve for elite angst, with the Iliad serving as a sort of “safe space” within which the traditional nobility were able to confront criticisms and concerns in a way that was less threatening. 95 Ultimately, what we can glimpse in the texts is a society that, despite a surface appearance of stability, seems to be facing genuine structural challenges. For Raaflaub (1998a), the nascent beginnings of important social and political changes can be glimpsed in the texts, and as Donlan (1973) has argued, there was a powerful current of anti-aristocratic sentiment in early Greek poetry. We have already seen in our brief initial examination of the archaeological evidence that major developments were already underway by this time. Taken all together, these various factors go a long way to explain why we see attempts to validate aristocratic authority appear so frequently in the elite-centric Homeric epics. Finally, we have seen how control over resources plays a significant role in establishing and perpetuating power relationships, as well as how this phenomenon seems to have been an additional source of tension. Economic Activity in Homer As I stated at the outset, my interest is not only in the specific forms of economic activity that are manifested in the texts, but also the “rules of the game” within which such activity takes place. This approach is fundamentally grounded in the premise that specific forms of economic behavior do not arise in a vacuum, but rather they are shaped in response to systems of rules. In examining these institutions, I 94 2.274. 95 I believe that the Odyssey offers something very similar in the Aeolus episode (X.34-50), in which Aeolus, lord of the winds, binds up all the contrary winds in a bag and gives it to Odysseus so that he will have a constant following wind back to Ithaca. Believing that the bag contained gifts for Odysseus that he was not willing to share with the rest of them, his crew decides to open the bag, thereby loosing all of the other winds and driving them away from the shores of Ithaca. In the Aeolus and Thersites episodes, all of the same elements are present: resentment of elite prerogatives, accusations of greed, criticism of elite control of resources, and open challenges to elite primacy; but most of all the end result of the challenge is punishment. Thersites is beaten and ridiculed by Odysseus for his outburst, while the sailors of Odysseus are driven away from the shores of their homeland, which they will never see again, because they failed to trust in Odysseus. Thus, the epics allow for the introduction of anti-elite sentiments, but in a way that discredits and delegitimizes them. 120 want to be particularly attentive to questions related to the prevailing social hierarchy in terms of how various institutions, and the specific forms of economic activity they encourage, benefit or disadvantage certain social groups, as well as how the institutional structure might be prone to challenge and how it might be defended in the face of protest. In doing so, I have no interest in attempting to re-invent the wheel, so I will utilize the familiar model followed by most authors, traceable back at least as far as Marx (2005 [1939]), who saw three fundamental aspects to economic systems: production, consumption, and distribution. 96 With this as a foundation, I will examine the mechanical processes of economic activity, and then contextualize these activities within the institutions as they appear in the epics. Production Agricultural Production Due largely to the constraints of the narrative, the Iliad is largely silent on aspects of production. It is in the Odyssey that we find by far the majority of available evidence, and throughout the text, there are abundant indications of the fundamental importance of farming to the very survival of Homeric society. The cultivation of olives and vines is mentioned at several points, but it is the production of grains (karpon), both wheat (pyros) and barley (krithē), that is represented as being absolutely essential, so essential in fact that the consumption of bread is presented as a fundamental aspect of what it means to be human. 97 This emphasis is probably not surprising, given the degree to which labor in the archaic Greek world was devoted to food production. By some estimates, roughly 80% of the population was involved in agriculture (Migeotte, 2007). Also, with the technology available at the time, farming involved a 96 In his Grundrisse, an unfinished manuscript that Marx set aside in 1858 and that was published posthumously in 1939, Marx actually identifies four different categories of economic activity: production, consumption, distribution, and exchange. However, the conventional view regards exchange as a part of the process of distribution. 97 Men are often described as “bread eaters” (VI.7-8, X.100-101, IX.190-192), and wheat and barley flour are referred to as the “marrow of men” (XX.106-108). 121 substantial investment of time and effort (Weisdorf, 2005). The labor demands of ancient agriculture may have at times been exaggerated in the modern scholarship, but the general impression that it constituted the single largest demand on the aggregate labor of archaic Greek society seems unassailable. In light of the apparent importance of grain production to Greek society of the late 8 th /early 7 th c. BCE, one would expect to find numerous and detailed descriptions of various aspects of agriculture in the Odyssey, including agricultural techniques, the distribution of resources, and the social status of farmers, among others. However, this could not be further from the truth. What we do have in the text are numerous suggestive passages indicating, consistent with our expectation, that agricultural production was widespread and its techniques well-understood. For instance, several passages demonstrate conventional farming techniques. For instance, we discover that animal manure is used to enhance the fertility of farmland (XVII.297-301). Also, we are told that there are draft animals, presumably used in farming, activities that are typical of what anthropologists typically refer to as “intensive agriculture” (Bodley, 2011). In fact, there is evidence for several different technological developments that are hallmarks of more developed, intensive agricultural strategies, including irrigation (VII.129-130), terracing (XXIV.223-225), the use of draft animals (VIII.124-125, XVIII.371-375) in which the text clearly states that both mules (hēmionoi) and oxen (boes) are used to plow fields, and the application of natural fertilizers (XVII.298-299). 98 98 At the same time, numerous questions about the nature of archaic Greek agricultural practices remain unanswered. This may very likely be due, at least in part, to the commonly observed tendency of the epics to pass over aspects of the contemporary society that are commonplace and mundane. Unfortunately, this means that any number of fundamental issues remain unresolved. For example, the passing references to practices that are characteristic of intensive agriculture indicate only that they existed, but do not provide any reliable sense of whether they were widespread. Also, agricultural practices were almost certainly far more complex than can be glimpsed in the brief textual references in the Iliad and the Odyssey. Amouretti (1994) highlights both the indications of the likelihood of widespread variability in practices in ancient Greece, due in part to differences in geography, weather, and other environmental factors, as well as the prevailing tendency until recently of modern scholarship to overlook such complexity. For the case of ancient Greece, see also Isager and Skydsgaard (1992) and Wells (1992). From the anthropological perspective, Johnston (2003, with references) argues forcefully against the over-simplification that tends to arise from reliance on the evolutionary model of Boserup (1965). 122 In addition, there is ample evidence in the text to demonstrate the prevalence and importance of grain farming. For example, when Telemachus and Nestor’s son Peisistratus travel from Pylos to Sparta (III.495), they cross a plain full of wheat (pedion pyrēphoron). Also, Doulichion is described at least twice (XIV.335, XIX.292) as being particularly distinguished for its abundant production of wheat (polypyros). So fundamental is agricultural production, in fact, that when Nausithous led the Phaeacians to settle in Scheria, the first three things that he accomplished were to build a wall around the city, build housing, and allot plough lands (VI.7-10). It is clear that agricultural production, most particularly of grains but also vines and olives, plays a central role in the contemporary society. 99 Nevertheless, given the apparent central importance and prevalence of agricultural production visible in the Odyssey, we might understandably expect to find numerous mentions of those individuals engaged in agricultural activities. However, this is anything but the case. At most, we are offered only the briefest indications about the way that agricultural labor is organized. For instance, in the house of Alcinous there are fifty female servants, some grinding grain (karpon), some weaving, and others pressing olves for oil (VII.103-107). In a similar passage, twelve serving women in the house of Odysseus are engaged in grinding wheat flour (alphiton) and barley flour (aleiata) with hand mills (XX.105-109). It would appear, from these excerpts, that ploughing and reaping was largely the responsibility of men (XVIII.366-380), whereas the processing of the grain into flour was undertaken by slave women. Beyond that, however, there is virtually nothing. This absence is so conspicuous that it seems to be more than just the omission of what would have been common knowledge to the contemporary audience. The elision of grain farming is so complete and systematic, in fact, that I believe it must have been a deliberate omission. We are told that bread is a 99 The temenos (land set aside by the community for the usufruct of the basileus) typically includes land for plowing; see 6.194-195, 9.578-580, 12.313-314, 20.184-185, XVII.297-299, and esp. 18.550-557, where the harvesting of grain is being carried out in the temenos basilêïos explicitly under the direct supervision of the basileus; see also Donlan (1989d) for a more extended description of the characteristics of the temenos. 123 virtually ubiquitous part of the diet, so important that its consumption is conceived of as an essential quality of what it means to be a human being. Descriptions of the territories of the Phaeacians, the Cyclopes, the Peloponnese, and Ithaca all indicate that land use for cultivation of cereals is of primary importance and widespread. There are even several remarks that describe the use of different farming techniques and equipment. Amidst all of this, there is only one mention of individuals directly engaged in the cultivation of grain, olives, or grapes. 100 Based on this scant textual evidence, it does seem clear that the basileis and the demos both participated actively in agricultural activities. However, the ways in which this was organized is opaque, and more importantly, it is not at all clear whether the bulk of the agriculture was produced by elite households. It is virtually impossible to form any firm conclusions regarding the ways in which agricultural labor was organized; we do not know if the majority of farming was conducted by small free landholders, larger kinship groups, free farmers with the aid of slaves or casual laborers, 101 larger proprietors, or some combination of some or all of the above. Ultimately, we can only conclude that such agricultural activities involved the participation of various social groups, while the lack of mention (in contrast to the emphasis on horticulture) can only reasonably be explained as the result of a more deliberate effort to exclude farming, and more specifically farmers, from the narrative. 100 In the famous description of the Shield of Achilles, there are two passages describing agricultural activity, one describing ploughmen (arotêres) tilling a field (18.541-549) and another of a temenos in which hired laborers (erithoi) are harvesting grain with sickles, and binders (amallodetêres) are binding the sheaves with the assistance of boys (paides), supervised directly by the basileus (18.550-557). The land in first passage is described as generic “fields/plow land” (arourai), while that in the second is specifically classified as a temenos. Meanwhile, unlike in the passage describing the work taking place in the temenos, there is no mention of any supervision of the ploughmen. The implication is that although both passages deal with aspects of grain farming, the work is organized and supervised in distinctly different ways. The agricultural labor in the temenos is clearly under the direction of the basileus, while labor in the plow lands seems to be a communal effort. 101 In addition to the description of hired laborers working in the temenos of a basileus on the Shield of Achilles, we are told that the slaves (dmôes) of Odysseus are the ones who spread the dung to fertilize his fields. This could imply that the rest of the agricultural labor in his household was carried out by slaves as well. 124 Household Production In contrast to the dearth of references to those engaged in farming, the Odyssey abounds with descriptions of elite household production. In particular, the text is fixated on animal herding and horticulture, 102 two activities that appear to be strongly associated with elite households. In fact, the two most notable examples of elite households, those of Alcinous and Odysseus, bear striking similarities in the ways that they are structured and organized. For instance, we are told that the orchard/garden (alōē) and plot of land (temenos) that belong to Alcinous lie outside the city (VI.293). Likewise, Odysseus has to venture out to the well-wooded fields (polydendreon agron) in order to visit his father Laertes (XXIII.360), who is tending to the orchard (orchatos) “far from the city” (XXIV.212). Indeed, an extended description of the house of Alcinous provides a substantial level of detail: ἔκτοσθεν δ᾽ αὐλῆς µέγας ὄρχατος ἄγχι θυράων τετράγυος: περὶ δ᾽ ἕρκος ἐλήλαται ἀµφοτέρωθεν. ἔνθα δὲ δένδρεα µακρὰ πεφύκασι τηλεθόωντα, ὄγχναι καὶ ῥοιαὶ καὶ µηλέαι ἀγλαόκαρποι συκέαι τε γλυκεραὶ καὶ ἐλαῖαι τηλεθόωσαι. τάων οὔ ποτε καρπὸς ἀπόλλυται οὐδ᾽ ἀπολείπει χείµατος οὐδὲ θέρευς, ἐπετήσιος: ἀλλὰ µάλ᾽ αἰεὶ Ζεφυρίη πνείουσα τὰ µὲν φύει, ἄλλα δὲ πέσσει. ὄγχνη ἐπ᾽ ὄγχνῃ γηράσκει, µῆλον δ᾽ ἐπὶ µήλῳ, αὐτὰρ ἐπὶ σταφυλῇ σταφυλή, σῦκον δ᾽ ἐπὶ σύκῳ. ἔνθα δέ οἱ πολύκαρπος ἀλωὴ ἐρρίζωται, τῆς ἕτερον µὲν θειλόπεδον λευρῷ ἐνὶ χώρῳ τέρσεται ἠελίῳ, ἑτέρας δ᾽ ἄρα τε τρυγόωσιν, ἄλλας δὲ τραπέουσι: πάροιθε δέ τ᾽ ὄµφακές εἰσιν ἄνθος ἀφιεῖσαι, ἕτεραι δ᾽ ὑποπερκάζουσιν. ἔνθα δὲ κοσµηταὶ πρασιαὶ παρὰ νείατον ὄρχον παντοῖαι πεφύασιν, ἐπηετανὸν γανόωσαι “Outside the courtyard, next to the gate, there is a large orchard 102 It should be noted that for the purposes of this discussion, I am using the terms “agriculture” and “horticulture” in a somewhat narrower sense, according to which horticulture denotes the cultivation of domesticated plants without the use of draft animals. This is in contrast to agriculture, which applies more intensive techniques to cultivation, including the use of draft animals, fertilizers, and irrigation. This seems to me to be the most effective way to convey the clear distinction in the text between the cultivation of fruits and vegetables carried out in a garden/orchard (orchatos, alōē, kēpos) and the cultivation of cereals which takes place in plowed farmland (arourai). 125 Of four measures of land; a fence had been built around it on both sides. Within grow large, lush trees, Pear trees, mulberry, and apple, with shining fruit, Sweet fig trees and verdant olive, And the fruit of these trees never goes bad or fails to grow Either in winter or summer, lasting through the entire year, and always The west wind blows, making some fruits grow and others ripen, Pear upon pear, apple upon apple, Grapes upon grapes, and fig upon fig, all flourish. Also within had been planted a fruitful vineyard, In a flat part of which, open to the sun, Some grapes are drying, while men gather more, And crush others with their feet. In front of these are others that are not yet ripe, Some only just flowering, and others turning color. Also within, garden beds are laid out next to the last of the vines, And they produce all kinds of vegetables, vibrant and abundant.” VII.114-128 Compare this description of the orchard of Alcinous with that of Odysseus: εἰ δ᾽ ἄγε τοι καὶ δένδρε᾽ ἐϋκτιµένην κατ᾽ ἀλωὴν εἴπω, ἅ µοί ποτ᾽ ἔδωκας, ἐγὼ δ᾽ ᾔτεόν σε ἕκαστα παιδνὸς ἐών, κατὰ κῆπον ἐπισπόµενος: διὰ δ᾽ αὐτῶν ἱκνεύµεσθα, σὺ δ᾽ ὠνόµασας καὶ ἔειπες ἕκαστα. ὄγχνας µοι δῶκας τρισκαίδεκα καὶ δέκα µηλέας, συκέας τεσσαράκοντ᾽: ὄρχους δέ µοι ὧδ᾽ ὀνόµηνας δώσειν πεντήκοντα, διατρύγιος δὲ ἕκαστος ἤην: ἔνθα δ᾽ ἀνὰ σταφυλαὶ παντοῖαι ἔασιν— ὁππότε δὴ Διὸς ὧραι ἐπιβρίσειαν ὕπερθεν. “Come, I will recall the trees in the well-built orchard That you gave to me, when as a boy I asked about each And every one, as you tended to the garden. As we passed among them As we reached each one you name and described it, And you gave me thirteen pear trees, ten apple trees, And forty fig trees; in this way you also pointed out the rows of vines That you were going to give me, fifty in all, each of them bearing fruit, one after the other. All kinds of grape clusters were on them, Whenever the seasons of Zeus on high weighed them down.” XXIV.336-344 The similarities between the two are striking. In both cases, the garden/orchard (referred to at various points as an alōē, kēpos, or orchatos) consists of three types of crop: fruit trees, vines, and vegetables. Of 126 the fruit trees, there are several types (fig, pear, apple, olive) that are common to both, with only the pomegranate (rhoia) being unique to the orchard of Alcinous. Both have vineyards as well that produce grapes year-round. Finally, they both have garden beds (prasiē) in which “all sorts” of vegetables are grown. Given the degree of similarity, we may safely assume that we are dealing with a type, or perhaps an ideal, of the hortus of an aristocratic household. The same kinds of crops are grown, and they are tended by numerous others, and it is clear that slaves formed at least some part of the labor force, since we are told that the house of Laertes was ringed by a shelter in which the slaves (dmōes) who worked under compulsion (anankē) would eat and sleep. As for the orchard of Alcinous, we cannot tell precisely who is carrying out the labor, only that there are several who work there, as evidenced by the use of the third- person plural, indicative active verbs trygoōsin (“they gather”) and trapeousi (“they tread grapes”). In any case, the picture that emerges from the representations of the households of both Alcinous and of Odysseus is one in which horticulture, not agriculture, is the use to which land is put when controlled by elite households. Fruits and vegetables, grown in a garden or orchard, are the products of cultivation in the elite household which are emphasized in the text, whereas mentions of cereals, grown on a farm, are almost completely absent. 103 In fact, the only other product of the aristocratic household that is featured prominently is livestock. Although not mentioned in connection with the household of Alcinous, herding is a distinct feature of the household of Odysseus. In addition, livestock appears to be an analogue to wealth and status. For example, when recounting the wealth of Odysseus to the disguised head of the household himself, Eumaeus describes its holdings solely in terms of livestock (XIV.96-104). 103 As a point of clarification, I do not mean to suggest that agriculture itself was absent from elite households. In fact, numerous mentions in the text (see n. 100 above) make clear that land specifically for agriculture was an integral element of the temenos. What is largely absent, however, are direct mentions of agricultural activity, whether in connection with elite households or independent of them. 127 Finally, a third type of production appears in the text at several points, namely the processing of foodstuffs by the female slaves of the elite household. Two examples illustrate this activity. First, in the extended description of the house and holdings of Alcinous in Book VII, we are told that there are fifty female slaves (dmōai), some grinding grain (karpon), some weaving, and others pressing olives to produce olive oil (elaion). In a similar instance, we find twelve women engaged in grinding wheat flour (alphiton) and barley flour (aleiata) in the house of Odysseus (XX.105-109). Based on these passages, it would seem that the elite household also functioned on occasion as a sort of processing center for various goods, including wool, olives, grapes, and cereals. As we have seen, the orchard/garden of the elite household typically includes vines and olive trees, and we have already noted the strong aristocratic association with the ownership and pasturage of livestock. Thus, for these goods the elite household functions as a vertically integrated firm, within which the basic factors of production (land, labor, capital) are combined to produce raw goods (olives, livestock, grapes), and then these in turn are processed in order to produce finished goods (olive oil, textiles, flour). At the same time, it also serves as a processing center for cereals that are not produced within the confines of the firm/household. 104 So, it seems evident that in the society presented in the Odyssey, the aristocratic oikos functions as something much more than the simple, self-sufficient unit proposed by Finley (1999 [1973]), and that even by the late 8 th c. BCE, archaic Greek economic organization was more complex than the oikenwirtschaft described by Weber (2013 [1922]). In addition, it is likewise clear that the household depended on outside sources for many items that it did not produce internally, including raw goods (metals, timber, etc.) as well as a whole range of finished goods. We must therefore presume the existence 104 This is not to suggest that such food processing activities could not have occurred independently of elite households as well. In fact, given that all indications are that agricultural production took place both within the temenos and elsewhere, it seems logical to presume that the processing of foodstuffs would have followed a similar pattern. 128 of a level of economic integration in the broader society that was necessary for the elite household, at least as described in the Odyssey, to function. To this we must also add craft production, which is clearly an extra-household form of production. We are specifically told that “those who work for the people” (dēmiourgoi) such as a prophet, healer, singer, or skilled workman (tektōn dourōn), typically come from abroad (allothen), and that their fame extends across the boundless earth. These craftsmen must include carpenters, shipwrights, and blacksmiths, at a bare minimum. We might also consider that those involved in the mining and smelting of metals, as well as those who produce timber (who are mentioned frequently in many of the similes in the Iliad), should belong to this category of extra-household production. We might even include bakers, since we know that bread was such an integral part of the diet. In any case, there is little doubt that the Odyssey presents an image of production that a) is centered on the activities of the aristocratic oikos, and b) strongly correlates certain specific productive activities with the elite household. As a result, a vast range of productive activities are either denigrated or marginalized, if they are even mentioned at all. We can find mention of fields of grain, descriptions of the techniques involved in the cultivation of cereals, the ubiquitous presence of bread at meals, and even the processing of grains to make flour. However, there are strikingly few descriptions of farming taking place, or of any individual engaged in farming. Thus, the representation of production in the Odyssey appears to be more about the elevation of a particular lifestyle available only to a select few. Although some like Hanson (1995) have argued that arboriculture and viticulture in Homer are evidence for a new agricultural regimen instituted by ordinary farmers for sale into the emerging market, this is anything but the picture presented by the Odyssey. Instead, the image is much more consistent with the arguments of Donlan (1997), who contends that orchards are as much for display as utility, constituting a highly visible marker of wealth, high rank, and distinctive lifestyle. In addition, he also argues that large-scale stock raising (especially of cattle and horses) is a class divider in Homeric society (Donlan, 1989d), and that the Homeric epics distort the economy by foregrounding the huge ranching operations of the basileus and 129 other wealthy elites, and pushing into the background the agricultural economy and the small and middling farmer (Donlan, 1997). It can only be safely concluded that the image of production in the Odyssey is an ideologically driven representation designed to valorize certain activities that served to maintain and reinforce the elevated social and economic standing of the aristocracy, while minimizing those other activities that did not serve such ends. In other words, the epics are engaged in a program to establish norms of behavior (i.e., institutions) that promote certain activities and repress others. Consumption It is on this topic that the epics are largely silent. There are numerous descriptions of productive activities, while distribution dominates the representation of economic activity. However, consumption, and particular consumer choice, are almost never discussed. In fact, aside from foodstuffs, it would appear as if virtually nothing is ever actually consumed or acquired for its own sake. Women in elite households (and their slaves) are constantly engaged in weaving, but clothing is almost invariably represented as something to be stored away or used as a medium for gift-exchange. There are mentions of craftsmen, but their products and services almost never appear to be used, and certainly not outside elite households. In fact, the picture that is created is of a world in which the oikos of a basileus is a center of production and distribution, but where durable goods are stored, not used. The most notable example of consumption, namely the suitors devouring the flocks of Odysseus, is represented as a violation of the highest order. It seems that in this universe, goods are meant to be produced and exchanged, but almost never actually consumed/purchased. In fact, there is almost no discussion of any kinds of goods, either raw materials (wood, metals, wool, etc.) that would be used as factors of production, or of finished goods (pottery, tools, etc.) that would be commonly consumed by non-elites. When metals are mentioned, they are represented as a means of 130 storing wealth, 105 a medium of exchange, 106 or as a prestige item involved in gift-exchange, 107 but not as a factor of production. 108 The Odyssey in particular represents material goods almost exclusively as items to be owned in elite households, but never put to any practical use. For example, Athena counsels Telemachus that he must not allow the suitors to divide up and consume all his property (ktēmata), and while the suitors are attempting to marry his mother, he must be careful to ensure that no property (ktēma) leaves the house without his consent (XV.12-19). Penelope herself recognizes that Telemachus is worried that his property (ktēsis) is being consumed by the suitors (XIX.533-534), and Telemachus himself confesses to Menelaus that must hurry home, since he has no one to look after his possessions (ktear) and stored wealth (keimēlion) and is fearful that he might lose it (XV.32). Distribution It is on this aspect of economic activity that the epics have arguably the most to say. In particular, the Odyssey, with its emphasis on overseas travel, speaks at length about mobility, which is of course an essential component of shipping/maritime trade. This aspect of commercial activity, as I have stated from the outset, is a central tenet of my argument. For that reason, I will make it a central focus in this section devoted to an examination of distributional processes. In addition, there are other passages in the epics that describe the presence of other mechanisms of exchange, including markets and what I argue are nascent forms of proto-money, all of which seem to indicate increasing levels of connectivity and economic complexity. Admittedly, there is no way to establish a quantitative basis for the degree to which maritime transport formed an important aspect of life in the archaic Greek world, either on the basis of textual or material evidence. What can be construed, however, is a general sense from the text as to how 105 Ex., II.337-340, IV.72.73. 106 Ex., 7.473. 107 Ex., IV.128-129, V.38, IX.201-204, XXIV.273-279. 108 There is the lump of iron that Achilles offers as a prize at the funeral games of Patroclus (23.826-835) and which is clearly intended to be used as raw material. However, this is an isolated instance, and is in my view the exception that proves the rule. 131 seaborne transport was conceived. In other words, does it appear to be rare or commonplace, safe or dangerous, seasonal or perennial? Also, what seems to be the typical motive for overseas journeys (i.e., travel, trade, etc.)? On these points, the textual evidence is uniquely qualified to offer insights into the ways in which such activity was regarded and represented, even if it is limited in other ways. Maritime Activity I begin with the question of the prevalence of maritime transport during the early archaic period. As I have argued already, the relative lack of frequency with which something is mentioned, or perhaps if a phenomenon is described only sparingly, is often a measure of the fact that it is so commonplace as to be considered obvious, and therefore undeserving of mention or lengthy explanation. 109 In many ways, this is a defining feature of maritime transport in the Odyssey. At numerous points, trade and travel are mentioned only in passing, as part of a larger discussion. I will deal with many of these in turn, but for now, let us examine first some of the ways in which it is overtly represented. For one, the capability to move overseas seems to be a mark of distinction and a source of pride. For instance there is the example of the Phaeacians, whom Athena describes to Odysseus as confident in the speed of their ships and able to cross the deep water (laitma mega) with ships that move swiftly as thought or winged creatures (VII.34-36). This seafaring skill is clearly a distinguishing feature of the Phaeacians, since they are referred to at various points as “oar-loving” (philērhetmos) 110 and “famed for ships” (nausiklytos). 111 In fact, it seems to be such a source of pride that Alcinous boasts to Odysseus that the Phaeacian ships and rowers are the best (VII.327-328). Also, this mark of distinction is not limited to the Phaeacians. For example, the Taphians are also referred to as “oar-loving” (philērhetmos). 112 In fact, it 109 See also Redfield (1975), Raaflaub (1997, 1998a, 1998b), and I. Morris (1986), among others, who all express the same general notion. 110 Ex., V.387, XI.349. 111 Ex., VII.39, XVI.227. 112 Ex., I.181, I.419. 132 would seem that the presence of ships is an almost essential feature of any settlement by the sea. For example, as Odysseus observes the city of the Phaeacians, he admires, among other things, their harbors and ships: θαύµαζεν δ᾽ Ὀδυσεὺς λιµένας καὶ νῆας ἐίσας αὐτῶν θ᾽ ἡρώων ἀγορὰς καὶ τείχεα µακρὰ ὑψηλά, σκολόπεσσιν ἀρηρότα, θαῦµα ἰδέσθαι. “Odysseus was equally amazed both by their harbors and ships, As well as the agoras of the heroes and broad, lofty Walls, fitted with palisades, wonders to behold.” VII.43-45 Lest we regard this feature of the Phaeacians’ polis as unique, we should regard Odysseus’ admiring description of Telepylos, the citadel of the Lastrygones: ‘ἔνθ᾽ ἐπεὶ ἐς λιµένα κλυτὸν ἤλθοµεν, ὃν πέρι πέτρη ἠλίβατος τετύχηκε διαµπερὲς ἀµφοτέρωθεν, ἀκταὶ δὲ προβλῆτες ἐναντίαι ἀλλήλῃσιν ἐν στόµατι προύχουσιν, ἀραιὴ δ᾽ εἴσοδός ἐστιν, ἔνθ᾽ οἵ γ᾽ εἴσω πάντες ἔχον νέας ἀµφιελίσσας.’ “’Then we reached the excellent harbor, which is ringed By sheer cliffs all around on both sides, With lofty headlands facing each other, And they protect the mouth, and the entrance is narrow, And they all keep their oared ships inside.’” X.87-91 In fact, the presence of a well-protected harbor seems to be virtually indispensable, even when it is not exploited. For instance, in describing the island near the land of the Cyclopes, Odysseus remarks on all of the natural advantages of the territory, including meadows where grapes could be grown, smooth land with rich soil for plowing, a fresh water spring, and a well-protected harbor (IX.131-141). Therefore, at least on the basis of this evidence, it seems necessary to regard seafaring in the early archaic period as something that was often valued highly and that was a prominent feature of life in the Mediterranean even by the late 8 th c./7 th c. BCE. 133 Taking a different perspective, Helms (1988) argues that the traditional anthropological approach to long- distance travel has been dominated by considerations of trade and traders, and in so doing has ignored the importance of what she calls the “political-ideological” aspects of long-distance contacts in many pre- industrial societies. For Helms, the knowledge of distant lands and peoples that is associated with long- distance travel also frequently holds a prestige value akin to Bourdieu’s cultural capital, and thus has the potential to serve as a means of distinction. 113 If we accept this premise, then the expansion of maritime networks in the Archaic period takes on additional significance and constitutes an even greater threat to elite primacy on several levels. For one, increased long-distance travel for the purpose of trade represents a subversion of “appropriate” travel for the sake of establishing contacts and inter-elite relationships. Second, as non-elite traders participate more actively in long-distance travel, they encroach upon elite prerogatives, and in so doing either compromise the prestigious quality of the activity, or garner some of that prestige themselves (or both). In either case, their activity poses a challenge to elite primacy above and beyond the commercial sphere. Navigation Navigation in the archaic period would have of course taken place without the aid of any equipment. However, the Odyssey makes clear that this was accomplished by means of a detailed knowledge of the stars, including an understanding of the seasonal variations in the position of the various constellations. We find Odysseus completing part of his journey from the island of Ogygia: αὐτὰρ ὁ πηδαλίῳ ἰθύνετο τεχνηέντως ἥµενος, οὐδέ οἱ ὕπνος ἐπὶ βλεφάροισιν ἔπιπτεν Πληιάδας τ᾽ ἐσορῶντι καὶ ὀψὲ δύοντα Βοώτην Ἄρκτον θ᾽, ἣν καὶ ἄµαξαν ἐπίκλησιν καλέουσιν, ἥ τ᾽ αὐτοῦ στρέφεται καί τ᾽ Ὠρίωνα δοκεύει, οἴη δ᾽ ἄµµορός ἐστι λοετρῶν Ὠκεανοῖο: “[Odysseus] skillfully guided the vessel, seated 113 See also Malkin (1998) for the strong heroic connotations of sailing in the Homeric epics, especially the Odyssey. 134 At the steering oar, nor did sleep cover his eyes, But they were fixed on the Pleiades, and the late-setting Boötes, And Arctus, which they also call the ‘wagon,’ That turns in place and watches Orion, And it alone never dips beneath the Ocean…” V.270-275 First of all, this passage seems to indicate very little fear of sailing at night, contrary to the conventional image of ancient seafaring as occurring exclusively, or nearly so, during daylight hours. It seems evident that the stars by which Odysseus is navigating correspond to those described by Ptolemy in his Almagest. This would mean that “Arctus” refers to the star Arcturus in the constellation Boötes, the brightest star in the northern celestial hemisphere, visible in the late spring in the northern hemisphere (Schaaf, 2008). Navigation by such a star indicates a detailed knowledge of the night sky, including the changes according to the time of year. The journey Telemachus makes between Ithaca and Pylos also is undertaken at night (II.434), and we are told that several times, Odysseus sailed without stopping for several days at a time, once for six days and nights (X.80-81), again for nine days and nights (X.28), and again for seventeen days (V.278), reinforcing the notion that sailing at night was not assiduously avoided. To be sure, I think it is safe to say that sailing continuously for seventeen days is an exaggeration of proportion, common in the Homeric epics. Nevertheless, there is nothing in these passages to suggest that sailing at night was eschewed as a matter of course. If we accept this notion, it then also becomes much more probable that sailing over open ocean was also much more common than typically believed. The traditional view of ancient seafaring was that it rarely took sailors out of sight of land. Instead, the presumed practice was a “cabotage,” involving short trips that rarely ventured out of sight of the coast (Nieto & Santos, 2008). However, this type of sailing would have been exponentially more dangerous at night, given the impossibility of sighting reefs or other threats at or just below the surface that become more common the closer one is to the shore. The only type of route that would have been practicable at night would have been one that ventured out into the open 135 ocean. This is not to say that much, if not most, of navigation in the 8 th c. BCE did tend to hug the shoreline, but if we put any credence into the picture presented here of nighttime sailing, we have to accept that not all seafaring took place near the shore. In addition, the Odyssey describes several instances in which individuals demonstrate a strong grasp of the geography involved in completing a journey overseas. According to Morton (2001), an extensive and detailed knowledge of geographical conditions was an essential part of Greek success in seafaring. This is manifested in the text at several points. For example, Nestor tells Telemachus that when he and Menelaus set out from Lesbos on their return home, they debated over the best route to take, whether to sail between Chios and Psyros, or to pass by Chios next to Mimas (III.169-175). As part of the fleet was driven towards Crete, Nestor also describes coastal conditions of the island, such as the existence of a cliff near Gortys on the way to Phaistos with a nearby rock that acts as a breakwater (III.291-296). Of course, local conditions are well-known, as evidenced by the proposal of Antinous to ambush Telemachus on his way back from Pylos. He suggests that the suitors attack Telemachus on the straits between Ithaca and Samos (IV.670-671), hiding in the small island between Ithaca and Samos where there is a good anchorage and ships can be hidden easily (IV.844-846). Based on the evidence from the text, navigation requires a substantial body of knowledge, including a thorough and detailed grasp of the conditions, physical hazards, sea conditions, winds and availability of shelter, the available routes (as well as what to avoid) and the markers by which one identifies the location, and other environmental factors in sailing routes, sailing directions, and navigational techniques. 114 Also, this cannot be limited solely to the particulars involved in a given route from the 114 This is a major point of emphasis for Morton (2001), and has also been extensively covered by Arnaud (2005, 2014a, 2014b), both of whom underscore both the impact of numerous varied environmental conditions on the practice of seafaring, as well as the high level of technical ability of ancient seafarers. This point is further underscored by Cuisenier (2001), who chronicles the encyclopedic knowledge of local conditions among mariners in the Aeolian Islands as well as their ability to predict changes based on detailed observations of winds, currents, and 136 point of departure to the destination. In light of the apparent frequency with which vessels were blown off course, navigation demands knowledge of a much broader expanse that would allow sailors when driven off course both to identify their present location, and to determine an appropriate revised route to their destination. In addition, if nighttime sailing was, if not common, then at least not extraordinary, this would have called for a strong familiarity with astronomical conditions as well. All told, as Taub (2011) points out, sailing even at this time entailed a high level of expertise and familiarity with detailed conditions across a broad geographic area that would have been acquired and accumulated over many generations. Sailing Although this may seem to be essentially the same topic as navigation, which we have just discussed, it involves a very different type of knowledge and skill. I should point out here that by “sailing,” I mean to say those activities that are directly involved in the physical operation of the vessel. There is ample evidence to indicate that there is considerable skill and training involved in these operations. As Arnaud (2005) has argued forcefully, the level of expertise involved is often grossly underappreciated. At the same time, the passages in the Odyssey that describe the actual processes are remarkably casual and offhand, suggesting that the associated skills were so common that they deserved no more than a passing mention, and yet the ability for numbers of men to row in unison, or to reef and furl sails (especially when sailing without a following wind), involves extensive practice and teamwork. Arnaud (2011) even suggests that while sailing with a following wind was ideal, it was not uncommon for ancient mariners to volcanic activity. He even finds numerous specific details in the Odyssey that are surprisingly accurate representations of some of the features of the islands, and concludes that ancient seafarers relied on much the same kinds of local knowledge as that of modern fishermen observed today in order to navigate, predict changes in weather, and make decisions regard when to sail (or not), seek shelter, etc. See also Beresford (2013), who argues that Greek seafaring occurred much more frequently throughout the year, as opposed to the conventional view that sailing was limited predominantly to certain times of the year when the weather was more optimal. 137 sail 90° from the wind, an important but highly technical skill that required both sufficiently sophisticated naval architecture as well as a trained crew. In the Odyssey, it seems that sailing often involved both rowing and sailing solely under wind power. In some cases, ships are both rowed and sailed, depending upon the circumstances. For example, rowing is the most common method by which ships are taken from the harbor out to sea. 115 This only makes sense for larger vessels, which could more easily exit and enter harbors or navigate tighter quarters under oar power. At the same time, it would appear that some vessels operated entirely without oars. The description of the journey of Telemachus from Ithaca to Pylos, for instance, mentions that in setting out from the harbor, his crew raises the mast pole and furls the sails from the very beginning, never mentioning the use of oars (II.422-428). On the return journey, the situation is the same, with the crew setting the mast and furling the sails as they set out, again seemingly without rowing (XV.287-291). This is not surprising for smaller vessels, which are by nature more easily controlled and thus would not require the use of oars. Likewise, the schedia of Odysseus makes use of sails alone, and is guided only by a pēdalion, a rudder or steering oar (V.255). Of course, this does not discount the complicated nature of sailing a larger vessel, such as those capable of accommodating crews of more than one hundred men that are mentioned throughout the famous “Catalogue of Ships” in Book 2 of the Iliad. To operate these ships would have necessitated a highly trained crew, well-versed in various complex operations, especially if we are to believe that it was not unusual to sail without a following wind. At the same time, the fact that the text largely takes such skills seeming for granted would suggest that they were commonplace. Ultimately, the image that is produced by an examination of the text is one of widespread technical skills that were necessary to operate comparatively large vessels. 115 Odysseus and his crew row out to sea and then set the sails and let the steersman control the ship under sail (XII.144-152); the Phaeacians row the ship out to sea, then sail with the wind (XII.76-78). 138 Maritime Networks In both ancient history and classical archaeology, interest in networks has been growing in recent years, 116 doubtless spurred in part by the development of network analysis. It is clear that a growing awareness of the ways in which networks develop and function has the capability of adding to our understanding of the ancient world. Although the Homeric epics probably do not offer the kind of information that would lend itself to the more traditional forms of network analysis, there are nevertheless indications that maritime networks, even by the late 8 th c. BCE, were more prevalent and fully developed than has typically been believed. 117 First of all, the Odyssey indicates the presence of local transportation networks. For example, when reciting the holdings of the household of Odysseus, Eumaeus includes twelve herds of cattle, twelve flocks of sheep, twelve troops of pigs, and twelve flocks of goats on the mainland along with herdsmen and pasture for them (XIV.100-102). The fact that Odysseus controls herds, herdsmen, and territory on the mainland strongly suggests the existence of a transportation network capable of maintaining connections between Ithaca and the holdings on the mainland. This would seem to be confirmed by the appearance of the cattle herdsman Philoetius, who has brought cattle for the suitors from the mainland, and we are told that they were conveyed by a ferryman (porthmeus), and that such men ferry whoever comes to them (XX.187-188). Thus, it seems safe to posit the permanent presence of a local network connecting the island to the mainland. Long-distance networks also would appear to be attested in the text. For one, when Athena appears to Telemachus in Book I, she comes in the guise of Mentes the Taphian, who is on his way to Cyprus to 116 Ex., Capdetrey & Zurbach (2012), Mobilités grecques : Mouvements, réseaux, contacts en Méditerranée, de l'époque archaïque à l'époque hellénistique; Malkin (2011), A Small Greek World: Networks in the Ancient Mediterranean, Knappett (2011), Network Analysis in Archaeology: New Approaches to Regional Interaction; Collar (2013), Religious Networks in the Roman Empire: The Spread of New Ideas. 117 As Arnaud (2014a) argues, the Greek port of Massilia constituted an important hub for maritime commercial activity and a central node for maritime networks at least as early as the 7 th c. BCE. 139 trade iron for bronze (I.182-184). For Tandy (1997), this suggests an ongoing trading relationship with the east that would have been necessary to foster the distribution of metals as described in this passage. If true, such a network would map very conveniently onto the apparent distribution patterns of metals suggested by the Bronze Age cargoes of the Cape Gelidonya (Bass, 2005) and Uluburun (Pulak, 2001) shipwrecks. 118 Additionally, the existence of east-west trading networks connecting Ithaca to the west is indicated by the suggestion of one of the suitors that they put the troublesome guests (including Odysseus) in a boat and take them to Sicily where they would fetch a good price (XX.382-383). Of course, we know from the patterns of colonization discussed earlier that such connections were in place by this time, and also that the locations of subsequent colonial settlements can only reasonably be explained by a desire to protect these existing routes (Boardman, 1999). Thus, it seems that even by this time, much of the Mediterranean was connected by overlapping transportation networks that integrated long-distance routes within a system of local and regional ones. This is a vital point, and one that we will explore further in our examination of the archaeological evidence. The Rules of the Game To this point, I have tried to point out the ways in which the epics illustrate a wide variety of economic activities, especially in the means by which goods are distributed, but also in some of the ways in which production is organized and carried out. In so doing, it has already become clear that the Homeric epics over-represent certain forms of economic activity. For example, the productive activities that are clearly associated with elite status (horticulture, herding) are prominently featured in the texts, especially the 118 To be clear, both the Cape Gelidonya Wreck (late 13 th c. BCE; see Bass 1967, 1973) and the Uluburun (late 14 th c. BCE; see Pulak [1988, 1998, 2002]; Manning, et al., [2009]) are much earlier than the period in question, so we must be careful about drawing too close a connection between these Bronze Age wrecks and maritime networks of the Archaic Period. However, the cargoes of the two wrecks both vividly attest to a flourishing trade in metals already by this time, and the Phoenician expansion into the western Mediterranean was also clearly motivated, at least in part, by a desire for access to metals. 140 Odyssey, whereas cereal production, despite substantial evidence of its importance to Greek society, figures only minimally. However, it has been a major point of my thesis that these specific manifestations of economic activity can only be understood within a context of rules that govern such behavior. Therefore, I wish now to examine various questions in reference to these institutions. Of course, first and foremost we need to identify what those rules appear to be, but above and beyond that are several important related questions, including whether or not the rules appear to favor certain groups, and if so, which ones. Also, what are some of the strategies (i.e., compliance, manipulation, etc.) that have been developed in response to the existing institutional matrix? And perhaps most importantly, I wish to examine the process by which institutions appear to have been involved in shifting power relationships, either as an impetus of change or as an effect of it. Gift-Giving To be sure, this category could easily have been incorporated into the section on distribution. In fact, that is probably where it should belong according to a more conventional interpretation of pre-modern economic activity. It is, after all, clearly a form of reciprocity, one of Polanyi’s (1944, 1957, 1963) three principal modes of exchange. 119 However, it is clear from the descriptions of elite gift-giving in the texts that the emphasis is far more on its potent symbolism than as a mechanical process for circulating goods. 120 In fact, one might argue that the defining feature of Homeric gift-giving is the degree to which it 119 The most famous work on this type of exchange continues to be the model of “systèmes totales de prestation” developed by Mauss (2002 [1923-1924]), in which certain types of prestige items are exchanged with carefully chosen trading partners in a ritualized form of culturally embedded exchange that establishes, reifies, and validates the interpersonal relations between the participants in the transaction. Most importantly, the symbolic significance of the thing or things exchanged is of primary importance, of far greater significance than the material object itself. See also Sahlins (1972) on his distinctions between the various forms of reciprocity. 120 This is not to say that ritual gift-exchange did not have a distributive function, but such a function would have necessarily been limited, first because only a narrow range of goods appears to have been appropriate for gift-giving, and second because the exchange partners were restricted to a narrow stratum of Greek society. In fact, this 141 follows a rigid, established script according to which both giver and recipient are ascribed highly elaborated roles. We should nevertheless be cautious about drawing too fine a distinction between goods transferred by the mechanism of gift-exchange versus those transferred via market exchange. There was doubtless some degree of overlap, and as Appadurai (1986) points out, the same commodity can be imbued with varying significance and symbolism depending on the context of the exchange, a concept that he describes as “regimes of value.” We should expect some degree of permeability between gift- and market-exchange, and in fact this helps to explain the persistent emphasis in the epics, which was likely at least partly motivated by a desire to minimize this ambiguity. It would be pointless to attempt an exhaustive review of this topic, so let me just make some essential points here, above and beyond the obvious conclusion that gift-giving in Homer takes place within a complex system of rules. First, gift-giving in Homer is clearly part of a much more extensive system of rules governing intra-elite relationships, 121 and it is played out in various arenas, including hospitality rituals, 122 feasting, 123 and more. Second, it is also clear that only a certain set of items was appropriate for ritualized gift-giving, including metals (incl. bronze and gold), tripods, horses, unique and precious items (such as gold and silver vessels), livestock, weapons, armor, and slaves. 124 There can be little doubt that limitation is one of the centrally important aspects of ritual gift-exchange since it functioned as a means of exclusion for archaic elites, allowing them to foster intra-elite relationships while prohibiting access to non-elites. 121 The elite-centrism of Homeric gift-giving is made more than obvious simply by the fact that the exchange partners are invariably members of the aristocracy (ex., Menelaus and Telemachus IV.590-592, Odysseus and the Phaeacian basileis V.38, Glaucus and Diomedes 6.216-220, Hector and Aias 7.299-305, etc.). 122 The definitive text that deals with both the extensive system of rules in place governing hospitality scenes in Homer and the Homeric Hymns, and the intelligibility of the social script to the contemporary audience, is Reece (1992). 123 The anthropological literature on the status implications of feasting is far too extensive to attempt an even cursory review here. Sahlins (1972) remains a seminal work, Weissner & Schiefenhovel (1997) and Dietler & Hayden (2010) are excellent surveys, Bray (2003) offers an archaeological perspective, and Rosman and Rubel (1971) is a notable ethnographic study of the phenomenon in a specific society (the Pacific Northwest Coast peoples). 124 See the exchange between Glaucus and Diomedes (6.216-220), the promise of Agamemnon to Teucer (8.287- 291), the exchange between Hector and Aias (7.299-305), and the gifts that Achilles gives to Eumelus (23.548-552); also the gifts presented to Menelaus by Polybus (IV.128-129), the gifts presented by Alcandre, the wife of Polybus, to Helen (IV.130-132), and those which Menelaus gives to Telemachus (XV.115-119). The correlation of particular types of goods with specific forms of personal relationships between exchange partners has been well-chronicled 142 the participants in these exchanges were keenly aware of the implications of these exchanges, not only on an ad hoc basis, but were even conscious of the genealogy of certain prominent gift items. 125 All of these factors combined to create a complex structure of rules and meaning, an institutional matrix against which such well-informed actors were capable of framing their actions in ways that provided enhanced meaning to their exchanges. As a result, skilled and knowledgeable actors could adopt diverse strategies designed to leverage the system to greater effect. The potential for exploitation and manipulation of this system can be seen most vividly in Agamemnon’s efforts to utilize the institution to his personal advantage, with varying degrees of success. Agamemnon presses his gifts upon Achilles (19.140-144), who in turn rebuffs him (19.146-150). Agamemnon, undeterred, urges him a second time to accept the gifts (19.188-195), and is rejected yet again (19.199- 202). It is only after yet a third urging, this time from Odysseus, that Achilles relents and the gifts are brought out (19.243-248). The desperate insistence of Agamemnon to press his gifts upon Achilles, and the latter’s persistent rejection of those gifts, only makes sense in the light of their recognition of the significance of the exchange. However, Achilles’ persistent rejection of the offer only highlights Agamemnon’s desperation, and in fact serves to invert the power relationship consummated by the transaction, producing precisely the reverse outcome of what Agamemnon hoped, and further illuminates the potential for polyvalent meanings within the system of ritual exchange. We know that Homeric elites were more than willing to exploit the existing institutional environment for their own personal advantage, as when Menelaus suggests that he and Telemachus make a tour of local poleis specifically in order to elsewhere, as with the “exchange spheres” observed among the Tiv (Bohannon, 1955, 1959, 1963) and Kabre (Piot, 1991) in West Africa. Also of note is the way in which gifts between elite females mirror the gifts between elite males (ex., the exchange between Menelaus:Polybus = Helen:Alcandre), which lends even greater weight to the assertion that the essential element of Homeric ritual gift-giving is the elite standing of the exchange partners; for the role of elite women in the hospitality ritual, see Pedrick (1988). 125 Many of these items could acquire complex life histories related to the prominent men through whose hands they passed, such as the armor of Areïthoös (7.137-142), Meriones’ helmet (10.261-271), and Agamemnon’s corselet (11.19-28). Such layering of prestige over time is a frequent feature of ritualized gift-giving, most famously observed in the kula exchange of the Trobriand Islanders (Malinowski, 1984 [1922]). 143 collect guest-gifts (XV.79-85), or when Odysseus reveals his plan to recoup his losses suffered by the predations of the suitors by replenishing his flocks with gifts from other Achaeans (XXIII.355-360). 