Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Responsibility and rights: a search for a principled distinction between criminal law and tort law
(USC Thesis Other)
Responsibility and rights: a search for a principled distinction between criminal law and tort law
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Responsibility and Rights:
A Search for a Principled Distinction between
Criminal Law and Tort Law
Aness Gaeyoung Kim Webster
PhD in Philosophy
Conferred by the USC Graduate School in August 2016
University of Southern California
1
Table of Contents
Acknowledgements ...................................................................................................... 2
Chapters
Chapter 1 – Introduction ........................................................................................ 4
Chapter 2 – Ripstein’s Answer .............................................................................. 26
Chapter 3 – Mental States, Defences and Different Accounts of Responsibility ...46
Chapter 4 – Blameworthiness for (Criminal) Negligence ...................................... 69
Chapter 5 – Responsibility and Tort Law ............................................................. 95
Chapter 6 – Rights and Responsibility ................................................................. 106
References ................................................................................................................ 118
2
Acknowledgements
This dissertation is dedicated to Jonathan McKeown-Green. He was my teacher, my
mentor, a collaborator, my biggest cheerleader, a philosophical ally and most
importantly, my friend. He saw something in me that I couldn’t see for myself. He was,
at times, my harshest critic. He saw versions of drafts that I would never dare show
anyone else and they were met with predictable insight and probing questions. He
showed me that a good prose can still be rigorous. He affirmed in me the importance
of being able to explain what your project is, no matter the audience. I would not be
the writer, the teacher, or the philosopher that I am without him. Jon, I miss you so
much. I only cried three times writing this and for that, and all the other things that
you’ve helped me do, both near and far, I hope I’ve made you proud.
My philosophical training and my love for philosophy began at the University of
Auckland. My fascination with ethical theories was sparked by the very different ethical
orientations of John Bishop, Rosalind Hursthouse, Tim Mulgan, Christine Swanton
and Vanya Kovach. Others I met at Auckland have also been crucial in my
development as a philosopher. For that, I thank Fred Kroon, Justine Kingsbury, Ben
Miller, Glen Pettigrove, Denise Vigani and Konni Woods. Many of you are my
teachers as well as my friends and for that I’m so grateful.
That I made the right decision to come to USC for grad school has been confirmed
many times. I feel extremely fortunate to have been one of Gary Watson’s students. He
is a wise philosopher from whom I have learnt so much. He pushed me to think about
the implications of my views for other philosophical issues and to see (and challenge)
the connections between them. He was able to make issues in criminal and tort law
come alive as issues in philosophy of action, agency, and responsibility. Thanks to Greg
Keating for teaching me tort law. I gained a greater appreciation for the intricacies of
tort law from his insightful analyses of tort cases and how they connect with wider issues
in political philosophy. Jon Quong came onto my committee fairly late into my
dissertation process. But nonetheless (and unsurprisingly, if you know him) he has given
me so many helpful comments. He urged me to not only consider those perspectives to
which I wasn’t particularly sympathetic, but made them compelling. Gideon Yaffe was
one of the first philosophers that I encountered who applies sharp analytical skills to
the understanding of the law. I thank him for opening my eyes to the kind of precision
with which we can describe not only the law, but the normative terrain in general.
Thank you all for pushing me to be a better philosopher and for all the support and
guidance you’ve given me throughout my graduate career..
This dissertation would not be in the shape it is today without the help of many
others who have read drafts of various chapters. For this, I thank Erik Encarnacion,
Steve Finlay, Ben Lennertz, Shyam Nair and Alex Sarch. Special thanks to Steve Bero
who has read more of the dissertation that anyone else other than Jon and my
3
committee. Your feedback has been invaluable to me. I look forward to sharing our
work for many years to come.
I’ve benefitted hugely from numerous conversations and seminars with the faculty
at USC. It’s been amazing to be supported so well by the faculty. Thanks, in particular,
to Scott Soames who inspired us “to do good and have a good time doing it”. Thanks
to Mark Schroeder and John Hawthorne for their invaluable support and guidance. I’d
like to say a special thanks to Kadri Vihvelin who has read so much of my work. Thank
you for your encouragement and your interest.
The amazing graduate community has been a constant source of engaging
philosophical conversations, encouragement, and fun. I could name almost everyone
who has been at USC at the same time as me. But I’ll risk missing some out and thank
Ara Astourian, Matt Babb, Renee Bolinger, Justin Dallmann, Kory DeClark, Alex
Dietz, Maegan Fairchild, Marina Folescu, Keith Hall, Nick Laskowski, Indrek Reiland,
Johannes Schmitt, Sam Shpall, Kenneth Silver, Justin Snedegar, and Jonathan Wright.
My time in LA would not have been as great without the company of James
Ashenhurst, Brian Blackwell, Lindsey Cordova, Emmy Feamster, Neetu Khan,
Hannah Lau, Marcy Lascano, AJ Moorehead, Clo Pazera, Amanda Prunesti, Jason
Raibley, Beth Snyder, Ryan Walsh and Ilana Winter. Thanks for enormous amounts
of fun and often, much needed distraction from philosophy.
To Deborah Milnes, who is one of my longest friends, I hope you know how much
your friendship means to me.
I thank my parents-in-law, Brian and Sherree Webster, for all their support and
encouragement. I hope they know how much I appreciate them.
I also would like to thank my siblings, Savina and Jun Kim. Sav, you are an
inspiration to me how you take on challenges with such tenacity. Jun, I love the fact
that we share the same intellectual curiosity and I look forward to many more
conversations about all sorts of things.
There are two people that I need to acknowledge more than any other. First, my
mum. I love the fact that she can understand how I’m feeling by the way that I say
‘hello’. I hope she knows how much it has helped me knowing that she believed in me
and felt my struggles with me. I’d like to thank her for teaching me the value of hard
work and perseverance. I love you so much and thank you for everything that you’ve
done for me.
And last, but definitely not least, I thank Myles Webster. Thank you for believing
that I could do this and for all the little and big things that you’ve done to support me.
I know that you know how much you mean to me. It’s been a great adventure. I hope
you agree!
4
Chapter 1: Introduction
A Search for a Principled Distinction between
Criminal Law and Tort Law
1 Demarcation Question and Some Candidate Answers
Many situations in which a crime has been committed are also situations where a tort
has been committed. For instance, a person who has been convicted of murder might
also be liable for a tort of wrongful death. But of course the criminal sanctions faced by
a person who is convicted of a crime are much more severe than the consequences
faced by that person who is found liable for a tort for the same behaviour. Both are laws
of wrongs but there are differences between the two domains of law. Tort law
compensates those who are wronged whereas criminal law punishes the wrongdoers,
and tort law requires that the plaintiff (the party who brings forward the suit) prove on
the balance of probabilities (also known as the preponderance of evidence) that the
defendant committed the tort whereas criminal law requires that the prosecution (who
acts on behalf of the public) prove beyond a reasonable doubt that the defendant
committed the crime. Is there a principled and coherent framework that makes sense
of the differences while upholding the similarities? Call this the Demarcation Question.
Many scholars have abandoned the task of answering the question. They argue
that the differences are arbitrary and products of historical accidents for which no
principled rationale can be given. The thought is that there is no principled distinction
between the two domains and all we can hope for, in order to understand the
differences between the two domains, is to provide some laundry list of the difference
between the two. Tort law and criminal law have different consequences for those
found liable and different procedures. Moreover, some conduct is both criminal and
tortious whereas others are not.
Despite these challenges, some have tried to establish the unity of tort law and
criminal law and have highlighted the elements that are common to both domains of
the law. Both are laws of wrongs and Oliver Wendell Holmes in The Common Law has
argued for an objective understanding of liability that underlies both domains (1881).
Of course, noting some similarities does not rise to giving a principled distinction
between the two domains and hence answering the Demarcation Question. Hence,
some have tried to show that the laundry list of differences can be unified by a
principled distinction between tort law and criminal law. One such attempt takes
seriously the idea that tort law is in general about compensation and criminal law is, in
general, about punishment. This has led some to argue that a principled distinction
between tort and criminal law can be explained by appealing to the distinction between
corrective justice and retributive justice. In Section 3, I explain further the answer to
the Demarcation Question that appeals to the distinction between the principles of
5
corrective and retributive justice. I conclude that this distinction cannot provide an
adequate answer to the Demarcation Question.
Another attempt to answer the Demarcation Question that has been proffered in
the literature distinguishes between public and private wrongs. The idea is that crimes
are public wrongs whereas torts are private wrongs. This view also asserts that it can
make intelligible certain procedural differences between the two domains of law. In
particular, the claim that crimes are public wrongs is supposed to be able to explain the
fact that state prosecutors bring forward a criminal suit against the one who is accused
of the crime (the defendant) on behalf of the public and the fact that the victim does not
have a formal say in whether or not the wrong she suffers is prosecuted.
1
In contrast,
the claim that torts are private wrongs can explain the fact that the parties to a tort suit
are private individuals and the fact that the person who suffered (the plaintiff) the wrong
has the discretion to sue the wrongdoer (the defendant). In Section 4, I elaborate on the
distinction between public and private wrongs and how that distinction constitutes an
answer the Demarcation Question. However, I shall argue that the distinction cannot
provide us with an adequate answer to the Demarcation Question partly because many
accounts of what makes a wrong public deliver the wrong result about what is criminal.
In particular, none of these accounts seems to have the resources to explain the fact
that performing one action can make us both liable for a tort and guilty of a crime. I
also examine one understanding of what makes a wrong public that has recently been
defended by two different writers, but I shall argue that this understanding collapses
into the answer to the Demarcation Question that appeals to the distinction between
principles of corrective and retributive justice.
Despite the inadequacies of these two attempts at answering the Demarcation
Question, I think that progress can be made by taking clues from the substantive
doctrines of tort and criminal law. I argue the we will find that different concepts of
intention are employed in tort and criminal law once we pay careful attention to the
elements of intentional torts and contrast those with what is required to satisfy the
intention or purpose element of crimes. Additionally, when we pay attention to
substantive doctrines of the two domains, we see that what counts as an excuse differs
in the two domains. (Insanity, for instance, is an excuse in criminal law, but not in tort
law.) In Chapter 3, I argue that these differences that we find suggest that two different
understandings of responsibility are implicated in the two domains. In particular, I
provide a prima facie case for the claim that that criminal law implicates blameworthiness
whereas tort law implicates an understanding of responsibility that is weaker than
blameworthiness.
1
Often the victim does have some say in whether a particular case is prosecuted; if the victim is unwilling
to testify or cooperate with the state prosecutor, then the prosecutor may use her discretion not to
prosecute the case. However, it is within the prosecutor’s powers to compel the victim to testify by issuing
a warrant for court appearance and whether or not the prosecutor compels the victim to testify, it is in
her discretion to prosecute the case.
6
This chapter, however, is focused on the two attempts at answering the
Demarcation Question mentioned above. In Section 3, I explain and argue against the
adequacy of appealing to the principles of corrective and retributive justice to answer
the Demarcation Question. In Section 4, I explain and argue against the adequacy of
the answer to the Demarcation Question that appeals to the distinction between public
and private wrongs. In Section 5, I briefly reflect on the lessons learnt from these two
sections and argue that an adequate answer to the Demarcation Question must provide
some principled way of distinguishing between criminal law and tort law while allowing
for the contingency and evolution of what counts as crimes and torts. I also outline a
methodology of discovering an answer to the Demarcation Question. In Section 6, I
outline the structure of the rest of the dissertation and what I hope to achieve in the
next five chapters.
2 Explanation versus Justification
Before I examine the attempts at answering the Demarcation Question, it is important
to note that the question with which we are concerned is an explanatory question and
not a justificatory question. That is, I am not asking what justifies preserving two distinct
legal domains of tort and criminal law and what justifies the distinctive features, if any,
of these two domains. This is important to point out because some who have attempted
to answer the Demarcation Question have the justificatory question in mind. For
instance, Antony Duff and Sandra Marshall appeal to the notion of public wrongs and
claim that crimes, but not torts, are public wrongs. However, they make it clear that
this claim is not a descriptive explanation of crimes, but a normative one. That is,
according to them, publicness of crimes is not a unifying story about what is
criminalised, but what should be criminalised. Their concern is “with the question, not
of how we can identify what is criminal, of what now falls within the scope of the
criminal law, but of how legislatures and polities should set about deciding what ought
to be criminalised” (2010: 70).
They contrast their approach to the approach that textbook writers on criminal
law might take. The latter might introduce the subject of criminal law by providing a
definition of a crime. They provide, as an example, the definition of a crime given by
Gerald Gordon in his The Criminal Law of Scotland: “The criminal law is probably,
therefore, sufficiently defined as that branch of the law which deals with those acts,
attempts and omissions of which the state may take cognisance by prosecution in the
criminal courts” (2000: 7). For this “descriptive or analytical” purpose, Duff and
Marshall claim, “a formal definition is the most we can hope for”. I agree that Gordon’s
definition might be a good definition of crimes, but I think that we can provide intelligible
explanation of crimes which provides more than a mere formal definition. To see the
distinction, consider the following question: “Why is Aidan a bachelor?” We might
7
answer this by stating that Aidan is a bachelor because he is an unmarried man. After
all, being an unmarried man may be all it takes for Aidan to be a bachelor; just as being
an item of conduct for which the state prosecutes may be all it takes for the item of
conduct to be a crime. But there is a deeper explanation that could be given for why
Aidan is a bachelor – perhaps he has not yet found a partner, or perhaps, he has an
objection to the institution of marriage. Similarly, we could ask for a deeper explanation
or rationale of why those items of conduct count as crimes.
Moreover, this deeper explanation need not be a justification. That Aidan has
strong objections to the institution of marriage may explain his bachelorhood, but this
may not justify it, especially if it turns out that his particular objections are unfounded
or misguided. To clarify further this distinction, it may be helpful to contrast an
intelligible rationale for some action with a justification for that action. Suppose
Beatrice, who is normally very mild-mannered, yells at Celia for not having cooked
dinner on time as she promised. We might be able to explain Beatrice’s action that is
very much out of character by appealing to the fact that with a fussy newborn, Beatrice
is extremely sleep-deprived. But although this makes intelligible Beatrice’s action (and
makes intelligible, in particular, the coherence of her action with her temperament),
this does not justify her action. Or suppose we find out that we have certain implicit
biases. That we have these biases may explain our action (such as thinking that a
particular resume with the name ‘Dean’ is better than an identical resume with the
name ‘Elaine’). So when we want to understand why the company decided to hire Dean
rather than Elaine, we could appeal to the implicit biases of the people involved in the
decision, but it certainly does not justify the decision.