126 It is important to note, however, that the rules of the game are not always and inevitably prone to usurpation or made to serve individual self-interest. In fact, they often serve to support precisely the kinds of outcomes for which they are ostensibly intended. There are numerous instances in the texts in which gifts are exchanged in ways that are not nearly so self-interested, and that appear to forge reciprocal bonds between the exchange partners. Thus, it is a mistake to presume, like Donlan (1993), that the Homeric system of gift-exchange is unavoidably competitive, simply by the fact that it can be used in such a way, or that all such systems are innately agonistic. 127 In the Iliad and the Odyssey, we find a system that ostensibly serves to forge and maintain reciprocal relationships among the elite, but can be made to serve the specific aims of individual actors who are well-versed in the nuances of the social system. In fact, for Rose (1992), the manifold ways in which the process of gift-giving is invested with various, even contradictory symbolism (i.e., as a means of competition versus a means of forging and maintaining elite relationships) is a symptom of an emerging split between achieved and ascribed bases of social standing. As I have already argued, it is one of the inherent features of institutions that they produce the framework within which culturally fluent actors are afforded a substantial degree of room to negotiate. The rules create a set of stable expectations, but “actors do not behave or decide as atoms outside a social context, nor do they adhere slavishly to a script written for them by the particular intersection of social categories that they happen to occupy. Their attempts at purposive action are instead embedded in concrete, ongoing 126 For additional examples of the ways in which the system can be open to manipulation, see Rose (1992) for some of the ways that power relationships are negotiated and validated through the division of war spoils. 127 There are exempla in the ethnographic literature that document the dynamic of gift-giving as a forum for competition, as among the Siuai of New Guinea (Oliver, 1955), the Yanomami of South America (Chagnon, 2009) or the famous potlatch of the Northwest Coast Peoples (Rosman and Rubel, 1971). On the other hand, there is abundant evidence for societies in which gift-giving generally appears to fulfill the objective of reinforcing social bonds, such as the hxaro of the !Kung people of the Kalahari (Lee, 1979, 2003) and the kula exchange of the Trobriand Islanders (Malinowski, 1984 [1922]). 144 systems of social relations…” (Granovetter, 1985). Seen from this perspective, the apparent inability of Homeric elites to agree on a mutually intelligible meaning associated with the practice of gift-giving would seem to suggest conditions in which traditional social institutions are in flux, to the extent that symbolic acts no longer have the same meaning and cultural actors are engaged in a process of re- negotiation in order to establish a new semiotic script in their wake. Of additional significance is that ritual gift-exchange created a closed structure, reserved only for the elites of Greek society. To participate in such a system required a minimum property qualification that by its nature would have excluded a substantial portion of society. It is no accident that many of the items that carry elite associations, like metal objects, precious metals, and horses, all require substantial resources, and a certain minimum level of disposable income in order to be able to participate in the ritual. Although there are indications that genealogy was a component of elite status, it does not appear that anything like formal titles were heritable in Homeric society – but property was. Given that a minimum level of wealth was a necessary condition for participation in the elite sphere, and that it could be exploited in order to enhance social standing, wealth enabled status to be maintained along blood lines. In addition, as Donlan (1993, 1994) points out, the reciprocal nature of most elite gift-exchanges had the effect of ensuring that such transactions took place between individuals of roughly the same economic standing. At the same time, the system of ritualized gift-giving provided a cover for the conversion of economic capital into cultural capital. The circulation of prestige items that were by their nature accessible only to a limited minority, occurring within a closed sphere of exchange, allowed for comparative advantages in wealth to be used toward the pursuit of status, but for this process to be obscured. As Bourdieu (1986) explains, when wealth is a basis for social standing, it becomes increasingly difficult to maintain the position that social standing is a function of qualities inherent in the elites instead of the result of random chance. As a result, cultural capital serves as a “non-economic form of domination and hierarchy, as classes distinguish themselves through taste…” (Gaventa, 2003), and the 145 shift away from material capital to symbolic and cultural capital conceals the underlying causes of inequality. Thus, we see that this system of exchange is highly ideologically malleable, and in fact may be one of the areas in which we can most clearly see indications the first cracks in the social order in which the traditional rules are failing to keep pace with the changing conditions brought about by the early phases of expansion of the Greek world at this time. So, we must ask ourselves: why then does this form of exchange feature so prominently throughout both texts? I believe that the most reasonable explanation for the almost obsessive fixation on elite gift-exchange in the epics is the product of underlying anxiety. I have already made the argument that the kinds of changes visible in the archaeological record with the initial phases of westward colonial expansion would have almost inevitably placed strains on the simple low-level chiefdom political structure of Homeric society as depicted in the epics. Also, I have made the rather uncontroversial claim that for Homeric elites, control of resources was an element in maintaining the traditional social hierarchy. So, in the event of increasing socio-political instability, it is reasonable to expect that a text that favors elite prerogatives would support the perpetuation of values and norms that advocate forms of exchange favorable to the aristocracy. Thus, we find at a time of increasing strain on the social fabric, a strategy developing in the Iliad and the Odyssey whereby specific institutions are promoted in order to consolidate elite control over resources, which they could in turn leverage to fortify their status. Market Exchange If we accept the premise that by the later 8 th c. BCE the traditional social order was becoming more and more difficult to maintain as the Greek world expanded, then it is probably not surprising to see the aristocracy adopting a regressive strategy to resuscitate the established hierarchy by clinging to conventional strategies that had allowed them to consolidate their position in the first place. We might very well interpret the constant the continuous attention on gift-exchange in the Homeric epics in the light of a general sense of unease as Homeric elites perceived their place of privilege was no longer as secure 146 as it once was. However, I believe that this anxiety has a more specific source, and that the repeated on the elite-specific exchange makes much more sense if we keep in mind the seemingly obvious reality that ritualized gift-exchange, by its very nature, excludes all but a narrow minority of a society. Because it is inherently exclusive, how can it be considered in any way normative or a dominant mode of exchange for the society as a whole? The mechanism of gift-exchange is by its nature limiting, and therefore the disproportionate level of attention paid to such transactions in Homer must necessarily over-represent the importance of this particular form of exchange. We must, therefore, presume the existence of other forms of exchange that are either not mentioned, or are under-represented in the text. In fact, once we begin to search more carefully for examples of other forms of exchange, we find subtle but unmistakable mentions in the texts to indicate the presence of non-elite forms of exchange. For example, trading is mentioned as one of the primary motives for sailing across long distances (III.71-74; IX.252-255). Also, Odysseus relates a story when he stayed in Egypt, growing wealthy, and then was engaged in a business partnership with an unnamed Phoenician trader, ultimately bringing a cargo to Libya (XIV.285-297). Unfortunately, the nature of the cargo is ambiguous, and there is no discussion whatsoever of the Libyan consumers or the transaction itself. Similarly, Eumaeus tells of Phoenician traders who came to Sicily selling “trifles” (athyrmata), but the identity of the exchange partners is unmentioned here as well (XV.415-416, 455-456). Still, even though the consumers are not described, the goods that are traded could provide some additional insight. As we have already seen, elite gift-giving is limited to a set of items that are appropriate for the culturally specific ritual of exchange. The description of the trade goods as “trifles” is yet one more way to delineate this type of transaction from the morally beneficial gift-exchange. Based on these examples we can safely posit the type of exchange between partners without a personal connection and for so whom the primary motive of the transaction was the item changing hands rather than the relationship of the parties to the transfer, what Polanyi (1944, 1957) would consider a “market” exchange. 147 However, the most vivid illustration of the co-existence of different modes of exchange occurs in the Iliad. According to the text, νῆες δ᾽ ἐκ Λήµνοιο παρέσταν οἶνον ἄγουσαι πολλαί, τὰς προέηκεν Ἰησονίδης Εὔνηος, τόν ῥ᾽ ἔτεχ᾽ Ὑψιπύλη ὑπ᾽ Ἰήσονι ποιµένι λαῶν. χωρὶς δ᾽ Ἀτρεΐδῃς Ἀγαµέµνονι καὶ Μενελάῳ δῶκεν Ἰησονίδης ἀγέµεν µέθυ χίλια µέτρα. ἔνθεν οἰνίζοντο κάρη κοµόωντες Ἀχαιοί, ἄλλοι µὲν χαλκῷ, ἄλλοι δ᾽ αἴθωνι σιδήρῳ, ἄλλοι δὲ ῥινοῖς, ἄλλοι δ᾽ αὐτῇσι βόεσσιν, ἄλλοι δ᾽ ἀνδραπόδεσσι: τίθεντο δὲ δαῖτα θάλειαν. 7.467-475 Many ships were there, bringing wine from Lemnos. Euneus, the son of Jason, sent them, He whom Hypsipyle bore to Jason, shepherd of the people. Specifically for Agamemnon and Menelaus, the sons of Atreus, The son of Jason sent a thousand measures as a gift. The long-haired Achaeans also bought wine from them, Some with bronze, and some with gleaming iron, Some with hides, and some with cattle, And others with slaves; and then they set up a great feast. Certainly, the portion of the cargo set aside especially for Agamemnon and Menelaus is a kind of gift between elites, consistent with the practice of gift-exchange that we have already seen. However, the remainder of the cargo is clearly exchanged on an ad hoc basis with exchange partners who do not appear to have any personal ties to the sellers, nor is there any indication that the transaction fulfills any objective other than the transfer of goods. 128 In fact, it seems that the text is very careful to draw a clear distinction between the gift-exchange on the one hand, and the market transactions on the other. 129 In this case at 128 See Sherratt & Sherratt (1991), who argue that regional trade in the Bronze Age, which has traditonally been characterized as emblematic of a command economy, exhibited substantial mercantile activity. According to their argument, inter-regional trade among the palace centers established connections and fostered the development of distribution networks that in turn created ample opportunities for much more explicit forms of market exchange. As they explain, “royal trade and individual enterprise…went hand in hand…” See also Pulak (2001, 2009) for an analysis of the Uluburun wreck as evidence for the trade in raw materials, especially metals, along with luxury items in the late 14 th c. BCE. 129 This episode is also a vivid illustration of Appadurai’s (1986) concept of “regimes of value,” where we can see the same commodity (wine) acquiring dramatically different symbolic significance depending on the exchange partners and the dynamic of the exchange. The work of Gregory (2015) has also been highly influential, particularly 148 least, it is the transaction that takes priority over social or cultural considerations, and is unmistakably carried out according to the logic of market exchange. For Tandy (1997), this is an isolated instance, an example of what he calls a “one-off” form of market exchange. However, there is abundant evidence to suggest much more regular trading activity, and to show that it is a mistake to regard the wine shipment from Lemnos as a unique event. For one, there is further mention of wine brought from abroad, as revealed when Nestor advises Agamemnon to hold a feast among the chiefs (9.70-72). In order to supply this feast, Nestor reminds him that their tents are full of wine that is brought by the Achaean ships across the sea from Thrace every day. Although the precise manner of this exchange is not mentioned, it seems extremely unlikely that it could be a gift. The quotidian nature of the supply is inconsistent with ritual gift-exchange. Second, except for the example above of wine sent by Euneus, there is only one other instance in either the Iliad or the Odyssey in which wine is a prestige item in a gift-exchange. 130 As we have already noted, there are specific types of items that are appropriate for such gifts (arms and armor, horses, slaves, precious metals, metal goods), and wine is not among them. Third, the fact that the Achaeans themselves send their own ships to Thrace in order to obtain the daily shipment of wine marks it out as something different. The repeated exchange with a single trading partner, or at least territory, is at odds with the practice of xeinia, in which the symbolism of the gift is enhanced by the fact that it occurs as a discrete, special occasion. In light of all these factors, the most reasonable conclusion is that the daily shipment of Thracian wine is being purchased, and therefore constitutes an example of what game theorists refer to as an “iterated”, as opposed to a “one-off,” market exchange, and just as importantly, we can clearly see such activity taking place side-by-side with elite exchanges. among anthropologists, in calling into question the traditional distinction between “gift” and “commodity,” arguing that the terms themselves reflect tacit assumptions about value, and that exchange items must be understood both within a broader social context and within the context of the specific exchange itself. 130 IX.196-207. 149 There is at least one more commodity which appears to be exchanged frequently, based on evidence in the text. There are numerous mentions in the Iliad that suggest an ongoing trade in slaves. For example, there is the story of Lycaon, whom Achilles had previous captured and sold in Lemnos. As Lycaon himself recounts, ἤµατι τῷ ὅτε µ᾽ εἷλες ἐϋκτιµένῃ ἐν ἀλωῇ, καί µ᾽ ἐπέρασσας ἄνευθεν ἄγων πατρός τε φίλων τε Λῆµνον ἐς ἠγαθέην, ἑκατόµβοιον δέ τοι ἦλφον. 21.77-79 On that day when you captured me in the well-built garden, And you shipped me off to sell me, taking me away from family and friends To sacred Lemnos, 131 and I brought you a price of a hundred oxen. The “pure” commercial nature of this transaction is suggested by the terminology. First, the verb peraō means “to carry beyond seas for the purpose of selling, to export for sale.” What is important is not whether or not this particular exchange is commercial, however, but more that the very existence of the word is clear evidence that already by this time, the practice of exporting commodities for the purpose of sale was sufficiently common to become part of the language. Nor is this an isolated instance. Achilles himself admits that he has made a practice of selling those Trojans whom he has captured, shipping them overseas in order to sell them. 132 In each case, the verb used is pernēmi 133 or one of its cognates (peraō 134 , pernaō 135 ), all of which indicate export for the purposes of sale. In only one of these instances are the captives ransomed (luō), 136 especially given that ransom would seem to vitiate the need to ship them abroad. In addition, the casual 132 21.100-102; see also 22.44-45, 24.751-753. 133 22.45, 24.752. 134 21.40, 21.58, 21.78, 21.102, 21.454. 135 22.44. 136 21.180. 150 mention of various locales to which captured slaves are shipped and sold strongly suggests that permanent, or at least frequent, markets existed in which captured Trojans could be sold. 137 It could be argued that the example of Lycaon is one of intra-elite exchange since he was ransomed, and also because of the involvement of Euneus, the son of Jason, but I do not believe that this is consistent with the evidence in the text, for several reasons. First, the transaction described by Lycaon is not described in such terms. Instead of using the term for ransom (luô), Lycaon specifically uses the word to describe export for the purpose of sale (peraô). Also, the text suggests that the transaction between Euneus (the purchaser) and Achilles (the seller) was commercial in nature. The term used to describe the amount paid for Lycaon is an ônos, 138 (“price paid for a thing”) and not one of the terms typically used to describe a ransom (lysis, apoina). Second, Lycaon explicitly states that Achilles took him far away from family and friends, which would naturally make it more difficult rather than easier to ransom him. There is nothing to suggest that Achilles has any expectation that Lycaon will be ransomed, in fact quite the reverse. For example, Priam bemoans the loss of so many of his sons that Achilles has either killed or sold in faraway islands (“ὅς µ᾽ υἱῶν πολλῶν τε καὶ ἐσθλῶν εὖνιν ἔθηκε/ κτείνων καὶ περνὰς νήσων ἔπι τηλεδαπάων…”). 139 Both the context and the use of the perfect tense (etheke) indicate that those of his sons whom Achilles captured and sold have not been ransomed back to him, suggesting that Lycaon is a unique case. Third, Lycaon is not the sole individual mentioned whom Achilles sells as a captive abroad. For instance, Achilles himself tells Lycaon that prior to Patroclus’ death, he took many captives and sold them abroad (“καὶ πολλοὺς ζωοὺς ἕλον ἠδ᾽ ἐπέρασσα”). 140 Finally, the connection with Euneus and the island of Lemnos 141 is not determinative. We are told by Priam in the passage just mentioned that many of 137 I use the term “market” here to designate a locale where goods are exchanged, rather than the set of conditions in which price is a determination of the supply curve. 138 21.41. 139 22.45. 140 21.102. 141 The mention of Euneus in connection with both the sale of captives and with the sale of wine to the Greeks, however, does present an intriguing possibility. I believe that these mentions in the text are indicative of some of the 151 his sons were sold in various faraway islands, not just in Lemnos. Also, Hecuba mourns the loss of her children, explaining: ἄλλους µὲν γὰρ παῖδας ἐµοὺς πόδας ὠκὺς Ἀχιλλεὺς πέρνασχ᾽ ὅν τιν᾽ ἕλεσκε πέρην ἁλὸς ἀτρυγέτοιο, ἐς Σάµον ἔς τ᾽ Ἴµβρον καὶ Λῆµνον ἀµιχθαλόεσσαν… Swift-footed Achilles sold my other sons, Selling whomever he captured across the barren sea In Samos, or Imbros, or harsh Lemnos… 24.751-753 While Lemnos was the location to which Lycaon was taken, it was clearly not the only location in which captives were sold, and there is nothing to indicate that there was any kind of elite connection in Samos, Imbros, or wherever else the captives were taken. Instead, everything points to a purely commercial enterprise, and that numerous locales served as marketplaces for this traffic. Once again, what is important is not whether or not such markets actually existed as they are described in the text, but if we return to Redfield’s (1975) test of plausibility, the significance is that such manifestations of markets must have existed and been familiar enough to the 8 th c. BCE audience in order to be plausible. Admittedly, it could be argued that the existence of such markets was influenced by an unusual set of circumstances, namely the decade-long siege of Troy. However, the passage nonetheless reflects a rational, efficient response to those conditions. Even if one could demonstrate that the appearance of slave markets was contingent upon the ongoing siege, the existence of such markets still ways that archaic elites could have engaged in profit-driven transactions while still maintaining the appearance of adherence to the prohibition against such mercantilism. In the case of Od. 8.159-164, the elite gift-giving between Euneus and the Atreidae provided a pretext for the more mundane “market” exchange of wine to the Greek forces. Meanwhile, war captives could have constituted a commodity that was considered exempt from the typical aversion to trade since they were so directly connected with warfare, which itself was associated strongly with the heroic code (see n. 144 below). This attempt to engage in maritime commerce might therefore indicate a realization on the part of the elites of the potential profits offered through seaborne trade, while the attempts to disguise it or legitimize it only when in connection with elite-centric activities indicate the recognition of the threat that the introduction of maritime trade on a widespread basis could pose to the traditional social hierarchy. 152 represents the ability to recognize commercial opportunities and the willingness to exploit them. As a result, the assertion of Tandy (1997) that the market exchange of wine mentioned in Book 7 is a one-off instance of market exchange, and the only such mention in the Iliad, does not seem to correspond to the evidence. Admittedly, it is possible to argue that the use of pernēmi and its cognates is actually indicative of a more limited meaning, specific to the act of selling someone into slavery. This is of course one of the meanings of the verb, and if used only in this way, it would make the connection to overseas commerce of a more general nature more tenuous. However, this is not the case, as Hector relates: πρὶν µὲν γὰρ Πριάµοιο πόλιν µέροπες ἄνθρωποι πάντες µυθέσκοντο πολύχρυσον πολύχαλκον: νῦν δὲ δὴ ἐξαπόλωλε δόµων κειµήλια καλά, πολλὰ δὲ δὴ Φρυγίην καὶ Μῃονίην ἐρατεινὴν κτήµατα περνάµεν᾽ ἵκει, ἐπεὶ µέγας ὠδύσατο Ζεύς. 18.288-292 In the past, mortal men would speak of the city of Priam As a place rich in gold and bronze; Now, however, the beautiful treasures of my house are laid to waste, And great riches have been sold off in Phrygia And lovely Maeonia, because great Zeus grew angry. Ultimately, it seems clear that for the audience of the poem, the notion of permanent or semi-permanent markets, as well as the overseas export of goods for the purpose of sale (not gift-exchange), were familiar concepts. We can see in the text that overseas commerce did in fact exist, and that it was not rare or unusual. However, just as we have seen that elite gift-giving was featured, even valorized, in the epics, the reverse is true of market exchange. This moral valuation is accomplished in several ways. First, profit-driven 153 commerce, and especially overseas trade, are associated with Phoenicans, 142 who are almost invariably described in negative terms. 143 Thus, by extension, maritime commerce acquires negative connotations by association. The central role of the Phoenicians in seafaring and colonization of the Mediterranean is well known (Bunnens, 1979; Winter, 1998; Aubet, 2002; Burkert, 2005). However, the early and active participation of many Greek cities in this expansion has also long been recognized (Dunbabin, 1948). As for the association between Phoenicians and maritime commerce within the context of dishonesty, I will say little other than to agree with the proposals of Winter (1998), who concludes that the Homeric epics deliberately distort the image of the Phoenicians, presenting a picture that is “flattened” and “one- dimensional,” as a way of minimizing Greek participation in overseas trading and Mediterranean expansion, and of casting such activities in a negative light. However, this representation of the Phoenicians in Greek sources such as the Odyssey is inaccurate. As she says, The picture of Phoenician shipping in the Mediterranean reconstructable from archaeological and historical evidence is one that includes complex urban organization and production in the homeland, a high degree of state control, varied cargoes and sophisticated targeting of differing socio-economic markets, extensive coastal installations along routes, the establishment of permanent settlements as 'daughter'-colonies, and a strong likelihood of competitive interactions with neighboring communities…Phoenician activity in economic and social spheres is far more varied and nuanced than is suggested by Greek sources… However, the strongest rebuke of maritime commerce occurs during Odysseus’ visit to Phaeacia. Odysseus is asked to participate in the athletic contests among the Phaeacian nobility. Laodomas, the son of Alcinous, invites Odysseus to join them in the games, and he initially demurs. When he does so, he is mocked by Euryalus, who compares him to a merchant ship-captain: ‘οὐ γάρ σ᾽ οὐδέ, ξεῖνε, δαήµονι φωτὶ ἐίσκω ἄθλων, οἷά τε πολλὰ µετ᾽ ἀνθρώποισι πέλονται, ἀλλὰ τῷ, ὅς θ᾽ ἅµα νηὶ πολυκλήιδι θαµίζων, 142 Phoenicians are mentioned in connection with trade between Phoenicia and Libya (XIV.287-2970, and with sailing to Ortygia to sell “trinkets” (athyrmata) and generate a substantial income (polys biotos) specifically through trading (empolaō); see XIV.415-456. 143 Phoenicians are at various points referred to as trôktai (“gnawers/nibblers”), a term of disparagement for merchants (XIV.289, XIV.416), “skilled in deceit” (XIV.288). 154 ἀρχὸς ναυτάων οἵ τε πρηκτῆρες ἔασιν, φόρτου τε µνήµων καὶ ἐπίσκοπος ᾖσιν ὁδαίων κερδέων θ᾽ ἁρπαλέων: οὐδ᾽ ἀθλητῆρι ἔοικας.’ “’I don’t make you out to be a man skilled in Athletics, the kind that are common among men, But rather like a man who shuttles back and forth in a large boat, A captain of sailors who are themselves merchants, A man who thinks only of his cargo, whose primary responsibility is his merchandise, Greedy for profits – not like an athlete.’” VIII.159-164 The scornful and accusing tone in this passage is obvious. It seems clear that Euryalus intends to ridicule Odysseus for lacking the kind of skills that are appropriate for a proper man (read: aristocrat). By way of contrast with what is fitting for an aristos, Euryalus sets up a sea-captain of merchantmen as an opposition. The unmistakable implication is that while skill in athletics is desirable, even necessary, for members of the elite, profit-seeking behavior and maritime trade are unsavory and inappropriate. If there was any question about the negative associations with such mercantile behavior, they are immediately dispelled by the reaction of Odysseus, who becomes angered at the insult and confronts Euryalus, calling him a fool (atasthalos) and accusing him of being handsome but shallow and obtuse (VIII.166-185). What is more, Odysseus’ reaction is seconded by Alcinous himself, who agrees that Euryalus’ words were an affront, and that Odysseus’ vehement reply was proportionate to the insult (VIII.236-249). There can be no doubt, on the basis if this exchange, that according to the aristocratic ethos of the Odyssey, profit-taking and plying the seas for the sake of mercantile exchange are activities that are diametrically opposed to the honor-seeking, agonistic athletic rituals practiced and endorsed by the Homeric elites. 144 144 The sole exception to this moral objection to maritime commerce may be seen in the trade for slaves (see 21.77- 79, 21.100-102, 22.44-45, 24.751-753). As we have already seen, several highly suggestive passages in the Iliad imply market places for trading slaves, and Achilles for one has regularly sold slaves to these markets. While this could be interpreted as a hole in the argument that maritime commerce is disparaged by Homeric elites, I would 155 It is impossible to deny the importance of sailing in the Greek world of the late 8 th c. BCE, 145 so important in fact that it is represented as a defining feature of human activity: οὐ γὰρ Κυκλώπεσσι νέες πάρα µιλτοπάρῃοι, οὐδ᾽ ἄνδρες νηῶν ἔνι τέκτονες, οἵ κε κάµοιεν νῆας ἐυσσέλµους, αἵ κεν τελέοιεν ἕκαστα ἄστε᾽ ἐπ᾽ ἀνθρώπων ἱκνεύµεναι, οἷά τε πολλὰ ἄνδρες ἐπ᾽ ἀλλήλους νηυσὶν περόωσι θάλασσαν: “For the Cyclopes lacked red-cheeked ships And they had no skilled shipwrights who might build Well-benched ships that would enable them To travel to all the cities of men, in the way that men so regularly Visit each other, crossing the sea in ships.” IX.125-129 In the above passage, ships are more than just a common feature of contemporary Greek life. Given that it is an excerpt from a longer one detailing the ways in which the Cyclopes differ from human beings, the implication is that overseas travel is an essential element of human culture. However, the critical point is that the motive for such travel is not economic gain, but rather in maintaining social connections. When figured in this way, it then becomes possible to represent maritime commerce as a perversion of proper behavior. Of course, this should sound familiar, because it is precisely the same type of rhetorical strategy that is used to characterize the difference between elite gift-giving, the supposedly morally upright form of exchange, and self-interested trade for profit, the debased, morally enervating form of exchange. It was during this time that the Greek world witnessed the beginnings of an expansion across the Mediterranean, with far-flung colonies established both east and west. The archaeological evidence on this point is undeniable, and along with this increased mobility of people came a surge in the mobility of argue the reverse, namely that here, too, we can see another way in which economic activity in the epics is represented in a way that is distinctly favorable to the aristocracy. Because war captives are won by the sword, they may represent an unusual commodity that elites might trade freely without risking the loss of prestige. 145 I wish to emphasize here that I do not equate sailing for the purpose of trade with sailing more generally. There are obviously many possible motives for taking to the sea. Sailing is a necessary but not sufficient condition for maritime commerce. I point out the prevalence of sailing here in order to demonstrate that the conditions conducive to widespread maritime commerce were already in place by this time. 156 goods (Boardman, 1999). The 8 th c. BCE was a period of major growth in imports of prestige items and artistic ideas (and quite possibly craftsmen) from the east and Egypt (Donlan, 1997), and in particular, a dramatic expansion of Phoenician trade and influence beginning as early as the 10 th c. BCE (Karageorghis, 1976) as far west as modern-day Spain (Aubet, 2001). In fact, evidence strongly suggests that Greek traders in the west began by following trade routes established by Phoenician sailors (Burkert, 2005). Thus, we can see that by this time the Greek world had expanded significantly, accompanied and driven by a substantial increase in the mobility of both goods and people along with increased contact with diverse groups of people. Meanwhile, the Phoenicians provided a vivid example of the potential benefits of overseas trade. In short, nearly all the elements conducive to growth in maritime commerce were in place. The Institutions of Long-Distance Exchange It is clear that the epics present a vision of exchange in its various forms that is highly moralizing, in which certain forms of economic activity confer an ethical benefit, whereas others are morally enervating. In particular, trade undertaken for the sake of profit is clearly represented as unseemly. It is presented as being antithetical to the cultivated elite lifestyle, as well as directly associated with immoral, greedy, untrustworthy Phoenicians. At the same time, elite gift-giving, despite some apparent problems, is constantly emphasized and presented as an essential part of elite interactions. We already know that gift-exchange as described in the Homeric epics is an element in the system of xenia, an extensive structure of rules and norms that was central to Homeric society as presented in the epics. 146 This of course is perfectly consistent with one of the central tenets of my argument, or rather my adoption of the neo-institutionalist argument that economic activity invariably takes place within the context of institutions. However, the rules that apply to market exchange are not nearly so clear. This is 146 See Herman (1987), Rose (1980), Reece (1992), Kurke (1991, 1999). 157 an important consideration, for as North (1981) has pointed out, underlying even the even the most basic transactions are complex systems of law and enforcement, whereby uncertainty is reduced by an accepted structure of property rights and their enforcement arrangements. This is made vastly more difficult and complicated in the event of long-distance trade, since the challenges to ensuring credible commitments are rendered much more difficult by the separation of the trading partners (North 1981, 1990, 1991). Sadly, the poems are almost completely silent on this topic. Yet, unless we deny the existence of long- distance trade during this period, which seems impossible in light of both the textual and the archaeological evidence, then we should presume the existence of institutions and enforcement arrangements in order to define property rights, produce enforceable contracts, and ensure credible commitments. However, because of the paucity of references in the Iliad and the Odyssey and the association of maritime trade with foreigners, especially Phoenicians, the specific forms of these institutions are never provided. There is, nonetheless, a tantalizingly evocative passage in the Odyssey that I believe indicates the kind of arrangements that may very well have served to create the necessary institutional structure that would have allowed long-distance trade to take place on a more frequent basis. It is a passage that recounts an episode in which Hephaestus captures his wife Aphrodite and Ares in flagrante delicto, trapping them in an unbreakable net and hanging them up for the amusement of all, at which point Poseidon urges Hephaestus to release Ares, promising Hephaestus that Ares will be compelled to compensate him. However, Hephaestus is reluctant to do so, claiming that he is not in a position to enforce the surety (engyē) for the debt (chreos) once he releases Ares. In response, Poseidon pledges himself as guarantor for the debt (VIII.266-360). What is fascinating about this exchange is that it is precisely the type of arrangement that would be particularly conducive to fostering commercial exchanges and the enforcement of obligations, as well as an understanding of the potential problems and pitfalls associated with ensuring the fulfillment of a debt 158 obligation if the debtor is not physically compelled, especially in a case in which the payment is deferred. Although this particular passage is not strictly speaking a commercial transaction, I believe that the mechanisms in place here are certainly those that could very well have been effective in ameliorating the kinds of difficulties facing long-distance trade, and suppressing the attendant risks and transaction costs sufficiently to make such exchange feasible. To be sure, we must always be careful not to read too much into the meanings of specific words, and especially not to assign modern economic functions to ancient institutions. 147 Still, there are other indications that the terms for debte and surety in this passage signify something very similar to commercial guarantees of contracts. The word for “debt” (chreos) occurs at several other points in the text. It can mean something more general, like a “matter” or “affair,” 148 but there are other instances in which it clearly indicates a debt to be collected. For example, we are told that the famous bow of Odysseus was given to him Iphitus the son of Eurytus, and that the two met in Messene when Odysseus “came to collect a debt owed to him by the entire demos…” (ἦλθε µετὰ χρεῖος, τό ῥά οἱ πᾶς δῆµος ὄφελλε). 149 Also, when Athena, disguised as Mentor, travels with Telemachus to the house of Nestor, s/he claims to be headed to collect a debt from the Cauconians. 150 Meanwhile, the word for “surety” (enguê) does not appear elsewhere in the epics, but by the 5 th c. BCE it clearly means a pledge made by someone who takes responsibility for another’s performance. 151 Also, the way in which these words are used in the above passage suggests that the concepts of debt and surety were already well established by this time. Poseidon offers to stand in as a guarantor for the debt of 147 This of course is the heart of the primitivist/substantivist criticism of modernism/formalism, what Polanyi calls “economic solipsism” (1977). 148 Ex., I.409, II.45. 149 XXI.17. 150 III.367. 151 See Antiphon, First Tetralogy 2.2.12; see also Demosthenes 33.10, 53.27. In Plato’s Laws (774e), it also means a “betrothal,” which is another kind of pledge/exchange. 159 Ares, and there is no indication that the terms in question require any sort of explanation or clarification. In fact, the arrangement proposed by Poseidon seems to be quite familiar and commonplace. Indeed, given how efficiently these arrangements conform to precisely the kinds of requirements that would have been necessary to create conditions amenable to a robust level of ongoing overseas trade, it does not seem unreasonable to posit that engyē really does mean something like “surety” or “collateral” in order to secure a debt, or that chreos here is a debt obligation in the sense that we would recognize it, and the fact that they do not appear in a commercial sense in the epics may very well be attributed to the overall suppression of commercial activity in the epics more generally. Conclusions I began this chapter from the basic premise that the simple framework of the low-level chiefdom society depicted in Homer would have been structurally insufficient to deal with the order of complexity indicated by the archaeological evidence as the Greek world expanded westward into Sicily and southern Italy. There is simply no ethnographic parallel of which I am aware in which such a political structure would have been able to maintain any degree of cohesion among such numerous and geographically disparate settlements. From this assumption, I examined the Homeric epics with an eye to what indications of social stress could be discerned, and what responses had been taken to address them. At the same time, I argued that the epics are motivated by a strong elite bias, which has significant implications for my analysis. First of all, this bias tends to produce a truncated image of Homeric society as consisting of two large groups, the aristoi and the laos. However, numerous indications in the text would appear to reveal a society that was much more complex and stratified than such an oversimplified picture 160 indicates. 152 The implication of this additional degree of complexity is that it represents a greater potential for the formation of class identity and for social fragmentation. I have also argued that, contrary to the position of Tandy (1997) and Rose (1992) that the Homeric epics depict a society experiencing a transition from one in which wealth follows status to one in which status follows wealth, that the epics demonstrate that wealth was clearly a fundamental and essential component of the formation of aristocratic standing from the beginning. It is much more accurate to say that class and status were tightly bound together in a self-reinforcing relationship, whereby Homeric elites gained and reinforced their positions by the mobilization of material resources, and at the same time were able to justify their preferential access to resources by virtue of their elevated status. However, it is important to recognize that in doing so, the aristoi were extremely careful to obscure the role of economic capital in the formation of status, instead preferring a narrative that depicted wealth as a function of status or a result of worthy achievements and martial success. To admit openly the role of wealth in the creation and perpetuation of status would be to allow some measure of validity to potential challenges to elite primacy through the acquisition of wealth. As a result, Homeric elites were keen to obfuscate such a connection, even if they could not obliterate it entirely, and the constant protestations to the contrary suggest an acute awareness of the potential of wealth as a tool for political and social advancement. 152 There are some hints in the epics that suggest a more complex ranking system than the bipartite social structure that Finley (1954) sees in Homeric society. For instance, the apoina Agamemon promises to pay consists in part of seven inhabited citadels whose income will accrue to Achilles (9.148-156). Such an offer clearly implies that these citadels are under the control of Agamemnon, so much so that he has the authority to transfer their rule to another by fiat. In addition, he maintains that the settlements are are located in the lowlands of Pylos, near the sea, which is a considerable distance from Mycenae where Agamemnon is the ruler. In short, Agamemnon appears to control an extensive redistributive network. This ability to maintain control over multiple sites separated from one another and from the ruling center implies a structure much more similar to the kind of chiefdom described by Sahlins (1963) and emblematized by the pre-contact Polynesian kingdoms that maintained control over varied populations dispersed across a broad geographic area. For Sahlins, it was an integral feature of the chiefdom that allowed for a complex hierarchy to be established and maintained, and which provided the means for controling and coordinating a diffuse population. Similarly, among the Homeric aristoi, there appears to be significant room for differentiation and the establishment of internal hierarchy. For instance, in addition to the example of Agamemnon above, we are also told that the Phaeacians are governed by a group of a dozen basileis along with Alcinous, who holds the supreme power (XI.354), suggesting yet another instance of a hierarchical arrangement among the basileis. 161 On a separate front, I have also proposed the necessity of understanding economic activity within the context of systems of rules, or “institutions.” This perspective implies several other important assumptions. First, these institutions are created and perpetuated by human beings, and therefore they are always susceptible to change, challenge, and renegotiation. Second, they are not innocent. There is a frequent tendency to view economic institutions as products of an evolutionary process, and therefore to see them as inevitably contributing to economic efficiency. However, there are any number of empirical examples that disprove this thesis, and instead demonstrate that rules are the product of human agency, and therefore the specific forms that institutions take are largely subject to the aims and desires of those with sufficient power to make their wishes manifest. Consistent with this larger argument, I have pointed out the ways in which the Iliad and the Odyssey promote strategies of production and distribution that are inherently conducive to maintaining the traditional social hierarchy, what Tandy (1997) calls “tools of exclusion.” Specifically, the epics distinctly favor two productive activities, horticulture and herding, that are both potent symbolic markers of elite identity. 153 Gardens and orchards are significant as much for their ability to indicate wealth and status as much as for their productive capabilities (Donlan, 1997), and large-scale herding activity also acts to exclude non-elites (Donlan, 1989c). As part of the pro-elite bias in the texts, the Iliad and the Odyssey present a distorted image of economic production, emphasizing these elite forms of productive activity and pushing agriculture, and consequently small and middling farmers, to the margins, despite the numerous indications throughout the epics of the central importance of cereal production to Greek society. 153 As Hilditch (2015) explains, gardens retained powerful and polyvalent symbolic associations in the ancient sources. For one, she notes that the figure of the garden in the Odyssey functions as a powerful metaphor, with the carefully tended, flourishing garden as a stand-in for the well-ordered, prosperous kingdom when cared for by the kindly and attentive basileus. Meanwhile, she also argues that archaic gardens likely were concrete manifestations of aristocratic control over resources, functioning both as productive physical spaces and elite identity markers, a function that was even further magnified by the association of gardens with eastern luxury. 162 In a similar fashion, the epics offer a pro-elite vision of exchange as well. There can be no disputing the prominence of gift-exchange, as it appears continuously throughout both texts. However, it is also clear that this too is a distortion. For one, the only partners to gift-exchanges depicted in the Iliad and Odyssey are aristocrats, and only certain items were seen as appropriate as gifts between elite exchange partners. 154 It is inconceivable that other commodities would not have exchanged hands on a frequent basis, and thus their near-complete absence is an indication of just how limiting a perspective is presented in the epics. Meanwhile, even when other forms of exchange are described, they are almost always presented in a distinctly negative light. More to the point, trade for the sake of profit, particularly long-distance commerce, is nearly always directly associated with people who are characterized in disparaging terms. There is an especially strong connection between profit-taking and Phoenicians, who are represented as nefarious and untrustworthy, and the clear implication is that the corrupt nature of the Phoenicians is an analogue for the kind of commercial activity in which they engaged. Taken together, there can be little doubt that the Iliad and the Odyssey present an ideologically loaded vision of economic activity that clearly favors the basileis at the expense of non-elites. In addition, I believe that this lopsided emphasis is best understood as a reaction to perceived threats to the traditional social order. But more than simply a response to generalized conditions of uncertainty, I would suggest that the perceived threat was more specific, namely challenges from non-elites who were exploiting the changing conditions in order to enrich themselves. As mobility increased and seaborne commerce became more common, the opportunities from maritime trade surged. In response, Homeric elites adopted a two- pronged strategy, one that both favored specific forms of production and distribution that supported their position, and at the same time sought to discredit those economic activities which were perceived as a threat. In short, the Iliad and the Odyssey operated as instruments in a program to establish behavioral 154 The association of certain specific types of items with specific types of transactions is well chronicled in the anthropological literature; for example, the phenomenon can be seen in the “spheres of exchange” observed among the Tiv (Bohannon, 1955, 1959, 1963) and Kabre (Piot, 1991) in West Africa. 163 rules favorable to those in power. In other words, they represent one side of a struggle to promote a particular set of institutions. 164 Chapter Five: Hesiod Introduction Admittedly, much of what I have said about the social and economic conditions based on my interpretation of the Iliad and the Odyssey is speculative, at times inferred more from what appears to be missing rather than what is there. My argument clearly would be strengthened from the introduction of a different perspective on contemporary archaic Greek society, and fortunately, the poetry of Hesiod, particularly the Works and Days, offers just such material. Despite its comparatively limited length of just slightly over 800 lines, it represents a rich historical source that it is vitally important historically, for several reasons. First, it represents a point of view on Greek society far different from the elite-centered attitude of the Homeric epics. It therefore allows us to confirm, or at the very least corroborate, our suspicions that the Iliad and Odyssey present a truncated, over-simplified view of the contemporaneous society that disenfranchises small landholders by omitting them from the narrative, and promote a set of values that valorizes economic strategies beneficial to the aristocracy and simultaneously denigrates those which appear to present opportunities for non-elites. In a related vein, Hesiodic poetry offers the capacity to enhance our understanding of the social background by offering a fuller account of social groups who are under-represented in the epics. Perhaps most importantly, Hesiod offers a starkly different economic ideology from Homer, one that not only seems largely indifferent to the kind of relationship-based elite gift-giving in the Iliad and the Odyssey, but that clearly has no objection to precisely the kind of profit-driven, impersonal, long-distance trade that the Homeric epics strive so hard to disparage. He seems to speak on behalf of the middling farmer who is so conspicuously absent from the Homeric accounts, and advocates strongly in favor of certain economic strategies beneficial to non-elites, including production specifically for export, maritime commerce, profit-driven trade, and more. In other words, the poetry of Hesiod directly challenges the norms promoted in the Homeric epics, and offers a competing set of rules in response, thereby indicating that the kind of generalized reciprocity which many have argued was emblematic of exchange in Greek 165 society in the latter half of the 8 th /beginning of the 7 th c. BCE 155 was not nearly so universally accepted or practiced. This is not to say that Hesiod presents a complete or even more comprehensive account of early archaic Greek social conditions than does Homer, in fact quite the reverse. Hesiod has his own agenda, particularly in the Works and Days, that presents its own sort of forced perspective just as much as Homer. The difference, and thus a great deal of the historical value of the poem, lies in the fact that the perspective is complementary to that of the Iliad and the Odyssey. Indeed, as R. P. Martin (1992) argues, from the standpoint of cultural history, Hesiod provides a contribution of inestimable value, since his poetry offers a perspective that is complementary to, albeit far different from, the elite-centered perspective on Greek social norms and values that dominates the Homeric epics. I should state from the outset that I explicitly reject the notion that either Hesiodic poetry or the Homeric epics are more reliable, valuable, or informative than the other as historical sources. To be sure, comparing the works attributed to Homer to those of Hesiod has a long and distinguished history. In fact, the predilection for considering the two in conjunction with one another has a tradition that extends well into antiquity. 156 Among modern scholars, however, despite the assertions of some who characterize Hesiod as “the most important of ancient authors…” (Lamberton, 1988), Homer clearly occupies a position of far greater veneration, whereas Hesiod has long languished in the shadow of Homer (Tandy & Neale, 1996). This trend also extends to the modern historiographical evaluation of Hesiod as well. As a source for archaic Greek history, Hesiod has been used by modern scholars most frequently as a supplement to the 155 See in particular Donlan (1989a, 1989b, 1989c, 1994, 1997, 1999), Tandy & Neale (1994), Tandy (1997). 156 In addition to the famous imagined contest between Homer and Hesiod, identifiable at least as early as the 4 th c. BCE (Koniaris, 1971; Mandilaras, 1992) this tendency appears to have been a prominent feature from the very beginnings of literary scholarship even among the ancients. For instance, Konig (2010) identifies no less than a half- dozen ancient texts devoted to a comparison of Hesiod and Homer. 166 Homeric epics rather than a “stand-alone” source in his own right, although in recent years this trend does appear to be reversing itself. 157 No doubt, this phenomenon is due, at least in part, to the modern valuative assessment of the two. This view of Homeric poetry as qualitatively superior to that of Hesiod has the potential consequence of regarding Homer as a more authoritative historical source. However, I would argue that neither can reasonably be given primacy of place over the other, at least insofar as they are used as sources for contemporary social history. This approach may sound rather obvious, but it is an important starting point, since it necessarily means that in those instances where the evidence of Hesiod conflicts with that of Homer, we cannot simply dispense with the Hesiodic passage as a matter of reflex, since there is nothing inherent in either that automatically relegates one or the other to inferior or superior status from a historiographical point of view. Therefore, I will give the evidence from Hesiod equal weight to that of Homer, and avoid the assumption that the evidence from Hesiod is deficient where it conflicts with the depictions in the Homeric epics. As in the previous chapter, my primary interests are twofold: 1) the mechanical processes of production, consumption, and distribution, and 2) the institutional matrix, or the rules of the game, within which these activities take place. In so doing, my examination is not limited solely to what those specific rules are, but in the ways that economic agents position themselves in regard to the established institutional matrix, with particular attention to the formation and propagation of these rules, as well as how actors navigate within them, without ignoring the fact that the texts have an ideological motivation. Finally, I wish to consider the evidence from Hesiod explicitly in the context of the evidence from Homer, especially from the standpoint of how they both reflect specific yet differing responses to changing social, political and economic realities. In particular, I wish to be alert to the role of economic development in the context of 157 For example, see Edwards (2004) and Konig (2010), as well as the recent Brill’s Companion to Hesiod (Montanari, Rengakos, & Tsagalis, 2009); several new translations have also appeared in recent years (Most 2006; Powell, 2017). 167 social and political change, and how different groups in the contemporary society appear to be adapting to these changing conditions, with particular emphasis on the role of economic activity. Hesiodic Questions As in the previous chapter, my argument rests on several fundamental premises that deserve mention, even if concision prevents a more thorough discussion. 158 Obviously, establishing a date for the texts is vital. Also important is the question of the historicity of the society as described in the texts, as well as the dating of that society and its usefulness for drawing broader conclusions about Greek society writ large. In addition, the selection of texts that have been attributed to Hesiod deserves at least some mention. Finally, the social perspective of Hesiod has important implications for my argument, and therefore my position on that issue should be made clear. First and foremost, my argument rests on a basic assumption that the Homeric and Hesiodic texts themselves, as well as the social background that they describe, are roughly coeval, separated by little more than a half-century. 159 Thus, the different perspectives they offer cannot be dismissed as the result changes over an extended period of time, especially since the traditional dates assigned to Homer and Hesiod would place them both at the beginning of a period of sweeping changes in Greek society during the archaic period, minimizing differential social, political, and economic development potentially 158 As before, I state these positions with full awareness that they may not be settled issues, and in fact frequently are part of much larger debates that have lengthy and contentious histories. Nevertheless, it is not possible to re-argue each and every issue, but where possible, I have included bibliographic references wherein may be found fuller treatments of these points. 159 The question of the dating of Hesiodic poetry, as well as the relative dates of Homer and Hesiod, has been a matter of controversy not only among modern scholars, but seemingly from the very earliest literary attestations in antiquity. As Most (2006) points out, ancient authors frequently adopted the position that Homer and Hesiod were contemporaries, although arguments in favor of Hesiod’s priority, or of Homer’s were both well-represented in the ancient sources. For instance, the Suda (η 583) states that Hesiod is probably older but admits that he could be later; Tzetzes (Scholium on the Works and Days) contends that Hesiod is later; Aulus Gellius (3.11.1-5) says that there is no consensus regarding the age of Homer and Hesiod, with Philochorus and Xenophanes arguing that Homer was prior, while others such as Lucius Accius and Ephorus say the opposite, and Varro saying that they both were alive at the same time; Pausanius (9.30.3) states that it is impossible to determine the relative dates of Homer and Hesiod, and these are just a few of the ancient sources. Kõiv (2011) even attempts to date Hesiod based on the various ancient literary attestations. 168 reflected in the texts. 160 The dating of the society is far less problematic than that of Homeric society, however, given that Hesiod appears to describe the specific social, political, and economic conditions in which he lives, and I see no compelling reason to doubt this picture. Because Hesiod is so clearly concerned with his own immediate present, the challenges posed by the need to distinguish those elements which are products of an “archaizing” effort from those which are more reflective of contemporary social conditions are vitiated. At the same time, I would adduce many of the same arguments applied to the Homeric epics, especially those concerning the constraints of verisimilitude. As Millett (1984) points out, "It is difficult to imagine any reason why the poet should want to confuse his audience by deliberately archaizing or otherwise misrepresenting social institutions…" In addition, my focus is not strictly on the accuracy of the specific descriptions in Hesiod. Just as with the Homeric poetry, my argument proceeds from the fundamental premise that Hesiod’s depictions are representational and that, as Tandy (1997) explains, they both reflect and refract the social world to which they refer. Consequently, I fully accept the notion that Hesiod does not offer some transparent window onto the early archaic world, and that it is precisely the “distortions” that are themselves both historical phenomena and objects of study. As Edwards (2004) explains, what Hesiod does not say is sometimes more important than what he actually says. Put another way, the “texts are neither a mere passing on of old stories, nor are they a (Gehrke, 2001) depiction of conditions in their time; they are 'intentional history'…" (Ulf, 2009). 