In addition, the kind of explanation that I am after is an explanation that is internal
to the law. Consider the jury’s verdict in the criminal case against O.J. Simpson. Here
are two, not necessarily competing, explanations of the verdict: (i) the jury believed the
gloves found at the crime scene did not fit Simpson’s hands and so believed that it was
reasonable to doubt that Simpson killed the victims; and (ii) the jury did not want to
convict a beloved football star. The former explanation is internal to the law in the
sense that it speaks to issues that explicitly appear in legal judgements. However, the
latter claim is not endorsed as a relevant consideration by the law. I am not claiming
that the latter is not a true explanation of what happened, but I am interested in the
explanations of the former kind and what might unify these internal explanations.
Indeed, the O.J. Simpson example is particularly pertinent as he was not found guilty
of homicide, but he was found liable for the tort of wrongful death for the same deaths.
Here is a claim that may explain these verdicts: the criminal case was sensationalised
in the media in a way that swayed the jurors, but the tort case was not so sensationalised
and so the jurors of the tort case were not so affected. Even if this can explain the
different verdicts in the tort and criminal cases of O.J. Simpson, this is an external
explanation and not the kind of explanation in which I am interested. An adequate
answer to the Demarcation Question should, by appealing to considerations that are
8
internal to the law, help us explain how it is possible that a person can be charged with
a crime and be sued for a tort for the same act and how it is possible there could be
different verdicts in such a case.
3 Distinction between Corrective and Retributive Justice
The claim that tort law is governed by the principle of corrective justice has been
championed by Ernest Weinrib (1995) who argues for the unity of private law (which
includes tort law as well as contract law and the law of equity) by appealing to this
principle. In addition, Jules Coleman (2001) argues that the principle of corrective
justice can provide an explanatorily adequate account of tort law. The principle of
corrective justice is a second-order principle in the sense it tells us about what happens
when somebody breaches another’s (first-order) right. It tells us that “individuals who
are responsible for the wrongful losses of others have a duty to repair the losses”
(Coleman 2001: 15).
2
Moreover, since the core idea behind retributive justice is about
when punishment is justified, it is natural to think that criminal law is governed by
something like the principle of retributive justice.
3
Indeed, according to Jeffrie Murphy,
“a retributivist is a person who believes that the primary justification for punishing a
criminal is that the criminal deserves it” (2007: 11).
4
In this section, I present the reasons
for appealing to this distinction between principles of corrective and retributive justice
in order to answer the Demarcation Question. I ultimately conclude that this distinction
cannot provide an adequate answer to the Demarcation Question.
3.1 Corrective-Retributive Answer
In this section, I focus on how the distinction between corrective and retributive justice
can provide an answer to the Demarcation Question.
2
Exactly what course of action is required by the principle of corrective justice is up for debate. One
understanding is that in order to repair the wrong and make the plaintiff whole, the defendant must
compensate for the wrong. This view is defended by Mark Geistfeld (2014). Even if the principle of
corrective justice does not entail that the tortfeasor must compensate the plaintiff, the claim that tort law
is governed by the principle of corrective justice can explain the fact that the default or the paradigmatic
legal outcome of a successful tort suit is the award of compensatory damages. This is because
compensating the plaintiff is the next best alternative to true repair and making the plaintiff whole.
3
Although I appeal to the principle of retributive justice, we should note that there are many different
versions of the principle although there is some convergence in the literature. See John Cottingham
(1979) for nine varieties of retribution.
4
Usually, retributivism is understood to entail the claim that punishment is, at least sometimes, justified.
But there is a weaker version of retributivism available according to which there is some positive value
in punishing a wrongdoer that does not consist merely in the good consequences of punishing the
wrongdoer.
9
CORRECTIVE-
RETRIBUTIVE ANSWER:
The claim that tort law is governed by the
principle of corrective justice while criminal law
is governed by the principle of retributive justice
can explain the differences between the two
domains of law.
This answer to the Demarcation Question can help explain that tort law is, in general,
about compensation and criminal law is, in general, about punishment. Moreover, the
difference between the legal outcomes in successful tort and criminal cases can explain
other features of the two domains. It could explain the difference in the standards of
proof: punishment, a deliberate infliction of pain, seems to require better justification
than demanding compensation. If tort law is about compensation, it makes sense that
the party who must bring forward the suit is the party that wants to be compensated,
namely the plaintiff. Moreover, the Corrective-Retributive Answer could be further
justified by noting that principles of corrective justice and retributive justice are
incompatible with each other in the sense that these principles make incompatible
demands of the defendants.
5
This incompatibility lends support to the idea that the
differences can be explained (at least, in principle) by the differences between the two
principles.
Of course, the claim that tort law is about compensation and that criminal law is
about punishment is a claim about what is typical or paradigmatic about each domain.
Indeed, there are other consequences that attach to the verdict that a tortfeasor is liable
or that the accused is guilty of a crime. Tort law sometimes requires tortfeasors to pay
punitive damages, the purpose of which, as the name suggests, is to punish the
defendants who violated the plaintiffs’ rights rather than merely compensating for the
violation. In addition, criminal law sometimes refrains from punishing the defendant
but instead requires the defendant to undergo psychiatric treatment.
6
However, the
Corrective-Retributive Answer could be an adequate answer to the Demarcation
Question even if it cannot accommodate every single feature of the two domains of
5
After all, corrective justice requires compensation that equals the amount of harm caused by the
wrongdoer whereas retributive justice requires punishment that is proportional to the desert of the
wrongdoer. For arguments for this incompatibility claim, see Coleman (1974) and (1982), Guido
Calabresi (1970) and (1975), and David Owen (1985), among others. The wide acceptance of this claim
is noted by Larry Alexander (1987: 5).
6
In some jurisdictions, a defendant who is deemed to be insane or mentally ill at the time of the crime is
found not guilty but may be required to undergo treatment. These jurisdictions include California (with
the verdict of ‘not guilty by reason of mental disease or defect’), Canada (with the verdict of ‘not
criminally responsible’), New York (with the verdict of ‘not responsible by reason of mental defect’) and
New Zealand (with the verdict of ‘not guilty by reason of insanity’). In some jurisdictions including
Colorado, District of Columbia, Michigan, North Carolina and Texas among others, the verdict of ‘not
guilty by reason of insanity’ carries with it a mandatory treatment order. (North Carolina has 120 days
of a mandatory treatment and Texas has mandatory treatment for violent crimes, but treatment order is
discretionary for nonviolent crimes.) In other jurisdictions, however, a defendant who was insane or
mentally ill at the time of the crime can be found guilty but the defendant is not punished and may be
required to undergo treatment. Examples of such jurisdictions include Alaska, Denmark, Montana,
Norway, Sweden and Utah.
10
law.
7
Accordingly, I do not argue that the mere fact that there are legal consequences
other than the one suggested by the principle of corrective justice in tort law and the
one suggested by the principle of retributive justice in criminal law establishes the falsity
of the claim that tort law is governed by the principle of corrective justice and that
criminal law is governed by the principle of retributive justice. However, the question
remains whether principles of corrective justice and retributive justice are the best
explanations of tort law and criminal law, respectively. After all, if these principles do
not adequately explain the two domains, then we have reason to doubt the Corrective-
Retributive Answer to the Demarcation Question. I examine two different objections
to the claim that principle of corrective justice governs tort law and that the principle
of retributive justice governs criminal law, with an eye to how this establishes the
inadequacy of the Corrective-Retributive Answer to the Demarcation Question.
3.2 Objection: Differences in Rights
One objection to the claim that the principle of corrective justice provides the best
explanation of tort law is that given the fact that the principle of corrective justice is a
second-order principle, it is not able to explain the first-order rights that are recognised
and protected by tort law. We can see how this objection might establish the
inadequacy of the Corrective-Retributive Answer to the Demarcation Question. If the
principle of corrective justice is a second-order principle and cannot explain first-order
rights of tort law, then a distinction between the principles of corrective and retributive
justice would not be able to account for the fact that there are differences in what is
criminalised and what is tortious. That is, given that the two principles are concerned
with what ought to be done when some first-order right is breached, the differences
between the principles cannot explain the differences between the rights/duties that
are recognised and protected by the two domains.
Although Coleman does not proffer the distinction between corrective and
retributive justice as an answer to the Demarcation Question, he anticipates a similar
objection to his conception of tort law that the principle of corrective justice is not an
adequate theory of tort law because it is incomplete (2001: 32). This incompleteness objection
is that the principle of corrective justice does not explain all of tort law because a
complete theory of tort law ought to provide an adequate account of the first-order
rights/duties as well as the second-order right to redress or duty to repair.
8
As Coleman
7
This follows from a claim about explanations in general. The best explanation of a patient’s numerous
symptoms may be such that it cannot accommodate one of the many symptoms. It would still be an
adequate explanation if it can make intelligible all the other symptoms. (See Josephson and Josephson
(1994: 9-12).)
8
Rights and duties are correlative in the sense that if I have a duty to you to φ, then you have a right
that I φ. I will note the correlativity of these by writing “rights/duties”.
11
puts it, the objection is that the principle of corrective justice cannot “provide a theory
of what counts as a wrong of the sort that gives rise to a duty of repair” (2001: 32).
Given the connection between the incompleteness objection to corrective-justice
theory of tort law and the incompleteness objection to the Corrective-Retributive
Answer to the Demarcation Question, it is productive to examine Coleman’s responses
to the incompleteness objection and see how, if at all, they can help answer the
objection to the corrective-retributive justice answer to the Demarcation Question. In
the end, I conclude that Coleman’s two responses are inadequate and that a corrective-
justice based theory of tort law is inadequate because it is incomplete. This also means
that the Corrective-Retributive Answer to the Demarcation Question is inadequate
because it is incomplete.
3.2.1 Coleman’s First Response
Coleman’s first response is that the principle of corrective justice, albeit second-order,
is substantive. This is because the principle circumscribes the kinds of first-order
rights/duties that are recognised by tort law in a way that makes sufficiently clear what
those rights/duties are. Coleman accepts that the principle of corrective justice does
not provide an account of the first-order rights/duties, but he insists that this does not
render the principle purely formal, or lacking in content. This is because the principle
of corrective justice is not compatible with just any set of first-order rights/duties (2001:
32). According to Coleman, corrective justice is only intelligible as a principle of justice
if the first-order rights/duties are such that it is plausible for the breach of those rights
to give rise to a second-order duty of repair. Moreover, he appeals to a set of paradigm
first-order rights/duties that are compatible with the principle corrective justice and
claims that this is sufficient to make corrective justice “an appropriately complete
account” of tort law.
My response is two-fold. First, in claiming that intentional torts are paradigmatic
first-order rights/duties, he relies on the claim that intention torts form a core of tort
law. However, what constitutes the core of tort law is controversial. There is a widely
(if not universally) accepted tripartite taxonomy of three distinct types of torts:
intentional torts, negligence and strict liability and Coleman denies that strict liability
partially constitutes the core of tort law. However, other theorists have regarded strict
liability as the core of tort law.
9
Second, it is not clear that a theory of tort law that can only account for the set of
paradigm first-order rights/duties is a sufficiently complete theory. If there are non-
paradigmatic first-order rights that are nonetheless protected by tort law, then a
sufficiently complete theory should explain why these rights give rise to the duty of
repair. This is especially so given that there are other first-order rights that are protected
9
Richard Epstein (1973) is an example. Stephen Perry (2001) also attempts to defend a theory of tort law
that can explain strict liability torts. (Perry’s account will be discussed in detail in Chapter 5.) In contrast,
Arthur Ripstein claims that negligence is the primary basis of tort liability (1999: 48).
12
by other domains of law, but not protected by tort law.
10
Suppose Ephraim and Gina
sign a (valid) contract. Gina, has a right that the terms of the contract are satisfied and
Ephraim violates that right by breaching the contract. However, Ephraim does not
commit a tort by breaching the contract even though he now has a duty of repair. This
is related to the claim that Weinrib makes that the principle of corrective justice governs
private law and that it is not specific to tort law. Rights protected by other realms of
private law (such as contract law) also give rise to the second-order right to redress.
Moreover, unjust enrichment gives rise to the right of redress even though it does not
involve a right-violation (since we do not have a right that others are not enriched at
our expense). This means that the principle of corrective justice cannot help in
demarcating tort law from other legal realms, and more specifically, other first-order
rights that are protected by those legal realms.
11
There are also rights that are protected by other domains of law but do not give
rise to a second-order duty of repair. For another example, consider our legal duty to
pay taxes. Given the correlativity of rights and duties, we have a right that the tax-
paying residents pay their taxes. However, your failing to pay your tax does not amount
to a tort.
12
Hence, there are certain rights that are not protected by tort law. This means
that an adequate theory of tort law ought to provide an account of which rights are
those that are protected by tort law. Moreover, it is an important part of tort law that
one commits a tort by detaining somebody without legitimate authority, and by
announcing to the public something false about somebody that damages that person’s
reputation. Since the principles of corrective and retributive justice do not come
equipped with accounts of the first-order rights that are protected by tort and criminal
law, plausibly, the Corrective-Retributive Answer to the Demarcation Question is not
10
We should also note that the notion of a wrong appealed to by the principle of justice is not a pre-
theoretical notion. An intuitive notion of wrong (if there is one) might classify an act of breaking a promise
without justification as a wrong. However, not all such wrong-doings count as tortious wrongs. Hence,
without an appropriately theory-internal understanding of what counts as a wrong in tort law (and
therefore, what counts as a first-order right), we do not have all the elements necessary for a sufficiently
complete theory of tort law.
11
This means that there is a more general demarcation question about how to distinguish between
private and public law and perhaps another question about how to distinguish criminal law from ‘redress’
law. The extent to which the distinction between criminal and ‘redress’ law helps depends on what is
counted as ‘redress’ law. After all, one can seek remedies for discriminatory practices of a public
institution (say, a public office, or a public university). If this is part of ‘redress’ law, then the answer to
the demarcation question concerning criminal and redress law will come apart from the answer to the
demarcation question concerning private and public law. All this is to say that there are many different
demarcation questions that we can ask and the relationships between them are not straightforward and
too complicated to discuss briefly.
12
Here is another example. Citizens of the United States of America have a right that their laws be
constitutional. However, a breach of this right (by some legislature enacting an unconstitutional law)
presumably does not amount to a tort. Also, Barack Obama (as the President of the United States) can
nominate a justice to the Supreme Court and has nominated Merrick Garland as a justice to the Supreme
Court. We may have a legal right that the Senate provides Garland a fair hearing and put him up for a
vote. Suppose this right is being violated. Yet we seem to lack any civil redress against the Senate. Thanks
to Erik Encarnacion for this second example.