161 160 Put simply, I adhere to the influential arguments of Janko (1982) who analyzes the linguistic evidence to conclude that the Homeric texts belong to the second half of the 8 th c. BCE, while Hesiod dates approximately to the turn of the century. He has since altered his stance slightly (Janko, 1996, 1998) on the basis of the arguments of Ruijgh (1995), pushing the dates of the Homeric texts back by a quarter century or so, but the consensus still regards them as roughly contemporary, with Hesiod perhaps a generation or so younger (see Tandy and Neale, 1996; Tandy, 1997; Edwards, 2004; Most, 2006; Ulf, 2009). Although Tandy (1997) and Donlan (1994, 1997) propose that the Homeric epics actually describe social conditions more reflective of a generation or two prior to that of the audience, Morris (1986) has compiled an impressive amount of comparative ethnographic evidence suggesting that it is very difficult for oral poetry to retain outdated elements for so long, and that in fact oral poetic descriptions are almost invariably reflective of prevailing conditions. There is also West’s (2012) recent argument that would put the Iliad down into the seventh century. 161 See also Gehrke (2001), Raaflaub (1998). 169 Also, much like Homeric poetry, I believe that Hesiod is shaped by an emerging pan-Hellenic outlook, and thus represents an excellent counterpoint to Homer. But whereas the pan-Hellenic nature of Homeric poetry has long been accepted, this is perhaps a more controversial statement regarding Hesiod. 162 Nevertheless, I suggest that the enduring and widespread appeal of Hesiod is testament to the fact that diverse ancient audiences and readers found the poetry relevant, and that the poetry itself helped to foster a growing sense of pan-Hellenism. Although some of the advice in the Works and Days is ostensibly directed toward a single individual (Perses), much of the advice in the second half of the poem is clearly intended for a more general audience. Also, both the Theogony and the Works and Days both show strong connections to eastern literary traditions (West, 1978, 1997; Burkert, 2005), and although they are unique in many ways, they must be viewed as representative of and working within literary genres that were widespread across the Mediterranean at the time of their composition (Most, 2006). As for the specific texts themselves, I focus primarily on the Works and Days, but I occasionally refer to the Theogony where it seems to clarify or further develop ideas expressed in the Works and Days. For instance, the view of the basileis is much more positive in the Theogony than in the Works and Days, a phenomenon that could suggest differing aims of the respective poems and perhaps the conditions under which they were produced. As for the other poems sometimes attributed to Hesiod such as the Catalogue of Women or the Shield of Heracles, I make no use of them, partly from a desire to avoid getting trapped in questions of dating, authenticity, authorship, etc., but mostly because they do not offer near the level of insight into questions of economic behavior as the Works and Days. 163 The various other works ascribed to Hesiod in antiquity which are clearly spurious, although of interest for entirely different reasons, have no place here. 162 The most notable arguments in favor of the pan-Hellenic nature of Hesiodic poetry have been made by Nagy (1982, 1999); see also Burkert (2005) and Most (2006). 163 The general consensus is that the Shield and the Catalogue are later works that were subsequently added to the Hesiodic corpus, with both dated from the end of the 7 th c. to the first half of the 6 th c. BCE (Most, 2006). 170 As for Hesiod himself, I have little interest in the question of the social standing of Hesiod, and even less in the question of the historicity of the poet. Any number of scholars have speculated on the specifics of Hesiod’s biography. 164 However, recent scholarship has begun to swing towards the point of view that “Hesiod” is less an actual writer than a poetic persona. 165 Ultimately, I accept in general terms the conclusions of Marsilio (2002), who contends that Hesiod should be understood as the combination of historically genuine elements and generic traditions, and of Tandy and Neale (1996), who accept the supposition that Hesiod may take on a poetic persona (and describe conditions not specifically his own), but conclude that the conditions described are real, reflecting the world as Hesiod/”Hesiod” perceived them. From this point of view, whether Hesiod may or may not have been an actual person is irrelevant. What is important is the poet’s perspective on contemporary society, and It makes no difference, for example, whether a real Hesiod won a tripod at the funeral games of Amphidamas, or if he had a brother named Perses with whom he quarreled over their inheritance, etc. For Edwards (2004), it makes no difference to the historicity of the final product whether what is represented actually happened or is fictional from the outset. The World of Hesiod Although I have little interest in engaging in a discussion of the biographical particulars of Hesiod, what is important is that he represents a particular outlook, and what kind of society he describes. One of the more vexing questions regarding the poetry of Hesiod is whether or not the society described in Works 164 For instance, Bravo (1977) argues that Hesiod, Perses, and their father were impoverished aristoi, whereas Starr (1977) sees Hesiod as a “semi-aristocrat.” 165 R. P. Martin (1992) is one of the more influential advocates for this position, in which he finds support from Rosen (1990), Nagy (1973), Hunt (1981) Griffith (1983), and Lamberton (1988), among others. See Tandy and Neale (1996) and Tandy (1997) for overviews of this debate. 171 and Days is representative of a “peasant society.” With some exceptions, 166 the weight of scholarly opinion has regarded Hesiod as a peasant, and the social world of Ascra as described in Works and Days as representative of a peasant society. 167 I disagree with this characterization for several reasons. For one, consistent with the points made by Edwards (2004), I find that the world of Hesiod’s Ascra is much more politically and economically independent, and one in which the challenges to the basileis are much more overt and contentious, than what one would expect of a so-called peasant society, especially if we accept Redfield’s (1956) insistence that subservience is a defining feature of peasant societies. Also, the term itself has been used in so many different contexts that it has lost a great deal of its heuristic value. For example, Edelman (2013) identifies four different kinds of definition of “peasant” (historical, social scientific, activist, normative), with degrees of variation even within these broader categories. As he explains, “the question of how to define ‘peasant’ and ‘peasantry’ has a long, complicated and contentious history…” The term has also frequently acquired pejorative connotations, and is so malleable that it can be applied to virtually any non-ruling, agriculture-based social group, leading many recent scholars to avoid the term altogether. 168 Even if we choose to avoid use of the term ‘peasant,’ we are still left with the question about just what kind of society Hesiod seems to describe. For Tandy (1997), one of the most important features of the Works and Days is that it conforms to a peasant society and thus sheds light on the universal plight of the peasantry throughout human history. Given our reluctance to employ this terminology, we are left in a more difficult position in finding an adequate way to describe and contextualize the social background depicted by Hesiod. But the most important point to be made here is that whereas the Iliad and the 166 Ex., Starr (1977), Mele (1979), Whitley (1991), Burford (1993), Hanson (1995). 167 This view stems from Redfield (1956) and its strongest advocates are Millett (1984) and Tandy (1997); Tandy (1997), Marsilio (2004), and Hall (2007) each offer a concise survey of the debate as well as a bibliography, while a more thorough review of the discussion can be found in Edwards (2004). 168 Nelson (1996, 1998) eschews “peasant” for the sake of “small farmer” and Magagna (1991) employs the term “communities of grain” instead of “peasant society.” For a recent argument that addresses many of these questions and defends the use of the term, see Owen (2005) 172 Odyssey present an aristocratic point of view, Hesiod represents a perspective that is clearly distinct from the Homeric epics, and at points is in explicit and direct opposition to the basileis. 169 As such, the Hesiodic poetry offers a tremendous opportunity to examine economic practices and ideology of a non- elite segment of society as a counter-balance to the elite bias of the epics. So, what exactly does Hesiod say about his social environment? As we know, Hesiod claims to live in Ascra, a remote, inland settlement in Boeotia. Until recently, outside of the Works and Days, we knew very little about the conditions of the region during this time, making analysis difficult, and arguments based on the textual evidence potentially circular. However, thanks to the systematic and methodical work of the Boeotia Survey project, the growing body of archaeological data has considerably enhanced our understanding of this region and its development over time. 170 What the archaeological evidence does not show, however, is the degree to which Ascra may have been integrated economically and intellectually within a much broader, cohesive cultural koinē. It is on this point that the Hesiodic poetry is particularly valuable, as it provides a level of detail regarding the socio-economic relationships between Ascra and a larger Greek world. To be sure, Ascra itself was clearly a small, remote settlement. However, this does not necessarily mean that it was cut off from all contact. Nevertheless, this is an argument that has been influential. For instance, Edwards (2004) argues that the society described by Hesiod is much simpler and less stratified even than the “low-level chiefdom” (Donlan, 1994) that many see represented in the Homeric epics. Even while admitting that the residents of Ascra appear to exist (at least to a degree) under the political aegis of Thespiae, Edwards on occasion even refers to “the society of Ascra” as if it 169 The depiction of the basileis is more explicitly negative in the Works and Days than in the Theogony; see “Hesiodic Questions” above. 170 Since 1978, the Boeotia Survey’s annual survey project has covered more than fifty square miles of southwest Boeotia, including the territory of Thespiae of which Ascra was a part; see (Bintiliff & Snodgrass, 1985; Bintliff J. L., 1997; Bintliff, Howard, & Snodgrass, 2007). On the basis of the archaeological evidence, it seems that Ascra was a small, unwalled, rural settlement that at the time of the composition of the Works and Days was likely under the political aegis of the nearby urban center of Thespiae. In short, the material evidence corresponds in all respects with the picture presented by Hesiod (Edwards, 2004), and thus lends additional credence to the text. 173 were some sort of self-contained community. But Ascra is clearly connected politically, socially, and economically to a more extensive and complex social, political, and economic network, and it makes little sense to make general claims about the relative simplicity of the prevailing social structure based on the consideration of Ascra in isolation, removed from its larger context. Also, the social structure described by Hesiod hints at a greater level of stratification than many have previously held. According to Edwards (2004), there are three distinct statuses described in Works and Days: slave (dmōs), free landless laborer (thēs), and landed free man. However, both the Theogony and Works and Days also mention the presence of basileis, frequently in their capacity as judges in resolving disputes. 171 At one point, Hesiod even addresses a portion of the Works and Days explicitly to the basileis. 172 It may be the case that no basileus resides in Ascra. Their identification as judges, and the association of public dispute resolution with the agora, would suggest as much. Nevertheless, it is undeniable that their dominion extends to Ascra, as is evident by the frustration Hesiod expresses over the perceived corruption of the “gift-eating basileis” (WD, 37-39). As disenchanted as he is with their decision regarding the inheritance dispute with his brother Perses, Hesiod never questions their authority to render such decisions. In fact, his frustration clearly stems from precisely the fact that he is subject to their jurisdiction. Also, this categorical schema does not appear to leave room for various craftsmen, who may or may not own real property, but do not earn a living through wage labor. We are told of the existence of potters, carpenters, singers, blacksmiths, and woodcutters, 173 for whom there is no place in Edwards’ blueprint. Thus, it seems that no fewer than five distinct socio-economic statuses are visible in the text. At the same time, the reduction of social hierarchy in Hesiod even to five status categories almost inevitably ignores the fact that there is clearly ample room for negotiation within them. For instance, 171 Ex., Theogony 80ff., WD 27ff., 219 ff., 248 ff. 172 WD, 202 ff. 173 WD 25-26, 493, 807. 174 Hesiod tells us about two different kinds of strife (eris). Both are children of Night – the first is cruel and honored by mortals only because of obligation, but the second is good for men: “ἥτε καὶ ἀπάλαµόν περ ὁµῶς ἐπὶ ἔργον ἔγειρεν. εἰς ἕτερον γάρ τίς τε ἰδὼν ἔργοιο χατίζει πλούσιον, ὃς σπεύδει µὲν ἀρώµεναι ἠδὲ φυτεύειν οἶκόν τ᾽ εὖ θέσθαι: ζηλοῖ δέ τε γείτονα γείτων εἰς ἄφενος σπεύδοντ᾽: ἀγαθὴ δ᾽ Ἔρις ἥδε βροτοῖσιν. καὶ κεραµεὺς κεραµεῖ κοτέει καὶ τέκτονι τέκτων, καὶ πτωχὸς πτωχῷ φθονέει καὶ ἀοιδὸς ἀοιδῷ.” She rouses even the helpless man to work. For when a man is idle, seeing another man Who is wealthy, and who hustles to plow and to plant In order to improve his estate, the one neighbor envies the other neighbor Who is quickly growing wealthy. This is the Strife who is good for mortals, And potter strives with potter, carpenter with carpenter, And beggar vies with beggar, singer with singer.” WD, 20-26 Given the kind of competition suggested here, it is clear that there are differential levels of wealth and success among craftsmen and among landholding farmers. In fact, for Hesiod, this inequity is beneficial, since it motivates those who are less successful to work harder in order to emulate their wealthier counterparts. But what is notable is that even among groups within the community, wealth is a means of distinction. It is not at all controversial to assert that control over resources is an important component of power, but as we have seen in the Homeric epics, among the elites there was a strong impulse to deny the connection between status and class, but in the Works and Days, the relationship is much more explicit. Also, it appears that social stratification is more pronounced than many have argued, such as Edwards (2004), who sees the population of Ascra as largely homogeneous and undifferentiated in their social standing. Ultimately, it would appear that in its broad strokes, the social background described by Hesiod aligns very conveniently with the presentation in the Homeric epics. Of course, the narrative represents a 175 markedly divergent point of view from the elite-centered perspective of the Homeric epics, and as we shall see, Hesiod attempts to establish an ethical program that is significantly different, but where the two representations overlap, there is a high degree of consistency. First of all, slaves would appear to be a common feature of the kind of agricultural household that Hesiod describes, 174 although there is no attempt to forge an obvious demarcation between free and slave as in the Odyssey, 175 with several mentions of agricultural laborers who are ambiguous, possibly slaves, but perhaps hired labor. 176 Also, craftsmen and artisans are mentioned throughout the Works and Days, although whether or not a significant portion of the population was engaged in craft production is a thorny question. 177 Although traders appear several times in Homer, they are not mentioned specifically by Hesiod. However, their existence is strongly suggested by the frequent references that Hesiod makes to trading activity, particularly maritime trade. 178 Combined with the evidence from Homer, the commonplace nature of maritime trade in the Works and Days would seem to require the presence of professional merchants. For example, Hesiod describes his own father thus: αὐτὸς δ᾽ ὡραῖον µίµνειν πλόον, εἰσόκεν ἔλθῃ: καὶ τότε νῆα θοὴν ἅλαδ᾽ ἑλκέµεν, ἐν δέ τε φόρτον ἄρµενον ἐντύνασθαι, ἵν᾽ οἴκαδε κέρδος ἄρηαι, ὥς περ ἐµός τε πατὴρ καὶ σός, µέγα νήπιε Πέρσῃ, πλωίζεσκ᾽ ἐν νηυσί, βίου κεχρηµένος ἐσθλοῦ: 174 There has been a lengthy debate over slavery in archaic Greece spanning numerous aspects, including whether or not Greece at this time was a “slave society,” what the defining features of a so-called slave society are (the mere presence of slaves is usually thought to be a necessary but not sufficient condition), and even whether or not the dmōs in Homer and Hesiod can be accurately described as a slave (Morris, 1989); Finley (1980) is a seminal treatment of this question, and Thalmann (1998) provides an extensive review of the debate. 175 The degree to which the Odyssey is concerned with justifying the brutal reality of slavery, and its implications for our understanding of early archaic Greek society, is covered in depth by Thalmann (1998). 176 WD 436 ff. 177 While both the Homeric epics and Hesiod seem to agree that craftsmen compose a narrow minority of the population, this picture may very well be due in part to the respective foci of the different works. As we have seen, both the Iliad and the Odyssey make almost no mention of free, landowning farmers like Hesiod. At the same time, these free farmers are a central focus of the Works and Days. I suspect that if we were to discover poetry written by a slave, a carpenter, or a sailor, then our picture of archaic Greek society would be very different. 178 WD 630-632, 641-645, 671-672, etc. 176 “You [Perses] yourself wait until the sailing season has arrived, Then drag your swift ship to the sea, and put your Assembled cargo in it and make it ready so that you can bring home your profit, Just as our father, you great fool Perses, Used to take to the sea in ships, in need of a good livelihood…” WD, 630-635 While Hesiod most frequently characterizes maritime trade as a sort of sideline in which farmers engage as a way of supplementing their income, it seems clear in the above passage that for some, Hesiod’s father included, 179 maritime trade was their primary means of income. Unless his case was unique, we can assume that there were others who also made their living as full-time, professional traders. Several very interesting features emerge from an examination of Hesiod’s social setting. First of all, I would suggest that there is a much greater degree of stratification than what might first appear. Among free landowners, craftsmen, and even hired laborers, there are internal hierarchies. More significantly, these distinctions are made largely on the basis of wealth. Second, the citizen farmer whom Hesiod and his neighbors represent is vastly different from either the landless laborers and the slaves on one end of the social spectrum, or the basileis on the other, who are described in the Homeric epics. The existence of such agricultural producers is inconsistent with the picture of archaic Greek society given in the Iliad and the Odyssey, from whose narratives their presence is eliminated almost entirely so that Finley (1954) concluded that Homeric society was dominated by a fundamental “cleavage” between mass and elite. In fact, if Edwards’ (2004) interpretation of the Works and Days is to be believed, Ascra was largely independent politically, even if it technically fell under the aegis of Thespiae. If that is true, then the landowners of Hesiod’s ilk are the de facto elites of Ascra. I am not ready to insist on the degree of Ascran autonomy that Edwards does, in part because I believe that Hesiod has his own motives for 179 I should point out that Edwards (2004) rejects the view that Hesiod’s father was a professional, full-time trader, but the consensus opinion follows the arguments of Mele (1979) that he was a professional trader who relied on commerce either entirely or substantially for his income; see also Tandy (1997) and Marsilio (2002). 177 minimizing the presence and influence of the basileis in much the same way as the Homeric epics clearly marginalized the role of free landowners. Nevertheless, there can be little doubt that such free farmers do not fit conveniently into the bipartite society envisioned by Finley on the basis of the Homeric texts. 180 In fact, given that their status is clearly based on their control over the means of production and their access to resources, it is safe to characterize this group as a class, and in light of their apparent position in the social hierarchy between the basileis on the one hand, and the landless laborers and slaves on the other, that they occupy a middling social stratum, distinct from both the aristoi on one hand, and thetes and slaves on the other. Yet even if we can realistically apply this admittedly loaded term to the free, landowning agricultural producers depicted by Hesiod, one wonders whether or not they were themselves aware of their common interests. In other words, can we perceive evidence of a class consciousness? To be sure, self-awareness or class consciousness is not a necessary condition for a social sub-group to exist and function as a class (Rose, 2009, 2012). However, class consciousness can be a powerful element in collective action, and as Marx (1994 [1852]) noted long ago in famously comparing the 19 th c. French peasantry to a sack of potatoes, its absence can frequently inhibit members of a class from recognizing their shared conditions and acting cohesively in their collective interest. Although there are no overt calls to action on the basis of class in Hesiod, there are indications of a nascent class sensibility. For instance, Hesiod frequently refers to the basileis collectively 181 (as does Homer), confirming the idea that they were perceived to have enough in common with one another to be regarded as a single, coherent group. In addition, the fact that the basileis were regarded as a class both emically (from the elite perspective of the Homeric epics) and 180 Finley does recognize the existence of peasants, and he places them on the bottom side of the horizontal cleavage that he sees, along with thetes and slaves. In any case, I maintain that Hesiod and his companion free farmers represent a more elevated socio-economic group that may be subject to the basileis, but possess far more autonomy, both politically and economically, than what would typically be expected of a peasant society. 181 Ex., WD 37, 202, 248; Theogony 79-88. 178 etically (from the decidedly non-elite view of Hesiod) further serves to reinforce this notion, and we might therefore see in the Works and Days a nascent class consciousness. But is the awareness of class among the elites relevant to the development of a class consciousness among the non-elite landowners? In short…maybe. By constituting a distinct social group distinguished in large part by the control of resources, the basileis served as a point of reference from which Hesiod and his counterparts could frame their own identity through the process of oppositional identity formation. In other words, the non-elite agriculturalists could find some common ground in what they were not. For example, we can discern this process at work in the following passage in which Hesiod speaks of Dikē, the personification of justice: αὐτίκα πὰρ Διὶ πατρὶ καθεζοµένη Κρονίωνι γηρύετ᾽ ἀνθρώπων ἄδικον νόον, ὄφρ᾽ ἀποτίσῃ δῆµος ἀτασθαλίας βασιλέων, οἳ λυγρὰ νοεῦντες ἄλλῃ παρκλίνωσι δίκας σκολιῶς ἐνέποντες. “Immediately sitting down beside her father Zeus, son of Cronus, She sings out the unjust mind of men, so that he Punishes the people for the sins of the mischief-minded kings, who with destructive thoughts Turn judgments astray, pronouncing them crookedly.” WD 259-262 It is clear from the above passage that a) the basileis are contrasted explicitly with and separated from “the people” (demos), b) the basileis have the authority to make decisions and render judgments (dikai) affecting the entire populace, and c) when these judgments are corrupt, the people pay the price. For our purposes, the most important consideration is the classification of the basileis into a single group, distinct and separate from the rest of society. But Hesiod and his fellow free farmers would likewise seem to represent a distinct class. If we accept the notion that control of the means of production is an essential element of class differentiation, it is clear that they are qualitatively different from landless laborers and slaves, who are both alienated from both 179 the means of production and from the products of their labor. In addition, there is textual evidence to indicate that these free, land-owning farmers at least tacitly recognized a sufficient degree of common interest to act collectively in order to further their own self-interest. In so doing, they act in concert bound solely by bonds of cooperation: τὸν φιλέοντ᾽ ἐπὶ δαῖτα καλεῖν, τὸν δ᾽ ἐχθρὸν ἐᾶσαι: τὸν δὲ µάλιστα καλεῖν, ὅς τις σέθεν ἐγγύθι ναίει: εἰ γάρ τοι καὶ χρῆµ᾽ ἐγχώριον ἄλλο γένηται, γείτονες ἄζωστοι ἔκιον, ζώσαντο δὲ πηοί. Invite your friend to feast, but leave your enemy be; But above all invite him who lives nearby, For if some urgent need should arise on your estate, Neighbors come ungirt, but in-laws dress themselves first. WD, 342-345 To be sure, these lines are not an overt expression of class unity, but the underlying sentiments can easily be regarded as the foundation in the initial formation of such a view, this time through an aggregative rather than oppositional process of identity formation. If we accept that a class is “a group of persons in a community identified by their position in a whole system of social production, defined above all according to their relationship…to the means of production…and to other classes…" (Ste. Croix 1981), then we can see that Hesiod and his ilk are distinguished by their relationship to the means of production in that they own farmland as well as livestock (including draft animals) and slaves. They are also identified in relation to other classes, since they are clearly distinct from the basileis at one end of the social spectrum as well as from the landless laborers and from slaves at the other. It seems clear that Hesiod’s neighbors share a common economic condition, and his advice directed to them throughout the Works and Days is consistent in its emphasis on certain forms of agricultural production. In addition, it is clear that they recognize mutual obligations and interests, so much so that they are willing to come to their neighbor’s aid in an instant. Above all, Hesiod demonstrates that these 180 obligations are even more compelling than the bonds of kinship. Taken in conjunction with the various passages in both Homer and Hesiod in which elites are described as a distinct group, largely on the basis of their control over means of production and preferential access to resources, the Works and Days provides evidence for an ongoing process of identity formation that would ultimately find expression in what Morris (1996, 1999b, 2004) refers to as a “middling ideology” both on the basis of shared characteristics (who they are), and on the basis of elimination (who they are not). Perhaps not surprisingly, concurrent with this development of a nascent middling sensibility can be seen growing class tensions. For various reasons that I have already presented, the Homeric epics present a picture of remarkably stable social hierarchy which is rarely questioned, and even more rarely challenged overtly. When it is, those who do so, whether slave (Melanthius) or free (Thersites), are castigated immediately and brutally. In addition, they are represented as morally deficient, so that their punishments are seen as less about maintaining control by force than they are about restoring order and balance to an ethical universe. Thus, I believe that these episodes may well be read as indications of underlying anxieties over challenges to elite primacy. It is in the Works and Days that we can witness such challenges from a non-elite perspective. It has long been recognized that Hesiod in the Theogony offers a much more sympathetic portrayal of the basileis than in the Works and Days. 182 For Tandy (1997), this is the result of different conditions during the composition of the respective poems. According to this argument, the Theogony was produced under the powerful influence of the rulers of the area in which Hesiod lived, but for some reason, the Works and Days was immune from this coercion (Tandy & Neale, 1996). Regardless of the cause, it is clear that the Works and Days is much more openly critical of the basileis. Whereas the Theogony represents the kings 182 Ex., Tandy & Neale (1996), Tandy (1997), Edwards (2004). 181 as the source from which all good things flow, 183 the Works and Days accuses them of subverting justice (dikē) in rendering their judgments for the sake of their own personal gain. For example, in one famous passage, he rails against the “gift-devouring” (dōrophagoi) kings whom his brother corrupted in their dispute over their father’s estate (WD, 37-39). While dorophagoi can be read as “bribe-devouring,” I would suggest that it takes on additional potency as a critique of elite practices if we retain the literal association of dora 184 with “gifts.” In light of the continued emphasis in the Homeric epics on ritualized gift-exchange as an essential element of aristocratic standing, the overt demarcation between Hesiod and the other free farmers from the basileis on the basis of their association with gifts creates a powerful division in which the symbolism of gift-exchange is used as a means of distinction both by those within the group, and by those without. However, when presented from Hesiod’s point of view, the ritual of gift- exchange is characterized in distinctly critical and negative terms, a sharp difference from the depictions in the Homeric epics. This is not to say that the Homeric epics are devoid of social criticism. Raaflaub (1988), for one, sees in the texts hints of latent dissatisfaction of the heroic value system that seems structurally doomed by its tendency to produce a perpetual cycle of violence for the sake of individual honor. Donlan (1973), too, finds subtle critiques of the system. However, as Thalmann (1998) argues, a goal of the epics is to defuse and ultimately invalidate these negative arguments. By contrast, the Works and Days admits that a good ruler is instrumental to the success of his polity, 185 but focuses far more attention on those who ignore the gods and render crooked judgments, bringing myriad ills to their subjects. 186 In fact, this is the central 183 Theogony, 84-92. 184 As Herman (1987) points out, there is a significant degree of ambivalence in the ancient literature regarding the potential of xeinia to forge bonds among elites that fostered loyalties that potentially conflicted with duty to one’s own polis. Consistent with this sentiment, it appears that dora frequently acquired a negative connotation, specifically in connection with anti-elite criticisms, and were often characterized as bribes instead of gifts. For example, the term adôrodokêtos that is often translated as “incorruptible” is more accurately someone whose loyalty lies with the polis and who refuses to engage in gift-exchange with xenoi. 185 WD, 225-237. 186 WD, 37-41, 219-224, 238-247, 248-273. 182 lament of the Works and Days. In his famous account of the races of man, Hesiod recounts the four previous races (gold, silver, bronze, and heroes) and bemoans the fact that he now lives in in the time of the fifth race of men, that of iron, when all manner of ills are upon the earth due to impiety and injustice that brings the retribution of the gods (WD, 174-201). Unlike the Odyssey, in which the disruption and disorder in Ithaca are ultimately resolved and balance restored by the intervention of the rightful and righteous basileus, the Works and Days presents a world in which chaos reigns, and in many cases the people suffer precisely because of the impiety and hubris of their leaders. This brings me to what I consider one of the more important aspects of archaic society as depicted by Hesiod: the potential for social mobility, and the degree to which such mobility is grounded in the acquisition (or loss) of personal wealth. It is on this point that the Works and Days differs perhaps most fundamentally from the Homeric epics. In the Iliad and the Odyssey, the social hierarchy is largely static and fixed. Slaves are slaves, and kings are kings. 187 As for the basileis, they hold their positions as a result of their inherent superiority (recall the numerous instances in the texts in which kings are recognized by their appearance) or by merit (see Sarpedon’s response to Glaucus at Il., 12.310-328), but never because of their wealth. We have already seen how the Homeric epics seek to elide the connection between wealth and elite standing. However, for Hesiod, wealth is a means to advancement. Nowhere is this more clear than in the following passage: ἐξ ἔργων δ᾽ ἄνδρες πολύµηλοί τ᾽ ἀφνειοί τε: καὶ ἐργαζόµενοι πολὺ φίλτεροι ἀθανάτοισιν… εἰ δέ κε ἐργάζῃ, τάχα σε ζηλώσει ἀεργὸς πλουτεῦντα: πλούτῳ δ᾽ ἀρετὴ καὶ κῦδος ὀπηδεῖ… 187 The tales of the beggar Odysseus, when he tells Eumaeus that he was the bastard son of a wealthy Cretan (XIV.199-355), or when he tells Penelope that he was the son of a Cretan lord (XVIII.172-189), could be seen as evidence for social mobility. However, in the first instance, Odysseus admits that he got involved in business dealings with a Phoenician, which could suggest that his own profit-driven engagement in commerce in some way karmically justifies his present circumstances. Also, he was not a true aristos by birth. In any case, the fact that it appears even in a lie implies that such a reversal of fortune was at least plausible, which in turn suggests that some degree of social mobility was possible, despite the attempts in the Odyssey to minimize it. 183 “It is through work that men acquire great flocks of sheep and become rich, And those who work are much more beloved by the gods… If you work, the idle man quickly will become envious As you grow rich: virtue and renown follow wealth…” WD, 308-313 There are several points in the Works and Days in which Hesiod praises the acquisition of wealth for its own sake, but in the above passage he makes it unequivocally clear that it is not heroic achievement that leads to wealth, as the Homeric epics would suggest, but rather that wealth is an achievement in and of itself, and that this achievement in turn leads to acclaim. What we see here is a radically different vision of what constitutes appropriate conduct, and what kinds of actions produce gains in status. The connection between wealth and status that Homer tries so hard to obscure has become explicitly acknowledged and actually encouraged. It would seem that for Hesiod, the elite grip on power is open to challenge on the basis of control over resources. In short, economic flexibility is creating a degree of social mobility that threatens the traditional power structure, and these class tensions are visible in the disquiet of the Homeric epics and the elite criticisms of Hesiod. 188 If we anticipate some of the arguments made in the Theognidea, one of the principal sources of social friction that would emerge later in the Archaic Period was the tension between birth/family status and wealth as determinants of social standing. It is here in Hesiod, and also to some extent in the Iliad and the Odyssey, that we can see the first outlines of that ideological conflict begin to take shape. As Rose (2012) 188 The elements of social criticism in Hesiodic poetry have long been recognized. For some, the anti-aristocratic elements in Works and Days are seen as a reaction to elite oppression through debt and the loss of land tenure (ex., É. Will, 1957; Detienne, 1963; Austin & Vidal Naquet, 1977). For others, the period was characterized by an erosion of elite power brought about by the growing importance of hoplite warfare (i.e., the so-called “hoplite revolution”; see in particular E. Will, 1967; for an influential contrary view, see Raaflaub, 1997, 1999; a recent overview of the debate can be found in Kagan & Viggiano, 2013) or the rise of the “yeoman farmer” (Hanson, 1995). For elite criticisms in the context of a broader tradition of anti-aristocratic thought, see Donlan (1973). 184 argues, in the Homeric epics we can see, along with the nascent beginnings of the polis, class tensions and competing class ideologies, particularly among the basileis. 189 At the same time, he sees in Hesiod an indictment of the basileis, especially in the arbitrary application of their power and in the malleability of wealth and status. I believe that Hesiod is giving voice to the concerns of a distinct class of free, land- owning farmers who have begun to differentiate themselves from other non-elites on the basis of their control over agricultural production, while also reaping the financial benefits of expanding maritime commerce. In so doing, they pose an increasing challenge to elite primacy, for several reasons. First, long-distance trade offers a means of acquiring wealth that is difficult if not impossible for the landed nobility to monitor and control. Second, the frank and open acknowledgement of the connection between wealth and status risks exposing the reality of wealth as a constituent element of elite standing These factors help to explain the antipathy in the Homeric epics to long-distance trade as well as the conspicuous absence of agricultural producers, and also foreshadows the social tensions that would only increase later in the period, as manifested in the Theognidea. It is perhaps for this reason that the presence of the basileis in the Works and Days is so restrained. While Edwards (2004) regards this relatively minimal interference in local Ascran affairs depicted by Hesiod as evidence of absence, he does not consider the possibility that Hesiod would have ulterior motives for minimizing the scope of their authority. Edwards himself notes that Hesiod appears to be one of a number of influential free farmers who are intent upon a greater degree of autonomy in their local affairs. If indeed Hesiod is arguing in favor of autarky, then it is not surprising to find that he would deny the importance of extra-local elites to the quotidian governance of Ascra. In the same way that the elite- centric Iliad and Odyssey wipe virtually all traces of free farmers from their social background, we should 189 Specifically, he sees the conflict between Agamemnon and Achilles as representative of two contradictory class ideologies, with Achilles representing a nostalgic view of the past when basileis earned their positions through merit, and Agamemnon as one whose power rests on the basis of inherited wealth and status in direct conflict with the welfare of the polis. Meanwhile, he sees Odysseus as an embodiment of status gained through accomplishments, in direct contrast with the indolent suitors, whose standing is entirely based on their birth. 185 be alert to the potential for Hesiod to eclipse those aspects of his society that are inconvenient to his goals. At the same time, we should be able to extend this dynamic of social tensions well beyond the limited borders of Ascra. As we have already noted, both the Works and Days and the Theogony had strong pan- Hellenic appeal, 190 and we must assume that the social conditions depicted in Hesiodic poetry had sufficiently widespread relevance to support this broad appeal. As a result, the strain on the social structure in Ascra that Hesiod describes is very likely a much more common and diffuse phenomenon in the archaic Greek world. Economic Activity in Hesiod Of course, if Hesiod is to be believed that changing economic circumstances had such pervasive political and social repercussions, then the natural next question to ask is what does Hesiod say specifically about economic activity? We have already seen the degree to which the Homeric epics appear to be consumed with establishing and ratifying an informal institutional framework within which certain forms of exchange are given the seal of approval and others are characterized as ethically deficient. However, this economic morality (or, if you prefer, moral economy) is strongly consistent with an elite outlook, and naturally favors economic activities that help perpetuate the prevailing social hierarchy. Therefore, we are left with several questions. First, how does Hesiod describe contemporary economic practices and institutions, especially those that seem to be under-represented in Homer? Second, does Hesiod also advocate a particular institutional framework? If so, how does it differ (if at all) from that which Homer promotes? Production Hesiod not only has more to say on this subject than on the other forms of economic activity (i.e., distribution and consumption), but it may even be argued that this topic is at the center of the Works and 190 See Marsilio (2002), Rose (2009), and especially Nagy (1982). 186 Days. As a result, the poem tells us a great deal about production, and agricultural production in particular, albeit in very general terms. 191 What develops is a fascinating picture of productive activities in the early archaic period. More than anything else, the Works and Days presents agricultural production as fundamental to daily life in Ascra. In a sharp departure from the Homeric representations, however, grain farming is a major focus of attention in the Works and Days, while other crops are mentioned only in passing, if at all. References to grain abound, both in general terms 192 and specific kinds of grains. 193 Without question, cereal cultivation takes center stage in the Works and Days. What is consistent with the Homeric texts, however, is the application of various forms of intensive agriculture, which we discussed in greater detail in the preceding chapter. In particular, the use of draft animals, especially oxen, appears to be common. 194 Also, slave labor is clearly essential for the properly functioning household in various capacities, including plowing, planting, and threshing the grain. 195 Also, while debates concerning the anthropology of agriculture continue to be vexing, 196 I will simply point out that the use of both draft animals and slaves is also attested in Homer, and the degree to which the farming techniques in Hesiod, although described at greater length, overlap and coincide with the descriptions of farming in Homer lends a degree of confidence in their reliability. However, in contrast to the testimony of the Homeric epics, the various crops that are mentioned so prominently in the Iliad and particularly the Odyssey are virtually absent from Hesiod. The only crop 191 Although the Works and Days is rife with advice on agricultural practice, the general consensus is that the advice offered in the Works and Days is so general that it cannot be seriously regarded as an actual handbook for farmers (see Isager & Skydsgaard, 1992; Amouretti, 1994; Most, 2006). 192 Various words used for grain: sitos at WD, 146; aktē at WD, 466, 597, 805; karpos at WD, 117, 172, 237, 576, 775. 193 Summer wheat (zeidōros aroura) at WD, 117, 173, 237; wheat (pyrophoros) at WD, 549. 194 Oxen as essential equipment for the oikos (436-437, 405-407, 795-797, 814-816); oxen associated with the plow (436-440, 580-581, 795-797, 814-816); mules are also mentioned (42, 794). 195 Setting out with both slaves (dmōes) to plow (458-462); have a slave follow the plow to plant seeds (469-471); slaves involved in the cultivation of vines (571-573); slave labor used in the threshing of grain (597-599). 196 Edwards (2004) is a strong proponent of Boserup’s (1965) influential population pressure model and its applicability to Hesiod, but Johnston (2003) offers a highly critical assessment of the oversimplification that reliance on Boserup’s model has produced. 187 aside from cereals that is mentioned with any regularity is the grape. 197 This is a striking contrast with the various descriptions of horticulture that are so prominent in the Odyssey, from the description of Alcinous’ lush gardens (VII.114-128) to the detailed inventory of Odysseus’ garden (XXIV.336-344), with pear, apple, fig, and olive trees, as well as grape vines. In fact, Hesiod mentions the wild fig (olynthos) but not the domestic fig (sykeē). But especially curious is the almost complete lack of any mention of olive cultivation. In fact, nowhere does Hesiod mention olive trees at all, nor the gathering or processing of the fruit. The only mention of olives in the entire text is in a passage in which he describes olive oil as a lotion. 198 The only other type of productive activity described at any length is the rearing of livestock. However, they are discussed almost exclusively in terms of their function as draft animals necessary for the cultivation of cereals. There is a brief mention of oxen being fattened indoors, 199 which suggests that they were occasionally sources of food, and some discussion of sheep shearing. 200 In addition, various items of clothing made from animal hide and wool are mentioned. 201 Thus, livestock were clearly sources of food as well as raw materials for clothing, as one would expect. However, the overwhelming predominance of passages in which animals are described is in the context of the value of their labor. Also, there are no mentions of pigs in Hesiod, whereas they appear frequently in both the Iliad and of course in the Odyssey. This is likely due to the fact that pigs have no value as draft animals, and therefore do not appear in the non-elite households that are the focus of Hesiod’s advice. Once again, this stands in sharp contrast to the descriptions of livestock in Homer, in which animals function as currency, a means of wealth storage, a source of food, and even as sacrificial offerings, but almost never as draft animals. 197 Pruning and trenching of vines (WD, 564-572); gathering and processing of grapes (WD, 609-617). 198 WD, 519-524. 199 WD, 452. 200 WD, 74-775. 201 Woven cloak (chlaima) and tunic (chiton), presumably made of wool (WD, 537-538); ox-hide boots lined with felt (WD, 541-542); rain slicker made from calfskin (WD, 544-545); wool cap (WD, 545-546). 188 This is not to suggest that there is no agreement between Homer and Hesiod. For example, both sources suggest that production was predominantly organized at the level of the household (oikos). However, in the Odyssey elite households are depicted as processing centers for agricultural products, embracing wool carding and weaving, wine-making, the pressing of olives, and the grinding of grain. 202 By contrast, such secondary production in Hesiod is almost completely absent, limited to the threshing of grain and the shearing of sheep. At the same time, it seems unlikely that the entire range of agricultural production mentioned by Hesiod (viticulture, cereal cultivation, livestock raising) would be practiced within a single non-elite household, especially if we accept the estimates of a typical household as described by Hesiod to consist of seven to ten permanent members (Tandy & Neale, 1996; Wees, 2009), and this even ignores the production of other agricultural goods that are largely ignored in the Works and Days. As a result, there seems to be no other option but to conclude that the non-elite households of the Works and Days were much more dependent upon an integrated system of local production than has been traditionally believed. In addition to various activities necessary for the processing of agricultural goods like the milling of grain, pressing of olives and grapes, etc., we must also include the services of craftsmen, including blacksmiths, potters, carpenters, and more. Therefore, it appears impossible to harmonize this evidence with the vision of the self-sufficient household that is central to the traditional notion of the archaic Greek economy as a “household economy.” The only possible conclusion that we can draw from these descriptions of farming and livestock 203 is that Hesiod’s representations of archaic agriculture are just as strongly characterized by a sort of forced perspective as are the Homeric depictions. As we have seen, in the elite-centric Homeric epics 202 Od., VII.114-128, XX.105-109, etc. 203 The degree to which agriculture and animal husbandry were integrated is a matter of continuing debate. Some contend that they constituted separate spheres of economic activity (Isager & Skydsgaard, 1992), while others argue for a more integrated system of farming and livestock rearing (Halstead, 1987; Tandy & Neale, 1996). Unfortunately, I can see nothing in the Works and Days that would conclusively support either position. 189 horticulture is emphasized almost to the complete exclusion of agriculture, whereas the opposite is true of the Works and Days. Similarly, the value associated with livestock in Homer is completely consistent with elite practices, while according to Hesiod they are used almost entirely in the service of agriculture. Whether this forced perspective is deliberate or simply a function of Hesiod’s point of view is difficult to ascertain. In the case of Homer, the exclusion of cereal cultivation does appear to me to be too systematic to be simple coincidence, whereas it is not so clear in the case of Hesiod, but this is perhaps not the point. What is more significant is the fact that the two sources present vastly divergent visions of contemporary productive activities, and it is clear that they represent the disparate priorities of markedly divergent social groups. Consumption Unfortunately, on this subject, the testimony of Hesiod is largely silent. On the rare occasions in which he does mention consumption, it is exclusively from the perspective of the dissolution of limited resources. As we know, Hesiod does not think in economic terms that are consistent with modern approaches. While that is not in and of itself an insurmountable barrier to analysis, it does mean that he often declines to mention certain aspects of economic behavior that are of particular interest to economic historians. This is particularly evident in the case of consumption, where Hesiod’s interest seems limited entirely to the consumption of non-durable or “soft” goods, particularly grain, and occasionally costs associated with hired labor 204 as well as the feeding of livestock. 205 Hesiod does spend some time describing the various tools and implements necessary for farming, 206 although he says next to nothing about how such tools are acquired. Some tools are apparently made at least partially within the oikos, as Hesiod describes the process of finding timber appropriate for 204 WD, 441-443; 559-560; 600-603. 205 WD, 559-560, 591. 206 Ex., WD, 424-436. 190 constructing a plow, but it is nevertheless assembled by a “servant of Athena” (Athēnaiēs dmōs). In fact, there is ample evidence to indicate that many other implements are produced by specialists. For example, we know that there is a blacksmith in Ascra, 207 and Hesiod also counsels his audience to assign certain tasks to a carpenter. 208 Unfortunately, the text is silent regarding the decision-making processes behind the acquisition of these items. We are told that craftsmen vie with one another, an indication that some are superior to others. If nothing else, this would indicate that consumers have choices available to them, but we have no information about how these consumption choices were perceived or made. Thus, we have no way of knowing what impact, if any, consumer choice played in production and distribution activities. Distribution Localized Exchange We have already seen that the Homeric epics offer a highly inflected vision of economic activity that paints commerce in general, and maritime trade in particular, in an ethically negative light. At the same time, highly ritualized exchanges of prestige goods between elite partners are presented either as normative, morally invigorating, or both. Local trade almost always takes the form of these reciprocal exchanges of prestige goods between elite partners which are motivated by the desire to forge and maintain relationships, or on occasion to enhance one’s social standing. The picture of trade in Hesiod, on the other hand, is vastly different, albeit frustratingly incomplete. For instance, there is virtually no mention of local trade in the Works and Days, despite the fact that we know that some goods were produced outside the household but within Ascra. As for long-distance exchange, Hesiod makes frequent references to maritime trading activity, 209 but is disappointingly vague regarding the specifics of the process, failing to describe the mode of exchange, the items exchanged, the places 207 WD, 493. 208 Have the woodcutter cut ship timbers and beams for building a room (WD, 807-809). 209 WD, 618-645; 663-677; 689-693. 191 where trade takes place, or even the trading partners involved. Nevertheless, I believe that what he does say, even if rife with lacunae, provides invaluable glimpses into many aspects of exchange, in particular the underlying logic supporting archaic trade. Concerning local trade, the little that Hesiod does say would seem to support the traditional view supported by Donlan (1994, 1997), Tandy (1997), and others that exchange was regulated by overriding concerns of social obligation and reciprocal relations. However, our understanding of institutions helps to lend nuance to this picture. In particular, I believe that Hesiod demonstrates the ways in which prevailing institutions can provide strategic opportunities for actors fluent in the rules of the game to manipulate them to their own individual advantage. A couple of passages from the Works and Days help to make this point clear: τὸν φιλέοντ᾽ ἐπὶ δαῖτα καλεῖν, τὸν δ᾽ ἐχθρὸν ἐᾶσαι: τὸν δὲ µάλιστα καλεῖν, ὅς τις σέθεν ἐγγύθι ναίει: εἰ γάρ τοι καὶ χρῆµ᾽ ἐγχώριον ἄλλο γένηται, γείτονες ἄζωστοι ἔκιον, ζώσαντο δὲ πηοί. “Invite your friend to feast, but leave your enemy be; But above all invite him who lives nearby, For if some urgent need should arise on your estate, Neighbors come ungirt, but in-laws dress themselves first.” WD, 342-345 εὖ µὲν µετρεῖσθαι παρὰ γείτονος, εὖ δ᾽ ἀποδοῦναι, αὐτῷ τῷ µέτρῳ, καὶ λώιον, αἴ κε δύνηαι, ὡς ἂν χρηίζων καὶ ἐς ὕστερον ἄρκιον εὕρῃς… καὶ δόµεν, ὅς κεν δῷ, καὶ µὴ δόµεν, ὅς κεν µὴ δῷ. δώτῃ µέν τις ἔδωκεν, ἀδώτῃ δ᾽ οὔτις ἔδωκεν. “Get good measure from your neighbor, and pay him back well, To the same degree or even more, if you can, So that if you may be able to rely upon him in the future if you are in need… And give to him who gives, and not to him who does not, For one gives to a giver, but no one gives to the non-giver.” WD, 349-355 192 Although the above passages seem to correspond with a certain kind of reciprocity involving social equals or near-equals, and even taking place within the ritual of the feast, the underlying logic of the exchange is vastly different from what we would expect from the idealized model. For one, there is no mention of the kind of items involved in the exchange, whereas in the Homeric epics as we have seen, the ritual of reciprocity typically demands specific items from a very limited range of objects, such as slaves, precious metals, textiles, metallic objects (tripods, drinking cups, bowls, etc.), and livestock. Granted, this could easily be explained as a non-elite imitation of elite practices, involving partners who are unable to afford the kinds of objects typical of elite reciprocities but who nevertheless wish to emulate the form of the exchange itself. However, there is one salient difference in these exchanges, specifically those mentioned by Hesiod are motivated by physical necessity, whereas those in Homer involve luxury items and are driven by concerns over status and prestige. This distinction between exchanges prompted by practicality and those motivated by status is a powerful one, and informs much of elite Greek attitudes towards economic activity. 210 Certainly, as Finley (1999 [1973]) has pointed out, there is a persistent current in Greek thought that disdains banausic activity as menial and ignominious, and exchange that takes place for the mundane goal of profit is juxtaposed in highly negative terms with exchange that is prompted by loftier motives of prestige. Of course, such an ethical frame inevitably favors elites who, by virtue of their preferential access to resources, are largely free from such concerns. 210 This negative characterization of behavior undertaken for the sake of physical need can be glimpsed, for example, in the numerous references in the Odyssey to the stomach (gastēr), which is almost invariably represented as a symbol for physical need, and which consistently drives men to take actions out of compulsion that they would otherwise reject as shameful and unseemly (see VII.215-224, XVII.226-232, XVII.286-289, XVII.470-476, XVIII.52-54, XVIII.357-364). I might also point out that several of these instances (XVII.286-289, XVII.470-476, and XVIII.52-54) are when Odysseus, disguised as a beggar, invokes the belly by way of explaining his actions as a result of compulsion rather than choice. 