13
sufficiently complete as it cannot explain the similarities and the differences between
first-order rights/duties that are recognised by tort law and criminal law.
13
3.2.2 Coleman’s Second Response
Coleman supports his conclusion that the principle of corrective justice amounts to an
“appropriately complete” theory of tort law by drawing an analogy with criminal law.
He claims that just as retributivism is a good justificatory and explanatory theory of
criminal law, the principle of corrective justice is a good justificatory and explanatory
theory of tort law. He claims that no one objects to retributivism (as an adequate theory
of criminal law) on the grounds that it fails to provide a list of criminal conduct.
Retributivism is a view about when and why punishment is appropriate, namely,
punishment is justified when the person being punished deserves it. Moreover,
punishment is meant to be proportional to the desert. Given this, a retributivist theory
of criminal law seems to be able to explain (and justify) the excuses that are available in
criminal law. For instance, it can explain why a defendant who committed an act that
constitutes a crime under duress should not be punished or should be punished less
severely than someone who committed the crime without being under duress.
However, it is not a good explanatory theory of criminal law for the very reason
appealed to by the incompleteness objection to corrective justice theory of tort law.
Plausibly, an adequate theory of criminal law should provide an adequate account of
the kind of acts that count as criminal.
14
Indeed, this echoes Douglas Husak’s lamenting
the neglect of the issue of what ought to be criminalised (2008, especially pages 58-60).
I should note that Husak is concerned with the normative question about what should
be criminalised. This is a normative question that must answer to morality and political
philosophy. But he also devotes a chapter on what he calls “internal constraints”,
constraints on what ought to be criminalised that come from the criminal law itself. I
take it that an adequate explanation of criminal law must explain and account for these
‘internal’ constraints on what is criminalised, or, what first-order right are protected by
criminal law.
In sum, the Corrective-Retributive Answer to the Demarcation Question is
incomplete because it does not pay adequate attention to the first-order rights that are
13
One potential response to this objection is to supplement the Corrective-Retributive Answer with an
account of first-order rights that are protected by tort law and an account of rights that are protected by
criminal law. As we shall see, this is similar to the kind of answer to the Demarcation Question that I
defend which is a two-pronged account where one of the prongs is an account of rights that are protected
by the two domains. However, supplementing the Corrective-Retributive Answer with an account of
rights is inadequate as there are objections to the principles of corrective and retributive justice
themselves (discussed in Section 3.3).
14
Again, an adequate theory of criminal law need not provide a complete list of items that constitute
criminal conduct. However, it should, at least, have some general groupings under which we can put
most criminal acts and an account of why those general types of conduct are such that criminality
attaches to them. Alternatively, it could provide the general principles underlying what counts as
criminal conduct.
14
protected by the two domains and hence, cannot explain the differences in the first-
order rights between the two domains.
3.3 Objection: Other Responses
Another objection to the Corrective-Retributive Answer is that the principle of
retributive justice cannot explain the other legal responses available in the criminal law.
I noted above that the mere fact that the principle of retributive justice cannot explain
all of the consequences of being found guilty of a crime need not be fatal to the theory
that retributive justice explains criminal law. However, in the next subsection, I outline
some of the criminal responses that cannot be explained by appealing to the principle
of retributive justice and why it is not plausible that these criminal responses are not a
core part of criminal law. Then, I turn to tort law and run an analogous argument
against principle of corrective justice.
3.3.1 Criminal Responses and Retributive Justice
I mentioned above that a criminal defendant may be required to undergo psychiatric
treatment. Plausibly, we cannot explain this requirement by appealing to the principle
of retributive justice since civil commitment is not punishment. However, one who
wishes to defend the claim that retributive justice best explains criminal law can argue
that in many jurisdictions, the defendant is not found guilty of the crime and this is
compatible with retributivism since a mentally ill defendant is not deserving of
punishment.
15
However, an adequate explanation of criminal law should explain under
what conditions non-punitive consequences are faced by a non-guilty criminal
defendant and this may be a problem for a retributivist theory of criminal law since it
does not speak to non-punitive consequences.
One kind of non-punitive consequences that can be faced by a defendant who is
found guilty of a crime is being subject to a compensation or a restitution order. The
court may require a guilty defendant to pay a sum of money to his victim to compensate
for the harm caused and the court may require a guilty defendant to return the stolen
goods that are still in the hands of the defendant. A proponent of the retributivist theory
of criminal law could argue that this is not a problem as this is simply a case of criminal
law stepping in and doing the work of tort law. After all, if the victim was compensated
by a criminal compensation order, she will not be eligible to seek civil redress by suing
for compensatory damages.
However, there are other consequences that a guilty defendant may face that are
more problematic. One such consequence is a community order. There are different
kinds of community orders. The court can order the defendant to receive alcohol or
15
As mentioned above, some jurisdictions deliver the verdict of ‘guilty but insane’ or ‘guilty by mentally
ill’ but perhaps a retributivist can argue that the ‘not guilty’ verdict is more accurate to what happens to
those defendants.
15
drug treatment or to do unpaid work (also known as ‘community service’). Another
court order that a guilty defendant can receive is an exclusion order which prohibits
the defendant from going to certain places. The exclusion requirement is designed to
prevent re-offending by keeping the offender from places where she is likely to commit
the same type of offence (of which she was found guilty). For instance, a shoplifter who
has been found guilty of theft on numerous occasions may be banned from going to a
particular shopping centre.
16
Another court order that is available is disqualifying a
defendant from driving when she has been convicted of a driving offence, such as
causing harm while driving under the influence of alcohol. Despite the differences
between these court orders, one thing is clear; they are not intended as punish the
defendant but to prevent reoffending and so these court orders cannot be explained in
terms of retributive justice.
Another consequence that every guilty defendant must face is having a criminal
record. Some have interpreted this as a punitive measure – designed to inflict harm on
the guilty. It may be true that this kind of public record and accompanying censure or
condemnation may be harmful, but it is not obvious that this is a punitive measure.
Moreover, to regard having a criminal record as punishment is problematic on
retributivist grounds since the sentence is itself supposed to be proportional to desert.
After all, the punitive effects of having a criminal record is not considered when
deciding what sentencing is proportional to desert. Hence, punishing someone by
keeping a criminal record is punishment that is inflicted over and above what is
proportional to desert. Furthermore, the use and maintenance of a sex offender registry
(which is a particular type of a criminal record) serves a purpose other than punishment.
It allows government authorities to keep track of the residence and activities of sex
offenders and this tracking is continued when the offenders have completed their
criminal sentences and so supposedly have been punished sufficiently.
17
(In many
jurisdictions, this registry allows authorities to impose and enforce some of the
restrictions mentioned above (restrictions on place of residence, for instance).) Although
being on a sex offender registry, or not being allowed to live in a certain place may be
harmful, it is not clear that punishment is the intention of these measures. More
plausibly, these measures are in place in order to prevent re-offending or making the
community feel safe that those who may reoffend are being tracked by governmental
agencies.
In sum, there are a variety of non-punitive consequences that may be faced by a
guilty defendant. Moreover, it is not clear that these measures lie outside the core of the
criminal law. In particular, the keeping of a criminal record is, plausibly, is central to
our criminal law. Of course, retributivists can respond by arguing either that these
16
Another example of an exclusion requirement is a prohibition of those found guilty of a sex offence
involving a minor from living in the proximity of schools or day care centres.