193 Not surprisingly, the non-elite Hesiod rejects this logic, and his representation of exchange explicitly embraces practicality as a driving force. It is clear from the text that the fundamental logic of these exchanges is different from the kind of generalized reciprocity described by Sahlins (1968). First and foremost, there is an overt consideration and recognition of a return. In fact, the “gift” itself is motivated entirely by the return, contrary to the basic premises of generalized reciprocity. Second, the calculation of status plays no part in the kind of exchange that Hesiod describes here. What is missing is the competitive element of ritualized gift-giving that is so vital to the broader analysis of Mauss (2002 [1923-1924]) as well as to the more particular arguments of Donlan (1989b, 1989c, 1993, 1994, 1997) and Tandy (1997), specifically relevant to archaic Greek society. According to these arguments, elite gift-giving incurs an obligation by the recipient to reciprocate, and an inability to do so results in a loss of status for the receiver and a gain in status for the giver. Thus, generosity is something that must be publicly displayed, and it is fundamentally motivated by considerations of status. Yet, there is nothing like that sort of rationale offered to explain reciprocal exchange in the Works and Days. Certainly, there is a sense that receiving a gift entails an obligation, but aside from that, we see absolutely no suggestion that the exchange is involved in the negotiation of the relative status of the partners. In fact, it would appear that reciprocity in Hesiod is conducted according to a very different set of rules. The outward appearance of the gift may be similar, but the principle behind it is far different. Instead of the elite gift-giving described in Homer which can be agonistic, 211 or at least disinterested, the reciprocal exchange in the Works and Days is explicitly represented as an investment with an expected return. From an institutional perspective, what we can see here is the development of a different strategy 211 For example, see the gifts offered to Achilles by Agamemnon that come with the explicit requirement that Achilles recognize Agamemnon as his superior (Il., 9.160-161), although we must keep in mind Agamemnon’s apparent social ineptitude, and thus the possibility that his attempt to use the system of exchange to extend his own political influence might be regarded as his clumsy attempt to usurp the institution of gift-exchange in an inappropriate way. 194 formulated in response to the prevailing institutional context. For Hesiod, who is fluent in the rules of the game, the ritual of gift-giving provides an opportunity to exploit them to his own advantage. 212 Long-Distance Exchange This brings us to long-distance trade, a topic that Hesiod mentions several times in the text. 213 Given the geography of the Mediterranean and the available transportation technologies, long-distance trade appears to be predominantly a maritime activity, as one would expect. Unfortunately, despite the numerous references to the movement of goods abroad by ship, Hesiod tells us very little about the actual mechanisms by which transactions were realized and maritime commerce was conducted, or the existence (if any) of maritime networks, but the lack of detail has left a great deal open to interpretation. 214 What we can safely say is that the kind of seaborne trade described in the Works and Days unequivocally corresponds most closely to Polanyi’s market mode of exchange. This is evident in the following lines: καὶ τότε νῆα θοὴν ἅλαδ᾽ ἑλκέµεν, ἐν δέ τε φόρτον ἄρµενον ἐντύνασθαι, ἵν᾽ οἴκαδε κέρδος ἄρηαι… “Then drag your swift ship to the sea, and put your Assembled cargo in it and make it ready so that you can bring home your profit…” WD, 631-632 212 For exchanges in which self-interest appears to be at least implied, see WD, 342-345 and also WD, 349-355. 213 WD, 45 (put the oar up in the smoke), 244 (Zeus extracts his vengeance on ships on the sea), 618 (best time for sailing), 629 (put the rudder up in the smoke) 630, 641 (best time for sailing and loading cargo for profit), 663, 678 (best time for sailing), 689 (warning not to put all one’s livelihood in ships). Tandy & Neale (1996) contend that the mentions of putting the oar or rudder up in the smoke refer to the practice of drying the wood out so that it will not rot. 214 Donlan (1997) and Cartledge (1983) minimize the frequency and importance of long-distance, market exchange, while Tandy (1997, 2001) recognizes the development of market exchange, but still denies the existence of what he calls “self-regulating” markets. For him, the self-regulating market requires that production can be shown to be responding to demand and that most people can be shown to depend on the market for their incomes. However, by neglecting the crucial role of institutions, he fails to realize that the self-regulating market is an illusion even (and especially) in the modern world. As a result, he is predisposed to regard archaic long-distance trade as something altogether different, conducted through ports of trade, or “one-time” markets, where in point of fact, there is nothing in the literature to support this thesis. 195 τύνη δ᾽, ὦ Πέρση, ἔργων µεµνηµένος εἶναι ὡραίων πάντων, περὶ ναυτιλίης δὲ µάλιστα. νῆ᾽ ὀλίγην αἰνεῖν, µεγάλῃ δ᾽ ἐνὶ φορτία θέσθαι. µείζων µὲν φόρτος, µεῖζον δ᾽ ἐπὶ κέρδεϊ κέρδος ἔσσεται, εἴ κ᾽ ἄνεµοί γε κακὰς ἀπέχωσιν ἀήτας. “As for you, Perses, keep in mind that all labors Have their season, but especially sailing. Speak well of the small boat, but put your cargo in a big one – The cargo will be larger, and there will be greater profit on top of profit, So long as the winds hold back their evil gales…” WD, 641-649 The message in these passages is clear: maritime commerce is carried out for the express purpose of making a profit, and as big a profit as possible. 215 Hesiod urges his brother to take to the sea precisely because it is profitable, although obviously not without risk. There is also not the slightest trace of any motivation from obligations of reciprocity, 216 nor any suggestion that the movement of goods is directed by any redistributive mechanisms. In fact, Hesiod’s free farmers appear to be remarkably free to dispose of their surplus as they see fit, without any outside influences (Edwards, 2004). There are, however, some indications that maritime trade was an enterprise that was undertaken when other options had been exhausted. For instance, Hesiod says at one point: εὖτ᾽ ἂν ἐπ᾽ ἐµπορίην τρέψας ἀεσίφρονα θυµὸν βούληαι χρέα 217 τε προφυγεῖν καὶ λιµὸν ἀτερπέα, δείξω δή τοι µέτρα πολυφλοίσβοιο θαλάσσης, οὔτε τι ναυτιλίης σεσοφισµένος οὔτε τι νηῶν. 215 It seems impossible to characterize this as anything but the kind of maximizing, rational agency that some have contended was not a feature of ancient trade. 216 There are those like Finley (1999 [1973]) who have explained maritime trade (at least in part) as motivated by a desire to maintain inter-elite connections. However, the shipwreck cargoes of the period are almost completely absent of the kind of prestige goods that are typically associated with ritualized gift-exchange. 217 Here again we see the appearance of the word chreos in a context that seems to indicate debt. It can mean “obligation” in a more general sense, but the fact that Hesiod recommends that his brother engage in trade in order to free himself of his obligations strongly suggests a financial aspect to the burden which would support the translation as “debt.” 196 “And if you ever turn your foolish heart to commerce, Wishing to break free of debt and miserable hunger, I will show you the measure of the thundering sea, Although I have no expertise in either ships or sailing…” WD, 646-649 And elsewhere, speaking of the endless benefits that accrue to the society that lives in accordance with justice, he mentions that: θάλλουσιν δ᾽ ἀγαθοῖσι διαµπερές: οὐδ᾽ ἐπὶ νηῶν νίσσονται, καρπὸν δὲ φέρει ζείδωρος ἄρουρα. “They flourish constantly with endless bounty; nor do they wander In ships, since the life-giving fields bear up fruit…” WD, 236-237 In both of these passages, the suggestion is that taking to the sea is a last resort. After all, Hesiod admits that he himself has done it only once, and not for trade. Nevertheless, the implication is that it is still an effective way to generate wealth, as Hesiod advises his brother to take to the sea in order to make an income, just as we are told their father once did. 218 Also, the passage that claims seafaring is something that righteous men have no need of is part of a more extended musing in which those who live by the principles of justice lead an idyllic existence, free from virtually all cares, and for whom the earth provides its bounty seemingly without any effort. 219 Thus, it seems clear that while Hesiod is well aware that seafaring brings with it some risk, such that there are safer ways to make a living, in the real world it is still a way to make an income, and most importantly, recalling the distinction that we made between the motives for exchange in Homer and those in Hesiod, maritime trade is undertaken out of physical need, although there is no suggestion that this concession to reality attaches any sort of shame. In fact, Hesiod 218 WD, 634-635. 219 A similar sentiment is expressed at WD 43-47, which states that had the gods not hidden the way to make a living (bios) from men, that they would neither work in the fields nor take to the sea. In this passage, it would seem that both seafaring and farming are equivalent, at least in so far as they are undertaken out of necessity. 197 exhibits none of the shame or embarrassment about engaging in commerce for the sake of turning a profit that we would expect if the depictions of mercantilism as unseemly in Homer were as universal as they are made to appear. Nowhere is there anything remotely similar to the disparaging tone of Euryalus regarding maritime trade, and this represents a major point of departure from the epics. In fact, the only caveat he makes is that trusting one’s cargo to the ocean involves a substantial degree of risk. There can be no mistake that we are observing a radically different economic morality here, in which trade is not stigmatized and profit is not some distasteful subject, but in fact quite the reverse. The confidence with which Hesiod assures his brother that overseas commerce will return a profit, however, has caused some concern among scholars. For example, Tandy (1997, 2001) sees this as evidence for the existence of what Polanyi (1963) calls “equivalencies,” or pre-arranged exchange rates for certain types of goods. Yet, I believe that Tandy’s predilection to see controlled markets in the ancient world as opposed to “free” or “dominant” markets in the modern world has colored his view, prompting him to propose the existence of equivalencies for which we have no actual evidence. It seems just as reasonable to suggest that this certainty of profitability might stem from the existence of maritime networks that were sufficiently developed to provide professional traders with adequate information about the most profitable markets for certain goods. In fact, the new colonies, especially in the west, which were often located in positions that offered excellent command of the sea rather than of nearby arable land, could very well have offered excellent potential markets for imported foodstuffs. In fact, other (admittedly later) literary sources support the profitability of maritime commerce. For example, according to Herodotus, the first Greeks to sail beyond the Straits of Gibraltar in the late 7th c. BCE were from Samos, and made a profit of 360,000 drachmas in a single voyage (Hdt. 4.152). To be sure, Herodotus mentions this figure precisely because it is unusually high, but it occurs as part of a discourse in which he is describing in more general terms the financial success of many voyages to the western Mediterranean, and he also points out that even the tremendous returns of the Samian traders didn’t match the profits made by Sostratus of Aegina. Of course, this only supports the intuitive premise 198 that profit margins must have been extremely high to support the risk associated with seaborne commerce. 220 Ultimately, the available evidence does not offer much support for the existence of equivalencies. In fact, the evidence would appear to suggest the effects of consumer demand on strategies of production and distribution, an element that Tandy (1997) claims is absent at this time. Sadly, we are told very little about what constituted the cargoes that Hesiod mentions. Given that Hesiod’s advice is almost exclusively directed to other free, landowning farmers like himself, it seems safe to assume that they are exporting grain, despite the objections of Snodgrass (1983), who rejects the idea there could have been a regular, long-distance, large-scale trade in grain in the archaic period. 221 Also, there is almost nothing to indicate what they received in return by way of profit. Given the presence of a blacksmith in Ascra, we might expect metals to be a desirable commodity. It is also possible that finished goods could have been taken in return, such as tools or certainly ceramics. Although pottery was certainly produced locally, the archaeological evidence clearly demonstrates the presence of imported ceramics in the region (Bintliff & Snodgrass, 1985; Bintliff, 1997). Nevertheless, the Works and Days specifically mentions only two items that are likely to have been acquired by Ascrans through exchange: olive oil 222 and biblinos oinos 223 , most likely from Thrace (Edwards, 2004). Unfortunately, we can only guess at the rest. Hesiod’s free farmers must have received something in exchange for their exports, but he is lamentably silent on this subject. Also interesting is the fact that Hesiod seems to regard it as commonplace for free landowners to take to the sea and even own their own boats. Despite his own confession that he has only once ever sailed on the 220 See Möller (2000); Bresson (2000); Wees (2009). 221 The question of the degree to which grain was imported by Greek cities, especially Athens, in the archaic period has been hotly debated; see Garnsey (1988); Arafat and Morgan (1994); Moreno (2007); Alston & van Nijf (2008). 222 WD, 522. 223 WD, 589. 199 open ocean, 224 we know (or at least suspect) that his father made his living through maritime trade, and there are numerous references in the text that give the impression that this was not unusual. Certainly, some passages could indicate trusting one’s produce to the ship of another, 225 but many others clearly refer to farmers gathering wood for ship-timbers and hiring carpenters to have ships built. 226 Also, in the passages in which Hesiod provides advice on the appropriate times to take to the sea, he explains the best way to store a boat during the times of the year when it was inadvisable to sail. This may even indicate that Hesiod assumes ownership of one’s own boat. After all, if farmers are entrusting their goods to the boat of another, then it seems highly unlikely that they would be responsible for maintaining the vessel during down times. The suggestion that they are in fact expected to maintain a boat implies that they are the owners. In any case, the frequency of passages mentioning ships and shipping is highly suggestive of the ready availability of maritime transport. Also, the confidence with which Hesiod assumes that maritime commerce will produce a profit 227 could be indicative of a developed maritime distribution network capable both of identifying the most advantageous markets and of delivering goods to those markets. After all, such profit-maximizing at the very least implies a high level of mobility of both information and of goods. Unfortunately, beyond simply presenting maritime commerce as a relative commonplace, Hesiod offers frustratingly few specific details as to how various processes are actually transacted. In fact, much of what Hesiod does say raises many other questions that he fails to answer. For instance, we do not know if Hesiod’s farmers bring their produce in their ships directly to consumers or perhaps to local markets where they are sold, or if they are transferred to professional traders who in turn resell the 224 WD, 648. 225 Ex., WD, 641-645; 665-666; 689-690. 226 WD, 622-625; 630; 671-672; see also 807-809 (the right time of year to build boats). Unfortunately, Hesiod does not describe the process of ship construction in anywhere near the kind of detail as we find in the Odyssey, aside from the fact that it apparently requires the services of a professional woodcutter to cut the planks. 227 WD, 630-635, 641-649. 200 produce either to other traders or consumers. We do not know if producers engaged in repetitive trade with the same trading partners in what game theorists would call an “iterated game,” or if each transaction involved a new trading partner and thus a new negotiation, nor do we know if prices were fixed or fluctuated on the basis of supply and demand. Likewise, in the event that markets existed to facilitate exchange, it is not possible to determine if these markets were fixed or mobile, permanent or temporary, etc. In fact, in spite of the frequent references to trade as a legitimate economic strategy, even to the point of dedicating all of one’s resources (WD, 689-693), many have discounted this activity as negligible. 228 However, there is no reason why trade would have needed to be “large-scale,” whatever that means, for it to have been an important, even transformative, economic development for free, non-elite landowners. There are several reasons why this could have been the case. First of all, the presumption that individual transactions must be high volume to be profitable is at best only partly true. To be sure, higher volumes per transaction typically have the advantage of reducing transaction costs, but this is only one single element in determining profitability. In fact, if by this time sufficiently sophisticated maritime networks had developed that were capable of distributing not only goods but information efficiently and rapidly, then such networks would have been able to increase returns both by reducing transportation costs and by targeting those markets in which demand (and there for price) was highest, thereby mitigating or even canceling out the higher transaction costs associated with smaller individual transactions. Also, scale is not simply a matter of the size of single transactions. Multiple sellers and buyers engaging in repeated, small-scale transactions over time are still capable cumulatively of generating a high volume of overall trade. Not to mention, this pattern of trade would have had the additional impact of producing a 228 The traditional view of Hesiod and his fellow farmers is that they exist on the edge of subsistence, under constant threat of starvation, and that exporting surplus produce was a rare occurrence; these scholars play down references to wealth as very much secondary to the supposed main message of the Works and Days, which is to prevent the household from falling below subsistence level or to aim for a marginal increase of wealth at best; see Cartledge (1983); Millett (1984); Tandy (1997); Edwards (2004); contra Starr (1974); Hanson (1995). 201 more diffuse impact, spread out over a larger swath of the population, which is a pattern entirely consistent with the broader social impact of maritime commerce that is a core feature of my argument. Finally, the returns from seaborne trade need not have been large on an individual basis to have had significant effects. We must keep in mind the lessons of the marginalists who demonstrated the impact of marginal benefit and marginal utility to our understanding of economic behavior. Put simply, the gains realized from long-distance trade need not have been great for their impact to have been extremely high. Trade-related profits could have, for example, constituted the difference between struggling for substance, perpetually at risk in the event of one or two years of poor harvests, and operating at a level that substantially ameliorated that risk from one year to the next, which is a dramatic change in circumstance. Perhaps free, non-elite farmers were typically dependent upon the local basileus in the event of a poor harvest, but that the additional income from maritime commerce reduced or eliminated that dependence. Such changing conditions would have had profound, far-reaching implications. Conclusions Institutions in Hesiod’s World Of course, we have already pointed out that Hesiod is no simple chronicler. The aims and ambitions of the Works and Days seem to go much deeper, to the point of establishing an ethical infrastructure, wherein certain activities are to be singled out for praise, and others condemned. This should not be surprising, especially in light of the fact that the Homeric epics pursue the same kind of goal. However, Homer depicts a moral universe favorable to the traditional elites, whereas Hesiod clearly represents a very different demographic and class perspective. As a result, the economic morality that he advocates differs substantially from that of the Iliad and the Odyssey. The moralizing quality of Hesiod’s so-called “didactic” poetry has long been recognized. We have already pointed out that his descriptions of agriculture are far too vague to constitute a practical guide to 202 farming, and the modern scholarly consensus is that both the Theogony and Works and Days serve distinctly ethical aims. 229 For the purposes of concision, suffice it to say that I concur with the majority on this issue, and I will operate under the basic premise that Hesiod’s writings are ethical treatises on some level. As such, the Works and Days offers a glimpse into both the central concepts that form the foundations of Hesiod’s moral outlook, and the particular instructions or advice that he offers on the basis of those assumptions. Specific Instruction In general, the majority of Hesiod’s advice is to his fellow free farmers. 230 There are some points at which he gives instruction to the basileis, but these rarely rise to anything more than exhortations to render straight judgements, and do not involve any specific reference to economic activity. However, Hesiod’s advice to the free farmers is far more detailed. We have seen any number of examples in the text that demonstrate Hesiod’s non-elite perspective, and on this topic he is no different. Whereas with the kings, his concern was generally limited to actions that affected the demos as a whole through their function as judges, with the farmers he describes a range of practices. In addition to numerous exhortations to engage in economic activities that he considers proper, there are also several specific behaviors that he cautions against. Broadly construed, the activities that Hesiod recommends are twofold. The first is hard work, presumably grain farming. The second is the pursuit of wealth. We have already seen several of the instances in which Hesiod encourages his brother to engage in hard work, and the benefits that accrue to it. As Edwards 229 Ex., West (1966, 1978, 1985, 1997), Heath (1985), Tandy & Neale (1996), Tandy (1997), Lamberton (1988), Nelson (1998), Edwards (2004), Burkert (2005), Osborne (2007), Ulf (2009), Konig (2010); notable contrary views have been expressed by Wilamowitz-Moellendorff (1962 [1928]) and Solmsen (1965); see Nelson (1998) for a brief survey of the debate. 230 I include those points in the text where Hesiod directly addresses his brother Perses with the advice to the free, landholding farmers in general. It is clear that he sees his brother as a social equal, and the exhortations to hard work, the pursuit of wealth, and especially farming are wholly consistent with the more general guidance on agricultural practices in the text. 203 (2004) explains, “Hesiod does not intend to instruct Perses in farming technique but to exhort him to lead a moral life by glorifying labor and its benefits.” More specifically, his advice is to emulate his own example by building and maintaining a functioning oikos. Hesiod’s advice throughout assumes that Perses will employ slaves and hired laborers (Wees, 2009). There is, of course, a range of functions that are involved with the operation of a household. Importantly, many if not most of them seem to involve the supervision of the labor of others rather than engaging in actual labor oneself. For instance, while Hesiod imagines the farmer himself driving the oxen (WD, 467- 469), the actual plowing is performed by a team of slaves (WD, 441-447), clothes are sewn by the women of the household (WD, 779), and slaves carry out the grain-threshing and storage (WD, 597-599; 607- 608). As van Wees (2009) points out, the commitment on the part of the farmer takes the form of energetic supervision of the laborers (rising before the slaves, reminding them of work to be done, issuing instructions, etc.). Thus, the strict work ethic advocated in Works and Days may involve manual labor, but also hands-on management. Indeed, once the various functions are taken into account along with the structure of the household, including the landowner, his immediate family, slaves, and hired workers, the oikos of Hesiod appears to function as a small firm, with multiple tasks integrated, organized, and supervised by management who delegate discrete tasks to various workers. It is this corporate structure that may partially explain Hesiod’s second goal, namely that of acquiring property and achieving greater wealth. We have already seen that at numerous points in the text, Hesiod explicitly encourages the pursuit of wealth as a goal unto itself. As he explains, πλούτῳ δ᾽ ἀρετὴ καὶ κῦδος ὀπηδεῖ. “Virtue and fame follow after wealth…” WD, 686 Here we have precisely the kind of sentiment that Tandy (1997) and Rose (2012) suggest is becoming more prevalent in Greek society. However, I maintain that this sentiment may have always been popular 204 among non-elites, for whom changes in financial circumstances were the only possible means to effect real changes in one’s social standing. In a society where birth is the sole basis for status, there is no means to improve one’s position within the social hierarchy. However, in a society in which wealth is valorized, birth becomes less of a barrier to advancement. To express it in neo-institutional terms, it is in the best interests of non-elites to promote institutions that 1) promote means of achieving wealth that are more open to the demos, and 2) promote a value system that more strongly connects status with wealth. In any case, this sentiment is particularly significant not only because it reinforces the various other passages in which acquisitive behavior is encouraged, but because it represents a pragmatism that diverges sharply from the attitude towards wealth that is reflected in the Iliad and the Odyssey. For the Homeric elites, wealth is a matter of privilege, and the pursuit of material gain is regarded as unseemly. In addition, poverty is a mark of shame. However, Hesiod and his brethren, who were not independently wealthy, did not have the luxury of being so openly scornful of material gain, and more importantly, recognized its potential to facilitate gains in social standing. Make no mistake, though – Hesiod does not advocate obtaining wealth by any and all means. As a matter of fact, he is insistent that there are appropriate means of pursuing wealth such as trade and especially farming, but that there are also many strategies that he strongly condemns. As Tandy (1997) notes, for Hesiod there is good wealth and bad wealth; both types can attract good standing in the community, but wealth must be properly acquired to deserve admiration among one’s neighbors. 231 Improper behavior includes deception, 232 false oaths and false judgments, 233 and the use of force. 234 Perhaps most importantly, Hesiod condemns idleness. In the passage concerning the difference between the two kinds 231 Ex., WD, 352. 232 WD, 190-195. 233 WD, 219-224; see also 280-285, 320-326. 234 WD, 320-326. 205 of strife, idleness comes from the bad Eris, while the good Eris rouses men to work. 235 Of course, famine is the companion of the idle man, but even more than that, the man who does not work is hated by the gods themselves. 236 The result of this position is a self-reinforcing, circular argument. If wealth is a marker of virtue, then poverty is a marker of immorality, and so wealth becomes its own justification. If I am richer, I am better, and I am richer because I am better. In an interesting way, Hesiod has presented his own form of essentialist discourse. Just as the Homeric epics connected the control of resources with the inherent nobility of the aristocracy, Hesiod connects wealth with the inherent nobility of hard work. In principle, what Hesiod is doing is no different than Homer, that is, advocating for specific moral values attached to particular forms of economic activity; the only difference is the activity that he promotes. The Works and Days as a Challenge to Elite Primacy We have already seen many of the ways that Hesiod represents a social background that is substantially different from that which appears in the Homeric epics. The focus on grain farming, the absence of any stigma associated with overseas trade, and the frank and open endorsement of the profit motive (when conducted appropriately) all represent significant departures from the society depicted by Homer. 237 In addition, the elements of social criticism in Hesiod are well known and widely acknowledged. Can we then include the Works and Days in the tradition of anti-arisocratic thought in archaic poetry described by 235 WD, 20-28. 236 WD, 302-305. 237 It is true that the word kerdos that I have chosen to translate as “profit” where it occurs in Hesiod also appears in Homer, especially in the Odyssey, where it can mean “cleverness/wit” (II.118, XIII.299, XVIII.216), “plans/schemes” (II.88, XXIII.217), or simply “benefit/advantage” (XVI.311, XXIII.140, 10.225). There is no doubt that self-interest and the ability to look out for one’s own advantage was an admirable trait among the elites in Homer. However, for economists, “self-interest,” and in particular “rational self-interest,” have a specific economic sense. In Hesiod the term kerdos clearly has this same sense, in which the notion of gain is grounded in the degree of material advantage one derives from a commercial transaction, hence “profit.” It is this particular sense of kerdos in connection with material benefit that is disparaged in Homer, and most especially when the transaction is conducted deliberately with the objective of maximizing the concrete gain (i.e., profit-taking). By contrast, Hesiod frankly and openly advocates precisely this type of maximizing behavior. 206 Donlan (1973)? To do so should not require evidence of direct and outright challenges. As Scott (1976) has demonstrated, subaltern groups frequently engage in more covert forms of resistance in the face of a dominant authority, and the Works and Days often seems to straddle the line between veiled criticism and direct confrontation. If nothing else, Hesiod clearly presents a picture of contemporary society that is distinctly separate from the elite experience, what Ulf (2009) calls “daily life stripped of heroism.” Without question, the Works and Days recognizes that the basileis are capable of contributing significantly to the well-being of the polis, but at the same time, the text demonstrates a keen awareness of the potentially negative consequences of elite power. Various elements in the text would seem to support this premise. For example, Hesiod offers a concept of competition that is far different from the heroic mode of competition. There are none of the traditional hallmarks of elite rivalry. There is not a single mention of athletic contests or raiding. The only descriptions of warfare represent it solely as a cause of suffering and destruction, and even gift-giving loses its agonistic aspect. Instead, an economic competition, with the pursuit of wealth at its center, replaces the heroic mode of competition. Along with this pursuit of wealth is the ambition for social mobility brought about through the acquisition of wealth. While it is not explicitly characterized as a challenge to the elite, I believe that this alternate form of competition constitutes a de facto opposition to that of the heroic code because the contest that Hesiod describes is fundamentally and explicitly formulated in terms of material gain. We have seen how Hesiod valorizes the acquisition of wealth, and at the same time the Iliad and Odyssey strive to obscure the connection between wealth and status. Thus, Hesiod’s advocacy of the pursuit of material gain as an avenue to gaining status is not just a non-elite alternative to heroic competition, but it is in direct opposition to the Homeric emphasis on status divorced from wealth, and even more so because the means by which Hesiod suggests that wealth can be pursued is through deliberate profit-taking, a notion that is diametrically opposed to the value system evident in the epics. Therefore, Hesiod’s concept of competition is not just 207 an alternative to the heroic model, but is directly opposed to the Homeric value system, an constitutes a fundamental philosophical difference. The Contribution of Hesiod Certainly, one of the historiographical virtues of the Works and Days is the text’s capacity to provide a direct counterpoint to the elite narratives in the Iliad and the Odyssey. All too often, scholars highlight the differences between the respective representations of archaic society as a way of fleshing out our picture of the Greek world at the end of the 8 th and beginning of the 7 th c. BCE. At the same time, we must recognize the ideological impetus behind Hesiod’s description of his contemporary world, and appreciate its importance as an attempt to participate actively in archaic Greek society. We know that the Works and Days offers a non-elite perspective on the archaic Greek world. We also know that the text presents a very different picture of contemporary economic activity, with agriculture at the center of production, and maritime trade as an apparently commonplace means of distribution. Thus, the mechanical aspects of archaic economic behavior are described very differently by Hesiod than by Homer, who emphasizes livestock raising and horticultural production, with ritualized gift-giving as the most prominently featured means of distribution. This much is clear. At the same time, we see indications of underlying potential social and class tensions. For one, Hesiod is at times openly critical of the failure of basileis to fulfill their duty as judges who choose instead to abuse their positions of power to serve their own selfish ends. More significant, however, are the indications in the text of a growing sense of class identity among the non-elite free landowners, who identify neither with the kings, nor with other groups such as landless laborers, craftsmen, or merchants. Instead, they recognize interests common to each other, and even appear to have developed novel strategies in response to prevailing institutions to further their own ends. What does seem difficult to reconcile with this behavior is the notion that Hesiod represents a peasant class. In fact, there are several aspects of the society that Hesiod describes that seem inconsistent with 208 traditional definitions of peasantry. 238 The first is the autonomy that Hesiod and those whom he advises seem to have in disposing of their surplus goods as they see fit. As Hall (2007) points out, one of the defining features of peasant societies is that they owe obligations to the state. However, there is no mention or even suggestion of interference from the basileis on this topic. Considering how outspoken Hesiod is regarding what he sees as inappropriate behavior on the part of the basileis, and how little he hides his resentment that they would interfere in his family dispute, it is prima facie unlikely that he would remain silent if they were to exert significant influence over the exports from Ascra. Another aspect that is problematic in characterizing Hesiod as a peasant is the apparent frequency of hired and slave labor. In general, peasants rely on the resources of their own household, although on occasion they may occasionally employ restricted slave or wage labor, but this is rare, and typically occurs only sporadically. However, all indications are from the text that Hesiod’s household includes several slaves and employs wage laborers throughout the year. Finally, peasant societies are typically subsistence-based, whereas Hesiod’s farmers aim specifically at production for surplus with an intent to trade that surplus and generate profit. It is clear that Hesiod’s goal is something well beyond subsistence production. In general, the view of Hesiod’s engagement in maritime trade has been that he sells his surplus agricultural products (bios), which given the mercurial nature of agricultural production would imply an infrequent, sporadic participation in maritime trade. However, there are several indications in the text that suggest to me that Hesiod’s production strategies were far more ambitious than simple subsistence. For example, Hesiod advises his brother: µηδ᾽ ἐν νηυσὶν ἅπαντα βίον κοΐλῃσι τίθεσθαι: ἀλλὰ πλέω λείπειν, τὰ δὲ µείονα φορτίζεσθαι. 238 Admittedly, there is widespread disagreement as to what constitutes a peasant society, what characteristics are most important, etc., although Hall (2007) provides a synthesis in which he provides several features that seem common to nearly all definitions of peasantry; see also Wolf (1966, 1969); Shanin (1973); Bernstein & Byres (2001); Bernstein (2003); Edelman (2013). 209 “Do not put all of your livelihood into ships – Leave the greater part behind, and put the smaller part on board…” WD, 689-690 There is nothing here to suggest that the agricultural produce that Hesiod recommends shipping overseas is conceived of as surplus. There is no indication that the bios that he refers to is unusual or irregular. To be sure, elsewhere in the text Hesiod does warn of the dangerous possibility of famine and recommends that steps be taken to ensure an adequate food supply from year to year. But in the above passage, the implication is that Hesiod automatically assumes that his brother will have some produce to sell, and his only admonition is not to sell all of it overseas because of the risk involved. I believe that this suggests a production strategy deliberately aimed at production specifically for the purpose of trade, above and beyond subsistence strategies. This notion is further reinforced when Hesiod recommends scrupulously sacrificing to the gods to ensure their favor, so that (among other things): ὥς κέ τοι ἵλαον κραδίην καὶ θυµὸν ἔχωσιν, ὄφρ᾽ ἄλλων ὠνῇ κλῆρον, µὴ τὸν τεὸν ἄλλος. “So they will be well-disposed towards you in heart and mind, In order that you may buy the farm of another, and not they yours…” WD, 340-341 Placed in the context of the other passages in which Hesiod speaks approvingly of efforts to become wealthy, it would seem here that he is advocating a production strategy that deliberately aims to acquire productive assets well beyond what would be needed to sustain one’s own household. Instead, he is promoting strategies engineered for the sake of consistent annual surplus that can be traded overseas. 239 239 The mechanisms by which these exchanges might have taken place in the absence of coinage are admittedly unclear. However, as Kim (2001) and Peacock (2013) have argued, archaic Greek society was not without various institutional arrangements that performed the essential functions that define money (unit of account, means of exchange, storage of wealth). Although Hesiod is largely silent about these aspects of archaic economic practice, the casual nature in which he treats trade is suggestive of the fact that such arrangements were well established and did not pose significant barriers to trade. 210 Thus, Hesiod and his other free farmers are much more representative of a sort of petit bourgeois than of a peasant society. Perhaps the biggest potential source of class tension, however, is the drive for increased personal wealth that Hesiod encourages. While critical of the abuses of the privileged basileis, the Works and Days offers the middling farmer represented by Hesiod a potential path to social advancement. Especially when considered along with the anti-elite current that runs through the work, there seems little doubt that the Works and Days represents at least the ideological foundations of a challenge to the primacy of the traditional aristocracy of the late 8 th /early 7 th c. BCE. At the same time, it seems very possible that elites would perceive this as a threat to their position, and this may help to explain the emphasis in the Homeric epics on validating the social hierarchy, and most notably elite prerogatives, through the advocacy of specific economic strategies. For Hesiod, profits (kerdea) are beneficial, because they bring virtue (aretē) and honor (kudos). But even more than the pure instrumentalist advantage of profit, there are good profits and bad profits, and their moral quality stems from the means with which they were acquired (Tandy, 1997). Ultimately, from an institutional perspective, what we are witness to in the Works and Days is the promotion of an institutional structure, a set of rules of the game, that govern economic activity. In and of itself, this is not terribly surprising, since it is a basic tenet of neo-institutional economic theory that economic behavior is always conducted within the context of institutions. What is significant from the perspective of our understanding of the history of the early archaic period is that we can compare the institutional system promoted by Homer to that advocated by Hesiod. When we do so, the picture that emerges is one of an institutional struggle in which particular forms of economic behavior (ritualized gift-giving, cattle raising, raiding, horticulture, etc.) that validate the prevailing power structure are valorized by the elite-centric Homeric epics, while other activities that are not so inherently favorable to the social hierarchy are condemned. Meanwhile, the non-elite moral agenda of the Works and Days promotes forms of production 211 and distribution (grain farming, maritime commerce) that are either ignored or disparaged by Homer, specifically because they represent a means of social advancement. As North (1990) has pointed out, during times of shifting power relations we often see attempts to re- draw social contracts and to re-negotiate the rules of the game. The competing economic strategies of Homer and Hesiod suggest that this is precisely what was happening in the early Archaic Period. According to Tandy & Neale (1996), autarky and kerdea are used as cultural and economic weapons in the Works and Days. The use of these weapons, and the apparent response to these perceived threats in the Iliad and the Odyssey, indicate the beginnings of a contest between the traditional aristocracy and the emergent class of free land-owning farmers. 240 As part of this contest, different forms of economic activity, as well as divergent economic ideologies and informal institutions that supported them, were central to the strategies of the contestants. As we shall see, this conflict would continue to reverberate and play out over the next two centuries. 240 I believe that this incipient struggle helps to explain the suppression of non-elite independent farmers from the narrative of the Homeric epics. It is this elision that Finley (1954) takes as evidence of absence and leads him to posit that the characteristic feature of Homeric society was a “deep horizontal cleavage.” For a more thorough examination of the shortcomings of Finley’s argument, see Thalmann (1998). 212 Chapter Six: The Theognidea Introduction So far, I have argued that the Homeric epics and Hesiodic poetry offer a glimpse into the early stages of a contest between traditional landed nobility and an emerging class of free, land-holding arrivistes. While this class tension undoubtedly has a distinctly economic aspect, it also incorporates a significant prescriptive element that is manifested in the specific rules that are promoted in the texts. Each text represents on some level an effort to establish a set of institutions that govern economic practices, and in particular, advocate for specific forms of economic activity that are most conducive to the respective groups. But if what we see in the poetry of Hesiod and the Homeric epics represents the initial phase of this class struggle, the natural next question is how this contest played out and resolved itself. In order to gain some insight into this aspect, the literature of the later Archaic Period is an obvious source. However, it is not feasible to conduct an extensive survey of the entire body of archaic Greek poetry. For that reason, I have chosen to focus on the body of elegiac poetry attributed to Theognis of Megara. I selected this corpus for several reasons: first, I believe that the corpus is reflective of a more broadly relevant elite viewpoint than simply that of the Megarian aristocracy, and is therefore more pan- Hellenic in its expression of elite values; 241 second, because the collection of poetry was compiled at a later date from various poetic fragments that were likely written largely over the course of the 7 th and 6 th 241 Theognidean poetry contains passages elsewhere attributed to other poets, but these are more than likely indicative of an assimilation of poetic traditions joined by common ideology (Cobb-Stevens, Figueira, & Nagy, 1985). In addition, the fact that they are written in the Ionic of elegiac poetry rather than in the Doric dialect native to Megara suggest wider diffusion. As a result, the Megara described by Theognis is more accurately a generalized Greek polis, reflecting the concerns and socio-political developments that were relevant to archaic poleis in general rather than events specific solely to Megara. At the very least, then, Theognidean poetry represents a “pan- Megarian” (i.e., relevant to Attic Megara, Sicilian Megara, and all of the associated colonies of both poleis) perspective (Figueira, 1985), and as Nagy (1982, 1985) argues, the very survival of the poetry indicates widespread, even pan-Hellenic applicability. This is further supported by the poetry itself, in which the author asserts his aim for his verses to be known and acclaimed by all men (Th. 23-24). Thus, we must regard the poetry of Theognis as describing events of broad significance. As Donlan (1973) explains, "the Theognidea may be seen…as an organic work-in-progress, reflecting the ongoing discourse within the pan-Hellenic aristocracy over their place and role in a changing society…" 213 centuries BCE, 242 and thus offers an amalgamated view of the changing socio-economic conditions during the Archaic Period; 243 and third, because the text speaks overtly about economic issues, in particular the complex relationship between ancestry, wealth, and status. In short, the poetry offers a perspective that is both geographically and temporally diffuse, while at the same time containing substantial material directly relevant to the class tensions observable in the Homeric and Hesiodic poetry. Also, the very size of the corpus constitutes an advantage for analysis from the perspective of social history. Consisting of nearly 1,400 verses, Theognidean poetry represents a sample size sufficiently large to form a reliable basis for analysis. Granted, one could expand the range to include a broader sampling of archaic elegiac or even lyric poetry, but this would exponentially increase the complexity of the investigation, particularly in terms of analyzing the ideological perspectives of each poet. In fact, the Theognidea represents one of the only collections of archaic poetry (aside from Homer, Hesiod, and the Homeric Hymns) to survive as a single corpus instead of fragments. And this brings me to my fourth and final point: the ideological consistency of the Theognidea. The poetry of Theognis can at times offer contradictory advice, recommending on the one hand duplicity and then on the other hand condemning deceit, for instance, but in its broader themes it retains a persistent cohesiveness. 244 In fact, in spite of many disagreements on various other aspects of Theognidean poetry, there is general agreement that the poems in the surviving text were included largely because they were congruent with a particular prevailing aristocratic outlook, and we should read the corpus as the product of an active engagement 242 As Cobb-Stevens et al. (1985) argue, verses 29-52 portray a political situation in the polis of Megara seemingly analogous to that prior to the rise of the tyrant Theagenes (@640-600 BCE), while verses 891-895 seem to correspond to war with Euboea in the second quarter of the 6 th c. BCE, and verses 773-782 refer to the Persian invasion of Megara in 479 BCE; see also Nagy (1985). 243 It has long been a point of almost universal agreement the text comprises a multi-author collection of archaic elegies (Figueira, 1985). Although some like Edwards (2004) discount the ability of the Theognidea to provide a reliable and uninterrupted connection between the social backgrounds described by Homer and Hesiod and that evident in the poetry of Solon, the date range of the verses suggests otherwise. 244 See in particular West (1974); also Fox (2003), van Wees (2003), and Lear (2011). 214 with contemporary social and political currents, unified by their ideological outlook. 245 It is for all of these reasons, then, that I have chosen to focus on the Theognidea as the principal complement to the Homeric epics and the poetry of Hesiod. As before, I wish to pay attention to the rules that are described in the text, especially those rules that apply to economic conditions. But in this chapter, my focus is slightly different. Whereas with the chapters on Hesiod and Homer, I wished to establish the outlines for nascent class tensions manifested in opposing systems of economic rules, it is my intent here to use the Theognidea as a source to understand how this contest developed. As a result, my emphasis will be on the evolving social hierarchy, and particularly the ways that changing economic conditions seem to have played a role of that evolution. In so doing, I realize that the Theognidea represents a particular elite perspective as well as a specific agenda. Therefore, my goal is to examine first how the text describes the current social conditions, then the argument that the text seems to be making, and finally, how we should interpret this in the context of what we already know about the elite bias of the text and history of the period. Theognidean Questions As always, every text presents unique challenges and raises particular questions. This is certainly true of Theognidean poetry. For example, as Campbell (2003 [1967]) claims, “The field of Theognidean studies is battle-scarred, strewn with theories dead or dying, the scene of bitter passions and blind partisanship.” Numerous issues are highly debated, and I will mention just a few. These include issues related to genre, the historicity of the poet himself, and the historiographical value of the text. To be sure, the question of 245 West (1974) is the most influential of those who regard a core of the collection (particularly verses 19-254, 319- 372, 805-822, 1171-1884b) as having been written by an actual Theognis, referring to those verses as the “Theognis blocks”; see also Fox (2003) and van Wees (2003). He also suggests that some of the passages in the Theognidea are excerpts from Solon and other elegiac poets (Ex., 153-154 = Solon fr. 8.3-4 GP; 227-232 = Solon fr. 1.71-76 GP; 315-318 = Solon fr. 6 GP; 585-590 = Solon fr. 1.65-70 GP; 719-728 = Solon fr. 18 GP; 795-796 = Mimnermus fr. 12 GP; 1003-1006 = Tyrtaeus fr. 9.13-16 GP; 1017-1022 = Mimnermus fr. 12 GP). The prevailing counter-opinion is most prominently expressed by Nagy (1985), who recommends that we abandon altogether the notion of Theognis as a historical figure. For a recent overview of the debate, the lines of which have actually changed very little, see Lear (2011). 215 genre is an important one, but touches on an enormous range of topics. 246 I accept the common view that elegy bears strong connections to the Greek symposion 247 which was, at least at the time of the Theognidea, a fundamentally elite practice. 248 The practice changed significantly over time, but it was essentially an aristocratic phenomenon and played crucial role in establishing and validating an individual’s place in the social hierarchy of his local community (Wecowski, 2014). There is also a significant amount of debate surrounding the existence of the eponymous Theognis. Even in antiquity, questions surrounded the identity of the poet, including his place of birth (Sicilian Megara or Attic Megara), and on the basis of the limited evidence they may very well be irresoluble. 249 As Donlan (1985) argues, no truly precise historical statements can be made about Theognis the poet. As Cobb- Stevens et al. (1985) point out, the figure of Theognis is perhaps best understood as a cumulative synthesis of Megarian political tradition, a self-representation of Megarian aristocrats. As a result, I believe that this makes the Theognidea even more valuable as a source for social history, since it speaks for an entire group, and therefore cannot be dismissed as simply the viewpoint of a single individual. As for the question of Theognis as a source for Megarian history, both Fox (2003) and van Wees (2003) make cogent arguments for Theognidean poetry as a source for contemporary class relations and ideology, with Fox preferring to see an early 6 th c. BCE date for the “Theognis blocks.” 250 I follow the arguments of Figueira (1985) that the corpus lacks a strong historical affiliation with either Sicilian or 246 West (1974) remains a classic study of Greek elegy, and Barron, Easterling, & Knox (1985) is a useful introduction. More recent surveys include Gerber (1997), Irwin (2005), Aloni & Iannucci (2007), and Aloni (2009). 247 See Levine (1985); Bowie (1986). 248 This view is most strongly associated with the work of Murray (1990). Catoni (2010) offers a thorough review of some of the more recent scholarship along with an attempt to combine textual with iconographic and archaeological evidence. 249 Even among ancient authors, biographical details about Theognis were controversial, with conflicting views regarding the dates of his birth and death as well as his actual birthplace. For example, Plato (Laws 1.630a) claims that he was from Megara Hyblaea in Sicily, a point on which the Suda (2.692.13) concurs, adding also that he lived in the middle of the 6 th c. BCE. However, Harpocration (Lexicon of the Ten Attic Orators) claims that he was from Megara in Attica, while the Scholiast balances both views, suggesting the possibility that he was born in Attic Megara and then later in life moved to Sicilian Megara. 250 For a contrary view, see Okin (1985). 216 Attic Megara precisely because it was meant to be more broadly relevant, and the arguments of Nagy (1985), who emphasize the pan-Hellenic qualities of the Theognidea rather than focusing on its narrow applications to the history of a single polis. Theognis on “Megara” As it turns out, “Theognis” 251 has a great deal to say about the social, political, and economic conditions of the world in which he lived. He is often deeply disturbed by what he perceives as the collapse of contemporary society that has, in his view, foregone the values and norms that are essential to a smoothly functioning polis. As a result, he frequently offers trenchant criticism of those whom he holds responsible for the decline of his society. In particular, I wish to examine the various verses that express three interrelated, recurring themes 252 throughout the text: first, the concern for a properly ordered polis and what elements are necessary for that to exist; second, the intensely pessimistic view of the current state of affairs; third, the view that economic factors, in particular the perception that wealth has become as important as birth as a determinant of social standing, have played an integral role in the collapse of society. At the same time, the Theognidea offers little in the way of direct commentary about specific economic practices. In contrast to the Homeric epics or the poetry of Hesiod there is almost no mention of specific economic processes (production, consumption, distribution) in Theognis. This is perhaps explained by the strong association of Theognidean poetry with the symposion, in which “non-economic” aristocratic modes of expression and identification play a prominent role (Wecowski, 2014). This may also be due, at least in part, to the idea that how an aristocrat makes his money is considered an unseemly topic for 251 Over the course of this chapter, I refer to the to the literary figure of Theognis, under whose name the collection of poetry was compiled, as an individual. This is more a matter of linguistic convention than a reflection of a belief in a historical Theognis. I maintain that the corpus of Theognidean poetry is united by a shared perspective and a set of values held in common, and it is simply more convenient to refer to the author of any given passage as Theognis. 252 This is not to say that there are not several other important recurring themes in the text, such as the dynamics of homoerotic relationships in the pederastic poetry, or proper drinking behavior, etc. 217 conversation among elites. In any case, it means that we cannot rely much on Theognis as a source for detailed information about specific contemporary economic strategies. The Good and the Bad First and foremost, in order to understand what Theognis says about a properly functioning society, we must remain acutely aware that the poetry represents a deeply aristocratic outlook. One might even say that drawing the distinction between the “noble” and “base” in Megarian society constitutes the single most prominent theme in the entire text (Cobb-Stevens, 1985). For example, Gerber (1999) notes that the words for “noble” (agathos, esthlos) and “base” (kakos, deilos) appear so frequently that it is not worth identifying each and every instance. In addition, these terms almost invariably appear in conjunction with one another as indications of social status. It is abundantly evident that the identification of and distinction between elites and non-elites in Megarian society is an overriding concern for Theognidean poetry, and that the aristoi define themselves in large part in relation to the kakoi. In the Theognidea, the poetry itself makes explicit claims to its function as an instrument in the service of an elite agenda: ταῦτά µοι ᾐνίχθω κεχρησµένα τοῖς ἀγαθοῖσιν: “Let these verses speak covertly to the nobles…” Th. 681 As Nagy (1982, 1985) explains, Theognis presents his verses as a riddle, comprehensible only to the nobility who are sufficiently versed in poetry to grasp their true meaning. Thus, the Theognidea does more than simply extol the virtues of the agathoi. The text itself is a device that is designed to distinguish the noble from the base, and the poet proposes to use his own poetry to serve as a means of instruction. Thus, it should come as no surprise to see throughout the text admonitions to avoid the kakoi at every 218 opportunity, 253 to associate with the agathoi 254 as much as possible, and to learn from them just as the poet himself did as a child. 255 Theognis views the connection between status and class in a way that we should now easily recognize as typical of archaic elites. Indeed, he makes arguments very similar to what we have seen in the Homeric epics in describing the qualities necessary for a man to be considered agathos. As Cobb-Stevens (1985) points out, these are: genos (“birth”), wealth, and aretê (“virtue”). However, these are necessary, not sufficient, conditions for aristocratic standing. Being born to an elite genos is not enough to guarantee elite status. As for wealth, it is something that a man has, as opposed to birth, which is something that a man is, but it is no more a guarantor of elite status than birth, and it must be both obtained and used in socially appropriate ways for one to be considered an agathos. As for true excellence, this is an immanent quality, revealed through a man’s actions: οὐδεὶς γὰρ πάντ᾽ ἐστὶ πανόλβιος: ἀλλ᾽ ὁ µὲν ἐσθλὸς τολµᾷ ἔχων τὸ κακὸν κοὐκ ἐπίδηλος ὅµως: δειλὸς δ᾽ οὔτ᾽ ἀγαθοῖς τις ἐπίσταται οὔτε κακοἰσιν θυµὸν ἔχων µίµνειν. “For no man is blessed in all things, but the noble man Stands strong even in the face of hardship, nor does he show any outward sign; The ignoble man does not know how to control his emotions Either in good times or in bad… Th. 441-444 253 Ex., 69-72, 101-102, 113-114. 254 Throughout the Theognidea, the terms kakoi and agathoi have a distinct class component in addition to their more general meaning of “good” and “bad.” It seems clear that Theognis tries to use them as terms of ethical valuation and as status markers. However, he consistently argues in favor of a social hierarchy in which control of resources is (or rather should be) mapped onto his ethical valuations (i.e., agathoi deserve to be wealthy, and kakoi do not), and that he recognizes the potential of wealth to affect status is beyond doubt. At the same time, Theognis, much like Homer, tries to efface the degree to which wealth and status overlap, and thereby elide the degree to which the status of the agathoi is contingent upon their preferential access to resources. 255 27-28. 219 For Theognis, the agathos must outwardly display the inner qualities that he is supposed to embody, including fortitude, moderation, equanimity, trustworthiness, and more. However, the clear suggestion is that the outward behavior is a reflection of the inner superiority of the aristocrat, and his actions merely confirm the unassailable reality of his preeminence. Meanwhile, the same principle holds true for non- elites, and the qualitatively inferior kakos will inevitably reveal himself as well. In truth, because quality is ultimately an inherent feature, the kakoi inevitably will find themselves unable to measure up to these standards, regardless of outward trappings. For Theognis, the underlying conditions of status are immutable, and there is no means by which a base man can become noble. In a prototypical display of class essentialism consistent with an aristocratic outlook, Theognis claims: εἰ δ᾽ ἦν ποιητόν τε καὶ ἔνθετον ἀνδρὶ νόηµα, οὔποτ᾽ ἂν ἐξ ἀγαθοῦ πατρὸς ἔγεντο κακός, πειθόµενος µύθοισι σαόφροσιν. ἀλλὰ διδάσκων οὔποτε ποιήσεις τὸν κακὸν ἀνδρ᾽ ἀγαθόν. “If good sense could be manufactured and installed in a man, Never would an ignoble son be born from a noble father, Since he would be won over by words of wisdom. But never will you Make a base man noble through teaching.” Th. 335-338 With logic that strongly echoes the essentialist discourse of the Homeric epics, Theognis maintains that a man may possess all the necessary conditions to be agathos (birth, wealth, character) but still come up short. The kakos lacks the capacity to become noble, regardless of what he might possess, or do, or say, because the true determinant is his inborn quality. There is an essential, qualitative difference between elite and non-elite in the logic of the Theognidea. The logical extension of this qualitative difference is that for Theognis, a well-ordered society is one in which the primacy of the agathoi is preserved and free from challenges, whether those challenges may come from non-elites or even from within the aristocracy. Throughout the text, we can see an antipathy towards elites who, in the view of Theognis, have betrayed their own class, but it is perhaps most evident 220 in the poet’s attitude towards the phenomenon of tyranny. It is generally accepted that the famous passage in which Theognis warns of the impending danger of social dissolution, resulting in the rise of monarchy, civil strife, and the spilling of kindred blood (ll. 39-52, esp. 51-52) reflects the rise of Theagenes, the first tyrant of Megara (floruit 640-600), 256 but the poet’s attitude towards tyranny is also made clear elsewhere: δηµοφάγον δὲ τύραννον, ὅπως ἐθέλῃς, κατακλῖναι οὐ νέµεσις πρὸς θεῶν γίνεται οὐδεµία. “Strike down the demos-eating tyrant however you may; No retribution will come from the gods.” Th. 1181-1182 We know that the particularly Greek form of tyranny during the archaic period was one in which the aspiring tyrants depended upon popular support for their ascendance (Andrewes, 1963 [1956]). We also know that the Theognidea is especially critical of those who are born into the aristocracy but fail to fulfill the dutiesof their position. It seems distinctly possible that these dishonored agathoi are those who courted the favor of the non-elites, either as tyrants or supporters of tyrants, or have otherwise declassed themselves by marrying their daughters to wealthy non-elites. For the poet, the only legitimate social structure is one that favors the traditional aristocracy, and any action that fails to sustain that structure is regarded as a betrayal. However, in contrast with the Homeric epics, elite standing appears to be significantly more unstable. As we have already seen in the Iliad and the Odyssey, there are numerous strategies by which aristoi can make claims to elite standing, but their status is a function of their character, and not open to challenge. To be sure, one aristocrat may stake a claim to greater authority than another. After all, this is the central debate that propels the entire narrative of the Iliad, but nowhere is there a suggestion in Homer that elite 256 See Nagy (1985), McGlew (1993), Gerber (1997). 221 status is so fragile. 257 Menelaus recognizes Peisistratus and Telemachus as sons of kings long before he ever asks their names, and Odysseus claims to perceive clearly the kingly mien of his father Laertes, despite the latter’s wretched and squalid condition. Even when Odysseus himself dons the guise of a beggar, his nobility is undiminished. The inability to recognize it is evidence of failure on the part of the suitors, not any shortcoming of Odysseus. By contrast, the inverse is true for Theognis. In fact, a strong sense of the tenuous nature of aristocratic standing emerges from the Theognidea. Life for the aristocrats of the Theognidea seems to have been much more subject to vicissitudes of chance: οὐδείς, Κύρν᾽, ἄτης καὶ κέρδεος αἴτιος αὐτός, ἀλλὰ θεοὶ τούτων δώτορες ἀµφοτέρων: “No man, Cyrnus, is responsible for his own success or failure, But the gods are the source of both…” Th. 133-134 µήποτε, Κύρν᾽, ἀγορᾶσθαι ἔπος µέγα: οἶδε γὰρ οὐδεὶς ἀνθρώπων ὅτι νὺξ χἠµέρη ἀνδρὶ τελεῖ. “Cyrnus, never talk big, because no one knows What day or night will bring to pass.” Th. 159-160 The representation of Homeric society as exceptionally stable could be viewed as an attempt to compensate for some of the major structural changes that had taken place in the archaic Greek world already by the late 8 th c. BCE, and thus evidence of anxiety over a lack of social cohesion and stability. 258 We must therefore be skeptical of the picture of constancy in Homer. Nevertheless, what we see in 257 There are the lying stories of Odysseus, who when disguised as a beggar claims to have been the second son of the Cretan lord Doulichion (XVIII.172-189), or the bastard son of another wealthy Cretan named Castor (XIV.199- 355). These episodes could be read as examples of the potential for elites to lose their standing, but they are exceptional, and do not invalidate the general impression in the Homeric epics that elite status is highly stable. 258 This is a central feature of the arguments of Thalmann (1998), who proposes that the Odyssey betrays an underlying ambivalence regarding the institution of slavery as constituting an ever-present threat to the status quo. 222 Theognis is now an overt recognition of the fragility of the social hierarchy as a feature of Megarian society that cannot be denied. If we take this argument a bit further, we might postulate that in the late 8 th c. BCE, challenges to elite primacy were still at such an early stage that they could be denied simply through willfully ignoring them.. However, in the latter half of the 7 th c. BCE and the 6 th c. BCE, these challenges had brought sufficient change to the social fabric that it was no longer possible to ignore them, but this new social reality was now something that demanded attention and had to be confronted. A Society in Crisis? This is in fact a topic that seems to emerge from the text, namely that the world of Theognis is one in which the traditional order has been threatened, and the society is now facing genuine threats to its stability. One of the more pervasive themes in the text is that of decline and social disruption. 259 It is tempting to see parallels here between the Works and Days of Hesiod and the poetry of Theognis, both of which treat the decline of the contemporary society at great length. This is the position, for example, of Donlan (1985), who finds similarities in both theme and diction between verses 1135-1142 of the Theognidea, chronicling crooked speech, impiety, and selfish pursuit of the property of others, with the Hesiodic description of the race of iron. However, for a number of reasons, I believe that this perceived similarity is superficial. For one, Hesiod’s vision of the debasement of the human race is much more pervasive. According to Hesiod’s view, the fundamental nature of his fellow human beings is now debased, and this has resulted in social and political disruption. By contrast, the Theognidea seems to suggest that the frustration over the current state of affairs is mostly anger regarding social and political conditions. Still, insofar as the terminological distinction between agathoi and kakoi implies a moral aspect, the inroads made by the kakoi represent a moral threat. However, the Hesiodic picture of the race of iron is one in which the entire race is corrupted, 259 See 39-52, 647-648, 667-682, 833-836 223 whereas for Theognis, the problem is limited to a segment of society, and therefore potentially soluble. Second, as we have already noted, Hesiod is much more critical of the basileis as a group, as well as their failure to discharge their duties. In addition, he actively counsels the pursuit of profit as a means to social advancement. In this regard, Hesiod and Theognis could not be more different. Finally, Theognis’ concerns center on the inversion of the social order at the expense of the traditional aristocracy and the concomitant rise of the kakoi, but he still insists that there are genuine aristoi who are capable of taking their city back. According to Theognis, it is not so much that all men are less fortunate, but rather that some who are deserving of good fortune have fallen on hard times, whereas many deceitful and immoral kakoi now revel in their success. This notion of the inversion of the proper social order finds expression in numerous passages, but is perhaps most cogently expressed in the pair that follows: οἳ πρόσθ᾽ οὔτε δίκας ᾔδεσαν οὔτε νόµους, ἀλλ᾽ ἀµφὶ πλευραῖσι δορὰς αἰγῶν κατέτριβον, ἔξω δ᾽ ὥστ᾽ ἔλαφοι τῆσδ᾽ ἐνεµοντο πόλεος. καὶ νῦν εἰσ᾽ ἀγαθοί, πολυπαΐδη: οἱ δὲ πρὶν ἐσθλοὶ νῦν δειλοί… “Those who formerly enjoyed neither courts 260 nor laws, But instead wore goat skins wrapped around them And lived outside the city like beasts Are now noble, while those who once were noble Are now base…” Th. 54-58 νῦν δὲ τὰ τῶν ἀγαθῶν κακὰ γίνεται ἐσθλὰ κακοῖσιν ἀνδράσιν: ἡγέονται δ᾽ ἐκτραπέλοισι νόµοις: αἰδὼς µὲν γὰρ ὄλωλεν, ἀναιδείη δὲ καὶ ὕβρις νικήσασα δίκην γῆν κατὰ πᾶσαν ἔχει. 260 I have chosen here to translate dikas as “courts” rather than the more conventional translation as “judgments” that is common both in Hesiod and in the rest of the Theognidea. I believe that the passage reflects elite concerns over the degree of success on the part of previously disenfranchised groups in adopting the existing formal institutions of their society, including the legal system, to achieve unprecedented influence and success. 224 “Now the vices of the nobility are virtues for wicked men; They revel in perverted laws, Reverence has perished, and shameless hubris Has conquered justice and now reigns over the land.” Th. 289-292 In these lines, it is clear that Theognis is concerned with a specific form of disruption, whereby the undeserving (i.e., the kakoi) are now enjoying privileges that once were reserved solely for the deserving (i.e., the agathoi). At the same time, many of those who should be responsible for the proper governance of the polis have failed to uphold the proper social order, and are themselves at least partly to blame for the present decline: πολλάκι δὴ πόλις ἥδε δί᾽ ἡγεµόνων κακότητα ὥσπερ κεκλιµένη ναῦς παρὰ γῆν ἔδραµεν. “Many times due to the wickedness of its leaders, the city Has run aground like a listing ship.” Th. 855-856. As Marsilio (2002) points out, Theognis represents the view of frustrated aristocrats who have become disillusioned with members of their own class. 261 For Donlan (1985), this frustration reflects a changing social order in which the traditional aristocracy is seeing their grip on power eroding and the conventional bonds are no longer effective, but they have been unsuccessful in finding or promoting an ideology that is capable of replacing the one that is losing its relevance. Of course, Theognis does not reserve his ire solely for those elites who, in his view, have abrogated their obligations. We have noted previously that the poet is a harsh judge of non-elites. This is particularly true of those kakoi who, in the poet’s view, have contributed to the inversion of the traditional and proper social order by overstepping their bounds. This notion finds its most cogent expression in the famous metaphor of the Ship of State, in which the citizens have deposed their “noble helmsman” (kybernētēs 261 See also Nagy (1982, 1985) and Fox (2003). 225 esthlos) even as the ship founders (ll. 667-682, esp. 675-676). Ultimately, the picture that emerges from the Theognidea is one of a world where the proper social order has been overturned and even inverted, at times by the non-elite members of society who ideally would have no active role in the direction of the polis, but often with the active participation of members of the aristocracy who pursue their own individual, selfish advancement, thereby betraying the members of their own class, and ultimately their society at large. Wealth as a Source of Social Disruption According to the traditional view of archaic economic behavior, propounded most famously by Donlan (1989b, 1993, 1994, 1997) who argues predominantly in favor of the formalist position, the naked pursuit of profit was inconsistent with the overriding social mores that denigrated commercial activity and regarded trade as a powerfully disruptive phenomenon. We have already seen this view expressed in the Homeric epics, where wealth is a means to an end and has only a limited range of socially acceptable uses, in particular in the ritualized gift-exchange that promoted bonds of amity among the landed aristocracy. However, we have also seen the countervailing position put forth in the Works and Days, where the acquisition of wealth and the pursuit of profit, especially through the active participation in maritime trade, is not only tolerated, but actively encouraged, and often seems to be an end in itself. For Theognis, there are rules that govern the distribution of wealth in society. Justice involves the appropriate distribution of wealth along aristocratic lines. As we have seen with Hesiod, wealth is mercurial and transitory. This is likewise true for Theognis, but with an important difference: whereas the increasingly fluid nature of wealth constitutes an opportunity for Hesiod, it is a source of upheaval for Theognis. In a properly functioning polis, wealth is in the hands of a select few who possess it because of their innate virtues. Even further, when wealth is held by the undeserving kakoi, it has a powerfully corruptive influence. By the same token, when the agathoi are stricken with poverty, the moral fabric unravels. 226 We should be careful to point out that Theognis does not adopt some puritanical notion that money is the root of all evil, nor does he argue in favor of some egalitarian principle – in fact, quite the reverse. Theognis fully supports economic inequality, but only so long as it supports and corresponds to his vision of the appropriate social hierarchy. In the Theognidea we can see the erosion of the seemingly self- evident principle that it is fitting for the agathoi to be wealthy and the kakoi to be poor (Cobb-Stevens, 1985). It is clear that the poet believes that in the world in which he lives, the distribution of wealth was once ordered along lines that made sense from an aristocratic perspective. However, it is equally clear that this is no longer the case. In the Homeric epics, we saw a vision of society in which the possession of riches formed part of a circular argument, whereby elites were wealthy by virtue of their essential character, and wealth was an outward manifestation and confirmation of that inner quality. In the Theognidea, however, the situation would seem to have changed dramatically: πολλοί τοι χρῶνται δειλαῖς φρεσί, δαίµονι δ᾽ ἐσθλῷ, οἷς τὸ κακὸν δοκέον γίγνεται εἰς ἀγαθόν: εἰσὶν δ᾽ οἳ βουλῇ τ᾽ ἀγαθῇ καὶ δαίµονι δειλῷ µοχθίζουσι, τέλος δ᾽ ἔργµασιν οὐχ ἕπεται. “Many men are small-minded but have good luck on their side, And for them what seems doomed to fail brings success; And then there are those of good will but followed by bad luck Who toil away but get nothing for their efforts.” Th. 161-164 πολλοὶ πλοῦτον ἔχουσιν ἀΐδριες: οἱ δὲ τὰ καλὰ ζητοῦσιν χαλεπῇ τειρόµενοι πενίῃ.. “Many fools have riches while others who strive For the good are beaten down by relentless poverty…” Th. 683-684 Obviously, Theognis is deeply concerned over the fact that in his view, the connection between nobility and prosperity has been severed (Cobb-Stevens, Figueira, & Nagy, 1985), or in Tandy’s (1997) terms, wealth previously followed status, but now status follows wealth. The prevailing sentiment is that the established order is disrupted. What once could be counted upon is being threatened, and the old rules no 227 longer seem to apply. The poet himself even laments the loss of his own possessions and how he is now afflicted with poverty. 262 But this is more than a personal tragedy for the poet. For Theognis, the disjuncture between wealth and the traditional social hierarchy is one of the factors most responsible for the overturn of the traditional values held by the archaic aristocracy (Donlan, 1985). Wealth as a Corruptive Influence on the Kakoi, and Poverty as a Corruptive Influence on the Agathoi According to Theognis, there are two ways in which the proper distribution of wealth has been distorted in the polis: kakoi who are wealthy, and agathoi who are poor. 263 In both cases, according to the poet, there are deleterious consequences. Above and beyond the fact that the distribution of wealth in the world of Theognis fails to reflect accurately the underlying difference in the essential quality of the members of the society, wealth in the wrong hands can be disastrous. Going beyond the simple class essentialist argument that the elites deserve to be rich and the ignoble deserve to be poor, the poet claims that the inversion of this principle not only violates the proper social order, but in fact causes harm even to those who are wealthy, so long as they are undeserving. According to this argument, wealth in the hands of non-elites corrupts them, because they do not possess the character that allows them to withstand the temptation to excess that comes with affluence: τίκτει τοι κόρος ὕβριν, ὅταν κακῷ ὄλβος ἕπηται ἀνθρώπῳ, καὶ ὅτῳ µὴ νόος ἄρτιος ᾖ. “Excess gives birth to hubris whenever good fortune comes to a base Man, and [anyone] who is not of sound mind.” Th. 153-154 262 Ex., 649-652, 667-668, 831-832, 1107-1108. 263 Ex., 149-150, 321-322, 865-867, 1117-1118. 228 For the poet, this phenomenon is inevitable, approaching the certainty of a natural law. As we have already seen, base men are simply incapable of controlling themselves in extreme situations, and affluence only serves to bring this out: εἰ δὲ θεὸς κακῷ ἀνδρὶ βίον καὶ πλοῦτον ὀπάσσῃ, ἀφραίνων κακίην οὐ δύναται κατέχειν. “If the god grants a good income and wealth to a base man, He is unable to resist disaster because of his foolishness.” Th. 321-322 It is a fairly simple equation: non-elites plus riches equals hubris. It is a toxic combination that inexorably produces negative results. As it happens, for Theognis the inverse of the equation is also true. In essence, Theognis is attempting to reconstitute the social conditions in which status, determined by birth, and class, determined by control over resources and the means of production, overlap perfectly. The problem for him is that control over resources is no longer exclusively the province of the aristocracy. According to this view, aristos and kakos formerly were terms that embraced both wealth and ancestry (i.e., were indicative of both class and status), but now, wealth is no longer distributed throughout the society strictly along status lines established by heredity, and so these terms have lost some of their significance as markers of class identity. As non-elites have made financial gains, the ability of the hereditary nobility to leverage their virtual monopoly on the resources of the society has dwindled. In other words, class has become a disruptive influence on status. Once again, we find a theme that is prominent in both Hesiod and Theognis, but here too, they each offer very different perspectives (Marsilio, 2002). Like Theognis, Hesiod acknowledges the potentially devastating impact of penury. In fact, this theme is featured so extensively that is has led many scholars to argue in favor of Hesiod as an impoverished farmer. Yet, unlike Theognis, Hesiod at times represents poverty, or at least the compulsion to hard work in order to avoid poverty, as producing ultimately 229 beneficial results. 264 In the Works and Days, poverty (peniē) can be a motivator, driving men to work hard and seek to improve their financial condition. In the Theognidea, poverty is never presented as anything but something to be avoided at all costs, and is even worse than death itself. 265 To be sure, poverty is miserable for all men, and this is repeatedly acknowledged in the text, 266 and in fact this is a common refrain throughout archaic poetry (Cobb-Stevens, 1985). But the unique detriment of poverty, according to Theognis, lies in its morally enervating effects. It compels one to compromise his morals and to sell his virtue, and even teaches evils such as deceit (exapatai), lies (pseudea), and deadly discord (oulomenai eridai), even against one’s better judgment. 267 In addition, it is the nobility who seem to be particularly vulnerable to its demoralizing effects. This is doubtless due to the fact that virtue, as we have already seen, is the exclusive province of the agathoi for Theognis. Because they are the virtuous, the nobility are naturally the ones most susceptible to corruption: ἄνδρ᾽ ἀγαθὸν πενίη πάντων δάµνησι µάλιστα καὶ γήρως πολιοῦ, Κύρνε, καὶ ἠπιάλου: “More than anything else, poverty crushes the noble man, Cyrnus, Even more than grizzled old age and disease…” Th. 173-174 Poverty, or more accurately, misplaced poverty, compels the agathoi to engage in immoral behavior. Even as they recognize its influence, the effects are so strong as to be seemingly irresistible. The divorce of wealth from status, and the succeeding attempt to reunite the two, drives the nobility to such lengths that they intermarry with non-elites, threatening their very genetic purity: “αὐτὸς ὁ τοιαύτην εἰδὼς κακόπατριν ἐοῦσαν εἰς οἴκους ἄγεται χρήµασι πείθοµενος, 264 See WD, 298-310. 265 181-182. 266 Ex., 181-182, 267-270. 267 See 389-391, 651-652. 230 εὔδοξος κακόδοξον, ἐπεὶ κρατερή µιν ἀνάγκη ἐντύει… “Knowing full well that she is low-born, He takes her into his home, seduced by money, Although his reputation is noble and hers is ignoble, because powerful compulsion Drives him…” Th. 193-196 The negative influence of poverty on the agathoi is apparently so great they are driven to compromise one of the characteristics that has ostensibly made them unique. Using the metaphor of animal husbandry (ll. 183-192, esp. 183-184), the poet explains that financial need has led many of the nobility to intermarry with the daughters of the kakoi, and also for the daughters of noble men to marry kakoi (ll 185-189). For Theognis, this intermingling of the noble (esthla) with the base threatens the very foundations of the polis and weakens the people. Here the poet makes an even more forceful condemnation, accusing members of the aristocracy of betraying their class for the sake of filthy lucre. As Kurke (1989) has argued, this is a truly stinging rebuke, combining both the denunciation of class betrayal and the association of marriage with the traditional aristocratic disdain for sordid commerce. In short, Theognis accuses those aristocrats of commodifying themselves and their daughters, a double reproach that expresses the poet’s disdain in the strongest possible terms. In short, because of the disconnect between nobility and wealth, the entire social order is being attacked from within and is at risk of collapse. The Pursuit of Profit and the Destructive Influence of Greed So, we can see in the Theognidea a pervasive sense of alarm and impending doom brought on by changing economic conditions in which the landed nobility no longer appear to have a monopolistic grip on the resource base of the society. This of course leads us to the question as to how and why this state of affairs has come to pass. If it is so clear, as it seems to be for Theognis, that an improper distribution of wealth in the polis has threatened its very survival, what could have persuaded its citizens to allow this to happen? For Theognis, the reason is simple: wealth and hereditary status have become divorced from one another, to the detriment of the entire polis. When the aristocracy exercised complete control over the 231 resource base, the connection between their wealth and their status could be elided, and their status could be justified on the basis of character or birth. However, now that non-elites seem to be threatening this economic order, the possession of property has become the yardstick by which men have come to be measured: πᾶς τις πλούσιον ἄνδρα τίει, ἀτίει δὲ πενιχρόν: πᾶσιν δ᾽ ἀνθρώποις αὐτὸς ἔνεστι νόος. “Everyone glorifies the rich man and shames the poor man; This attitude is the same among all men.” Th. 621-622 One of the reasons why this drive for personal enrichment is so problematic, at least according to the poet, is that it is a desire that cannot possibly be satisfied. Instead, the chase for wealth leads inevitably to excess: πολλῷ τοι πλέονας λιµοῦ κόρος ὤλεσεν ἤδη ἄνδρας ὅσοι µοίρης πλεῖον ἔχειν ἔθελον. “Greed has destroyed more men by far than famine Who wanted to possess more than their share.” Th. 605-606 268 For Theognis, this valorization of wealth has significant consequences. For one, it obscures the innate character of the man, as immoral men are now capable of hiding their moral turpitude if they are wealthy, while the virtue of others goes unnoticed if they are poor (ll. 1061-1062). But above and beyond the fact that affluence (or penury) can disguise one’s character, the allure of riches actually prompts men, both agathoi and kakoi alike, to engage in unethical and even self-destructive behavior. The kakoi, who don’t have the kind of innate virtue that is ostensibly a feature of the aristocracy, are susceptible to the seductive allure of riches. In their case, the pursuit of wealth drives them to engage 268 See also 227-229, 693-694, 1157-1158. 232 actively in immoral behavior, enriching themselves by unjust (adikē) means such as seizing property by force 269 and deceit. 270 This admonition is similar to the complaints of Hesiod, who likewise disparages those who seek personal gain through unjust means, but there is a critical difference. Whereas for Hesiod, acquisitiveness can be a powerful motivation for self-improvement, causing men to work hard and better themselves, for Theognis, profit (kerdos) is associated with deceit, greed, oath-breaking, and injustice. 271 Thus, the desire for riches is problematic on several levels, since not only does the attainment of wealth lead to the corruption of the kakoi, but the very pursuit itself causes one to act contrary to justice. Theognis and Institutions Once again, I wish to emphasize the ways in which our understanding of economic institutions can help to frame the analysis. As Nagy (1982, 1985) has pointed out, there are numerous convergences between Theognis and lawgivers like Solon and Lycurgus as exponents of dikē. According to this argument, the Theognidea represents something similar to the law codes attributed to Solon and Lycurgus, with the difference that unlike Solon, whose law code was independent of the poetry, or Lycurgus, who was credited with the creation of a law code but did not write poetry, the precepts of Theognis are contained within the poetry itself. Thus, if we accept Nagy’s contention, the Theognidean corpus is a product of a program to establish and promote a particular set of institutions, consistent with an intellectual tradition that was recognizable to the contemporary audience. At the same time, the Theognidea implies the existence of a very different institutional agenda that runs directly contrary to the aristocratic ethos. Although it is never expressed overtly, it seems clear that there is a competing agenda at work. We have seen how Hesiod advocated for the pursuit of profit, as well as exposing the role that wealth played in the formation of status, even to the point where he claimed that 269 Ex., 345-347, 675-679. 270 Ex., 65-68, 224-226, 389-391. 271 Ex., 145-148, 197-202, 224-226, 465-466. 233 fame and virtue follow wealth (WD, 686). Based on the way that the Theognidea describes the changing conditions of society, it would seem that this value has taken root, and that there are many non-elites who have been successful in both pursuing wealth and in leveraging it for their advantage in the social hierarchy. It is not clear how this is specifically manifested, but there are some indications. For example, recall the passage (ll. 54-58) in which the poet laments the rise of those who formerly knew neither courts (dikai) nor laws (nomoi) but who are now noble. The complaint about those who previously knew no laws would seem to suggest that now they do. The implication here is that the kakoi have now achieved positions of influence in society to the point that they are now involved in the legal and juridical processes of the city. This possibility is further supported by another passage (ll. 289-292) in which Theognis claims that the kakoi have traded vice for virtue, and “take pleasure in perverted laws…” (ἡγέονται δ᾽ ἐκτραπέλοισι νόµοις). I suppose that it is possible that the agathoi are the ones who are responsible for subverting the laws, but there is no such suggestion in these lines. The blame for the current plight of the polis in this passage is exclusively reserved for the kakoi, and we should assume that the blame for the disruption of the laws is laid at their feet as well. If so, we must conclude that non-elites have been able to make changes to the legal code. We should consider the description of the Ship of State in regards to shifting power relations, particularly where it states that “the kakoi now stand above the agathoi…” (κακοὶ δ᾽ ἀγαθῶν καθύπερθεν) 272 and that “they have deposed their noble steersman…” (κυβερνήτην µὲν ἔπαυσαν ἐσθλόν). 273 The clear message is that non-elites have taken political control in the polis. Even if we discount somewhat the account of Theognis as alarmist concerning the degree to which the social hierarchy has been transformed, the simple fact that non-elites could be capable of deposing a ruler indicates a major shift in political power. 272 679. 273 675-676. 234 Finally, another important way that non-elites have been able to leverage wealth as a means of challenging elite control is evident in the changes to elite marriage practices. As it appears in Theognis, marriage is an institution that serves to foster elite bonds while perpetuating elite domination by creating a closed system. 274 However, some of Theognis’ harshest criticism is reserved for those elites who have, in his view, betrayed their class by marrying non-elites. 275 Furthermore, the motive for this betrayal is for the sake of material gain. 276 As a result, it appears that non-elites have been able to take advantage of the existing institution of marriage and develop strategies that allowed them to penetrate the barrier to elite status by exploiting their wealth to their advantage. If my interpretation of these passages is correct, non-elites have achieved a substantial degree of control over the affairs of the polis, including the ability to shape the formal institutional framework by altering the laws, exerting a measure of control over who is in charge of the city, and undermining marriage patterns that fostered elite cohesion and excluded non-elites. These are major achievements and represent a significant turning point that has powerful implications. Up until this point, what we have seen in the competing institutional agendas of Hesiod and Homer has been a contest over the prevailing informal institutions that govern economic activity. However, if my interpretation of this passage is correct, what we can now see is that the non-elite petits bourgeois have not only succeeded in disrupting the traditional social order by taking advantage of new sources of wealth, but have even reached the point of influence where they are able to exert a degree of control over the formation and direction of the formal institutions of their society. 274 The role of marriage as an important component of the elite hierarchy can also be seen clearly in Homer, such as the offer from Agamemnon to Achilles to take one of his daughters in marriage (which would bind Achilles to him), or the offer of Alcinous to Odysseus to marry his daughter Nausicaa, and of course the eagerness of Penelope’s suitors. 275 Ex., 183-192, 193-196. 276 See p. 243 above. 235 What Does Theognis Say? Ultimately, the various themes that we have discussed (the appropriate social organization of a well- ordered polis, the current state of Megarian society, and the dangers of greed and the improper distribution of wealth) occur so frequently throughout the Theognidean corpus that they must be regarded as central to the text. When taken together, I believe that they form the outline of an argument, illustrating what the archaic elites view as the most pressing and worrisome socio-political problems, as well as the source of those problems. What we see in the Theognidea is as follows: first, for a polis to function smoothly and efficiently, there must be a social structure in place whereby the nobles (esthloi, agathoi) are in positions of authority and in firm control of the majority of the resources. To be considered an agathos, however, one must possess various qualities, including proper birth, wealth, and virtue. In addition, elite standing is neither inevitable nor guaranteed. For instance, it is possible for one to become poor; as the poet reminds us several times, he himself appears to have suffered serious financial setbacks, either through an unsuccessful sea voyage, the treachery of his friends, or the loss of his possessions. It is also possible for an agathos to become kakos through his actions, such as rendering false judgments or trading his virtue for money. Thus, aristocratic standing is tenuous, susceptible to challenges or even simply the vicissitudes of fortune. Second, the current state of affairs in Megara is one in which the traditional, ideal social order is being eroded. More accurately, the order is being inverted. Rather than total anarchy, the power structure of the polis has been reversed, with kakoi now in charge of aristoi. The blame for this decline, however, is not exclusively laid at the feet of the low-born. In fact, it appears that many recalcitrant agathoi have failed to fulfill their moral obligations, either driven by poverty to compromise their ethics, or simply through a self-interested pursuit of wealth. Of course, the kakoi bear some responsibility as well, since they apparently have overthrown their rightful rulers and even taken up positions of power in the polis. Finally, the underlying causes for the dissolution of Megarian society are economic. First, the citizens, both noble and base, have come to valorize wealth at the expense of virtue. This has several negative 236 consequences. For one, it has made it more difficult to differentiate the good from the bad citizens. Because everyone reveres the wealthy and derides the poor, public opinion is now shaped by financial circumstance rather than inherent character, birth, or even one’s actions. Second, the selfish pursuit of profit (kerdos) is blamed as the factor most responsible for the overthrow of the traditional values held by the archaic aristocracy (Donlan, 1985). 277 This drive for gain leads the agathos to compromise his morals, with some even going so far as to sell their daughters to the highest bidder. This same pursuit also drives the kakos to seek profits through unjust actions, including violence and deceit. In fact, upward mobility is equated with outright theft (Figueira, 1985). This profit-motivated activity is even more problematic because once begun, the drive for profit has no end, since it is not possible to be too rich. Also, in part because of these efforts, the citizens of the polis have effectively changed the distribution of wealth contrary to the “appropriate” pattern where it coincides with hereditary status. Some kakoi have become rich, but because of their deficient nature they are incapable of handling their new-found financial success. At the same time, some agathoi have become impoverished, and oppressed by poverty they are driven to sacrifice their ethics out of compulsion. While these themes most frequently occur individually and are scattered throughout the text, there are points at which they come together and crystallize this argument. Perhaps the most cogent expression can be seen in the famous metaphor of the Ship of State: νῦν φερόµεσθα καθ᾽ ἱστία λευκὰ βαλόντες µηλίου ἐκ πόντου νύκτα διὰ δνοφέρην: ἀντλεῖν δ᾽ οὐκ ἐθέλουσιν: ὑπερβάλλει δὲ θάλασσα ἀµφοτέρων τοίχων: ἦ µάλα τις χαλεπῶς σῴζεται οἷ ἕρδουσι: κυβερνήτην µὲν ἔπαυσαν ἐσθλόν, ὅτις φυλακὴν εἶχεν ἐπισταµένως, χρήµατα δ᾽ ἁρπάζουσι βίῃ: κόσµος δ᾽ ἀπόλωλεν, δασµὸς δ᾽ οὐκέτ᾽ ἴσος γίνεται ἐς τὸ µέσον: φορτηγοὶ δ᾽ ἄρχουσι, κακοὶ δ᾽ ἀγαθῶν καθύπερθεν: 277 See 50, 83-86, 197-202, 221-226, 227-232, 465-466, 523-524, 607-610, 833-836. 237 “…we are now being carried along with the white sails trimmed from the Melian sea through the murky night, And they refuse to bale the ship even though the sea is pouring over On both sides. Salvation is truly difficult, Because of their actions: they have deposed the noble steersman Who skillfully held to the safe course; They seize money by force, demolishing all order; There is no longer an equitable division of wealth in society; The traders are now in control, and the commoners have overcome the nobles.” Th. 671-679 In this one passage, many of the central themes come together to express the frustration and helplessness of the aristocrat who sees the social order that once gave him such comfort now disappearing before his eyes. The privileges that he once enjoyed unquestioned are now disappearing. The noble (esthlos) leader who was responsible for the prosperity and well-being of the state has been overthrown by the ungrateful mob. 278 Worse still, the rabble has taken control and maintains a course of action that continues to eat away at the very stability of the polis. They seize wealth by unjust means, they destroy the social order (kosmos), and they have redistributed wealth contrary to proper proportions. 279 Perhaps most significantly, the ultimate blame is laid at the feet of the phortēgoi, a term that I have chosen to translate here as “traders,” and who now have taken over the Ship of State. 280 As the expanse and degree of maritime trade was steadily increasing during this period (a phenomenon that we will confirm through an examination of the material evidence), the ability of the traditional aristocracy to maintain a quasi-monopolistic control over the available resources was eroded. I have already argued 278 The supposed lack of gratitude (charis) of the low-born is another recurring theme in the text; see 853-854, 955- 956. The term can also mean more broadly something like “reciprocal social obligations,” which of course is reminiscent of the reciprocities that form the basis of the elite-centric system of xeinia. 279 Gerber (1999) translates dasmos isos as “equal distribution,” but I believe that this does not accurately convey the sense of the statement. Here, isos should be taken to mean “equitable” rather than “equal,” since it is clear that the aristocratic voice of Theognis would never condone such a distribution of resources, but rather that the wealth of a society should be allotted along lines consistent with the conventional social hierarchy. 280 Although both Campbell (2003 [1967]) and Hudson-Williams (1910) contend that this term is more accurately translated as “porters” because of its clear etymological connection to phortos (“cargo”), I am more inclined to support the arguments of Bravo (1977), Cobb-Stevens (1985), Kurke (1989), and Coffee (2006), all of whom agree that it more likely should be taken to mean “traders” or “merchants.” 238 that from the very beginning, wealth was inextricably intertwined with status. The Iliad and the Odyssey demonstrate ample evidence of the ways in which Homeric elites leveraged their control over resources as a means of consolidating their position. Although the aristocratic ideal represented wealth as a function of status, it seems undeniable that both existed in a complex, mutually reinforcing feedback loop, but one that elites strenuously sought to elide through the conversion of financial capital into social and cultural capital. As we can see in the Theognidea, in their efforts to take advantage of the newfound potential to generate financial capital through long-distance trade and challenge elite primacy, archaic arrivistes laid bare that underlying reality, and in the degree to which they were successful in gaining wealth, their efforts disrupted the balance of power, and thus gave rise to the anxiety and frustration expressed in the Theognidean corpus. What Does It Mean? The poetry of Theognis is fundamentally concerned with the codification of an ethical and institutional system, favorable to the traditional landed nobility, in which economic activity plays a central role. By now, many of the features of this system should be familiar from Homer: the denigration of commerce (especially the vilification of profit as the primary motive for the distribution of goods), concentration of wealth exclusively in the hands of the aristocracy, and the preservation of a traditional social hierarchy. What is different between the picture presented by Homer and that of Theognis, however, is the degree to which this ideal has been achieved in reality. Whereas the Iliad and the Odyssey present a society that conforms in all respects to this ideal model (in spite of hints of tension beneath the surface), in the Theognidea we see a society that is vastly different. Whether or not we agree with the value judgment of Theognis that this departure spells doom for the polis, the conditions that the poet describes indicate a far greater degree of instability and contentiousness. Although this is clearly a source of great consternation for Theognis, the aristocratic poetry seems unable to offer any novel solutions to the dramatic changes taking place, or even to recognize that the conditions have changed. Instead, the proposed solutions are entirely regressive, suggesting nothing more original or 239 innovative than simply a wistful return to an idealized past when people knew their place. Certainly, this kind of outmoded thinking is not unusual during times of dynamic social, political, and economic change, particularly by groups who perceive themselves to be disadvantaged by the changing conditions. However, the Theognidea seems to reflect two things: first, that circumstances have fundamentally changed, and second, that the poet is either unwilling or unable to grapple with the new realities in a proactive way. Instead, Theognis clings to outmoded institutions and looks to make Megara great again simply by returning to social and economic conditions that in reality are impossible to recreate. The genie is out of the bottle. We have seen that the poetry is strongly connected to the symposion, the ritualized gathering by which elites attempted to accomplish the dual objectives of first, establishing a unique group identity, drawing a distinction between themselves and the rest of society; second, forging mutually reinforcing bonds of reciprocal obligations between members within the defined group (Catoni, 2010; Wecowski, 2014); and third, reproducing the values that support their economic practices and transmitting them to succeeding generations. More prominent in the poetry, though, is the emphasis on friendship (philia) among the elites. Unfortunately for Theognis, these bonds of friendship are dependent upon trust (pistis) that the poet laments is now largely absent from his world. 281 As Donlan (1985) points out, this perceived disappearance of pistis is more accurately a failure of outmoded social rules to retain their relevance to changing social relations: The poetic expression that presents the positive…visions of the harmonious symposium and lover…as symbols of the polis presupposes a fundamental reliance on the pistos philos hetairos as an eternally enduring institution [emphasis mine]. If these models seem to fail ultimately, it is because the poet 281 Ex., 65-68, 69-72, 73-74, 75-76, 77-78, 79-82, 283-284. 240 recognizes…powerful new social forces that threaten to invalidate, make obsolete, that basic presupposition. In other words, the institutional context that traditional elites perceived as being responsible for, or at the very least contributing to, their privileged position was quickly lapsing into obsolescence. During a hypothetical period when the agathoi both were the ruling class in the polis and managed more or less to embody the code of values that they espoused, the terms agathos and kakos could remain relatively straightforward in both their meaning and their application. However, when economic or political changes made it possible for these various marks of being agathos or kakos to occur separately or in varying combinations, then both the meaning and the application of the terms became as unsettled as the world that contained them (Cobb-Stevens, 1985). In short, Theognis and his fellow elites seem to have found themselves trapped by changing circumstances and unable to make the adjustments necessary to retain their elevated social, political, and economic positions. Of particular importance for the present discussion is the unequivocal and unmistakable fact that the poetic voice of the Theognidea recognizes that economic factors have been central to the contemporary lapsarian state of society, and that the ability to exercise control over resources, as well as the processes of consumption, production, and distribution, is crucial to aristocratic success (or failure) in the future. What this required was the establishment of a set of rules and institutions that could be effectively enforced, and that supported particular forms of economic activity and repressed others. Moreover, although it is the view of the Theognidea that self-interested profit-seeking should be sublimated to overriding concerns of reciprocal social obligations, the poetry likewise makes clear that this was not the case, contrary to the expected conditions if we accept the arguments of Donlan (1985, 1993, 1994) and others. It is true that this tenet of ideology is the one most prominent in the archaic literature, but we must acknowledge that it was not universally endorsed and in fact faced direct and serious challenges. Although we do not possess non-elite literary sources for this period equivalent to the Theognidea, by reading between the lines it also appears to be the case that non-elites were active in promoting a 241 competing ideology that stressed the importance of wealth as a means of exercising social and political influence. In spite of the attempts on the part of the aristocracy to obscure the process whereby financial capital was converted into social and cultural capital, it would seem that previously disenfranchised groups were capable of promoting a competing ideology involving the acquisition of wealth as a means of enhancing social standing. In addition, this effort was so successful that it not only caused great consternation among the landed nobility, but even prompted many of them to adopt this strategy themselves and even adopt non-elites into their ranks through marriage. Conclusions I do not mean to suggest that I accept the depiction of social transformation in the Theognidea literally. I believe that the text clearly advocates for a return to social and economic conditions that cannot be reproduced, but that is not the same thing as saying that archaic elites were completely unable to respond or adjust to changing conditions. We should be careful not to take the consternation expressed in the text as an analogue for the degree to which archaic society was transformed, and in fact, we should expect that the anxiety could very well be disproportionate, in the same way that many modern commentators decry a “war” on Christians in America, in spite of the fact that more than two-thirds of Americans identify as Christian in 2014 (Smith G. , 2015). At the same time, we should not dismiss the sentiments expressed as purely alarmist, but rather reflective of an underlying reality, even if that reaction is exaggerated. For example, to continue with the above example, while a “war” on Christianity is clearly an over-reaction in a country with such a solid majority of Christians, it is nevertheless a response to very real changes, with the percentage of Americans identifying as Christians declining from 81.6% in 2001, to 78.4% in 2007, and 70.6% in 2014 (Smith G. , 2015). With this in mind, I would suggest that the social and economic changes described by Theognis are very real, even if archaic Greek society might not have faced the kind of existential threat that the text might suggest. When juxtaposed with the evidence from Homer and Hesiod, an examination of the Theognidea reveals several important developments. First of all, given the date range of the text (that I addressed at the 242 beginning of this chapter), we can now gain a glimpse into the continuing development of the institutional conflict whose lines were laid out in the late 8 th c. and early 7 th c. BCE in the Homeric epics and the poetry of Hesiod. As a brief reminder, I suggested that the Iliad and the Odyssey are actively engaged in an attempt to reify a system of informal institutions advantageous to the hereditary nobility and deeply involved in economic behavior. They prescribe specific forms of economic activity, and discourage others. Horticulture and stock raising are exalted production processes, while ritualized gift-giving is the most highly regarded form of distribution. At the same time, market exchange is represented as morally enervating, and the pursuit of kerdea something to be avoided. By contrast, we see in the Works and Days a vastly different set of values promoted. Instead of horticulture, agriculture, especially grain farming, is presented as the dominant type of production, while maritime trade is accepted without any stigma, and self-interested profit-seeking is characterized as a laudable goal. Clearly, these texts present two competing contemporary economic ideologies, each one favorable to a specific class. What they do not show, however, is how these competing economic views subsequently developed. As North (1990) has explained, institutions are not ideologically neutral, nor do they arise of their own volition. They are human creations, and reflect the interests of those who are powerful enough to institute and enforce them. The explicit challenge to elite economic ideology that Hesiod represents should, therefore, be regarded as evidence in and of itself for the attainment of a certain degree of political autonomy. 282 Still, solely on the basis of the texts of Hesiod and Homer, the outcome of this institutional rivalry is still very much in doubt. As van Wees (2009) argues, The elite liked to stress its ability to forgo profit for the sake of honor occasionally, and looked down on those who could not afford to be so high- minded. Archaic poets lamented that profit-seeking was the root of all society's evils. But all this only confirms that competition for wealth was pervasive: it does not imply a pre-modern rejection of the profit-motive any more than 282 Although he accepts the primitivist view of archaic economic activity in Hesiod, Edwards (2004) also argues that one of the principal aims of the Works and Days is to assert the political independence of Ascra and the local farming landholders that Hesiod represents. 243 modern moralizing about greed implies that our own economy is not overtly driven by the search for the largest possible gain… It is here that the Theognidea offers a diachronic perspective on the development of economic ideology in the archaic period. Accepting the date range proposed by Nagy (1982, 1985) of the latter half of the 7 th c. to the beginning of the 5 th c. BCE, we see in Theognis that the battle has continued, and that elite prerogatives have eroded. No longer can they maintain a virtually monopolistic control of resources, and an emergent class of free, land-holding farmers seems to have been successful in acquiring a greater share of the resource base, and in turn leveraging that financial success into a share in the direction of the polity. To be sure, the degree to which elite prerogatives have eroded is an open question. It is difficult, if not impossible, to parse out completely the sense of alarm in the Theongidea from the reality of social disruption. There is ample evidence that aristocrats retained a disproportionate share of resources as well as a substantial level of political influence and control, even in the democratic Athens of the 5 th c. BCE. So, we should rightly be skeptical of any depiction of the social changes during this period that suggests a complete inversion of the social order, as Theognis occasionally appears to imply. Still, it would seem clear that we are dealing with a substantial shift in power dynamics, with non-elites now laying claim to a greater degree of wealth and even to legal and juridical authority. This is a significant development, prompting me to wonder what has changed to make this possible. I would suggest that long-distance trade seems to be the most logical means of achieving these goals. In the absence of some radical shift in land-use patterns that does not appear to be evident in either the literary or material records, commerce in all likelihood constituted the most effective means of financial, hence political success. This is supported, I believe, by the persistent negativity expressed by aristocratic voices 244 towards market exchange. 283 The most logical explanation for this disdain is that it stems from anxiety over the recognition that maritime commercial activity represented a means of wealth generation that was not vulnerable to elite domination in the same way as other forms of production and distribution, given the elite control over real property, especially in the early Archaic Period. By its very nature, long- distance trade was conducted in such a way as to render it practically immune from aristocratic control. 284 Finally, what the Theognidea also demonstrates is a measurable degree of progress in the degree to which previously marginalized groups managed to exploit existing institutions to their advantage. If my sense of the meaning of dikas and nomous in line 54 is accurate, 285 then not only have non-elites managed to attain a level of political control in the polis, but they have effectively incorporated themselves into the legal and judicial institutions in place at the time. This again is an important feature of institutions, according to North (2005). As he explains, institutions are not exclusively constraints, in fact often quite the reverse. Their existence frequently provides the opportunity for various strategies that would not be possible in their absence. In Theognis, I propose that what we see, in addition to the promotion of an economic ethos that is in overt opposition to that supported by the aristocracy, is a measure of success on the part of an emergent class in adopting and assimilating certain aspects of the prevailing institutional matrix. Admittedly, this interpretation is destined to remain speculative, although I believe that it is one that most accurately accounts for all of the available evidence that we have seen so far. If my interpretation of the literature is correct, then in the contemporary archaeological evidence we should be able to find some confirmation of my thesis. Specifically, if the growth of long-distance trade is central to the successful implementation of novel economic institutions favorable to previously marginalized groups, then we 283 Recall the metaphor of the Ship of State (ll. 671-679), in which Theognis blames the phortēgoi for usurping the control of the rightful captain. 