17
Perhaps we think that a particular sentence was not sufficient given the severity of the crime, but
plausibly, in the eyes of the criminal law, the offender who has completed her sentence has been punished
sufficiently and is not allowed to be punished more for that crime.
16
measures are not part of the core or that retributivist theory can explain these measures.
In the absence of such an argument, however, a retributivist theory of criminal law is
unable to explain a central part of the criminal law.
3.3.2 Injunction and Corrective Justice
I mentioned above that punitive damages can be awarded to plaintiffs in a tort suit and
these damages, as the name suggests, are intended to punish the defendant who violated
the rights of the plaintiffs. But I claimed that this is consistent with the claim that an
award of compensatory damages is the default outcome in tort law. However, I argue
that this kind of response is under additional pressure when we take into account
another kind of consequences that can be faced by a tortfeasor, namely, injunction.
The court may issue an order requiring the defendant to perform or refrain from
performing some action. In tort law, injunctive remedies have been awarded to
plaintiffs in successful trespass and nuisance suits.
We can see that these injunctive remedies are difficult to be accommodated by the
principle of corrective justice. An injunction order requiring the defendants to refrain
from coming onto the plaintiff’s (real) property does not repair the breach of the property
right, but prevents future breaches. Similarly, an injunction order that restrains the
defendant from repeating the activity that caused the nuisance (say, emitting loud noises
or objectionable odours) does not repair the breach of the right to quiet enjoyment of
one’s land.
Indeed, the availability of injunctive remedies is one of the reasons provided in
support of the civil recourse theory of tort law, a rival of corrective justice theory. Civil
recourse theory explains tort law, not in terms of the second-order right of the plaintiff
to have the injury repaired, but in terms of the second-order right of the plaintiff to
pursue legal action. That is, when there is a breach of a first-order right, the plaintiff is
legally entitled to a legal avenue or recourse against the defendant. The second-order
right that the plaintiff has, according to the civil recourse theorist, is sufficiently broad
to explain the award of compensatory damages, punitive damages and injunctive
remedies. However, if the principle of corrective justice is not sufficiently complete
because it does not provide an account of first-order rights that are recognised by tort
law, then civil recourse theory also faces the same objection. This is because the civil
recourse theory does not specify an account of rights that are protected by tort law and
so does not provide an account of when a plaintiff has a right of action. Furthermore,
civil recourse theory, as it has been objected, does not constrain what sort of redress is
available in tort law and so faces an additional kind of incompleteness objection (along
with the one faced by both civil recourse theory and the corrective justice theory).
18
18
A civil recourse theorist may respond by providing principles which specify which kind of redress is
appropriate when. But some have taken this response, not as a way of rescuing the theory from the
objection, but as a way of paving the road to a different account of a corrective justice theory according
to which, the duty to ‘repair’ is understood more broadly as a duty to respond to the wrongdoing
appropriately. However, it is unclear how to understand injunctive remedies as a way of requiring the
17
In sum, both the principles of corrective and retributive justice cannot explain the
wide variety of legal consequences that are faced by those defendants who are found
guilty of crimes and those who are found liable for torts. Moreover, as we saw in Section
3.1, principles of corrective and retributive justice cannot explain the similarities and
the differences between the rights that are protected by the two domains because they
are both second-order principles. An answer to the Demarcation Question that appeals
solely to a second-order principle about what should happen when there has been a
breach of a first-order right cannot explain the similarities and the differences in the
first-order rights between tort law and criminal law. In the next section, I explain a
different answer to the Demarcation Question that focuses on the similarities and
differences in the first-order rights of tort and criminal law.
4 Public-Private Distinction
One major flaw of the Corrective-Retributive Answer is that it is unable to account for
the differences in the rights that are protected by the two domains. This suggests that
an answer to the Demarcation Question must pay adequate attention to this. One
thought that has been dominant is that crimes are public wrongs. After all, we can
divide domains of law into public and private law. Public law includes constitutional
law, administrative law
19
, and tax law as well as criminal law, and private law includes
contract law, property law, family law, and corporate law as well as tort law. If we think
that one thing that criminal law and tort law has in common is that both are laws of
wrongs (à la Holmes) and crimes are public wrongs, we have a candidate answer to the
Demarcation Question.
PUBLIC-PRIVATE
ANSWER:
The claim that crimes are public wrongs and
that torts are private wrongs can explain the
differences between the two domains of law.
If crimes are public wrongs, then we make intelligible the extent to which the public is
involved in addressing violations of criminal rights, rights that are protected by criminal
law. The state prosecutes the crime (on behalf of the public) and utilises public resources
(such as the work of state police departments, and prisons overseen by the
government
20
). Moreover, the claim that crimes are public wrongs could also perhaps
explain the punitive aspect of criminal law. An agent who has committed a public
wrong is liable to a public response, and in particular, a response by the state on behalf
defendant to discharge the duty to respond to the wrongdoing appropriately. Moreover, this new
corrective justice theory still faces the original incompleteness objection as it does not specify or constrain
what rights are protected by tort law.
19
which governs the activities of administrative or regulatory government agencies.
20
Indeed, that crimes are public wrongs may be used as a reason for arguing that prisons (which are
instruments of the criminal law) should be run and owned, not by private corporations, but by
government.
18
of the public. In contrast, a tortfeasor, who merely commits a private wrong, according
to the Public-Private Answer, is not liable to such a public response. Indeed, some have
thought that thinking of crimes as public wrongs is necessary to justify punishing those
who have committed crimes. Duff writes: “A justification of criminalization will need
to begin by specifying some value(s) that can be claimed to be public, as part of the
polity’s self-definition; show how the conduct in question violates that value or threatens
the goods that it protects; and argue that that violation or threat is such as to require or
demand public condemnation” (2007: 143). To make sense of punishing criminal
defendants which involves public condemnation, we must think of crimes as threats to
public values, so the argument goes.
Furthermore, the Public-Private Answer could provide a unifying story about
which rights are the criminal ones and which rights are tortious ones. Even though the
rights against murder, rape, theft, and fraud seem disparate, if we think that all of the
violations of criminal rights are public wrongs, then we could explain why these rights
are protected by criminal law.
21
Moreover, if rights against battery, defamation, and
trespass are private rights (and so that the violations are private wrongs), then we could
explain why these rights are protected by tort law. In the next subsection, I discuss the
sense in which a wrong is supposed to be public. There are different accounts of what
makes a wrong public. I ultimately argue that none of these accounts provides a good
account of what makes a wrong public that can be used in the Public-Private Answer
to the Demarcation Question.
4.1 What Makes a Wrong Public?
To get a handle on the Public-Private Answer, we must understand the sense in which
crimes are public wrongs whereas torts are private wrongs. There are clear cut cases of
conduct that are public wrongs because they wrong the public. So on this account of
publicness, a wrong is public if and only if it wrongs the public. Obvious examples
include treason and tax evasion. Plausibly, tax evasion harms the community and the
victim of that crime is the community who has been deprived of its resources. However,
there are crimes that do not harm the community directly. Homicide is a crime even
though the direct victim of that crime is an individual who has been killed. So in order
for the explanatory claim that crimes are public wrongs to be true, it must be the case
that crimes like murder and theft wrongs the public. That is, there must be some public
interest that is affected by the conduct. Indeed, Robert Nozick argues that crimes not
only harm the particular victims of those crimes, but harm the community by making
the members of that community feel fear by viewing themselves as potential victims of
crimes (1974: 65-71). That is, wrongs are public when they cause fear among the public.