284 Seemingly the only possible strategy for traditional landed nobility to control long-distance trade would have been to participate directly, as Tandy (1997) has suggested. On the basis of both the literary and archaeological evidence, I find this to be unlikely, and it is a point on which I will elaborate in my conclusions. 285 Note 260 above. 245 should find corresponding physical evidence. On the production side, this should include, but may not be limited to, more centralized production, greater standardization in manufactured items, increased specialization, and production on a larger scale. On the distribution side, we might expect to see a more sophisticated transportation infrastructure, a greater geographic range of dispersion of goods, an increase in the sheer volume of trade goods across longer distances, and increased frequency in trading, all of which should produce a greater number of shipwrecks as well as larger vessels designed to accommodate increased demand. On the distribution side, this economic expansion would likely lead to a greater degree of integration that would enhance communication between producers, distributors, and consumers, which could then foster the potential of consumer demand to affect production and distribution strategies in turn, so that producers would begin manufacturing goods specifically for export, and even changing their techniques specifically for their consumers, as well as traders specializing in particular markets. 246 Section Three: The Archaeological Evidence Introduction As I hope has become clear by now, for the most part my argument hinges upon a neo-institutional analysis of the literary evidence. This reliance on the written sources is probably unavoidable, since the written sources are uniquely capable of offering insights into the rules that governed economic behavior in the Archaic Period. Unfortunately, the material evidence stands as a mute witness to this particular aspect of the period. That is not to say, however, that archaeology cannot provide a valuable contribution, indeed quite the reverse. First of all, my particular interpretation of the texts is based to a great degree on the initial picture of the early Archaic Period that is generated from an analysis of the archaeological evidence. For instance, I have noted the dramatic and rapid expansion of the Greek world resulting from the colonial expansion beginning in the 8 th c. BCE, and based on that evidence, I argue that the relatively simple political structure described in the Homeric epics would not have been sufficient to support the extensive networks and maintain the levels of connectivity that are suggested by the material evidence. Also, I have emphasized the apparent divergence between the express distaste for maritime commercial activity apparent in the Homeric epics and the relatively obvious indications in the archaeological evidence that the Greek expansion into the western Mediterranean was motivated, or at the very least conditioned, by considerations of long-distance trade. As a result, I believe that the persistent appearance of elite gift-exchange in the epics was crafted, at least in part, as a response to the increasing volume and frequency of other modes of exchange. Also, we have seen that the literary evidence, although indispensable, is ideologically charged, and this potential bias seriously complicates our analysis. The sources that I have examined (the Iliad, the Odyssey, the Theogony, the Works and Days, and the Theognidea) have all offered marked preferences for certain behavioral norms, and are indicative in one way or another of a distinct class perspective. As a result, they are much more concerned with presenting a picture of how economic behavior should be, 247 rather than offering a description of how it actually was conducted. It is here that I believe the archaeological evidence can again make a valuable contribution to our understanding of commercial activity, especially maritime trade. At least on the basis of these sources, it seems that long-distance trade was a transformative element, sufficiently disruptive to cause trepidation reflected in the elite view of Homer, powerful enough for Hesiod to recognize its potential to alter the prevailing balance of power, and ultimately for Theognis to characterize it as being responsible for the dissolution of the traditional social hierarchy. Rather than characterizing the expansion of maritime trade in the Mediterranean as simply an unintended consequence of broader social and demographic changes, then, I offer the hypothesis that the growth in trade provided an engine that contributed to those changes, ultimately producing some of the characteristic features of Greek social and political organization that are so familiar to us from the 5 th c. BCE onwards. It is in order to test this hypothesis that I propose an examination of the archaeological evidence, specifically the shipwrecks from the period, and a data-driven examination of ceramics excavated from several sites in the region. More than simply a counterpoint to the literary sources, the archaeology of the period is capable of providing answers to many questions that the literature cannot offer, such as the relative levels of trade over time, regional distribution patterns, the form and existence of maritime trade networks, and other insights into how long-distance trade was organized, carried out, and developed over the course of the 7 th and 6 th centuries BCE. In particular, I believe that the shipwreck evidence demonstrates several important aspects of archaic maritime trade. First and foremost, the revolutionary transformation of Greek naval architecture that began in the 6 th c. BCE can only reasonably be explained by a persistent and compelling demand for larger ships capable of transporting greater cargoes across longer distances, thus indicating a sharp increase in the overall levels of maritime commerce during the period. Second, the variability in size and ship types, visible in part through the archaeological evidence but also suggested by written sources, 248 would seem to suggest multiple levels of seaborne commerce occurring simultaneously. Third, the various wrecks demonstrate a wide range of cargoes, from small and highly heterogeneous to larger cargoes of which upwards of 90% of the entire cargo is a single commodity, and others in between. On the one hand, the transition in Greek naval architecture is reflective of powerful demand factors which compelled shipbuilders to adjust to accommodate these demands, and on the other hand, the variety of both ships and cargoes would appear to demonstrate the accommodations made by archaic maritime traders in order to increase the capacity and complexity of their distribution capabilities. Taken together, all of these features are indicative of increasingly sophisticated maritime shipping networks capable of transporting high volumes of goods across large geographic areas, and incorporating a wide range of previously excluded consumers who in turn placed ever-greater demands on those networks. At the same time, the ceramics data both confirm and enhance the picture generated by an examination of the shipwreck evidence. First of all, the regional variations in the composition of pottery assemblages is strongly contraindicative of the traditional, “supply-side” view of the trade in archaic ceramics. Instead, they suggest, along with the ships’ cargoes, the existence of persistent, overlapping trade networks that allowed traders to become familiar with the inclinations of their customers, who exercised their consumer preferences through the process of “dollar voting.” In turn, traders were capable of providing feedback to producers, who were then able to adjust production strategies to accommodate the information transmitted by the traders, who were thus able to provide products in order to satisfy specific consumer demands, and presumably maximize profits. All of this is consistent with the larger argument that maritime trade during the Archaic Period involved more intensive production, more complex distribution, and higher levels of consumer demand than has traditionally been assumed, all of which in turn imply the participation of numerous groups. Also, the evidence strongly implies widespread market exchange as opposed to the more limited, elite-centric gift- exchange emphasized in the Homeric epics. In addition, the dedication of extensive resources to long- distance trade implies high profitability, which in turn indicates the introduction of elevated levels of 249 wealth to Greek society through previously inaccessible channels. It may very well have been the injection of these capital flows that would have produced the kind of destabilizing effects indicated in the literary sources. 250 Chapter Seven: Shipwrecks as a Proxy for Maritime Trade Growth in the 6 th c. BCE Introduction Since my interest in the archaeological evidence is primarily motivated by a desire to discover additional information regarding the extent and nature of maritime trade, the decision to focus on the shipwreck evidence is an obvious one. For one, shipwrecks are perhaps the most direct and vivid evidence for long- distance trade in processu, constituting snapshots of maritime commercial activity at particular moments in time. At the same time, there are limitations to the capability of a single shipwreck to shed light on broader patterns of exchange. In particular, cargoes are the product of a narrow temporal window of activity and are therefore limited in what they can offer regarding developmental trends over a longer period. However, this is where the architecture of a ship can provide additional information. A ship’s construction is the product of a shipbuilding tradition of a lengthy trajectory, the culmination to that point of technology and experience gathered over many years and distilled into the particular form of an individual vessel. In addition, specific choices in nautical architecture can reveal much about the intentions of the builders, because different features offer certain advantages and disadvantages. Constructing a ship inevitably requires certain compromises and trade-offs. For example, greater cargo capacity requires the sacrifice of speed and mobility, and a shallower draft allows for the navigation of shallower waters like rivers and estuaries, but compromises the stability of the vessel in rougher seas. Thus, several aspects of shipwrecks provide different levels of insight into the prevailing conditions of the contemporary society. According to X. Nieto (2009), the relatively young discipline of nautical archaeology is currently undergoing a revolution, and is already entering what he calls the “third phase” of the discipline. In its earliest stage of development, a submerged wreck was investigated primarily, or even exclusively, for its cargo. Even then, the study of a cargo was often motivated predominantly in order to establish the date of the wreck. As the discipline developed and became more sophisticated, the wreck and the cargo were studied as a single unit. In this second phase of the discipline, the relationship of cargo and vessel became 251 important, offering new ways of considering how the architecture conditioned the cargo, the route, and commercial capability of the ship. Finally, in the present phase, wrecks are considered within the larger context of the historical situation on land, taking into account diverse factors like the profitability of the voyage, techniques of cargo stowage, weather conditions, navigation technology, trade routes, port infrastructures, and more. Even a comparatively simple wreck can present unique challenges, including analyzing often heterogeneous cargo elements (as well distinguishing between commercial cargo and items belonging to the crew), accounting for redeposition as a result of wave action or currents, analysis of the vessel itself (including the identification of various specific parts of a wooden ship), and attempting to place it within the broader context that Nieto describes as characteristic of nautical archaeology’s “third phase.” It is for this reason that the study of a single wreck often comprises an entire dissertation 286 or a substantial monograph. 287 Clearly, my goals are much more modest, and for practical reasons, I will study two aspects of the archaic shipwreck evidence: the technological shift in naval architecture that began in the 6 th c. BCE, and the assorted cargoes of several shipwrecks from the western Mediterranean during the same period. On the basis of both of these bodies of evidence, I argue that the most reasonable interpretation of the material is that it reflects a substantial growth in both volume and complexity of maritime commerce during the 6 th c. BCE. Ancient Greek Naval Architecture What we know, or more appropriately what we can guess, about ancient Greek shipbuilding is still a work in progress, in large part due to the paucity of evidence. 288 For the organic remains of a wooden vessel to 286 Ex., Krotscheck (2008), Vivar Lombarte (2013). 287 Ex., Nieto & Santos (2008). 288 The shipwreck evidence for the archaic period is frustratingly spotty. Parker (1992) is still an important source, but of course only catalogues shipwrecks discovered prior to 1992, to which we can add subsequent discoveries that include the Point Lequin 1A (Long, Miro, & Volpe, 1992), Grand Ribaud F (Long, Drap, & Gantès, 2002, 2003; 252 survive more than two-and-a-half millennia submerged beneath the surface of the ocean requires a highly fortuitous set of circumstances. This situation is further exacerbated by the fact that the discipline of maritime archaeology itself is a comparatively young one. As a result, our pool of evidence remains limited. In fact, to date there have been no Greek ships dated securely to earlier than the 6 th c. yet excavated. 289 Nevertheless, through a process of juxtaposing the material evidence with the rare written sources, as well as extrapolating the lines of development on the basis of archaeological finds prior to and after the period in question, our understanding of archaic Greek naval architecture, at least in its broader outlines, can be traced with some degree of confidence. Based on the evidence, it appears that the 6 th c. BCE ushered in the initial stages of a transformation of Greek naval architecture. In both its extent and rapidity, it is not an exaggeration to say that this transformation would be arguably the most dramatic and profound metamorphosis in Mediterranean wooden shipbuilding for nearly two thousand years. However, to appreciate the nature and significance of this revolution, some basic background in the principles and broader outlines of Mediterranean naval architecture are necessary. 290 Long, Gantès, & Rival, 2006), the Cala Sant Vicenç wreck (Nieto & Santos, 2007, 2008; Nieto, Santos, & Tarongí, 2004, 2005a, 2005b, 2006), the Pabuç Burnu wreck (Greene, Lawall, & Polzer, 2008), as well as the Punico-Iberian shipwrecks such as the Mazarrón I and II (Gómez-Gil Aizpurua & Sierra Méndez, 1996; Marín Baño & Zambrano, 1995; Negueruela, Pinedo, Gómez, Miñano, & Barba, 1995; Negueruela, 2004) and the Bajo de la Campana wreck (Mederos Martín & Ruiz Cabrero, 2004), the recent excavations of which have been completed (Polzer, 2014), but are still awaiting full publication. 289 The earliest excavated Greek shipwrecks are still the Giglio wreck, dated to approximately 600-580 BCE (Bound & Vallintine, 1983; Bound, 1985, 1991), the partially excavated and poorly published Miet 3 – 600-550 BCE (Krotscheck, 2008), and the Pabuç Burnu wreck – 575-550 BCE (Greene, Lawall, & Polzer, 2008). One continuing source of curiosity and frustration is the recent discovery of two wrecks off the Turkish coast that have been preliminarily identified as Greek vessels and dated to the end of the 8 th /early 7 th c. BCE, but that at present have yet to be excavated systematically (Greene E. , 2011). Unfortunately, through personal communication with Elizabeth Greene, it appears that due to recent changes to the legal environment and permitting process in Turkey, the likelihood that the two wrecks, one at Kekova Adası and the other at Tektaş Burnu, are destined to remain unexcavated for the foreseeable future. 290 What I offer here is a very brief outline of the subject. For a more thorough discussion, de Juan (2013) provides an excellent survey, although he focuses more intently on Hellenistic and Roman naval architecture; Mark (2005) is focused on the Archaic Period, but several important wrecks have been excavated and published since the publication of his work; Polzer (2011) is the most up-to-date survey, but Pomey (2005) remains the most thorough treatment. 253 First and foremost, all ancient vessels were built using the “shell first” method, by which the hull was assembled initially, and then the reinforcing cross-members were fitted to the hull at a later stage. This is in contrast to the “frame-first” method, which was ultimately to become the standard practice in wooden vessels, but does not appear until the 7 th c. CE with the Saint Gervais II. Among vessels built with the “shell first” method, it appears that two main types of construction can be identified: “sewn-plank” or laced construction, 291 and mortise-tenon-peg (MTP) assembly. 292 Broadly speaking, the sewn-plank method is precisely what the name implies, namely a system of joinery whereby the architectural elements of the ship are attached to one another by a system of stitching, where twine made of organic fiber is run through perforations made in the parts to be joined and then drawn tight, binding them together. By contrast, the mortise-tenon-peg technique employs a system of rectangular mortises that are cut from the pieces to be joined, into which are inserted rectangular tenons. After the pieces are fitted together, pins are driven in perpendicularly so that they penetrate both the architectural members being joined as well as the tenons, holding them fast (Figure 1). 293 291 Polzer (2009) argues in favor of the term “laced” rather than “sewn-plank,” although he admits that both terms are in common usage. His basic argument is that so-called sewn-plank ships employed lacing in numerous other applications (e.g., strakes to end-post, garboard to keel, frames to hull, etc.), and therefore to refer to these vessels as sewn-plank is somewhat reductive insofar as it elides the other various uses of ligatures. At the same time, there is ample evidence to indicate the existence of a regional shipbuilding tradition in the northwest Mediterranean that employed ligatures to affix the frames to the hull, but utilized pegged mortises and tenons to join the strakes (Marlier, 2005, 2016; Pomey, 2011a; de Juan, 2013). Referring to ships that employ lacing to join the strakes to each other as “laced” therefore risks conflating these very different shipbuilding traditions. As a result, I prefer the more specific term “sewn-plank” for the shell-first, ancient wooden vessels that employed ligatures to join the strakes together, despite the problematic nature of the terminology; on the terminological difficulty in distinguishing these various methods of construction, see Marlier (2007). 292 What I am calling here “mortise-tenon-peg” or “MTP” construction is more commonly referred to as simply “mortise and tenon” by most authors. However, as we shall see, both the Cala Sant Vicenç and Pabuç Burnu wrecks utilize a system of mortises and tenons without fastening pegs. In addition, although they are much later, one of the hallmarks of the transition from shell-first to frame-first construction beginning as early as the 4 th c. CE was the utilization of unpegged tenons and mortises (e.g., St. Gervais B, Yassi Ada 7). For the sake of clarity, then, I have preferred to use the more descriptive term for the full system in order to avoid any possible confusion. 293 It should be noted that the type of joint described here is not a pure mortise-tenon joint, in which a mortise is cut from one of the pieces to be joined, and a portion of the other element is shaped into a tenon that is fitted into the mortise. The type of joinery described above employs a “false” or “loose” tenon. My thanks are owed to Dr. Daniel Richter for his assistance in elucidating some of the finer points of carpentry. 254 Of these two methods, the sewn-plank method is by far the oldest, dating back to at least the middle of the 3 rd millennium BCE with the Cheops ship in Egypt. This vessel, dated to approximately 2600 BCE, already shows a fully mature shipbuilding technology, suggesting that the method of construction has its origins even earlier (Lipke, 1984). 294 This technique is also geographically widespread, with examples found in various parts of the world, including Israel, the Adriatic, Russia, Sweden, and Kenya (Marlier, 2005). In addition, this method of construction continues to be used to the present day as seen in the river boats of Egypt and the dhows of India, attesting to the viability of the method. Despite the efficiency of sewn-plank construction, it nevertheless produces a structurally inferior product to MTP joinery, all things being equal, and this is due to the respective advantages of each construction method in withstanding the types of stresses which any seagoing vessel must resist. In particular, any wooden plank vessel must at all costs avoid two structural issues. The first of these is horizontal or transverse displacement, where the planks separate away from one another. The second is lateral or longitudinal displacement, where the planks slip past one another. For obvious reasons, both of these issues are anathema to the structural integrity of a wooden plank vessel. Sewn-plank construction is extremely effective at avoiding the former. However, as one might suspect, this method is less effective in guarding against the latter, and it is in this respect that MTP construction finds its greatest advantage over sewn-plank construction. This structural superiority is critically important in the case of shell-first construction, because unlike in the case of frame-first vessels, the bulk of the structural load in boats constructed via the shell-first method is borne by the hull itself, and not the frames (Pulak, 2003; Gertwagen, 2014). The relative structural unimportance of the frames in shell-first boats can be seen by 294 We should point out that there are numerous substantial differences between the type of sewn-plank construction of the early Egyptian vessels and the Greek sewn-plank ships that are discussed here, including the absence of a keel in the Egyptian vessels, a transverse rather than longitudinal lacing pattern, and more (Pomey, 2011a), all of which suggest a separate and distinct shipbuilding tradition, despite the use of ligatures to attach the strakes to one another. In addition, the Egyptian sewn-plank shipbuilding tradition appears to have disappeared long before the Archaic Period in the Mediterranean (Pomey, 2011b). 255 comparing the size and density of frames between shell-first and frame-first boats. In the former, cross members are fewer in number and much less robust, whereas in the latter they are both more numerous and sturdier (Pomey, 2011a) It was once believed that sewn-plank boats had disappeared by the first millennium BCE. However, the discovery of a number of vessels constructed by this method has unequivocally confirmed the viability and prolonged existence of sewn-plank technology in antiquity. Most significantly, it now is clear that ships built using both methods shared the waters of the Mediterranean for much longer than once thought. For example, the presence of MTP construction is clearly attested in the Uluburun wreck, found off the coast of Turkey and dated on the basis of dendrochronological analysis to the last quarter of the 14 th century BCE (Pulak, 1996, 1998, 2003; Manning et al., 2009). 295 By contrast, the continued use of sewn- plank construction in Greek shipbuilding is visible at least as late as the end of the 5 th /beginning of the 4 th c. BCE in the Ma’agan Mikhael wreck (Kahanov & Pomey, 2004a, 2004b). Therefore, the archaeological evidence seems clear: instead of disappearing with the arrival of MTP construction, sewn-plank construction co-existed in the Mediterranean for a nearly a thousand years, if not more. The Literary Evidence In a famous and much-discussed passage in Book V, Odysseus constructs a vessel to take him off the island of Ogygia. Calypso provides Odysseus with an adze and a double-bitted axe (V.234-237), then leads him to the edge of the island where there are tall trees of poplar, fir, and alder. These trees are old and dry (well-seasoned), and will float well because they are light (V.237-240). Odysseus cuts down twenty trees and trims them with the adze, and then proceeds to assemble the ship itself: ξέσσε δ᾽ ἐπισταµένως καὶ ἐπὶ στάθµην ἴθυνεν. τόφρα δ᾽ ἔνεικε τέρετρα Καλυψώ, δῖα θεάων: τέτρηνεν δ᾽ ἄρα πάντα καὶ ἥρµοσεν ἀλλήλοισιν, 295 See Figure 2. 256 γόµφοισιν δ᾽ ἄρα τήν γε καὶ ἁρµονίῃσιν ἄρασσεν. ὅσσον τίς τ᾽ ἔδαφος νηὸς τορνώσεται ἀνὴρ φορτίδος εὐρείης, ἐὺ εἰδὼς τεκτοσυνάων, τόσσον ἔπ᾽ εὐρεῖαν σχεδίην ποιήσατ᾽ Ὀδυσσεύς. ἴκρια δὲ στήσας, ἀραρὼν θαµέσι σταµίνεσσι, ποίει: ἀτὰρ µακρῇσιν ἐπηγκενίδεσσι τελεύτα. ἐν δ᾽ ἱστὸν ποίει καὶ ἐπίκριον ἄρµενον αὐτῷ: πρὸς δ᾽ ἄρα πηδάλιον ποιήσατο, ὄφρ᾽ ἰθύνοι. φράξε δέ µιν ῥίπεσσι διαµπερὲς οἰσυΐνῃσι κύµατος εἶλαρ ἔµεν: πολλὴν δ᾽ ἐπεχεύατο ὕλην. τόφρα δὲ φάρε᾽ ἔνεικε Καλυψώ, δῖα θεάων, ἱστία ποιήσασθαι: ὁ δ᾽ εὖ τεχνήσατο καὶ τά. ἐν δ᾽ ὑπέρας τε κάλους τε πόδας τ᾽ ἐνέδησεν ἐν αὐτῇ, µοχλοῖσιν δ᾽ ἄρα τήν γε κατείρυσεν εἰς ἅλα δῖαν. “[Odysseus] skillfully planed and straightened them with a chalk line, While Calypso, divine among goddesses, brought augers. He bored holes in all of the timbers and fitted them together, Then hammered them in place with dowels and fasteners. As wide as a man marks the curve of the bottom of a vessel, A broad cargo ship, knowing well the art of wood-working, Odysseus made the broad raft to the same measure. Setting up the half-deck and fitting it to the many cross-braces, He built it. Then he finished it with long strakes. Inside he set the mast, and attached a yardarm to it. He added a steering-oar in order to guide it accurately. He shored it up with plaited willows throughout As a protection from the waves, and covered this with abundant brush. Then Calypso, divine among goddesses, brought cloth To make sails, and he fashioned them skillfully, Then attached rigging, reefing lines, and cleats to the ship And hauled it down to the ocean with levers.” V.243-261 Many opinions have been offered on the significance of this passage, and it has yielded numerous divergent interpretations. A great part of the difficulty lies in the fact that many of the terms used to indicate different types of fasteners have many potential meanings. For example, the term that I have translated here as “dowel” (gomphos) can mean “bolt” or “dowel,” but can also refer more generally to a generic bond or fastener. Also, the term that I have translated as “fastener” (harmonia) is usually simply a means of joining, but can also mean “framework” or even “stringing” in the case of musical instruments. Obviously, given the malleable nature of the terms involved, this passage is problematic at best, and 257 probably not sufficient, in and of itself, to serve as the basis for a reliable examination of contemporary shipbuilding techniques. And yet, many have tried. One of the more influential interpretations has been that of Casson (1964, 1971), who argued that the Odyssey describes a type of construction employing mortises and tenons (the harmoniai) that are in turn affixed with transverse pins (the gomphoi). The comparandum that Casson uses for this type of construction is the Kyrenia wreck, dated to the end of the 4 th c. BCE (Steffy, 1985). At the time, the dearth of available archaeological evidence at the time meant that the Kyrenia wreck was his only tangible evidence. Nevertheless, Casson’s interpretation faces some hurdles, most notably an oft-cited line of the Iliad: ἐννέα δὴ βεβάασι Διὸς µεγάλου ἐνιαυτοί, καὶ δὴ δοῦρα σέσηπε νεῶν καὶ σπάρτα λέλυνται… “A full nine years of great Zeus have elapsed, And the timbers of the ships have rotted and the cords given way…” 2.134-135 For Mark (2005), this line is supportive of the thesis proposed by Pomey (1985) that the ships described are “sewn-plank” vessels, constructed by attaching the various elements of the vessel together with woven cords. 296 Fortunately, a number of important discoveries in recent years allows us to test Pomey’s thesis that sewn-plank technology was employed by Greek shipwrights at least as early as the late 8 th c. BCE. 296 Wright (2008) is openly dismissive of this assertion, claiming that it comes on the basis of “flimsy evidence” that fails to explain how mortise-tenon technology, already attested by the late 14 th c. BCE, would have disappeared or been replaced in later centuries and then reappeared by the end of the 4 th c. BCE. This position rests on a tacit assumption that the appearance of pegged mortise-and-tenon joinery automatically rendered sewn-plank construction obsolete. Unfortunately, Wright seems either to have ignored or been unaware of the substantial amount of archaeological evidence demonstrating unequivocally that this was a method of ship construction in wide use in the centuries prior to the Kyrenia wreck, and her arguments represent an example of prioritizing one form of evidence (in this case, the written sources) over another (the archaeology). 258 The Archaeological Evidence So, to summarize briefly – in the ancient Mediterranean, there were two main types of ship construction, distinguished by the method by which the architectural members were joined: laced or “sewn-plank,” which is the older of the two, dating at least to the third millennium BCE, 297 and mortise-and-tenon, of which the earliest known example is the Uluburun wreck, dated to the end of the 14 th c. BCE. Based on the literary and archaeological evidence, it appears that Greek ships were constructed by means of sewn- plank joinery well into the Archaic Period. However, as we shall see, Greek shipbuilders abandoned sewn-plank construction in favor of mortise-and-tenon construction in the second quarter of the 6 th c. BCE, a process that would be complete no later than the end of the 4 th c. BCE. Unless we believe that sewn-plank construction disappeared during the Bronze Age with the advent of mortise-and-tenon construction and then was somehow reinvented by Greek shipbuilders sometime in the Early Iron Age, we must presume that both technologies co-existed from at least the end of the 14 th c. BCE The question that then presents itself is if sewn-plank technology was sufficiently effective to persist for nearly a thousand years after a seemingly superior alternative presented itself, what motivated Greek shipbuilders to abandon this method of ship construction when they did? That will, I hope, become more apparent, but the short answer is that in the 6 th c. BCE, we can see the early stages of the shift away from sewn-plank to MTP construction, and I believe that the conditions of the 7 th and 6 th centuries BCE, especially the growth in both the volume and complexity of maritime trade, prompted this transition. Although the lack of shipwrecks from earlier in the period is not ideal, there are no fewer than eleven shipwrecks with sufficient hull remains to allow for analysis that can be identified as Greek vessels of the Archaic Period, and which date to the 6 th c. or the very beginning of the 5 th c. BCE. On the basis of this 297 The earliest known examples of sewn-plank construction include the 14 ships found in the boat-grave cemetery at Abydos, conservatively dated to approximately 2675 BCE (with some arguing for an even earlier date) and the Khufu ships dated to approximately to 2550 BCE; see Ward (2003, 2006). 259 evidence we shall see that during the course of the 6 th c. BCE, Greek shipbuilders began to abandon a shipbuilding technology that appears to have had a long history of successful use and development, and adopt an entirely new approach. In order to examine this phenomenon, I will begin with a brief review of the relevant wrecks, in chronological order. 298 Since these have all been published elsewhere, I will say little more than a few sentences about each, instead highlighting the parts of the hulls that were preserved and what they reveal about the respective ships’ construction. 299 Giglio Wreck (600-580) 300 Although originally discovered by divers in the early 1960s, this early 6 th c. shipwreck discovered on a sloping shelf off the coast of Isola del Giglio in waters from 30-60 meters in depth and was not excavated until some two decades later. According to the principal excavator, even in the interval between initial survey and the initiation of formal excavations, the wreck was noticeably looted (Bound, 1991). The surviving hull remains include sections of a rabbeted endpost, 301 the garboard and two strakes from the port side, and the garboard from the starboard side, as well as fragments of frames. 298 This chronological ordering is more a choice of convenience than anything else. It is difficult to construct much of an argument based on the relative dating of the wrecks, for several reasons. First of all, as with a great deal of ancient archaeological evidence, the dating cannot be established more precisely than a range, in this case of fifteen years for even the narrowest date range. Second, the date ranges of several wrecks either overlap, such as the Pointe Lequin 1A (520-480), the César 1 (510-500), and the Gela 1 (500-480), or are even impossible to distinguish from one another, as with the Jules-Verne 9, Jules-Verne 7, and the Cala Sant Vicenç wreck, all dated from 525-510. Finally, and perhaps most importantly, the date of the wreck is not equivalent to the date of construction, nor to the stage of naval architectural development that it may represent. For example, I believe that the Cala Sant Vicenç wreck exhibits a slightly later stage of ship construction than the Jules-Verne 9, but earlier than the Jules-Verne 7, even though the actual wrecks are all dated on the basis of their cargoes to the same date range (525-510 BCE). 299 The study of naval architecture and shipbuilding is rife with specialized terminology that can often be confusing even to those familiar with the material. In order to make some of these terms more clear, it may be helpful to consult a glossary of terms such as Steffy (2011) available here: https://tinyurl.com/ycxyyupz. I have also included images to help visualize the architectural elements discussed in this chapter (see Figure 6 and Figure 7). 300 Bound & Vallintine (1983); Bound (1985, 1991). 301 Bound (1991) identifies the endpost as part of the keel, but Nieto & Santos (2008) as well as Kahanov & Pomey (2004a) argue that, on the basis of the angle of the rabbet, the curve of the piece, and the proximity to the stern-post, that it was an endpost. 260 The Pabuç Burnu Wreck (575-550) 302 Discovered in 2001 and excavated in 2002 and 2003 at a depth from 30-50 meters. Hull remains consist of the poorly preserved sections of six planks, the largest three of which are approximately 200 cm x 30 cm. Because no other architectural members have survived, it is not possible to identify with any certainty their location within the original ship. Bon Porté 1 (540-510) 303 The origins of this vessel were originally uncertain, but since its initial discovery in 1971 it has been identified as Greek from Massilia. As with the Giglio wreck, the site was heavily looted. The hull remains are in a poor state of preservation and include a nearly 3 m length of the keel, fragments of both garboards, one starboard strake and two portside strakes, five frames, and a mast step. Based on the keel dimensions, Kahanov & Pomey (2004b) suggest an original length of 10 m. Jules-Verne 9 (525-510) 304 and Jules-Verne 7 (525-510) 305 Found together during excavations off the coast of Marseille in 1993, the position of the remains and the complete lack of any associated cargo suggests that these two wrecks were intentionally abandoned. Dating on the basis of stratigraphic associations provides a date range towards the end of the 6 th c. BCE (525-510 BCE). The preserved section of the Jules-Verne 9 consists of roughly the aft half of the boat and is approximately 5 m long x 1.5 m wide. The Jules-Verne 9 preserves the most complete set of elements of a Greek sewn-plank vessel, including parts of the keel, endpost, strakes, and frames. It exhibits all of the prototypical features of Greek sewn-plank construction. The boat is thought to have been a small fishing vessel, approximately 10 m in overall length. The hull remains of the Jules-Verne 7 consist of an assemblage of fragments totaling some 14 m in length and 4 m in width, including full frames and several 302 E. Greene (2003, 2005); E. Greene, Lawall, & Polzer (2008); Polzer (2009). 303 Joncheray (1976); Pomey (1981, 2002a); Hadar & Kahanov (2004). 304 Pomey (1995, 1999a, 1999b, 2001, 2002b). 305 Pomey (1995, 1999a, 1999b, 2001, 2002b). 261 strakes that are preserved at their extremities, revealing the manner in which they were attached to the endpost. The strakes were attached via MTP construction, while the typical Greek lacing system (tetrahedral notches, tension pegs, double-helix lacing pattern) was used to attach the strakes to the keel rabbet and endposts. The full frames, consisting of floor timbers and futtocks attached by hook scarfs, were crenelated at their lower surface to provide access to the sewing system and trapezoidal in cross- section, although they were attached to the hull by iron nails. The Jules-Verne 7 is thought to have been a merchant vessel approximately 16 m in overall length. Cala Sant Vicenç Wreck (525-510) 306 Discovered in 2002 off the northeast coast of Mallorca near the town of Palma after an initial survey conducted under the joint auspices of the Centre d’Arqueologia Subacuàtica de Catalunya (CASC) and the local government of the island of Mallorca. In all, the surviving section measures roughly 6 m in length and 4 m in width, consisting of part of the keel and false keel, four cross-members (with traces of a fifth), parts of four floor boards, and parts of four strakes on one side and five on the other, including both garboards (Nieto & Santos, 2008b). Pointe Lequin 1A (520-480) 307 First discovered in 1985 off the coast of the Île de Porquerolles (approx. 50 km W/SW of Marseilles) at a depth of between 8-15 m, the Pointe Lequin was one of what was initially believed to be three ancient wrecks in the area (the Pointe Lequin 1, 2, and 3). During the excavations carried out from 1986 to 1993, it became clear on the basis of finds that the site of the Pointe Lequin 1 actually contained the remains of two wrecks, one of which is the Pointe Lequin 1A, dated on the basis of its cargo to the end of the 6 th 306 For preliminary results of the excavations, see Nieto, Santos & Tarongí (2004, 2005a, 2005b, 2006); see also Nieto & Santos (2007); the definitive publication is Nieto & Santos (2008). 307 Long, Miró, & Volpe (1992), Krotscheck (2008). 262 c./beginning of the 5 th c. BCE (520-480). Of the wooden remains, part of the hull and keel(?) with mortise and tenons, 7.5 cm pieces of strakes, and two parts of the rudder, still attached to the hull, were recovered. Grand Ribaud F (510-490) 308 Discovered in 1999 by submersible at a depth of 60 m near the island of Grand Ribaud (approximately 2 km west of the Île de Porquerolles where the Pointe Lequin wrecks were discovered), the site was partially excavated and also investigated in 2000 and 2001 using remote operated vehicles (ROVs) and photogrammetry. Hull remains consist of a 3 m x 1.2 m section that includes portions of the planking on one side of the keel, the keelson, a stanchion, and two frames. These remains were recorded in situ and not removed. The keel, if it survives, could not be observed since it was obscured by the keelson. The strakes were attached to each other by MTP joinery but sewn into the rabbeted sternpost. Only fragments of floor timbers remain of the frames, but these show hook scarfs to accommodate futtocks. They were also crenelated and trapezoidal in cross-section, even though they were attached to the hull with iron nails. The length of the boat has been estimated to be 25-30 m in length. Place Villeneuve-Bergemon/César 1 (510-500) 309 Found off the coast of Marseilles in 1997, the César 1 appears to have been deliberately abandoned just as the Jules-Verne 7 and the Jules-Verne 9. Dating on the basis of stratigraphy provides a date range at the very end of the 6 th c. BCE (510-500 BCE). Hull remains measured approximately 6 m in length by 1 m in width, including a section of the keel that was completely preserved to scarfs at either end, and fragments of and endpost, garboard, five strakes, and two frames. The overall length is estimated to have been roughly 10 m in overall length, similar to the size of the Jules-Verne 9, but in its construction details it bears a strong resemblance to the Jules-Verne 7. The strakes were attached to each other by MTP joinery 308 Long & Rival (2007); Long, Gantès, & Rival (2006); Long, Drap, & Gantès (2002, 2003); Long & Sourisseau (2002); Pomey & Rival (2002). 309 Pomey (2001). 263 and sewn into rabbets at stem and stern with round dowel coaks and the typical Greek lacing system. Full frames consisting of floor timbers attached to futtocks by hook scarfs, crenelated and trapezoidal in cross- section, were attached to the hull with iron nails. Gela 1 (500-480) 310 Discovered in 1988 off the southern coast of Sicily and partially excavated from 1989-1992. Substantial remains of the hull survive, preserving a length of 17-18 m and nearly 7 m in width. In fact, the ship is nearly complete, with portions of the endposts, keelson, keel (but no false keel), and mast step, as well as seventeen floor timbers and strakes from both sides of the hull. The vessel’s construction is uniquely mixed, combining the two major methods of joinery. Up to the third strake, the planking is attached by a combination of round dowel coaks and traditional Greek lacing. Above the third strake, the planks are attached by a combination of round dowel coaks alternating with MTP joints. Full frames consisted of (Panvini, La nave greca di Gela, 1993)floor timbers joined to futtocks with hook scarfs. The floor timbers were crenelated on their lower surface to accommodate the ligatures, and they were trapezoidal in cross- section although attached to the hull with metal nails. Greek Ship Construction of the 6 th c. BCE Characteristic Features So, what are some possible conclusions that may be drawn from these wrecks regarding the state of Greek naval architecture of the 6 th c. BCE? First of all, it is absolutely clear that these vessels are all products of a coherent, mature shipbuilding tradition. The specifics of their construction are simply too similar and distinctive to be explained any other way. Also, tremendous thought and care are evident in the Greek sewn-plank system, consisting of a combination of features that complement one another and work together to form an integrated whole. These shared features include: 310 Panvini (1993, 2001a, 2001b). 264 • Tetrahedral perforations for the ligatures (Figure 3) • “Crow’s foot”/”double-helix” stitching pattern (Figure 4) • Tension pegs inserted in the perforations to stabilize stitching (Figure 3) • Use of wooden coaks to counteract lateral displacement (Figure 5) • Trapezoidal cross-members with rounded interior surfaces (Figure 5) • Crenelated cross-members (Figure 5) • Hook scarfs between floor timbers and futtocks (Figure 5) • Longitudinal stitching • Seam wadding Perhaps most distinctive is the “crow’s foot” stitching pattern, facilitated by tetrahedral perforations 311 through which the ligatures were threaded and extended longitudinally along the length of the strakes, and then into which tension pegs were inserted to increase and maintain the tension of the lacing. Another shared construction detail is the trapezoidal cross members which were crenellated, with notches cut out at the points where they would have come in contact with the seams of the hull planks. In addition, fabric treated with pitch was placed over the seams and sewn in place in order to increase the tension of the ligatures and enhance water resistance at the joints, and then the entire interior surface of the hull was coated with pitch as well. Each of these features was clearly designed to improve the functionality of the sewn-plank vessel. For instance, the crow’s foot stitching helped to reduce the susceptibility of sewn- plank construction vessels to lateral displacement, and the tension pegs of course contributed to greater stability of the joints. Also, the crenellation of the cross members provided the dual benefits of both avoiding damage to the stitching where the cross members would have rubbed against them, and by facilitating repairs once the boat was assembled since one could access the stitching at the seams without having to remove the cross members entirely. Also, the use of wooden coaks to affix the strakes to one 311 As Pomey (1999a) points out, the tetrahedral perforations in sewn-plank shipbuilding in the Mediterranean have so far been discovered only in Greek contexts, and predominantly in wrecks from the Archaic period. 265 another as well as between the garboard and keel would have helped to stabilize the members during the process of stitching them together, and of course increased resistance to lateral displacement. These distinguishing features of the archaic Greek shipbuilding tradition are clearly innovations that are meant to work with and augment the effectiveness of other characteristic traits. This combination of details, when taken together, exhibits a complex, interworking technology that is reflective of a lengthy period of development which does not spring forth ex nihilo. 312 At this point, I would like to return briefly to the literary sources. As Mark (2005) has argued, the (admittedly brief) description of sailing vessels in the Iliad (2.134-135) suggests that they were sewn- plank vessels, a contention that is given even greater weight by the more extended description in the Odyssey (V.243-261) of Odysseus’ construction of his “raft” (schedia). I have already argued in favor of translating gomphos as “dowel,” even though it can also be used in a more general sense to indicate a generic bond or fastener. There is, I believe, a reasonable amount of internal evidence in the text to support this translation, since we are told that among the tools Calypso provided were teretra (“drills/augers”). To be sure, the use of augers is consistent both with the use of round dowel coaks that are a feature of Greek sewn-plank construction, as well as with the perpendicular wooden tree-nails that appear in mortise-tenon-peg construction. However, the mention of gomphoi in connection with the teretra suggests association with the former rather than the latter form of shipbuilding, a conclusion that is much more consistent with the archaeological evidence. Also, I would like to revisit the term harmonia, which I translated earlier as “fastener.” As I pointed out, this term can also mean “framework” or even “stringing” in the case of musical instruments. In the light of the archaeological evidence for the exclusive utilization of sewn-plank versus mortise-tenon construction in the Greek ships of the 6 th c. BCE, it seems very possible that the Homeric description 312 This argument is echoed by Pomey & Rieth (2005), Mark (2005), and Nieto & Santos (2008), among others. 266 refers specifically to the construction of a sewn-plank vessel. There are those who discount this possibility, either because the descriptions are considered too vague, or on the grounds that they are instances of deliberate archaizing. 313 Quite to the contrary, when juxtaposed with the archaeological evidence, the Homeric texts (especially the Odyssey) seem to contain a surprising level of detail and accuracy that is remarkably consistent with the material record. Thus, I would suggest that when taken together, the strong agreement between the literature and the archaeology, as well as the fact that the surviving hull remains of the 6 th c. BCE clearly seem to be products of a lengthy evolution, indicates that sewn-plank construction was the standard practice for Greek shipbuilders at least as early as the late 8 th c. BCE, even in the (as yet) absence of any direct physical evidence. The 6 th -Century Revolution It is clear that there are several vessels that, when taken together, form a core that exhibits most if not all the canonical features of Greek sewn-plank construction. 314 Included in this category are the Giglio wreck, the Jules Verne 9, the Bon-Porté 1, the recently excavated Pabuç Burnu wreck, and the Cala Sant Vicenç wreck. All of these wrecks share so many features that they are almost identical in their construction. In fact, the sole difference is the use of rectangular tenons instead of round dowel coaks in the Pabuç Burnu and Cala Sant Vicenç wrecks. In every other discernible respect, their construction methods are identical. However, we must also note a second group of vessels that, although still retaining many elements of the canonical Greek sewn-plank shipbuilding technique, display several features that indicate a shift away from sewn-plank construction and an increasing adoption of MTP joinery. This category includes the Jules-Verne 7, the César 1, the Gela 1 and 2, the Pointe Lequin 1A, and the Grand 313 See, for example, Wright (2008), who makes both of these arguments, and dismisses Mark’s (2005) interpretation. The persistence of counter-arguments is surprising, especially given that the continued existence of sewn-plank construction in archaic Greek ships has been a settled question for nautical archaeology for some time. The fact that this is even a matter of debate is another example (of many) for the need for greater collaboration between history and archaeology. 314 Pomey & Rieth (2005) refer to this type of construction as indicative of the archaic Greek family of vessels. 267 Ribaud F. 315 Together, they allow us to trace the broad outlines of what appears to have been a remarkable shift in Greek ship construction that was initiated during the course of the 6 th c. BCE. 316 Table 1: Construction details of 6 th c. BCE shipwrecks Ship- wreck Date Strake Attachmt Coaks Frame- Hull Attachmt Strake- Endpost Attachmt Frame Shape Frame- Futtock Joint Est. Length Giglio 600-580 Sewn Dowels Laced n/a Trapezoidal w/rounded upper surface; crenelated n/a n/a Pabuç Burnu 575-550 Sewn Tenons Laced n/a Trapezoidal w/rounded upper surface 317 n/a 17-18m Bon Porté 540-510 Sewn Dowels Laced n/a Trapezoidal w/rounded upper surface; crenelated Diagonal and hook scarf, tree nails 10m Jules- Verne 9 525-510 Sewn Dowels Laced Sewn Trapezoid w/rounded upper surface, crenelated Hook scarf, tree nails 9.5m Jules- Verne 7 525-510 MTP n/a Metal nails Sewn Trapezoid w/rounded upper surface, crenelated Hook scarf, tree nails 16m Cala Sant Vicenç 525-510 Sewn Tenons Laced n/a Trapezoid w/rounded upper surface, crenelated Hook scarf, tree nails 20-22m Pointe Lequin 1A 520-510 MTP Tenons n/a n/a n/a n/a 30m 318 Grand Ribaud F 510-490 MTP n/a Metal nails Sewn Trapezoid w/rounded upper Hook scarf, tree nails 30m 315 For Pomey & Rieth (2005), these ships exhibit a “transitional” phase of Greek shipbuilding. 316 See Figure 8 for a map illustrating the distribution and construction methods of the ships discussed in this chapter. 317 According to Polzer (2009), the frames of the Pabuç Burnu were trapezoidal in cross-section with a rounded upper surface. However, no parts of the frames appear in the catalogue of hull remains, and he does not explain how he reached this conclusion. 318 The calculations of the size of the Pointe Lequin 1A are based on comparisons of the rudder fragments with those of the Grand Ribaud F (see Long & Rival [2007]), as well as the size of the cargo (Krotscheck 2008). 268 surface, crenelated César 1 510-500 MTP n/a Metal nails Sewn Trapezoid w/rounded upper surface, crenelated Hook scarf, tree nails 10m Gela 1 500-480 Mixed (strakes 1- 3 sewn) Dowels & tenons Metal nails Sewn Trapezoid w/rounded upper surface, crenelated Hook scarf, tree nails 22-25m 319 Gela 2 450-425 MTP 320 Tenons Metal nails n/a Trapezoid w/rounded upper surface, crenelated Hook scarf, tree nails 17-18m Ma-agan Mikhael @400 MTP Tenons Copper nails Sewn Trapezoidal w/slightly rounded upper surface, crenelated Hook scarf, tree nails 13.5m Observations In the table above, several trends can be discerned. For one, there seems to be a move towards increasing vessel size. Admittedly, with a sample size of only ten or so examples, this is a tenuous assertion, but in the first group of traditional Greek sewn-plank vessels (the Giglio, Bon Porté, Pabuç Burnu, and Cala Sant Vicenç wrecks as well as the Jules Verne 7), only the Cala Sant Vicenç wreck was estimated to have exceeded 20m in length, and that only barely. 321 But of the next five which belong to the “transitional phase” (Pointe Lequin 1A, Grand Ribaud F, César 1, and the Gela 1 and 2), four are in excess of 22m, and both the Grand Ribaud F and Pointe Lequin 1A had cargo capacities of 40 tonnes or more (Pomey, 2008). 319 As Polzer (2009) points out, both Beniini (2001a, 2001b) and Winters & Kahanov (2004) estimate the original length of the Gela 1 to have been 17 m, but this is incompatible with the reported length of the preserved remains of 17-18m (Panvini, 2001; Kahanov & Pomey, 2004a); a more realistic estimate of 20+m is provided by Long et al. (2001), and Pomey (2008) suggests a length of 22-25m. 320 The surviving hull remains of the Gela 2 were initially assembled via MTP joinery, but a subsequent repair was carried out with laces and exhibits the tetrahedral notches characteristic of Greek sewn-plank construction (Panvini, 2001; Benini, 2001a, 2001b). 321 Earlier estimates of the length of the Giglio wreck put it at 25 m, but these were based on the measurements of a piece of what was presumed to be the keel, but which now appears to have been an endpost (Kahanov & Pomey, 2004b; Nieto & Santos, 2008). 269 At the same time, we see the persistence of certain distinctive forms, such as the trapezoidal frames with rounded upper surfaces, when the structural motives for these features no longer exist. As Polzer (2009) points out, the characteristic form of the frames in Greek sewn-plank vessels was developed specifically to accommodate the demands of sewn-plank construction. The trapezoidal cross-section allows for a narrower surface to come into direct contact with the interior of the hull, while the rounded upper surface provides a greater surface area for the lacing to hold the frames in place but without any sharper edges that would have caused excess wear to the laces. Also, the crenelations avoid direct contact between the frames and the seams between the strakes, which averts wear upon both the stitching and the caulking at the seams resulting from friction between the frames and the interior of the hull, and at the same time would have allowed for access to the seams in the event that repairs became necessary. However, once the frames began to be attached to the hull via metal nails, these characteristic features no longer had any architectural or structural function. Their persistence can only be explained by the conservatism of Greek shipwrights. 322 But most importantly, the move away from sewn-plank shipbuilding to mortise-tenon-peg construction is readily apparent. 323 Chronologically, five of the six earliest wrecks (the Giglio, Pabuç Burnu, Bon Porté, and Cala Sant Vicenç wrecks as well as the Jules-Verne 9) are all completely assembled by the traditional Greek lacing system, and the Jules-Verne 7 uses MTP joinery to attach the strakes to each other and the garboards to the keel, retaining the Greek lacing system only at the stem and stern in the keel rabbet and the endposts. From that point forward (from the very end of the 6 th c./beginning of the 5 th c. BCE), all of the surviving vessels follow the same general pattern as the Jules-Verne 7 (laced at stem and stern with 322 Among others, Nieto & Santos (2008) have pointed out the wealth of ethnographic evidence to indicate that a pronounced degree of conservatism and adherence to tradition are typical features of shipbuilders. 323 See Figure 10 for a rough timeline of the shift from sewn-plank to MTP construction in the Greek shipbuilding tradition. 270 strakes joined by mortises, tenons, and pegs) 324 with the only exception being the Gela 1 and its unique mixture (the first three strakes attached by lacing, the rest with MTP joinery, and lacing also at stem and stern). At the same time, there is also a concurrent shift away from lacing to attach the frames to the hull. In the earlier ships, the frames are all sewn to the hull. In the second, but after the Cala Sant Vicenç wreck (525-510 BCE), the frames are all affixed to the hulls by means of metal nails. I believe that the initial step in this shift can be seen in the Cala Sant Vicenç and Pabuç Burnu wrecks, which are distinctive for one shared feature, namely the presence of rectangular tenon coaks in between the strakes instead of round dowels. As we can see in the table above, all of the other vessels that belong to the Greek family of sewn-plank boats that are earlier than the Cala Sant Vicenç wreck use round dowel coaks in between the strakes. These dowels accomplished two main goals: first, to prevent the architectural members from slipping during the assembly process, and second, to inhibit lateral displacement of the strakes once the assembled vessel had entered the water. Given the way that they were employed in the Cala Sant Vicenç and Pabuç Burnu wrecks, the rectangular tenon coaks offer no structural advantages over round dowels. For one, they are placed too far apart to provide any appreciable benefit in avoiding horizontal displacement. For another, they lack the pegs inserted perpendicularly through the tenons that would provide the additional reinforcement against transverse pressures necessary to take full advantage of MTP joinery. As mentioned previously, the rectangular tenons have the distinct advantage over round dowels in that they provide sufficient surface area to accommodate the perpendicular pegs that complete the full MTP joint. However, in both wrecks, this additional step is lacking. What makes this especially puzzling is the fact that rectangular tenons are more labor intensive and require a greater level of skill and care to produce and employ than round dowels. Given that they occur here, and would have necessitated the additional time and expertise to affix them, but in such a way 324 The strakes of the Pointe Lequin 1A also employed MTP joinery, but it is not clear how the frames were attached to the hull or whether, as with the other contemporary vessels, the attachments at the stem and stern-post employed sewn-plank construction. 271 that their benefits are completely unexploited, their presence is puzzling, and ultimately can only be satisfactorily explained as the very first “baby steps” in the transition from sewn-plank to MTP construction. In examining the evidence of the hull remains of the Greek vessels from the 6 th c. BCE, several aspects emerge. First and foremost, it is clear that the 6 th c. BCE witnessed the initial stages of a fundamental transformation in Greek wooden shipbuilding, from sewn-plank construction to pegged mortise and tenon construction. This shift is readily apparent in the vessels that belong to the “transitional” group (the Jules- Verne 7, the César 1, the Gela 1 and 2, the Pointe Lequin 1A, and the Grand Ribaud F), 325 but it would seem to have begun somewhat earlier, as evidenced by the utilization of rectangular tenons in the Pabuç Burnu (575-550) and Cala Sant Vicenç (525-510) wrecks. This process would continue through the 5 th and 4 th centuries BCE, as evidenced by the Gela 2 326 and the Ma’agan Mikhael 327 wreck, culminating in the Kyrenia wreck at the end of the 4 th c. BCE, in which not a single element was joined with lacing. For most of the 5 th and 4 th centuries BCE, the more complex carpentry at stem and stern-post retained laced construction in Greek vessels, as Greek shipwrights were presumably reluctant to employ a construction method with which they were less familiar in more demanding applications. 328 The joinery at the bow and stern of wooden vessels involves strakes joined edge to edge and the ends of the strakes fitted into rabbets in the stem and stern-post which themselves curve upwards, while the strakes curve inwards. Therefore, these parts of a wooden ship represent the most complex and challenging carpentry. I believe that the persistence of sewn-plank joinery at the bow and stern that we can observe in vessels like the Jules-Verne 9, Grand Ribaud F, and Ma’agan Mikhael that utilize MTP joinery in the rest of the hull is best explained 325 Pomey & Rieth (2005). 326 Panvini (2001); Benini (2001b). 327 Kahanov & Pomey (2004a, 2004b). 328 As de Juan (2013) and others point out, the joinery at the stem and stern of a vessel is extremely delicate and complicated, with the keel joining stem and stern-post that in turn curve upwards while the strakes bend inwards, creating a highly complex junction of oblique angles, with the result that the carpentry at these particular points is by far the most demanding for shipwrights. 272 as an initial lack of familiarity with the technique and faith among Greek carpenters in their own abilities to execute a more unfamiliar system. As a result of this hesitance, Greek shipbuilders were more willing to experiment with an unfamiliar method in those applications that were less technically demanding, but retained the more traditional approach in the parts of the vessel that required the greatest technical precision. Over time, as their confidence grew, Greek shipwrights ultimately adopted MTP joinery throughout. We should also keep in mind that the date assigned to these wrecks applies to the apparent date of the shipwreck event, and is based on the relative dates from the items in the cargo. Clearly, the ships themselves were constructed at some time prior to the shipwreck event itself, but exactly how much time elapsed between the date of the vessels’ construction and their demise is impossible to determine with any certainty. In theory, continuous repairs could keep a boat functioning for a considerable length of time, and there is ample evidence for extensive repairs in numerous instances. Still, we must operate under the assumption that ancient wooden ships had a limited lifespan, as not only the hull would suffer degradation, but also fittings, sails, rigging, etc. would experience wear so that at some point, the necessary repairs to keep a ship afloat would become prohibitive. The fact that the Jules-Verne 7 and 9 as well as the César 1 seem to have been abandoned provide vivid testimony to this reality. Establishing a firm date range, however, offers serious challenges. We do have some hints in the literature, as we are told in the Iliad (2.134-135) that in the ten years that the Greek ships were beached at Troy, the cords have rotted away. However, all indications are that this is partly a result of simple neglect. In fact, it does not seem as if the Greeks are particularly worried that after nine years there is any cause to suspect that their ships will be unable to carry them home again. In fact, Gianfrotta and Pomey speculate that ancient wooden ships would have had a typical life span of approximately thirty years (1981). If this is true, it would certainly support the notion that maritime commerce was capable of producing truly significant returns. It is safe to presume that building a ship would have required a substantial investment, as the archaeological evidence for substantial repairs in numerous ancient wooden vessels would 273 suggest. 329 It is unlikely that extensive efforts to repair ships in order to keep them in service would have been undertaken unless the investment required to construct a new ship was high. Thus, we must believe that overseas trade was sufficiently profitable to justify the cost of construction and produce a satisfactory return on investment within no more than three decades. At the same time, if we are to take this estimate of the average life span of an ancient wooden vessel a norm, then we can back-date the Jules-Verne 7 (525-510) and César 1 (510-500) by three decades, assuming that they were indeed abandoned as it appears, and that this was done presumably because they had reached the end of their effective life span. We also might even push the date of the Pabuç Burnu vessel (575-550) back closer to 600 BCE. Thus, the evidence suggests that the earliest tentative shift from sewn-plank to MTP construction can be seen in the Pabuç Burnu wreck perhaps as early as the first quarter of the 6 th c. BCE, and the adoption of MTP joinery in all but the most challenging parts of the hull had already taken hold around the middle of the century with the Jules-Verne 7. Meanwhile, we must keep in mind that sewn-plank technology appears to have been the dominant construction method for Greek ships for hundreds of years prior, as evidenced both by the literary sources and by the fact that the particular form of Greek ship construction gives every indication of being an evolved, mature technology. The speed with which this transition took place is indeed remarkable, given the inherent predilection for conservatism among shipbuilders in general. We can even see some of the ways in which this tendency is manifested in the preservation of other features, such as the distinctive trapezoidal shape of the frames, well into the 5 th century and beyond. In light of this resistance to change, it seems more than likely that the adoption of an entirely new method of ship construction was prompted by external pressures rather than an organic development. For shipwrights, innovation is more often than not a dirty word. But most importantly, we must ask the 329 For example, the Gela 2 shows evidence of repair to the hull (Kahanov & Pomey, 2004b); also see the substantial repairs to the Cala Sant Vicenç wreck, in which the entire keel and both garboards were replaced, all of the surviving frames appear to have been repurposed from another vessel, and a large false keel was added, probably (at least in part) to enhance stability (Nieto & Santos, 2008). 274 question why this development occurred at this particular point in time. Sewn-plank technology was clearly effective enough to persist for several centuries, at the very least, all while Greek ships plied the same waters as others utilizing the structurally superior MTP construction method. In addition to the Cape Gelidonya (1240-1200 BCE) and Uluburun (1315-1310 BCE) wrecks, the Mazarrón 1 and 2 demonstrate unequivocally that MTP construction was being used in the second half of the 7 th c. BCE, or perhaps a century or so later. 330 When juxtaposed with the unassailable evidence from the Cape Gelidonya and Uluburun wrecks that MTP joinery was being employed in Mediterranean wooden ships in the 13 th and 14 th centuries BCE, we are forced to conclude that Greek sewn-plank vessels co-existed alongside ships in the Mediterranean that employed pegged mortise-and-tenon joinery for nearly a thousand years prior to the point at which Greek shipbuilders began the shift from MTP to sewn-plank construction in the second quarter of the 6 th c. BCE. At this point, I want to make clear that I am not suggesting that the use of ligatures to assemble various elements of wooden vessels completely disappeared at this time. There is ample evidence for the use of ligatures to attach frames to hulls, especially in the northwestern Mediterranean, 331 continuing at least into the first century CE (Marlier, 2007). However, this method of construction utilizes MTP joinery between the strakes, and clearly belongs to a separate, regional shipbuilding tradition that Pomey (2011a) and de Juan (2013) have called a “Punico-Iberian” tradition. What I am referring to here is specifically the shift in Greek shipbuilding from the use of ligatures to attach the strakes to one another (and to the stem and stern-post, and the garboards to the keel), to the use of pegged mortises and tenons in all of these 330 The excavators of the Mazarrón 1 and 2 have dated both wrecks to approximately 650 BCE largely on the basis of the pottery evidence (see Neguerela et al., 2000; Neguerela, 2005). However, de Juan (2012) argues that the pottery evidence is in fact also consistent with a much later date, and that the excavators have conveniently ignored radiocarbon dating evidence of wood samples and of Posidonia oceanica from the Mazarrón 2 that appears to have overgrown the hull almost immediately upon sinking, and he suggests a much later date of the last quarter of the 6 th c. BCE. 331 Ex., the Mazarrón I and II, Binisafuller, Dramont C, Cap Béar 3, Tour Fondue, La Roche Fouras, Cavaliére, Cap de Vol, and Cala Cativa wrecks. 275 applications. The archaeological evidence strongly suggests that this transition began in the first half of the 6 th c. BCE, and would be complete at least by the end of the 4 th c. CE, after which point sewn-plank construction appears to have disappeared completely. In fact, there is not a single identified use of sewn- plank construction in any ship excavated in the Mediterranean later than the Ma’agan Mikhael, dated to approximately 400 BCE (Kahanov & Pomey, 2004a, 2004b). The answer to why this shift occurred when it did may be seen, in part, in the apparently increasing size of the vessels in the 6 th c. BCE. According to Pomey (2008), the fact that such a substantial portion of the hull of the Cala Sant Vicenç wreck is preserved, and that it remains in such an excellent state, allow for estimates of the size of and cargo capacity of the vessel with a high degree of confidence. First, the portion of the hull that remains is substantial enough that it still preserves a great deal of the curvature of the hull, which permits the calculation of the diameter of the vessel with a very small margin of error. Second, the size of the keel cannot stray too far from accepted norms without adversely affecting the navigability of the ship. For these reasons, the estimated dimensions of 20-22m in length, 6.3m in width, and 2.1m in height can be relied upon confidently. The above dimensions, indicating a tonnage of approximately thirty metric tonnes, make the wreck of the Cala Sant Vicenç the largest ancient Greek completely sewn-plank vessel yet discovered. The next largest is the Pabuç Burnu wreck, at an estimated 17-18m in length. The size and cargo capacity of the vessel are significant, because it seems that the Cala Sant Vicenç ship may have approached the maximum feasible length for a cargo vessel constructed entirely via the sewn-plank method, due to the degree of flexibility in sewn-plank construction. As I have already noted, in shell-first construction the hull bears the greatest structural load, while the frames function solely in order to mitigate transverse stresses. As a result, any structural disadvantage in the construction of the hull is significant, and MTP joinery is measurably superior to sewn-plank construction, due to its increased rigidity and consequent resistance to horizontal displacement and torsion. 276 The greater flexibility of sewn-plank construction is not an issue for smaller vessels such as the Bon-Porté 1, with an estimated length of approximately 10m. However, as the hull increases in size, this flexibility becomes cumulatively greater, ultimately affecting the structural stability of the vessel and acting as a limiting factor on the maximum size of ships built using this technology. At different points in time, various approaches were devised to deal with this issue. For instance, the reliefs of Deir el Bahari show ships with tensors and thick cables attached stem to stern, allowing for a piece of wood to be used as a sort of tourniquet in order to adjust the curvature of the ship in different sailing conditions. 332 Ultimately, however, the increased rigidity that is a fundamental feature of MTP construction was the most efficient solution, and in all probability, the only way effectively to build larger cargo ships past a certain critical point. Thus, it would appear that during the 6 th c. BCE, shipbuilders began to respond to external pressures to build larger vessels capable of transporting bigger cargoes across greater distances of open ocean. Such changes would place enhanced pressures on those engaged in maritime trade to increase their capacity to transport a greater volume of goods across greater distances, which would naturally have an impact on ship size during the period. To be sure, there are wrecks like the Bon-Porté I, Jules Verne 9 and the Dattier wreck that appear to have been engaged in cabotage, or the coastal-hugging trade of a limited range and scale. The Bon-Porté I, for example, is estimated to have been about 10m in length, and the Jules Verne 9 about 9.5m in length with a total displacement of about 3 tonnes (Pomey, 2008). Vessels of this size would have been unfit for long-distance crossing of broad expanses of open ocean (“blue ocean” travel). This is consistent with the presumed pattern of cabotage, in which subsistence goods were distributed across limited geographic areas by small, coastal-hugging vessels (Nieto, 1997). 332 See Mark (2005); Nieto & Santos (2008). 277 However, the Cala Sant Vicenç wreck presents a rather different picture. The sheer size of the vessel makes it unlikely that its primary function was as a vehicle for cabotage. With a length of 20-22m and a cargo capacity of approximately 30 tonnes, such a vessel would almost certainly require more substantial harbor installations than the Bon-Porté 1 or Jules Verne 9, both of which were no more than half the size of the Cala Sant Vicenç ship. One of the necessary characteristics of ships associated with the small-scale cabotage type of activity is that they are of sufficiently small size and shallow draft that they are capable of adequately negotiating areas along the coast that often lack deep-water harbors and possess minimal docking facilities, if any (Hohlfelder & Vann, 2000). A ship with the displacement of the Cala Sant Vicenç wreck would have serious difficulty operating under such circumstances. Plus, the time, effort, and energy required to construct such a vessel would presumably be prohibitive were there not some additional incentive to produce such a significantly larger ship. It is for these reasons that we see in the middle of the 6 th c. BCE two simultaneous trends emerging in the construction of Greek wooden ships. On the one hand, we see the shift away from sewn-plank to MTP construction, and on the other hand, the appearance of larger vessels with far greater carrying capacities such as the Pointe Lequin 1A, Grand Ribaud F, and César 1, all estimated at 25m or more in length. I do not believe that this concurrence is the result of chance. Rather, I am convinced that both of these trends spring from the same impetus, namely growing pressures exerted by increased volume of maritime trade. This pressure induced Greek shipbuilders to construct vessels of greater size and cargo capacity that exceeded the structural limitations of sewn-plank technology, and it is the end result of this process that is visible in the naval architecture of the period. Cargo Evidence Unlike hull remains, many items that constituted ship cargoes, particularly metals and ceramics, have a much higher rate of survival. Of course, this is not to say that organic matter did not constitute a major component of many ships’ cargoes. The remains of the Cala Sant Vicenç wreck, for example, included a basket, the contents of which did not survive, but which almost certainly contained some sort of 278 foodstuffs (Nieto & Santos, 2008). Also, we are told in the Odyssey that some of Odysseus’s livestock were pastured on the mainland as well as on Ithaca (XIV.96-98), and they are seemingly rather easily transported when necessary: τοῖσι δ᾽ ἐπὶ τρίτος ἦλθε Φιλοίτιος, ὄρχαµος ἀνδρῶν, βοῦν στεῖραν µνηστῆρσιν ἄγων καὶ πίονας αἶγας. πορθµῆες δ᾽ ἄρα τούς γε διήγαγον, οἵ τε καὶ ἄλλους ἀνθρώπους πέµπουσιν, ὅτις σφέας εἰσαφίκηται. “Along with them came a third man, Philoetius, a leader of men, bringing a cow and fat goats for the suitors. Ferrymen brought them across, and they bring other men As well, whoever comes to them…” XX.185-188 Still, in spite of the virtual certainty that a wide range of goods were regularly transported in ships during the Archaic Period, the reality is that many of them have simply not survived to the present, and thus pottery provides a crucial form of evidence. To be sure, the ability of pottery to serve as an effective proxy for broader patterns of commercial activity is not universally accepted (Lawall, 2005; Orton & Hughes, 2013), nor even whether ceramics constituted an economically significant commodity. 333 Nonetheless, the ceramic evidence is invaluable, if for no other reason than it is what we have, but also because it offers a foundation on which to compare ship cargoes, and to identify tentatively the route of the vessel prior to sinking. However, in order to examine the evidence of cargoes from the period, some adjustments must be made to the data set. First of all, several of the vessels that were examined on the basis of their architecture (Jules-Verne 7 and 9, César 1) were not associated with any cargo, which has led excavators to conclude that they were intentionally abandoned. Therefore, they have been removed from this list. I have also 333 For the argument against pottery as a principal trade commodity, see Vickers (1984, 1994) and Gill (1987a, 1991); the opposing viewpoint is expressed by Boardman (1979, 1988) and Osborne (1996b, 2001, 2007). This is a topic that will be treated more fully in the next chapter. 279 removed the Pabuç Burnu wreck, since it was found off the coast of Turkey, and the cargo is therefore indicative of trading patterns in the eastern Mediterranean. At the same time, I have included several other wrecks from the period that did not retain sufficient hull remains to be included in the first list, but which have preserved sufficient portions of their cargoes to be introduced here. These include the Miet 3 (600- 550 BCE) and the Dattier (@550 BCE) and La Love (575-550 BCE) wrecks. Individual Cargoes Giglio Wreck (600-580) Items such as lamps with evidence of use were identified as belonging to the crew, and along with other shipboard items and tools indicate a crew of Corinthian or east Greek origin. The cargo included lead and copper ingots, two bronze Corinthian helmets and bronze arrow heads, and iron bars possibly used as currency. 334 Ceramics include amphorae (Etruscan, Punic, east Greek, and Corinthian) and fine wares (Etruscan, Corinthian, Samian, Ionian, and Laconian). Miet 3 (600-550) 335 Unfortunately, this wreck has only been partially excavated and remains poorly published, but the preliminary finds include approximately 100 Etruscan amphorae of which the vast majority are of the Py 3B type, with some Py 3A. Excavators also discovered 6-8 Etruscan bucchero nero vessels. La Love (575-550) 336 The cargo contained 183 amphorae, of which 173 (approx. 95%) were Py 3b type. The remainder consisted of Pye 3A type (2%) and several others of Corinthian manufacture (2%). Also among the finds were 92 pieces of fine wares, which were predominantly Etruscan manufacture, including 65 pieces of bucchero nero, as well as eight Etrusco-Corinthian pieces. 334 For iron bars/spits used as currency prior to the adoption of coinage, see Tandy (2007). 335 Liou (1975), Hesnard (2002), Long (2004). 336 Pruvot (1971); Bouloumié (1982); Long, Miró & Volpe (1992). 280 Dattier (@550) 337 Among the finds were approximately 17 amphorae, including 15 Ionio-Massiliote amphorae of the Bertucchi 1 type, one Etruscan Py 5 type, and one east Greek Clazomenian amphora. Part of a stone anchor and a grindstone were also found. Bon Porté 1 (540-510) 338 The ceramics include remains of some 40 amphorae (Chian, Clazomenian, Corinthian, and Greco- Massiliote). Cala Sant Vicenç Wreck (525-510) Several items showing signs of wear and/or use clearly belonged to the crew. All told, approximately 4,600 ceramic fragments were recovered during the excavations from 2002 to 2004, most of which correspond to amphora fabrics. These ceramics were scattered over the area of the vessel and beyond, making it realistically impossible to determine their original location. On the basis of diagnostically significant fragments (rims, handles, bases), the minimum number of individuals (MNI) is 123. Considering the number of pieces that cannot be identified or have been lost, the actual number would almost certainly have been much higher. The amphorae consisted of two main types, along with a scattering of others. The first main type was Ionian-Massiliote/Corinthian in form, with fabric suggesting production in southern Italy. The other main category was amphorae identified on the basis of their fabric as Iberian, consisting of several different forms, suggesting numerous points of origin. Finally, there was a scatter of east Greek amphorae (one Corinthian A, two Chian, and one northern Aegean). As for fine wares, they were almost all Ionian cups (type B2), produced in southern Italy. In addition to those items that could be securely associated either 337 Liou (1974); Calmes (1976). 338 Liou (1974, 1975); Pomey (1981) 281 with the crew or the commercial cargo, several unusual finds were also recovered. These include a wooden pyxis lid, glass paste beads, a bronze Corinthian helmet, decorated fine wares, and a stone jewelry mold. It is unclear whether these were part of the commercial cargo or perhaps destined for use by the crew. In addition to the ceramics, several metal objects were found, both raw metal and finished objects. Numerous tin scraps were discovered, along with a tin ingot weighing 30.6 kg. 339 Also, a large number (136 in total) of double-bladed iron picks were found. These were discovered in bundles, bound with cords made of grass. Finally, the remains of a woven basket, probably to contain some foodstuffs, were preserved among the wreckage. Pointe Lequin 1A (520-480) The vessel is notable for its cargo, of which all told approximately five metric tons were recovered. This includes more than 100 amphorae (about half of which were the “Milet-Samos” type and the rest an assortment of Clazomenian, Ionian-Massiliote, Chian, Corinthian, and Etruscan), 10 pithoi, bronze and terracotta statuettes, various small finds, and more than 2,000 pieces of ceramic tableware. In direct contrast to the typical view of fine wares as “saleable ballast,” fine wares comprised 94% of the cargo of the Pointe Lequin 1A. 340 Grand Ribaud F (510-490) The cargo is dominated by a strikingly homogeneous group of Etruscan (Py 4-type) amphorae, estimated to be 800-1000 in total, and stacked in four or five layers. Other ceramic finds include a few scattered fine wares as well as some coarse wares that are mostly Etruscan in origin. Based on the ceramics, Kahanov & 339 The Iberian peninsula was an important source of tin in antiquity for Phoenician traders (Fernández Jurado, 2002), and the presence of these tin objects on board the Cala Sant Vicenç suggest that Greeks were also interested in trade with Iberia in order to obtain metals, just as has long been supposed of the early Phoenician expansion in the western Mediterranean. 340 Krotscheck (2008, 2015). 282 Pomey (2004b) offer a date range of 515-470 BCE, but Long et al. (2003) narrow that date range to between 510 and 490 BCE, and also conclude on the basis of the ceramics (particularly the coarse wares that they believe belonged to the crew) that the crew was Etruscan. Given the size of the cargo, the length of the boat has been estimated to be at least 25 m. Gela 1 (500-480) The cargo consisted of wine, oil, and ceramics. A total of 55 amphorae were recovered, most of which were from the eastern Mediterranean (31 Chian, 10 Corinthian, 2 Attic, and 6 assorted “east Greek,” along with 4 Greco-Massiliote, and 4 Punic). Various Attic, Laconian, and Ionian fine wares were also presumably part of the cargo. Various other coarse wares, many of which show signs of use indicating that they belonged to the crew, were products of southern Italy/Magna Graecia, suggesting a Greek crew. Observations Upon examining the evidence from various cargoes dated to the 6 th c. BCE, several aspects deserve mention. First, it would appear that three interconnected Greek shipping networks operated in the western Mediterranean (Figure 9). The first connected southern Italy and Sicily with the Aegean, and is evidenced by the cargoes of the Gela 1 and the Giglio and Bon Porté wrecks. The second operated in the northern half of the western Mediterranean, and connected southern and central Italy with southern France. Evidence for this network can be seen in the cargoes of the Miet 3, Grand Ribaud F, Pointe Lequin 1A, and the Dattier and La Love wrecks. A third possible network may have also been operative, connecting southern France with the Balearic Islands and the Mediterranean coast of Spain, as evidenced by the cargo of the Cala Sant Vicenç wreck, although the precise outlines of this network are more hazy. 341 341 The precise details are unclear, and it may very well be the case that the route of the Cala Sant Vicenç ship, as indicated by its cargo that included substantial quantities of Iberian amphorae, was simply an extension of the route between the west coast of Italy and the southern coast of France. However, for the most part, the vessels that show evidence of the Italy-France route (i.e., the Miet 3, Grand Ribaud F, Pointe Lequin 1A, and the Dattier and La Love wrecks) do not contain elements in their cargoes that would indicate that they ventured further west than Massilia, 283 Second, the importance of Etruscan products to maritime trade in the western Mediterranean in this period is abundantly clear, as shown by the cargoes of the Miet 3 (100 Etruscan amphorae of mostly Py 3B type, as well as Etruscan bucchero nero vessels), Grand Ribaud F (800-1000 Etruscan Py 4 amphorae), and La Love wreck (183 amphorae of which 95% were Etruscan Py 3B type). On the basis of these cargoes, it seems evident that Etruscans played an active role in maritime commerce in the region, a fact that his been largely overlooked and unappreciated, especially if we accept the assertion that (on the basis of ceramics that appear to have belonged to the crew) that the crew of the Grand Ribaud F was Etruscan. But most significant for my argument is the range, variety, and complexity of trading patterns visible in the 6 th c. BCE cargoes. To be sure, there are several examples of small, heterogeneous cargoes (the Bon Porté, Giglio, and Dattier wrecks) that one would expect from highly localized, coastal-hugging cabotage of the kind typically assumed to have existed in the Archaic Period. But in addition to these, we also find evidence for larger, more homogeneous cargoes (Miet 3, La Love, and Cala Sant Vicenç wrecks) that would have almost necessarily traded in a few larger ports-of-call and relied upon the existence of trading networks rather than engaging in the geographically limited cabotage, a form of trade that some have referred to as “emporic” (Nieto & Santos, 2008). Finally, we see the appearance of large, highly homogeneous cargoes from Etruria (Grand Ribaud F) and southern Italy (Pointe Lequin 1A) that indicate a point-to-point form of trade that is is almost certainly dependent on high levels of production, and would have necessitated the existence of sophisticated and complex networks capable of distributing the contents of these cargoes across wide geographic areas from a central nodal point. Conclusions When taken together, both the naval architecture and ships’ cargoes from the 6 th c. BCE reveal a great deal about the evolution of maritime commerce during this period. For one, the cargoes offer a highly which suggests that trade routes to Emporion, the Mediterranean coast of the Iberian Peninsula, and the Balearic Islands were part of a more discrete regional trading network. 284 cosmopolitan picture of overseas trade, with producers of goods from the Aegean, Etruria, southern Italy, and the Iberian Peninsula, all connected by overlapping/interlocking networks that allowed goods to move throughout the Mediterranean (see Figure 2). At the same time, the shipwrecks suggest the logistical arrangements that would have supported elevated levels of exchange, with larger vessels allowing for high levels of production to be exported to nodal distribution points, from which they could be distributed by other vessels more adapted to regional distribution, and finally more localized trading patterns. 342 The variety exhibited in the hull remains also finds support in the literary evidence. For example, there are indications in the Odyssey of a degree of variability not only in the size, but also the types of sailing vessels in existence at the time. Odysseus, for instance, initially rejects Calypso’s offer to help him build a raft that will allow him to return home, saying: ‘ἄλλο τι δὴ σύ, θεά, τόδε µήδεαι, οὐδέ τι ποµπήν, ἥ µε κέλεαι σχεδίῃ περάαν µέγα λαῖτµα θαλάσσης, δεινόν τ᾽ ἀργαλέον τε: τὸ δ᾽ οὐδ᾽ ἐπὶ νῆες ἐῖσαι ὠκύποροι περόωσιν, ἀγαλλόµεναι Διὸς οὔρῳ.’ “’But you must be planning something else than my journey, Goddess, Who tells me to cross the broad depths of the ocean in a dinghy, A terrifying and dangerous prospect; for not even do the well- balanced Swift ships make such a crossing, taking advantage of the favoring wind of Zeus.’” V.173-176 342 Specifically, the larger vessels that we see at the end of the 6 th c. BCE would have been unable to navigate rivers, make port in locations without anchorage and port installations, or dock in shallow-water harbors. Thus, in order to distribute the goods that they carried beyond a limited range of seaports would have required the transfer of those goods to other vehicles, with larger ports functioning as distributional nodes. In addition, the larger, more homogeneous cargoes such as that of the Pointe Lequin 1A or Grand Ribaud F could not have been off-loaded piecemeal without destabilizing the vessel and compromising its navigability. Thus, they would have delivered their entire cargo to a single destination. 285 In this passage, Odysseus explicitly contrasts one type of vessel, that I have translated as a “dinghy” (schedia), with “ships” (nēes) that are swift and well-balanced. The typical translation of schedia is a “raft,” although it can mean a sea-going vessel more generally. 343 The reason why I translate it here as a dinghy is that the contrast Odysseus draws here is clearly on the basis of size, which is the quality that adversely affects its ability to brave the open ocean. Admittedly, it may seem rather obvious to point out that ships could vary in size, but it is a fact that is often ignored, and one that has powerful implications for the capability to transfer goods and people across the Mediterranean. Smaller vessels, such as the Bon Porté, Jules-Verne 9, and the Mazzarrón 1 and 2, would doubtless have been sufficient to connect nearby locales and would have been vital, especially considering the often rocky and difficult coastal terrain in this part of the world. At the same time, larger vessels like the Cala Sant Vicenç wreck, the César 1, and the Gela 1 and 2 would have been necessary to make longer journeys across greater distances, and it is clear in this passage that such ships capable of making these kinds of crossings did in fact exist. In fact, we know of the existence of ships with banks of fifty oars (pentekonter) from the Iliad, but we are also told in the Odyssey of smaller oared vessels, still referred to as “ships” (naus), such as that used by the twenty suitors and their attendants who set out to ambush Telemachus on his return from Pylos (IV.778- 779). A crew of this size would have been insufficient to man a pentekonter, but would still require something more sizeable than the kind of dinghy that Odysseus could sail from Ogygia by himself. Finally, the larger vessels like the Pointe Lequin 1A and the Grand Ribaud F could have provided the kind of transport that enabled the development of intensive, high-volume production and exploited the regional and local distribution capabilities of the smaller and medium-sized ships that were clearly active at this time. 343 Plato (Phaedo, 85d) uses the term to describe a raft and for Xenophon (Anabasis, 2.4.28) it is a raft made of skins, whereas for Herodotus (4.48) it is a more generic term for a boat. Iconographic evidence also supports the translation of schedia as a “raft.” For example, the so-called Catalogue of Ships in the Althiburus Mosaic from Tunisia provides images of nearly two dozen different watercraft along with their names. The boat identified as a schedia is a small, two-oared craft without steering oar/rudder or sails (see Gauckler [1905]). 286 But in addition to variations in size, the Odyssey also indicates variability in the types of vessels that were available. Although it would seem that most of the larger ships were equipped with oars, 344 it appears that even some larger boats, like the one in which Telemachus makes his voyage to Pylos and back, may have lacked oars, 345 in addition to the kind of smaller dinghy used by Odysseus in Book V. Also, some vessels may have been designed specifically to perform more specialized functions. For example, the wooden log that Odysseus and his men fashion into a stake to blind Polyphemus (IX.322) is described as being equal in size to the mast for a ship of twenty oars (eikosoros). This ship is also referred to as a “broad cargo ship” (phortis eurys), precisely the same terminology used in the passage in Book V that describes Odysseus’ construction of his schedia (V.247-248), where it is said that he marked out the size of the hull equal to that of a phortis eurys of the kind that crosses the open sea (mega laîtma, lit. “great depths”). Clearly, the term phortis is related to the Greek term for “cargo” (phortos). It is also telling that in both cases, the ship is referred to as “broad,” a feature that one expects of vessels designed specifically to carry large cargoes. Although a wider, flatter hull impairs the responsiveness of the vessel and makes it more difficult to maintain a stable bearing, it provides greater storage capacity, and the choice of such a hull profile is therefore a sacrifice that is made in the design of vessels deliberately constructed for enhanced cargo capacity at the expense of navigability. Thus, we can see diversity in both form (oared v. sail-driven vessels) and function (cargo ships v. war ships), as well as variations in size. In fact, in those cases in which the text is sufficiently unambiguous, it corresponds with everything that we know to be true of shipbuilding from the archaeological evidence. As a result, it seems that we may confidently rely upon its representations to help elucidate those areas that are left blank by gaps in the material record. Ultimately, the textual evidence in and of itself is too vague to support firm conclusions about certain aspects of shipbuilding technology of the late 8 th c. BCE. Fortunately, the archaeological evidence 344 Ex., IV.577-580, IV.780-784, XII.144-152, XIII.76-78. 345 See II.422-428, XV.287-291. 287 presents a far more complex, interesting, and continually evolving picture of archaic ship construction that permits us to re-examine the textual evidence from a much more informed perspective. In addition, archaic literature provides tantalizing hints as to the variety of goods that were transported overseas, as well as the frequency with which goods and people were mobilized. However, the oblique mentions in the written record are insufficient to allow firm conclusions to be formed. But once again, the material record supplies us with further evidence. In particular, the cargoes of archaic shipwrecks provide us with an additional point of comparison from which to make inferences regarding the existence and nature of maritime shipping networks, connections between producers and consumers, and relative levels of activity. Specifically, as we have seen in the archaeological evidence, the naval architectural remains of the shipwrecks of the 6 th c. BCE demonstrate clearly that Greek wooden shipbuilding underwent the initial stages of a fundamental transformation during this century. Greek shipwrights moved away from a highly evolved, mature technology that had proven reliable over centuries of development and refinement, towards an entirely different method of construction. In addition, they were so unfamiliar with this method of ship construction that they retained the more traditional sewn-plank construction in the more complex joinery at the stem and stern. Given this unfamiliarity along with the manifest conservatism of shipwrights, visible in the persistence of outmoded architectural features in Greek vessels, we should assume that the impetus for this transition was external, rather than simply innovation for its own sake. To reiterate, it is not my contention that all forms of laced construction disappeared at this time, nor that all wooden ships in the Mediterranean followed a clear, consistent, and uni-directional evolutionary path. The utilization of ligatures in various applications of wooden shipbuilding in the Mediterranean, especially to attach frames to hulls, continues for centuries (Marlier, 2005; de Juan, 2013), and sewn- plank ships continued to be produced the Red Sea and Indian Ocean (Pomey, 2011a). Instead, I wish to limit my discussion to the empirically observable transition in ship construction in Greek wooden ships from the 6 th c. to the end of the 4 th c. whereby Greek shipwrights ultimately abandoned completely the use 288 of ligatures to join the strakes in favor of pegged mortise-and-tenon construction. This transformation was total and swift, particularly if we consider that there is no evidence of the use of sewn-plank technology in a single Greek vessel after roughly 400 BCE. Compared with the transition to frame-first from shell-first construction that recent discoveries indicate took place over a period of roughly seven hundred years (Mor & Kahanov, 2006; Pomey, Kahanov, & Rieth, 2012), the adoption of MTP joinery by Greek shipwrights was rapid indeed. In addition, we must ask ourselves what motivated this shift during this time, especially in light of the fact that the Greek sewn-plank technique would appear to have co-existed, and even evolved, along with MTP joinery for hundreds of years. To be sure, MTP joinery is structurally susperior to sewn-plank construction, but in the Greek wooden vessels of the 6 th c. BCE we can see a refined, highly developed tradition of sewn-plank shipbuilding that gives every indication of having evolved over an extended period of time. Clearly, this technology was sufficient to serve the needs of Greek seafarers for many years, even centuries, before the 6 th c. BCE, so why abandon that method now? In part, I believe that the motive for this change can be seen in the appearance of larger Greek vessels than any that belong to an earlier date. This additional factor suggests that the dramatic change in Greek shipbuilding technology was prompted, at least in part, by a desire to build larger ships for which sewn- plank construction proved to be inadequate. This in turn seems to be explicable predominantly by a desire to move greater volumes of goods across larger distances, which corresponds to rising levels of maritime commercial activity. The means by which high volumes of goods could have been distributed across extended distances is also indicated by the variety of ship sizes in the archaeological record, as well as the diversity of types and sizes suggested by literary references, all of which would have provided a wide range of transportation capabilities able to perform a variety of functions and accommodate diverse maritime environments and thus reach a large number of potential consumers. The variety of vessel types and sizes would have produced the capacity to trade a local, regional, and inter-regional levels. Also, this diversity would have had the capability to support a complex, interdependent distribution system. 289 Meanwhile, the cargoes of the contemporary wrecks provide an additional level of insight into the maritime commercial activity of the period, and indicate precisely the kind of sophisticated distribution patterns that would have been possible with a diverse range of maritime transportation capabilities. First of all, they demonstrate multiple levels of seaborne commerce, from small, heterogeneous cargoes indicative of highly localized cabotage to large, highly homogeneous cargoes representative of point-to- point commerce which implies the existence of more extensive distribution capabilities. At the same time, we can see in the composition of the cargoes the effects of interlocking/overlapping inter-regional trade networks that connected large portions of the northern half of the western Mediterranean and the Aegean. First of all, they indicate that the Etruscans were highly active participants in archaic trade in the western Mediterranean, especially with Massilia. Also, the cargoes imply several different trade routes, one between southern Italy/Sicily and the Aegean, another between the west coast of southern/central Italy and the southern coast of France, and possibly a third between southern France and the eastern coast of Spain and the Balearic Islands. These trade networks would have been capable of moving a high volume of goods with great frequency, as well as integrating producers and consumers across a broad geographic spectrum, and even to distribute goods to previously isolated areas and to incorporate the demands of previously excluded consumers. In sum, all of the evidence so far is indicative of conditions conducive to complex maritime networks with substantial capacity for transport of high volumes of goods across long distances. What we see, in fact, are precisely the kind of sophisticated distribution capabilities that would have had the potential to provide new sources of wealth that could have produced the disruptive socio-political effects that we see in the contemporary literary sources like the Theognidea, and which appear to have been previewed in the earlier archaic literature, such as the antipathy towards seaborne commerce in the elite-centric Homeric epics, and by contrast, the overt advocacy of overseas trade from the free land-holding Hesiod. Admittedly, this argument depends on a series of speculative inferences, but the confluence of literary sources with the two different sources of archaeological evidence (naval architecture and shipwreck 290 cargoes) all seem to point in the same direction, namely that the mobility of people and goods, even fairly early in the Archaic Period, has been underappreciated. Also, maritime commerce expanded significantly in its breadth and volume during the period, and the impact of this expansion is manifested in various ways, including the archaeological evidence from the naval architecture and shipwreck cargoes of the 6 th c. BCE. When considered within the context of the broader political, social, and demographic changes of the period, it seems all too coincidental to find significant economic developments occurring at the same time. Instead, it would seem that we must reconsider the scope of overseas trade in the Archaic Period, and in particular, how economic growth played a role in the broader social and political changes that were taking place at the time. 291 Chapter Eight: Pottery Distribution as an Indicator of Developing Trade Networks Introduction To this point, I have argued that the contemporary literature strongly suggests that during the seventh and sixth centuries BCE, the profound political and social transformation of Greek society was highly contested, and that the battles were fought in large part along economic lines. On the basis of the literary analysis, it seems that 1) archaic elites were deeply concerned about the perceived threat posed by changing economic relations, specifically widespread redistribution of wealth (characterized as the disjunction of wealth and class), 2) an emergent class of petits bourgeois successfully seized on maritime trade as a means of acquiring wealth that was largely free of elite control and intervention, and 3) this economic contest played out as an institutional struggle over the establishment of an elite-centric ideology/set of norms that characterized commercial activity as unseemly and elite forms of gift-exchange as honorable, on the one hand, and a contrasting set of non-elite norms that validated commercial exchange and profit-driven activity as a means for social advancement on the other hand. Admittedly, this interpretation of the literature demands that we read between the lines to a certain extent. For example, the repeated insistence on ritualized gift exchange in the Homeric epics might be a fairly accurate representation of the dominant mode of exchange in the late 8 th c. BCE, or it might be the case that such formalized distribution of goods is given exaggerated emphasis in the literary sources, for reasons that I have already suggested. At the same time, the degree of social disruption due to the redistribution of wealth that is chronicled in the Theognidea may be reflective of contemporary economic realities, or it may also be an alarmist reaction stemming from a mis-perception of reality. There are no shortages of modern comparanda to illustrate some of the ways this might occur, such as recent survey results indicating that nearly half of all Americans believe that “reverse” racism in the U.S. is now as much a problem as conventional racism (Jones, Cox, Cooper, & Lienesch, 2015), despite the mountains of readily available data that conclusively point to the contrary. At the same time, the widespread concerns over the disadvantages supposedly facing white Americans, even in the face of ample 292 countervailing data, very likely stem from genuine social changes. While it may be fair to say that the anxiety is misplaced, it is nonetheless a reflection, however disproportionate, of these social changes. However, an obvious and salient difference exists between the white alarmism of Americans in the early 21 st century CE, and the elite hand-wringing of the 6 th c. BCE: namely, the lack of hard data for the ancient world. This lack of quantitative data presents several problems. First and foremost, we must be alert to the fact that the degree of social and political disruption depicted in the Theognidea may be grossly exaggerated, magnified by the latent fears of elite-centric authors. Second, as I have already pointed out, the contemporary phenomenon of the redistribution of wealth (and its attendant social impact) seems clear, but the actual means by which this redistribution takes place is never made explicit. In the absence of reliable econometric data, how are we to bridge the potential perception-reality gap? It is for precisely this reason that I propose a broad-based study of pottery assemblages in the western Mediterranean during the 7 th and 6 th centuries BCE. As is so often the case, the material remains can present a much different picture from that offered by the contemporary literary sources. In addition, the archaeological evidence can frequently provide insights into patterns of behavior, such as the mechanisms of archaic maritime commerce, that are completely absent from written accounts. This is not to say that I find that the material record occupies some place of primacy by virtue of greater objectivity, 346 merely that it often has the potential to answer some questions that written sources cannot, or to provide a complementary picture and therefore a more nuanced perspective when both are considered simultaneously. 347 For instance, we have already seen that the pattern of development of the Greek 346 I am well aware of the post-processual criticisms (e.g., Miller [1982], Hodder [1990], Shanks & Tilley [1992]) that cast grave doubt on the ability of archaeology to provide anything along the lines of “objective” answers, and even how problematic that very belief has been for the discipline. For a more thorough review of post-processual critiques, see Trigger (2006). 347 The profitability (no pun intended) of combining analyses of material and literary evidence, particularly for insights into economic behavior, has been amply argued elsewhere; see Berdan (1985, 1987), Berdan et al. (1996), Hodge & Smith (1994). 293 colonial expansion in the western Mediterranean, a phenomenon on which the written sources are largely silent, cannot be explained adequately by some sort of “land hunger,” but more properly by an attempt to expand and protect maritime trade routes (Dunbabin, 1948; Tandy, 1997; Boardman, 1999). Combining the historical record with archaeology is particularly important when the question revolves around ancient practices, like trade and commerce, that were highly colored ideologically in the literature. When the written sources concern themselves with what one should or should not do, the archaeological evidence can often provide a more salient counterpoint in terms of what people actually did, or at the very least, what was left behind in the wake of what they did. In particular, for the written sources that we have examined to this point, the figure of the trader, if mentioned at all, exists on the margins of society, such as the unscrupulous, mendacious Phoenicians in the Iliad, or the nameless, faceless individuals into whose boats Hesiod suggests one should load his generic cargo. Unfortunately, there is little that the material evidence can offer in the way of supporting or refuting the characterizations of archaic maritime traders, but what it can help show us is some indication of the level of activity of archaic maritime commerce, as well as its developmental path, both diachronically and inter-regionally. For the sake of clarity, I wish to be as transparent as possible about the questions I plan to ask of the pottery data. First and foremost, I wish to test the hypothesis that maritime trade may have been responsible for some of the anxieties expressed in the literature. In order to do that, I will examine the data for evidence of the expansion of trade in terms of geographic expanse, complexity, and overall volume. 348 Also, I wish to analyze the data with an eye to the economic theory which I proposed in the first section. As I have argued, neo-institutional theory demonstrates that repeated, long-distance trade requires the existence of supporting instutions and their enforcement arrangements. The greater the 348 I am aware of the objections to using pottery as a proxy for trade more broadly speaking (see Orton & Hughes, 2013). However, I believe that the ceramic evidence is capable of offering evidence for broader patterns of trade that are valid more generally. 294 distance between trading partners, the greater the opportunities (and incentives) for shirking. In the absence of sufficient institutional support, the risks involved with long-distance trade are simply too great to occur on a sustained, repeated basis (North, 1990). If, as I suspect, the pottery evidence demonstrates the existence of complex networks of trade capable of delivering specific products to regional consumers based on their express demands, as well as providing relevant information to producers that would prompt them to adjust their producton strategies in response to particular consumer demands, then I believe we must postulate the presence of institutions capable of supporting this ongoing commercial activity. Of course, I do not expect the pottery to be able to offer much if anything in the way of elucidating the specific forms that these institutions may have taken; that is a subject for the following chapter. Ultimately, what I hope to accomplish through an examination of the archaeological evidence is some way of testing the hypothesis that maritime commerce during the Archaic period underwent a significant expansion during the 7 th and 6 th centuries BCE, in terms of sheer volume, demographic penetration, and geographical scope. If this can be confirmed, or at least supported, by the physical evidence, such a conclusion would then lend credence to the supposition that emerging markets constituted attractive markets for Greek traders (among others), and thus provide support for the larger argument that seaborne trade played an important role in the social and political transformation of archaic Greek society. The Project Outline Of course, it is one thing to say that I am going to examine the pottery evidence as an indicator of expanding trade levels, but it is quite another thing to do it. As I said, my overarching goal is to examine the ceramic evidence for insights into patterns of maritime trade in the Archaic period, but just how does one go about that? Deciding on the type(s) of data to be used and how to organize them will have a powerful impact on the ability of the data set to answer the research questions (Orton & Hughes, 2013). First of all, the basic outlines of my approach take their inspiration from previous work by Archer Martin (2004, 2005, 2008). In a series of articles, Martin tested the hypothesis that variations in the composition 295 of different pottery assemblages, in particular the relative percentages of different functional groups (transport amphorae, fine wares, coarse wares, cooking wares, and lamps) could be correlated with differential degrees of openness to trade. Broadly speaking, he concluded a higher percentage of transport amphorae, which are generally assumed to be imported, 349 appears to be strongly correlated to higher overall levels of imports. At the same time, a greater proportion of fine wares, especially combined with a lower overall percentage of coarse wares, 350 is also indicative of higher levels of imports. In order to avoid the potential distorting effects of comparing assemblages from sites resulting from different site formation processes, I have chosen the filter the assemblages by site type (domestic, burial, sanctuary, etc.) and limit the data to those from domestic contexts. Also, because I am attempting to generate a picture of maritime trade, I selected only assemblages from sites that were on or very near the ocean. In addition, not to belabor the obvious, I am clearly not in a position to gather, study, analyze, and record the pottery from several different assemblages, so I have relied on published sources for the data. Even with these fairly basic filters, the pool of available sites that were coastal, had at least portions that were domestic and that had been excavated, and which had fully published ceramics data became very small very quickly. Even so, it was still necessary to make certain concessions to the limitations of the sources, as I will explain shortly. Ultimately, there were three sites that fulfilled most or all of my criteria (coastal location, domestic settlement, occupation levels corresponding to the 7 th and 6 th centuries BCE) and for which the pottery data had been fully published the excavated pottery): Salento, at the southern end of what is now Apulia in southeastern Italy (Figure 22); Tarquinia, an Etruscan settlement in central Italy; and the palaia polis of Empúries (Greek Emporion), a Greek settlement in northeastern Spain. 349 This premise is generally accepted, especially for Greek transport amphorae, albeit with some reservations and refinements (ex., Lawall, 2005, 2011). 350 Contrary to transport amphorae, coarse wares are widely considered to be produced locally for the most part (Rice, 2005; Orton & Hughes, 2013); however, some studies have demonstrated the potential for coarse wares to travel across great distances (Shepard, 1942), although this is still regarded as an infrequent occurrence. 296 The initial challenges of this project were several. First, I needed to decide precisely what categories of data I was to catalog. These needed to be narrow enough to provide some degree of specificity, but general enough that I could be confident that published sources for each of the sites would provide such data. The categories I selected were as follows: 1. Functional group (fine wares, coarse wares, cooking wares, transport amphorae, lamps, and miscellaneous). I followed the classification system of the Atlante delle forme ceramiche, which has become a standard for ancient pottery, and therefore facilitated cross-comparability of the data sets. 2. Class (Black Gloss/Varnish, Attic Red Figure, White Ground, etc.). This category applied almost exclusively to fine wares, but is important in being able to distinguish the place of origin, at least in general terms. 