21
We might also think that this distinction explains why there are victimless crimes.
19
However, this account of publicness is problematic. As have been noted, many
crimes do not generate fear among the public.
22
Grant Lamond provides, as examples
of crimes that do not generate fear, “crimes against public welfare, such as disclosing
classified information, or bribing public officials, or defrauding the revenue” (2007:
615). He writes: “These offences, while very serious, rarely generate any fear among
the public” (2007: 615). This is a good objection to the thesis that crimes are public
wrongs because crimes generate fear among the public.
Nevertheless we should note that the examples that Lamond proffers are crimes
that do harm the public. Disclosing classified information could threaten national
security, for instance. What we want is the commonality between these crimes that
harm or wrong the public and those crimes whose victims are individuals but where
the public is indirectly harmed or wronged. Duff and Marshall (1998) provide an
answer. According to them, crimes like murder are wrongs done to the public at large
(as well as wrongs done to particular individual victims). They wrong the public in
virtue of the fact that they attack values that are “central to a community’s identity and
self-understanding [and] to its concept of its members good” (1998: 20). To illustrate
how a crime could both wrong an individual but also threaten a public value, they
consider a racially motivated attack on an individual. This crime wrongs the victim of
the racially motivated attack, but it also wrongs the public by threatening the values of
justice and racial equality that the public endorses. Generalising from this particular
case, we can claim that crimes are wrongs done to the public or they are wrongs done
to individuals but are shared by the public because they threaten the values that the
community endorses.
23
However, it is not clear that this general claim about crimes is true. Take homicide
and theft, for example. What are the values that are shared by the community such that
homicide and theft are wrongs to the public? Plausibly, the community endorses the
importance of rights to bodily integrity and property. So, arguably, homicide and theft
threaten values that the community endorses and so are public wrongs. But this means
that any act that threatens these public values by violating the right to bodily integrity
or a property right is a public wrong and not a private wrong. However, these values
are also embodied in tort law. Wrongful death and trespass are torts and violations of
rights to bodily integrity and property, respectively. So simply pointing to a value that
the public endorses is not sufficient for distinguishing between crimes and torts.
Of course, a proponent of this understanding of publicness could respond by
claiming that it is a virtue of her account that there are some values that are embodied
by both criminal and tort law. After all, this can explain how the same act could
constitute both a crime and a tort. The same act could constitute a public and a private
22
See, for instance, Grant Lamond (2007: 615-616).
23
According to Duff and Marshall, crimes are shared by the community “insofar as the individual goods
which are attacked are goods in terms of which the community identifies and understands itself” (1998:
20).
20
wrong insofar as it threatens community values and it harms an individual. However,
this response leads to an impoverished understanding of what is wrong with a crime
like murder or rape. After all, what is heinous about brutal murders or rapes is that it
wrongs the particular victims. It is a mistake to think that wronging of the particular
individual makes it a private wrong and so that wrong falls within the scope of tort law,
and to regard only the threat to community values as falling within the scope of
criminal law. Moreover, there are differences between the tort of wrongful death and
the crime of murder. As we shall see in Chapter 3, the intention element that is required
for wrongful death and murder are different.
The same objection applies to a slightly different account of publicness of crimes
that is defended by Duff and Marshall (2010). There, they claim that a crime is “a kind
of wrong that properly concerns ‘the public’ – a wrong that is a matter of public interest
in the sense that it properly concerns all members of the polity by virtue simply of their
shared membership of the political community … it is a wrong in which we share as
fellow members of the political community to which wrongdoer and victim belong, and
a wrong to which we must therefore respond collectively” (2010: 71-72). They cite
domestic violence as an example of a public wrong because it is a wrong that “should
concern us all, and that should not be left to the couple to sort out for themselves as a
merely domestic affair” (2010: 72). It is true that domestic violence is a wrong that
should concern us all, but notice that the same occurrence of domestic violence can
constitute a tort of battery. Hence this account of publicness must provide which
wrongs concern the public but are also private wrongs and which wrongs concern the
public and are not private wrongs.
In addition, there is another objection to an account of publicness as what concerns
the public. This is because although the parties to a tort suit are individuals, tort law is
a legal, and therefore, a public institution. As Ambrose Lee puts it, “if private wrongs
are wrongs that do not properly concern the public, but only the plaintiff, then why do,
and should, we have torts against such wrongs?” (2015: 159). I take it that the objection
here is that if torts are private wrongs and wrongs that do not properly concern the
public, then we should not have a public institution that adjudicates these wrongs.
Lamond recognises these objections to the various versions of the thesis that crimes
are public wrongs in the sense that they are wrongs done to the public (by wronging the
public directly or by threatening values that are endorsed by the public or values with
which the public identifies). Accordingly, he rejects the claim that crimes are wrongs to
the public. Rather, he defends the thesis according to which crimes are “wrongs that
the community is responsible for punishing” (Lamond 2007: 621). This thesis is also
defended by Ambrose Lee who writes: “[P]ublic wrongs should not be understood
merely as wrongs that properly concern the public, but more specifically as which the
state, as the public, ought to punish” (2015: 156). On this view, crimes are those wrongs
that merit punishment by the state and this is distinctive of crimes and does not apply
to torts. Torts may be wrongs that should concern the public, but they do not merit
21
punishment by the state. This gives us a Public-Private Answer to the Demarcation
Question that “[focus] on the different responses that criminal law and tort law afford
to these wrongs, and conceive ‘public wrongs’ and ‘private wrongs’ in terms of these
responses” (Lee 2015: 162). What the differences are between the response of criminal
law and tort law are familiar to us. Criminal law responds to wrongs by punishing the
wrongdoers whereas tort law responds to them by requiring the wrongdoers to
compensate for the victims.
However, when publicness and privateness are understood in this way, we have a
different answer to the Demarcation Question: a Compensation-Punishment Answer.
This is different from the Corrective-Retributive Answer since the latter appeals to the
principles of corrective and retributive justice that tell us when compensation and
punishment are appropriate. That is, the Compensation-Punishment Answer is
entailed by the Corrective-Retributive Answer.
24
This means that the former faces all
of the objections raised against the latter.
5 Upshots and a Proposal
We have just seen that the best version of the Public-Private Answer faces all of the
obstacles faced by the Corrective-Retributive Answer. If you find the objections to both
of these answers compelling, then you may be inclined to think that there is no good
answer to the Demarcation Question. After all, the Demarcation Question presupposes
that there is a principled explanation of the similarities and differences between
criminal and tort law and you might want to reject such a presupposition. However,
one thing to notice about both the Corrective-Retributive Answer and the Public-
Private Answer is that they take seriously the structural differences between the two
domains and the fact that they have different responses to wrongdoing. We have also
seen that rights, such as the right to bodily integrity, are recognised and protected by
both domains. But we could pay attention to the substantive doctrines in tort and
criminal law and examine the elements of torts and crimes. That is, we can examine,
for instance, what exactly is required for a verdict that a defendant is liable for the tort
of battery and compare that with what is required for a verdict that a defendant is guilty
for the crime of assault. Of course, it is an important aspect of tort law that a tortfeasor
is required to pay compensatory damages to the plaintiff, but it is also a crucial aspect
of tort law that the tort of battery is committed when the defendant performs an act
that results in physical contact with the plaintiff, with the intention to make the contact,
and the plaintiff does not consent to the contact. I argue, in Chapter 3, that paying
attention to the elements of torts and crimes suggests an answer to the Demarcation
24
This is on the plausible assumption that one way to repair a loss for which one is responsible is to pay
compensatory damages.