3. Fabric/Production (i.e., S. Italy, E. Greece, Aegean). For pottery types other than fine wares where it is not possible to determine the region of production based on decoration, such as coarse wares or transport amphorae, the general area of production can often be estimated on the basis of form or fabric; in other cases, it is even possible to adduce quite specific points of origin based on characteristic forms, as in the case of the distinctive handles of Rhodian amphorae, for instance. 4. Provenience. For fine wares, it was often possible to distinguish local imitations of imported pottery on the basis of factors such as clay type. This is an important distinction to make in order not to fall into the trap of assuming that all black gloss pottery, for example, is imported. 5. Form (cup, jar, bowl, vase, plate, etc.). Especially for the fine wares, this was a way of distinguishing more specific types of often highly varied pottery. There was some difficulty arising from the heterogeneity of categorization methods from the different published sources, so I chose to base the categorical distinctions on ostensible use-patterns. For instance, specific pottery forms that were likely used for serving liquids I compiled under the “jar” category, 297 whereas those that were more probably used for storing liquids were assigned to the “vase” category. 351 6. Subform (lekythos, oenochoe, etc.): With fine wares, many of the forms are clearly standard, specific types, and I chose to retain this category in order to test the possibility of interregional and chronological variation at a level of greater detail. 352 For certain forms, such as cups, there are numerous different subforms that fall under the broader category, such as kotylai, kylikes, kantharoi, etc.. 7. Type (Ionian B1, B2, C; round aryballos, etc.): Sometimes it was possible to divide the subforms into even more specific types. In the case of Ionian cups, mostly because they are so numerous and geographically widespread, this was thought to be potentially important, although as it turned out, it was not especially helpful in establishing trends over time. 8. Date range. 9. Number of fragments. 10. Minimum number of individuals (MNI). This calculation has become a fairly standard measure in pottery studies, and is used to avoid artificially inflating the figures. Given the typically fragmentary nature of the evidence, it is often the case that multiple fragments belong to a single 351 I realize that this may seem a somewhat arbitrary category, but given the fact that among the various published sources there was sometimes no agreement as to some of these terms, certain editorial decisions were necessary. I realize that classification of pottery forms (especially fine wares) has been grounded in a basic assumption that form follows function, but in recent years, much attention has been paid to the problem of distinguishing between “intended use,” either by potters or primary consumers, and “secondary” and “tertiary” use; see Rice (1996a) for an overview from the perspective of anthropological archaeology; more recently among classical archaeologists, the important work of Peña (2007), although more specific to Roman pottery, has inspired others to address similar issues with Greek pottery (Lawall & Lund, 2011). See also Braun (1983), Skibo (1992, 2013). 352 Although certain subforms are typically associated with specific use-patterns, such conclusions are especially problematic when applied to imported fine wares. The argument is sometimes made that finding several different subforms specific to the Greek symposion gathered in one place is evidence for the associated cultural practice. However, recent studies have demonstrated the propensity for the cultural recontextualization of imported pottery. For instance, Osborne (2001) has demonstrated that Greek drinking and serving vessels typically associated with the symposion when imported to Etruria were frequently used in drinking and dining practices that were much more specifically Etruscan than Greek. 298 pot. Counting only fragments involves an inherent degree of imprecision that the calculation of MNI is intended to ameliorate. 11. Maximum number of vessels. This measure was adopted as a counter to the potential suppression of MNI. When multiple fragments of the same type/form/subform are found in a single stratigraphic unit, the application of MNI requires that they be considered as belonging to a single vessel if they cannot be definitively distinguished. This can obviously deflate the figures artificially. Perhaps not surprisingly, there were numerous challenges to this study, some of which were anticipated, while others became more apparent as the project developed. The most immediate hurdle to overcome was posed by the idiosyncrasies of the various published sources. As pottery specialists have increasingly tended to exploit the powerful computing tools that are now so readily available, but were unthinkable for previous generations, the issue of data inter-comparability has loomed ever larger (Orton & Hughes, Pottery in Archaeology, 2013). Especially when we begin to study pottery from a regional perspective, seemingly minor discrepancies at narrower levels become significantly magnified. This problem is further compounded by the fact that archaeologists have often demonstrated a lack of statistical sophistication (Rice, 1996b). Some of this is only to be expected in a field that, as Rice (1996a) points out, “has been expanding so fast that it is in danger of flying apart.” Where pottery once was used by archaeologists solely to establish relative dates, or perhaps as objets d’art (Rice, 1996a; Bresson & Callatäy, 2013), archaeologists have been asking increasingly sophisticated questions of the evidence, and as more powerful technological tools become available, this trend seems likely only to continue and even accelerate (Orton & Hughes, Pottery in Archaeology, 2013). However, as the range of options available to archaeologists increases, the complexity, and hence difficulty of establishing generally accepted standards, becomes far more difficult. This is in part reflected by terminological idiosyncrasies. In different regions of the world, the same type of pottery is often known by different terms. Among American archaeologists, for instance, a particular 299 type of early pottery with a burnished black slip is classified as “black gloss” ware, whereas in Spain or Italy, this same pottery finish is referred to as “black varnish” (berniz negra/vernice negra). While such a minor terminological difference may seem innocuous, when attempting to produce cohesive data sets, it is vital to assign the same exact term to all instances. A similar difficulty arises when the same production is attributed by some as “East Greek,” by others as “Ionian,” etc. But such challenges pale in comparison to that of finding standard statistical measures. The most common standard of measurement is the number of fragments. However, this measure is notoriously susceptible to bias (Orton & Hughes, Pottery in Archaeology, 2013). Different types of pottery are more fragile than others, leading to relatively inflated numbers of fragments. Also, for obvious reasons, larger pots will typically manifest more fragments than smaller ones, ceteris paribus. Finally, different depositional and post-depositional processes will lead to differential levels of fragmentation in the same types of pottery from site to site, and even within single sites. As a result, numerous other measures have been advocated at various points by different ceramologists, including sherd weight, surface area, and displacement volume, with varying degrees of success. 353 Also, combining two different measures has proven to provide more information than relying on single measures, 354 and some have even argued for the creation of statistical equivalencies that further reduce the distorting effects of extraneous factors. 355 353 Various attempts have been made to compare measuring techniques: Glover (1972) compared sherd count, weight, and surface area, concluding that any one is accurate as a measure of frequency; Hinton (1977) compared sherd count, rim sherd count, weight, and displacement volume, concluding that weight was the fastest but sherd count probably the most accurate measure; Millett (1979) compared sherd count, weight, adjusted weight (an estimate of surface area), and MNI, concluding that they were all highly correlated but for practical reasons weight was probably the best; however, Orton (1975, 1982, 1993, 2013) ruled out entirely sherd count and number of pots as biased, with weight as a respectable but less useful measure; by contrast, Py (1993) reaches precisely the opposite conclusion (and one that has been very influential among European archaeologists), namely that quantification of pottery by weight produces distorting effects that are too great to be an efficient comparative measure, whereas sherd count provides more reliable data. 354 For the benefits of the combination of sherd count and sherd weight, see Solheim (1960); see also Bradley & Fulford (1980); Orton (1985); Schiffer (1987). 355 These statistical measures include Estimated Vessel Equivalents (EVEs) and Pottery Information Equivalents (PIEs); for a thorough explanation of their advantages and how they are calculated, see Orton & Tyers (1990, 1991). 300 Ultimately, though, I am constrained by the information that is available in the sources on which I must rely. For better or worse, the published data are in large part a reflection of standard practices in the field, and of course are therefore heavily influenced by broader trends within the field of pottery analysis. This is not an appropriate forum to engage in a discussion of the history of pottery studies, 356 but some brief excursus is required by way of explanation. It is well known, for instance, that in the field of classical archaeology, pottery analysis has long focused on fine wares, especially figure-painted wares. 357 This is partly due to the continuation of what Rice (2015) calls the “art-historical” phase of pottery analysis, which she claims persists strongly even today within anthropological archaeology, especially in Britain. It is also partly due to the long shadow cast by Beazley, and the attempts of art historians to use Greek vase painting as a way to glean insights into ancient Greek painting (Bresson & Callataÿ, 2013). This is in contrast to anthropological archaeology, in which the cooking pot has become a central object of study in attempts to track the spread of cultural influences, with cooking techniques as a proxy, manifested in the distribution patterns of different forms of cooking vessels (Rice, 2015). This approach has prioritized fine wares almost to the exclusion of other classes of pottery. For instance, as Lawall (2011) explains, the study of transport amphorae was almost completely ignored until just the last couple of decades. In addition, at least within classical archaeology, the study of coarse and cooking wares still lags far behind. Also, as we have already noted, the larger field is both expanding and splintering, with different sub-fields and multiple directions of study, each asking different questions and utilizing different data. Of course, this presents great challenges for efforts to compare pottery assemblages on a regional level, and some have openly expressed an increasing pessimism for the very 356 For some background on pottery analysis from an anthropological perspective, see Rice (2005), Orton & Hughes (2013); Rice (1996a, 1996b) also provides a thoughtful and (relatively) recent survey of the state of the discipline; for a brief overview from a perspective that is more specifically from the point of view of classical archaeology, see Sinopoli (1991). 357 A recent example can be seen in Rasmussen & Spivey (1991), a collected volume ostensibly dedicated to Greek vases more generally, but in which every single contribution is dedicated to painted fine wares, and that defends the perpetuation and validity of the Beazley approach (Robertson & Beard, 1991). 301 feasibility of such projects (Lawall, 2005). This combination of a tradition of emphasis on fine wares, increasing disciplinary heterogeneity, the lack of a robust “middle theory” to bridge the gap between ethnographic, historical, and archaeological evidence (Rice, 2015), along with the overall lack of statistical sophistication among archaeologists (Rice, 1996b), produces an environment in which most archaeologists have little familiarity with or incentive to utilize more elaborate, data-driven analytics. As a result, even those few published sources which include more-or-less complete accounts of entire pottery assemblages frequently limit the quantitative data solely to sherd counts, or at best, sherd counts and weights. In the case of the present project, the only quantitative measure that exists as a common denominator among all of the three published sources is that of sherd count. As we have already noted, 358 the degree of variability of this measure, and the potential for distortion from extraneous influences, are greater than with some other quantitative metrics. Thus, in the absence of any other means of cross-comparison, we are compelled to ask the more basic question as to whether or not the imprecision associated with sherd counts is prohibitive. Ultimately, for several reasons, I believe that sherd count, even given the caveats already noted, is a satisfactory measure for the questions I am asking. First and foremost, the data sets involved are large (11,784 fine ware fragments in Salento, 13,814 fragments of all pottery classes in Empúries, and 3,056 of all pottery classes in Tarquinia). Such large data sets mitigate the potential distortion to which smaller sample sizes might otherwise be prone. 359 Second, I am looking most closely 358 See n. 353 above. 359 The issue of sample size is fundamentally important to virtually any statistical analysis that relies on samples instead of an entire population. Any analysis that examines a sample should take pains to see to it that the samples are representative, and for that reason, larger samples are generally superior. There are several problems that can result from small sample sizes, but the one to which I refer specifically here is the problem of variability. Obviously, multiple sherds from a single pot can skew the results through over-representation of the categories to which that single pot belongs, and with a smaller data set, the degree of bias is proportionally greater. By their nature, larger data sets are not as prone to the effects of anomalous findings (see Drennan [2004]). On a related note, this is also the motive behind including the measure of Minimum Number of Individuals (MNI) in the analysis, which is a statistical measure specifically designed to counteract precisely that kind of bias (see Orton & Tyers [1990], Orton [1993], Orton & Hughes [2013]). 302 at comparative diachronic changes in the assemblage composition rather than basing an argument solely on sum totals. Thus, it seems reasonable to posit that the site formation processes are relatively consistent, especially since I have chosen to focus on domestic contexts. There are no indications of dramatic changes in activity at the sites, except in the case of Empúries towards the end of the 6 th c. BCE when the nea polis was settled nearby, and all indications are that the occupation levels at the palaia polis dropped significantly. Aside from that, there is no evidence of changes that would introduce potentially distorting effects. Third, because the comparisons are made between assemblages rather than within them, some of the statistical deficiencies of sherd counts are inapplicable. For example, the tendency of fine wares towards greater levels of brokenness than other pottery classes, or of transport amphorae to produce more potsherds, is not problematic when comparing fine wares from one assemblage to fine wares from another assemblage. For these reasons, among others, I believe that sherd counts will be adequate to ascertain broader, more general trends in the pottery evidence, consistent with the research goals. The problem of sherd count as a statistical measure is of course not the only challenge to this investigation, however. We have already alluded to the challenge of terminological consistency between published sources, and another difficulty was posed by the need to establish a consistent standard of dating. In most cases, numbered date ranges were supplied by the authors, but at times the date range would be in less numerically specific terms, such as “end of the 6 th c. BCE,” or “middle of the 5 th c. BCE.” In the absence of more specific information from the author, I assigned 20-year date ranges to such descriptions. So, if a particular stratigraphic unit was described as belonging to the “beginning of the 6 th c. BCE,” the pottery from that SU was entered into my database with a date range from 600-581 BCE. Also, the choice of dating on the basis of the date of the stratigraphic unit versus the known production date ranges presented an additional challenge. 360 The basic assumption regarding pottery in a specific 360 The often complex use (and re-use) patterns of pottery are discussed in some detail by Orton & Hughes (2013); the interpretive problems posed by the “use life” of pottery, especially the degree to which those of different classes 303 stratigraphic unit is that one expects to find a certain percentage of earlier pottery within any given unit due to the fact that the pot was in use over a period of years, decades, or even longer. Also, the general rule of thumb is that the more expensive the pot, the longer the gap between production date and deposition (Peña, 2007). Thus, fine wares are presumed to have the longest use life, and transport amphorae the shortest. Therefore, the production date probably correlates more strongly with the date when a pot was traded than the date of the stratigraphic unit in which it was found, especially if we accept the premise that most pots were typically exported relatively quickly after they were produced. 361 Nevertheless, I have chosen to use the dates supplied by the respective stratigraphic units wherever possible, for several reasons. First and foremost, for obvious reasons I wanted to utilize a dating approach that could be applied to the entire data set universally. Unfortunately, in classical archaeology comparatively little attention has been paid to coarse and cooking wares. As a result, production date ranges for these functional groups are not widely available. Second, unlike most fine wares that can change frequently according to the vicissitudes of changing tastes, many classes of pottery such as cooking wares and transport amphorae often exhibit little stylistic development over time. Because the forms of many types of transport amphorae frequently are so static, the production date ranges for many transport amphorae are quite broad, spanning well over a century in some instances. Thus, the dates supplied by stratigraphy are often more precise in the case of these pottery classes. Finally, my argument is essentially one that advocates for an expansion in maritime trade from an early date. By using of pottery are more extended or abbreviated, is an important theme in Peña’s (2007) recent work, especially for ancient Roman pottery, and Lawall (2011) builds on Peña’s model and applies it to Greek pottery, noting that there are important differences in the observable use lives of different pottery classes of Greek pottery versus Roman. 361 This makes intuitive sense, since it seems unlikely for traders to purchase pottery for the sake of overseas trade if it had been in use for an extended period of time, especially for fine wares that adapt to changing style preferences with great frequency and rapidity. Also, both Boardman (1979, 1988) and Osborne (2001, 2007) clearly demonstrate that some potters manufactured pottery specifically for the purpose of export, and it is almost impossible to believe that they would have done so only to store it for a number of years prior to exporting it. Admittedly, Lawall (2011) has shown that graffiti and tituli picti on many transport amphorae clearly show that it was not unheard of for some transport jars to be fired well before they were eventually filled with goods for transport, but this interval is usually at most a year or two. 304 stratigraphic dates, which typically down-date the objects in the stratigraphic unit, I am adopting a more conservative approach. In a related vein, for the purposes of cross-comparison I wanted to establish consistent time horizons. The comparisons are understandably more apt when the identical time periods are used across all the assemblages. On this point, I initially incorporated the time horizons established by the excavations of the palaia polis at Empúries (Alquilué, Castanyer, Santos, & Tremoleda, 2001). On the basis of recent excavations at the site, the stratigraphic evidence indicated several occupation phases, with the dates as follows: • 650-625 • 625-600 • 600-580 • 580-560 • 560-540 • 540-520 • 520-480 However, this schema is far from perfect. For one, there are no occupation levels in the Archaic period that date earlier than the middle of the 7 th c. BCE. Also, the first two occupation phases cover 25-year intervals, whereas the remaining phases are in intervals of twenty years (with the exception of the last phase of forty years). Clearly, time horizons of different lengths make comparisons between periods on the basis of sherd totals difficult, if not impossible. So, for the sake of flexibility, I ultimately rejected this particular date breakdown, and instead opted to create a database that tracked dates and separated them into customizable units of a duration that I could select manually. 362 From there, the only remaining 362 I should point out that my level of expertise in database management is far from advanced. For that reason, and after wrestling with the data for an extended period, I ultimately consulted an outside consultant firm to have them 305 decision to be made was the length of the units. In order to do that, it was necessary to balance the competing desires of computational flexibility against complexity. In order to strike this balance, I assigned time units of five years which I then used in various combinations to fit whatever time horizons were appropriate for my analysis. 363 Finally, one additional problem concerned the availability of the data. As I have said previously, given the criteria, the selection of published sources was very limited. In reality, some concessions needed to be made to the fact that none of the sources precisely matched all of the criteria. The Preliminary Picture Visible in the Ceramics As an initial foray, I conducted a preliminary analysis of catalogued Corinthian pottery, 364 using data from Amyx (1988), in an attempt to establish at least a rudimentary snapshot of early exchange patterns during the period. The sample numbers, along with the number of different hands and a breakdown of provenances, were as follows: Sample Numbers Protocorinthian (and Transitional): 22 painters 177 vessels/ 129 w/ provenance (73%) write scripts that would allow me to produce reports which would allow me to enter various categories, and to select custom date ranges. 363 Ultimately, I chose to break down the date ranges into 5-year units which allowed a great deal of precision and flexibility, but which also added greatly to the number of data points, and therefore to the computational load. I felt that the increased versatility of the data was worth the effort instead of breaking down the dates into 25- or even 50- year intervals that would simplify the process greatly, but significantly sacrifice precision. In order to accomplish this goal, I used 5-year intervals to which I assigned a discrete integer for each. So, for instance, the period from 700-696 BCE would be represented by a “1,” the next five-year span of 695-691 BCE by a “2,” and so on. Of course, most of the pottery was assigned to date ranges that were longer than five years, so that these items would be associated with multiple five-year time units. As an example, a stratigraphic unit that the excavator dated to the “end of the 7 th c.” and to which I assigned a beginning date of 620 BCE and an end date of 601 BCE would be associated with four different temporal units (620-616, 615-611, 610-606, 605-601), each one represented by a unique integer. From here, I was then able to mix and match the data to produce any time horizon, so long as it was divisible by 5. In other words, I could compare pottery data from time horizons of 10, 20, 25, 50 years, etc., and then these numbers were used to produce the graphs and charts. 364 For this data, I was considerably aided by Prof. Angela Ziskwoski of Coe College, to whom I owe considerable thanks for her input and advice. 306 Early Corinthian: 10 painters 107 vessels/ 64 w/ provenance (60%) Middle Corinthian: 10 painters 145 vessels/ 79 w/ provenance (54%) Late Corinthian: 10 painters 105 vessels/ 49 w/ provenance (47%) On the basis of distribution data, it would appear that pottery from the Proto-Corinthian (PC) period is clustering in Cumae, Syracuse, Tarentum, Aegina, Delos, and Rhodes; pottery from the Early Corinthian (EC) period is clustering in Caere, Syracuse, Rhodes, Tarentum, and Delos; pottery from the Middle Corinthian (MC) period is clustering at Syracuse, Tarentum, Selinus, Delos, and Rhodes (with strong pockets in Syracuse/Megara Hyblaea, Tarentum, and Caere); pottery from the Late Corinthian (LC) period is clustering in Caere, Delos, and Rhodes, with more also going to Gela. It would seem from this very initial investigation that archaic trade was expanding further west, moving from central Italy into southern Italy and Sicily. Admittedly, it would be overly optimistic to draw any firm conclusions on the basis of such small sample sizes. Also, since Amyx catalogues only complete or nearly complete painted pottery, we are dealing with only a single functional group (fine wares) rather than a broader spectrum of ceramic evidence, and only those very rare pots that have survived mostly complete and cannot be remotely considered a random sample. Due to the nature of typical site formation processes, the overwhelming majority of complete ancient pots are recovered from burials, a consideration that needs to be factored into the analysis. Still, this pattern of westward expansion is certainly what we would expect to see from the pottery, given what else we know about the period and other types of material evidence. As such, it supports the general premise that Greek trade, and probably Greek traders, were moving steadily westward from an early date. Perhaps most importantly, it indicates that pottery distribution patterns correlate with broader observable trends, and therefore that the ceramic evidence can likely be an effective proxy for trade and migration more generally. 307 The ability of pottery to serve such a function has long been debated. According to one view, the economic importance of pottery (both plain and decorated) in the ancient world was minimal. According to this argument, painted pottery was not traded for its own sake. Instead, pottery served as “space-filler” or “profitable ballast.” 365 However, as Boardman (1988) points out, all cargoes are in some way “space fillers”; “saleable” or “profitable” ballast is a contradiction in terms. 366 More recently, an opposing view has been offered, pointing out that various factors such as the sheer volume of Greek pottery throughout the Mediterranean (but particularly in the western Mediterranean), the amount of work and effort required to stow fragile pots in order to avoid breakage during a journey, distinct regional preferences, and clear evidence of pottery specifically produced for export to particular consumer markets, all support the thesis that pottery could be and often was a primary trade item. 367 Williams (1989, 2009, 2013) has also argued that pottery evidence demonstrates not only the mobility of pottery, but of potters who set up shop in Italy and had already sufficiently influenced local potters to the degree that a distinctive “Etrusco-Corinthian style” was visible as early as the third quarter of the 7 th c. BCE. The Data If we are to put any faith in the pottery catalogued by Amyx, the preliminary ceramic evidence strongly correlates to larger trends of mobility of both goods and people. 368 However, there are some caveats that must be made with respect to the data that should be acknowledged. For the pottery from Salento, I utilized the data presented by Semeraro (Semeraro, 1997). This text collates pottery from nearly thirty 365 For the most prominent exponents of this view, see Vickers (1984, 1994), Gill (1987a, 1994); on the use of the term “saleable ballast” see Gill (1987b), Vickers (1985/6); Gill (1985) states that we have no evidence that fine wares were traded long distances on their own, occurring only as space-fillers in more valuable cargoes, often of oil- or wine-carrying amphorae. Of course, more recent discoveries such as the Pointe Lequin 1A shipwreck render this argument untenable. 366 Boardman attributes the “extreme and wrong view” represented by Vickers and Gill to studies dedicated to demonstrating the prime importance of vessels in precious metal in classical antiquity, especially silver (i.e., Vickers, 1985). However, as he also notes, the tendency to emphasize the importance of silver as the most valuable part of a ship’s cargo ignores the fundamental point that value does not directly translate to profitability. 367 In particular, see Boardman (1979, 1988), Osborne (1995, 2001, 2007). 368 I realize that “correlation” has a discrete statistical meaning, but it is one that I am not using here and want to be careful to avoid any such implication. 308 different excavations in the region rather than from one single site. Thus, some of the sites are nearer the coast than others, and would therefore be expected to demonstrate some degree of variability in the proportions and distribution patterns of the assemblages. It remains to be seen how that variability will affect the data. Also, Semeraro only catalogues the imported fine wares, thus making it impossible to employ the approach of A. Martin to the data. Instead, what I propose is to compare the fine ware distribution patterns with those at the other two sites in order to establish or refute a correlation between the various assemblages. I realize that this presents a weaker case, but one that I still am convinced supports the main thesis. Next, I have included the pottery data from Tarquinia, published by Chiaramonte Treré (Chiaramonte Treré, 1999) and Bonghi Jovino (Bonghi Jovino, 1999). The data in these volumes are from a single site, and include all of the functional groups. However, each particular class (i.e., black gloss, red figure, etc.) was presented in a separate chapter which was written by a different contributing author. As a result, some of the challenges already discussed above with respect to establishing consistent categories of comparison between the different assemblages even applied to this source. Also, the site is slightly removed from the coast, raising the possibility that the composition of the assemblage might be affected. Indeed, as we shall see, the percentage of transport amphorae recovered at the site is extremely low, a fact noted by the authors themselves, and more than likely an indicator that goods shipped in transport jars were unloaded and transferred to different containers immediately upon arrival at nearby Gravisca, 369 then sent on their way to Tarquinia and beyond. 369 I would have preferred to utilize the data from Gravisca, especially considering that it seemed to have served as an emporion for the region. Unfortunately, although nearly all the pottery has been published in its own dedicated volume (see Valentini [1993] on black gloss wares, Boldrini [1994] for Ionian pottery, Huber [1999] for Attic red- figure, Pianu [2000] for bucchero, Gori & Tiziana [2001] for coarse and cooking wares, Galli [2004] for lamps, Iacobazzi & Johnston [2004] for Attic black-figure, and Bruno [2009] for Corinthian and Etrusco-Corinthian pottery), the sole exception is that of the transport amphorae which still await publication. If and when this last volume appears, the site would likely represent an excellent case study for the kinds of questions asked here, but that will obviously have to wait. 309 Finally, the pottery from the palaia polis at Empúries was published by Alquilué Abadías, Castanyer i Masoliver, & Santos Retolaza (2001), from excavations carried out in the 1990s. This site matched nearly all of the criteria. It was located on the coast, the pottery catalogued all came from domestic contexts, and the entire range of pottery was published. The only caveat is that the earliest occupation levels dateable to the Archaic period begin around the middle of the 7 th c. BCE. There is evidence for previous occupation at the site, but there appears to have been an interregnum between these early settlements and the archaic activity. Thus, the site does not provide a view for the entire span of the 7 th and 6 th centuries BCE, although it does offer coverage for the majority of the period. In addition, the sites offer coverage across a large geographic area, from one end of the western Mediterranean to the other. Based on the evidence of the cargo of the Cala Sant Vicenç wreck, it seems that the vessel plied a circuit from western Italy to eastern Spain, most likely following the coast of the northern Mediterranean along the way. This is consistent with the conventional historical view that maritime commercial activity in the western Mediterranean was divided roughly into two regions, with Greek traders active in the northern half (with Spain to the west, France to the north, and the Italian peninsula to the east) and Phoenician traders associated with trading partners in the south (with Spain to the west, north Africa to the south, and Sicily and possibly Sardinia to the east). Pottery from Salento The Data Although the pottery data published by Semeraro is restricted to imported pottery, this limitation is mitigated by the fact that the data are compiled from excavations at numerous sites throughout the region (see Figure 22). 370 Of the total pottery catalogued, the greatest portion was recovered from Oria, with 6,746 fragments accounting for more than half (%57.3) of all the pottery from the region. This was 370 Admittedly, there are questions when attempting to generalize patterns of pottery distribution for all pottery from a single functional group, in this case finewares. 310 particularly surprising, given that Oria is landlocked. We would have expected to see settlements closer to the ocean such as Brindisi, Otranto, or Leuca present greater quantities of fine wares. Admittedly, we need to be careful about reading too much into raw totals, since differences can often be explained by greater or lesser intensity of excavation efforts. However, the disparity here is so striking that it seems impossible to attribute it to nothing more than a fluke of fieldwork. We can express this first in terms of the difference between mean and median. The mean number of fragments from the various catalogued contexts is 393. However, the median is just 19. Such a difference between mean and median is almost always an indication of one or two anomalous measurements distorting the overall picture. Yet another measure supporting this interpretation is that of the standard deviation. For this set of data, s = 1382, meaning that the total from Oria is 4.88 times the standard deviation. Thus, the predominance of pottery from Oria, although possibly exacerbated by differences in the application of archaeological method, is far too anomalous to be the result of simple distortion produced by variations of archaeological practice. As a result, it should be regarded as reflecting some underlying reality independent of the data collection methods. This does raise questions about the role of Oria during the 7 th and 6 th centuries BCE that would produce this distribution pattern of pottery. It does seem that Oria was a major city of the Messapians, one of three Iapygian tribes (Peucetians and the Daunians) to occupy modern Apulia (Herodotus, Histories, 4.99.2-5). According to the Hellenistic historian Nicander, the Messapians controlled the Salento peninsula, a view that finds support in the archaeology (Carpenter, Lynch, & Robinson, 2014). The prominence of Oria then becomes more understandable, since it is centrally located within the area of Messapian political control. Also, it is located atop a hill with excellent views of the surrounding countryside, a location with some strategic advantage. Later written sources, specifically Strabo (Geographica, 6.3.1) relate that the region was divided into two halves, with the Salentinoi occupying the southern portion part of the peninsula from Otranto to Leuca and Leuca to Manduria (see Figure 21), and the Kalabroi in control of the northern region on the Adriatic from Otranto to Egnazia. If this was true during the earlier period in question, Oria 311 was located along the mid-line separating the two groups, and thus could have served as a point of interaction. However, it seems too much to retroject the testimony of Strabo, writing in the 1 st c. BCE, back some five hundred years, especially when Carpenter et al. (2014) only date this division to the 4 th /3 rd centuries BCE. Also, the archaeological evidence shows that Oria was a significant ritual center at this time. It therefore seems possible that the site attracted worshipers from a wide area, adding to the importance of Oria as a hub of regional activity. Meanwhile, we can see in the data that there is a high proportion of cups among the fine wares (see Figure 22). This is especially true of the black varnish pottery, in which cups account for 89.24% of all black varnish pottery. The high proportion of cups at a site that functioned as an important ritual center seems to suggest that cups were featured in Messapian ritual practice. If true, this might help to explain the strong presence of black varnish pottery relative to the other sites in Salento. In any case, the evidence indicates that due to its central location within the territory under the political control of the Messapians, as well as its importance as a religious center, help to explain the predominance of pottery at the site. Regardless of the role of Oria in the region, we are still left with the task of interpreting the data, and the obvious danger of the aforementioned disproportionate distribution pattern is the potential for the pottery from Oria to distort the overall results. With such a predominance of the evidence coming from a single site, we must determine whether or not the composition of the Oria assemblage is consistent with that of the other sites. If so, then we can regard it as a representative sample and proceed accordingly, but if not, it can significantly skew the data. Thus, we must compare the pottery data from Oria with those of the other recorded sites (see Figures 6-8). As we can see in the charts, there are a couple of ways in which the pottery distribution from Oria differs from that of the rest of the region. First of all, the percentage of pottery that is unidentified is extremely low (.64%) compared to the region as a whole (26.82%). Of course, this does not present any interpretive dilemma. Having more detailed information rather than less only enhances our ability to drill down into the data. However, the second major point of departure 312 consists of the percentage of black varnish pottery from Oria, which is significantly higher (81.59%) than the rest of the region (40.93%). The standard deviation (s) of the percentage of black varnish pottery from the various assemblages throughout Salento is 29.73374. Even given such a high standard deviation, the percentage of black varnish pottery at Oria is 1.351 times the standard deviation, which suggests that it constitutes a statistical outlier. Still, when we compare the chart of the overall pottery distribution including Oria with the distribution chart with the data from Oria removed, the general outlines remain largely the same. Basically, whether we include or remove the data from Oria in the data from the entire region, the overall picture does not change dramatically. When the data from Oria are removed from our totals, the percentage of Attic Black Figure pottery lowers from 7.97% to 7.11%, Banded pottery decreases from 3.19% to 1.32%, Attic Red Figure pottery rises from .51% to 1.02%, Laconian pottery increases from 1.67% to 6.61%, while White Ground and Apulian Ware percentages remain unchanged. The most significant changes are to the categories of Black Varnish pottery, which drops from 76.83% to 37.9%, and Corinthian pottery, which rises from 6.88% to 19.11%, when the data from Oria are removed. Also, the percentage of unidentified pottery increases from 3.51% to 26.82% of the total. In short, Black Varnish pottery remains the single largest category, although it does nearly double as a share of the total when the data from Oria are added to the mix while Corinthian pottery changes from the second-largest to the third-largest category. In spite of the fact that Black Varnish pottery constitutes a significantly higher percentage of the pottery assemblage from Oria than it does in the region overall, adding these data to our pool does not appear to destabilize the results, and is at least mitigated by the fact that by reducing the percentage of unidentified pottery, the data from Oria reduce the statistical noise. So, it seems reasonable to treat the entire range of pottery from Salento as a (more or less) single assemblage. But what does this assemblage indicate about levels of exchange during the period in question? First of all, when we examine the pottery data diachronically, the sheer increase in numbers from the middle of the 7 th c. to the end of the 6 th c. BCE is striking. In fact, the comparative levels of imported pottery for the 313 second half of the 7 th c. BCE pale in comparison to those of the 6 th c. BCE, especially in the second half of the 6 th c. BCE (see Figure 14). Despite the caveats I made regarding the volatility of pottery data based solely on sherd count, the dramatic change over a century-and-a-half is significant enough that it cannot be dismissed as an accident of data collection. Total numbers in the second half of the 7 th c. BCE are less than 100 per five-year period, whereas by the end of the 6 th c. BCE the totals are in excess of 5,000 total fragments over the same time frame, an increase of at least two orders of magnitude. Meanwhile, the distribution of pottery by class shows further developmental trends over this same period. During the second half of the 7 th c. BCE (Figure 15), the limited pottery remains are dominated exclusively by Corinthian and Black Varnish pottery, at 23% and 77% of the total, respectively (although with such limited numbers it is difficult to draw any firm conclusions from the pottery distribution during this period). In the first half of the 6 th c. BCE (Figure 16), this picture changes considerably, as Corinthian pottery becomes the most numerous class, 371 accounting for 56% of the total, while Black Varnish is second with 40%. Meanwhile, Laconian (2%) and Athenian Black Figure (2%) make an appearance, albeit in rather limited numbers. Finally, in the second half of the 6 th c. BCE (Figure 17), we see further changes, the most obvious of which is the overwhelming presence of Black Varnish pottery, which at 93% of the total fairly dwarfs all of the other pottery classes. However, the volume of Black Varnish pottery potentially obscures other developments during this same time frame. In fact, if we look at the time series data (Figure 14), several other pottery classes increased or at least remained steady during this period. For example, Attic Black Figure, Laconian, and even Corinthian pottery reach their highest levels (by total sherd count) in the second half, and particularly the last quarter, of the 6 th c. BCE, but these increases are obscured by the dramatic surge in Black Varnish pottery. 371 This development is not at all surprising, and in fact is consistent with broader distribution patterns of Greek ceramics as Corinthian pottery dominated Mediterranean markets (Arafat & Morgan, 1989; Osborne, 1996; Thomasen, 2003) and was heavily produced during this period (Stillwell, 1948; Stillwell & Benson, 1984). Also, given the potential distorting effect of the concentration of Black Varnish pottery at Oria, the predominance of Corinthian pottery in the rest of the region is even more pronounced. 314 Given this substantial overall increase in pottery, as well as the dramatic growth in Black Varnish pottery (both in sheer numbers and as a share of the total), it seems essential to analyze these trends in greater detail. By drilling down further into the data, some interesting trends begin to emerge. First of all, when we examine the time series data for the pottery from Salento broken down by provenience (Figure 18), we find that a substantial portion of the pottery from the second half of the 6 th c. BCE was produced in Southern Italy. Not surprisingly, this trend is related to the other major development of the period, namely the large increase in Black Varnish pottery. In looking at this class of pottery, we must remember that unlike some other pottery classes (i.e., Attic Black Figure, Attic Red Figure, Corinthian), Black Varnish pottery was not location-specific, but was instead produced in various locations. When we examine the provenience of the Black Varnish pottery from Salento (Figure 19), what we find is that the majority (74.19%) was produced in southern Italian workshops, with a smattering from other regions in the eastern Mediterranean, and Attica (18.85%) accounting for the only other significant locus of production. Analysis of the Data What are we to make of these developmental trends? Based on our examination of the pottery data from Salento, there are certain features of the data set that are particularly compelling. First of all, the most obvious feature of the pottery data from Salento is the dramatic rise in overall volume of imported fine wares in the second half of the 6 th c. BCE. Second, concurrent with the overall increase in pottery volume is the surge in Black Varnish pottery, but several aspects of this trend emerge when we drill down further into the data. For example, the majority of the Black Varnish pottery in Salento was produced in southern Italian workshops. Meanwhile, Black Varnish pottery from Attica, which constitutes the second largest share of this pottery class in Salento, forms the single largest production of all Attic pottery imported to the region (Figure 20). At the same time, we must not ignore several other features of the data that might be overshadowed by the rise in Black Varnish pottery. For instance, several other classes of pottery continued to increase in the same period, including Corinthian pottery. 315 In any case, regardless of the source of the imported pottery at Salento, it would seem that the overall increase is, for one thing, strongly indicative of an openness to trade. As A. Martin (2004, 2008, and especially 2005) argues convincingly, higher proportions of fine wares correlate strongly with increased levels of participation in trade more generally. Also, even with the caveats made earlier regarding pottery and its ability to serve as a proxy for trade in other goods, the impressive surge in the levels of fine wares at Salento in the latter half of the 6 th c. BCE would seem to strongly suggest a greater level of commercial development and economic growth. Finally, the presence of such high levels of imported pottery necessarily implies increased levels of interconnectivity that facilitated the distribution of trade goods. Next, we must consider the issue of the increase in Black Varnish pottery in the latter half of the 6 th c. BCE. First and foremost, the sheer volume of Black Varnish pottery is striking in and of itself. While it would be irresponsible to base firm conclusions solely on this single feature of the data, such a volume of a particular class of pottery is at least suggestive of regional consumer demand affecting the distribution patterns of trade goods. 372 Several other aspects of this phenomenon also deserve attention as well, all of which are consistent with this basic premise. For one, of all the pottery found at Salento that was produced in Attica, Black Varnish pottery is by far the most numerous pottery class (Figure 20). In general, we might expect Attic Black Figure pottery to exhibit a greater presence, and indeed it is prominent. However, the predominance of Black Varnish pottery among the Attic products is consistent with the larger overall distribution pattern. More importantly, it reflects a dynamic known as “dollar voting,” whereby consumers manifest their preferences through their consumption choices. Significantly, this process requires both the presence of traders who engage with the consumer base with sufficient frequency to be aware of these preferences, and of producers who alter their production strategies to 372 This argument is echoed by Walsh & Antonaccio (2014), who contend that consumers in southern Italy and Sicily exhibited a preference for pottery types that echoed metal vessels, and that Athenian manufacturers adjusted their production in order to accommodate this regional preference. Similarly, Osborne (1996, 2001) finds sufficient evidence in the pottery from southern Italy to conclude that already by the 6 th c. BCE, customers had exhibited pronounced consumer preferences that impacted the strategies of traders and potters who produced for export. 316 accommodate specific consumer demands, and presumably generating greater profits in so doing. All of this is entirely inconsistent with the picture of archaic pottery trade as driven by supply-side decisions, and this degree of mobility of information and the malleability of distribution and production in response to information indicates precisely the kind of market dynamics that Tandy (1997), for one, insists were not present at this time. Yet another supporting piece of evidence is the proportion of cups among the entire assemblage of Black Varnish pottery (Figure 21). Across all pottery classes, cups account for 73.34% of the total. However, they constitute 89.25% of all identified forms of Black Varnish pottery. This proportion is the highest of all the property classes, and is above the standard deviation. 373 According to Walsh (2014), Greek cups were more highly sought after by consumers in the western Mediterranean because they were viewed as markers of wealth, and their disproportionate level of representation among Black Varnish pottery is yet one more indication of the effect of consumer preference. This is also evident in the rapid growth of southern Italian Black Varnish pottery, as it appears that regional producers sought to capitalize on existing demand by manufacturing their own, possibly cheaper, version of the same product. At the same time, we must still keep sight of the fact that other forms of imported pottery were also increasing at this time. Largely as a result of the huge demand for Black Varnish pottery, Attic imports rose significantly in the latter half of the 6 th c. BCE, but Corinthian pottery also reached its highest total numbers by the end of the century as well, although still declining substantially as a share of the overall total. Contrary to the general Mediterranean-wide pattern, Corinthian pottery at Salento still accounted for both more numbers overall and a greater share of the total. In fact, in the second half of the 6 th c. BCE, Corinthian pottery at Salento outnumbered Attic Black Figure by nearly four times during a period in 373 The mean of the entire assemblage is 55.51%, with s = 31.99; the maximum SD is 84.72%, which is exceeded by the 89.25% of all Black Varnish pottery composed by cups. 317 which in the rest of the Mediterranean, Attic Black Figure was displacing Corinthian pottery as the dominant fine ware export. Still, several questions remain. For instance, most of the pottery excavated in Salento was recovered from Oria, a site that is, relatively speaking, nowhere near the coast (Figure 22). Conventional wisdom tells us that sites which are closer to the ocean, and therefore preferential access to maritime trade routes, should have relatively higher levels of imports. Obviously, this is not the case at Salento. What makes this even more puzzling is that Oria is not only landlocked, but it lacks any other geographic advantages that would compensate for its relatively isolated position. It is not situated on or near an especially navigable river, nor are there any other physical features of the terrain that would immediately explain why this distribution pattern would exist. It may be that we should reconsider our belief in the degree of correlation between coastal access and volume of trade, but for the moment, I can think of no obvious explanation. Pottery from Tarquinia The Data In comparison to the pottery data from Salento, the data from Tarquinia (Bonghi Jovino, 1999; Chiaramonte Treré, 1999) fortunately derive from a single source, thereby vitiating the need to reconcile data generated from different excavations. Also, and again unlike the pottery data from Salento, the data from Tarquinia include all of the main functional groups (fine wares, coarse wares, cooking wares, and transport amphorae), with the exception of lamps, which are normally very few in number, and presumably none were recovered or identified at the site. The overall numbers from Tarquinia are lower, although it is not surprising that totals from a single site would be lower than those from an entire region. Still, of the three assemblages used for my analysis, the totals from Tarquinia are the lowest (3,056 sherds from Tarquinia, compared to 8,003 from Salento and 13,814 from Emporion). I have no intention of drawing any specific conclusions from the relative totals, merely to point out that with the fewest total sherds, whatever conclusions we draw from the data from the excavations at Tarquinia will be more tentative than those from the other two sites. 318 In looking at the data from Tarquinia (Figure 23), several observations are to be made. First of all, the percentage of amphorae is extremely low (~2% of the total from 700-500 BCE). Cooking wares are even more rare (<1% of the total from 700-500 BCE). No lamp sherds were identified, but lamps typically constitute an extremely small portion of any assemblage, and their forms during this period are not nearly as recognizable as in later periods, and so it is not unusual for them to be unidentifiable when in a fragmentary state. Meanwhile, coarse wares form a substantial component of the total assemblage (21% of the total from 700 to 500 BCE). According to the excavators, all of the potsherds classified as coarse wares were locally produced “impasto” wares. 374 Finally, fine wares constitute the vast majority of the pottery from the site (76% of the total from 700-500 BCE). Upon closer examination of the fine wares (Figure 25), we find that Etruscan Bucchero Nero accounts for the most substantial portion of the fine wares with 64% of the total. The next two most numerous pottery classes are Etrusco-Geometric, with 18% of the total, and Attic Black Figure at 10% of the total. Black Varnish pottery accounts for 4% of the total, and no other pottery class exceeds 2%. The elevated presence of Bucchero Nero pottery is not surprising, given that it was a well-known product of Etruscan potters that required great skill and care to produce. 375 Also, the Etrusco-Geometric pottery was a style developed in the 8 th c. BCE, imitating imported Greek pottery, in particular from Euboea, but also from the Cyclades and Attica as well, with Tarquinia as one of the main centers of production, especially in the 7 th c. BCE. 376 From a diachronic perspective (Figure 24), we can see that overall, pottery levels remain 374 Etruscan “Impasto” is a type of pottery characterized by a buff or light brown color and frequent inclusions of mica and silex. It is highly frangible, but the coarse fabric makes it highly resistant to heat shocks (Carafa, 1995); for this reason, it could be the case that a portion of the impasto pottery classified as coarse wares may very well have been cooking wares. 375 Bucchero Nero is a characteristic pottery from Etruria, characterized by a glossy black, metallic sheen, and required kilns capable of achieving firing temperatures between 900°-1050°C (see Hirschland-Ramage, 1970; Rasmussen T. B., 1979; Bouloumié, 1982). 376 See Martelli & Adembri (1987); Ambrosini (2013) offers an excellent recent survey of the scholarship. 319 low throughout the 7 th c. BCE, with a major spike at the beginning of the 6 th c. BCE, and then another less dramatic increase in fine wares in the last quarter of the 6 th c. BCE. Analyzing the Data Once we begin to analyze the data, a number of features deserve mention, particularly if we consider the pottery data from Tarquinia in juxtaposition to that from Salento. First of all, the earliest ceramic finds at Tarquinia are from a much earlier date than the pottery from Salento. The earliest pottery from Salento begins to appear in the middle of the 7 th c. BCE, and even that is but a trickle. By contrast, some of the pottery from Tarquinia was recovered from contexts that date as early as the second quarter of the 8 th c. BCE. On stylistic grounds, several other imported pieces are dated to the early 8 th c. BCE. 377 Also, local production of Bucchero Nero and Etrusco-Geometric pottery is found in contexts dating to the end of the 8 th c./beginning of the 7 th c. BCE. It seems that from a very early date, Etruscan consumers had developed a taste for Greek pottery styles and had begun producing their own imitations. 378 Still, relative levels of fine wares remain fairly low throughout the 7 th c. BCE, amounting to fewer than 100 sherds per five-year period until the last quarter of the 7 th c. BCE (Figure 24). At the same time, given the lower overall numbers of potsherds from the Tarquinia excavations, they nevertheless account for a more significant percentage of the total, and strongly suggest more intensive connections between Etruscan consumers and Greek producers from a much earlier date. 377 These include the aforementioned Euboean skyphos and kotyle, as well as a number of Corinthian cups and one oenochoe. 378 Williams (2013) identifies the work of several potters that he believes are Greek immigrants working in southern Italy as early as the 8 th c. BCE, and he argues that we can trace the in