22
Question that is an alternative to both the Corrective-Retributive Answer and the
Public-Private Answer.
One upshot of this approach to finding an answer to the Demarcation Question is
that it has a better chance of providing a principled way of distinguishing between
criminal law and tort law while paying attention to the ways in which tort and criminal
law have evolved. It is a contingent matter which particular elements are required for
various torts and crimes and the hope is that paying attention to these elements can
answer the Demarcation Question without presupposing an answer to the justificatory
question about whether tort law and criminal law should differ from each other in the
way that they do.
6 Structure of the Dissertation
In this section, I outline the aims of the next five chapters.
Chapter 2: Ripstein’s Substantive Approach
In the next chapter, I examine an answer to the Demarcation Question offered by
Arthur Ripstein. His answer is worth taking seriously because he uses the methodology
that I outlined in the above section. That is, he explores substantive doctrines of tort
and criminal law to answer the Demarcation Question. He argues that whether one
commits a crime or a tort depends on whether one makes a certain sort of choice.
According to Ripstein, committing a crime requires not merely violating (or risking the
violation of) a right that is protected by criminal law, but also choosing to violate, or at
least, choosing to risk violating, that right. However, I argue that this answer to the
Demarcation Question is incompatible with the fact that both criminal law and tort law
are objective: Boundaries of acceptable behaviour set by both domains appeal not only
to the mental states of the defendant but what is reasonable to expect. This objectivity
of the law is noted by Ripstein, but he fails to note the inconsistency between the claim
that crimes require choosing to violate a right and the claim that criminal law is
objective.
Chapter 3: Intention and Excuses
In this chapter, I provide a prima facie argument for the claim that tort and criminal law
implicate different accounts or standards of responsibility. In particular, I argue that
criminal law implicates blameworthiness or culpability and that tort law implicates an
account of responsibility that is weaker than blameworthiness. I examine the kinds of
mental states that are required by tort and criminal law. I argue that the mental states
that are required to satisfy the intention element in tort law are much weaker than the
mental states that are required to satisfy the intention element of criminal law. In
addition, I explore the differences in what counts as an excuse in the two domains. In
23
particular, I show that there are some conditions that count as excuses in criminal law
that do not count as excuses in tort law. I argue that this suggests that that the notion
of responsibility that criminal law implicates is stronger than the one that tort law has
in mind and that the kinds of excusing conditions allowed in the criminal law indicate
that criminal law implicates blameworthiness.
Chapter 4: Negligence: A Wrinkle?
In this chapter, I tackle one problem for one of the claims argued for in the previous
chapter, namely that criminal law implicates blameworthiness.
The thought is that the
existence of criminal negligence poses a problem for the claim that criminal law
implicates blameworthiness. I explore arguments for the claim that we are not
blameworthy for negligent conduct.
25
I argue that none of these arguments work. I also
discuss some positive accounts of blameworthiness that are put forward to show that
we are sometimes (non-derivatively) blameworthy for negligent conduct. I argue that
two accounts of blameworthiness (that of George Sher and Holly Smith) are not
adequate. Hence this chapter concludes on a negative note: We do not have a plausible
view of blameworthiness that can deliver the result that we are (non-derivatively)
blameworthy for negligence. This might spell trouble for the claim defended in this
chapter, namely that criminal law implicates blameworthiness if one can be guilty of a
crime for negligence and the tracing strategy is unavailable. However, I argue that this
may not be a problem for the general claim that criminal law implicates
blameworthiness since there are only two crimes which can be committed negligently
(homicide and harm that is caused by neglect of children). That is, given the marginal
role that negligence plays in criminal law, we can accept that criminal law typically, or
generally, or as a default implicates blameworthiness.
26
Chapter 5: Tort Law and Responsibility
The other claim that was argued for in Chapter 3 was that tort law implicates an
account of responsibility that is weaker than blameworthiness. This chapter is devoted
to identifying the account of responsibility (that is weaker than blameworthiness) that
would generate an obligation to pay compensatory damages and can be used to explain
tort liability. I also outline two desiderata for an adequate account of responsibility that
25
Actually, the claim that is the target of Chapter 3 is a qualified version of this claim. This is because
the tracing strategy has been successful in showing that we can sometimes be blameworthy for negligence.
The strategy involves tracing back to some decision or choice that caused the negligent conduct. If the
prior decision or choice is something for which the agent is blameworthy, then the agent is derivatively
blameworthy for the negligent conduct. Since, arguably, there are cases where tracing strategy is
unavailable, I examine arguments for the claim that we are never non-derivatively blameworthy for
negligent conduct.
26
One natural idea that comes out of this discussion is that whether or not criminal law should implicate
blameworthiness can guide us when thinking about justification of the criminal law. If we think that
criminal law should implicate blameworthiness, then perhaps this is a reason for thinking that we should
no longer criminalise negligence, or at least distinguish between tracing and non-tracing cases and
criminalise only the tracing ones.
24
is implicated in tort law. The first desideratum is that we are responsible for negligently
or inadvertently casing harm (while remaining neutral on whether or not we are
blameworthy for negligently or inadvertently causing harm).
The second desideratum is that we are responsible when we are liable for strict liability
torts. That is, the account of responsibility that is implicated in tort law must be such
that we are responsible when we cause harm even though we were not negligent. An
account of causal responsibility would satisfy these two desiderata, but in this chapter,
I seek the strongest account of responsibility that satisfies both desiderata.
I argue that Hart-inspired account of outcome-responsibility is the strongest
account of responsibility. I also examine Joseph Raz’s account of responsibility but
show that it cannot satisfy the second desideratum (although it is a novel way of
satisfying the first). However, as I shall argue, there is no account of responsibility that
can satisfy all three desiderata. This suggests that the distinction between criminal and
tort law cannot be explicated solely by an appeal to the distinction between
blameworthiness and responsibility. I end the chapter by providing some reasons for
thinking that an adequate solution to the demarcation problem must include an
account of rights that are protected by tort and criminal law. That is, one general lesson
we can draw from this chapter is that we need to supplement a responsibility-based
view with an account of rights that are protected by each domain. I develop such an
account in the final chapter.
Chapter 6: Responsibility and Rights: A Two-Pronged Account
In this final chapter, I argue that solely appealing to the distinction between
blameworthiness and outcome-responsibility cannot provide an adequate answer to the
Demarcation Question. This is because the notion of responsibility implicated in
criminal law was stronger than the notion implicated in tort law and if this was all there
was to the answer to the Demarcation Question, then all crimes should be torts.
However, it is not true that whenever one is criminally liable, then one is tortiously
liable. After all, attempting to injure or kill someone could constitute crimes even
though one cannot be liable in tort law for attempts, for instance. But the solution is
not to abandon a responsibility-based answer, but to supplement it with an account of
rights that are protected by each domain. I develop such an account in this final
chapter.
According to this two-pronged account, there are two different ways that tort law
and criminal law can be distinguished: (i) which rights are protected by each domain;
and (ii) what account of responsibility is required for someone to be held tortiously or
criminally liable. This two-pronged approach allows us to say that criminal law
recognises and enforces a duty not to attempt to harm others, for instance, whereas tort
law does not recognise or enforce such a duty. Moreover, there are rights and duties
that are recognised by tort law, but not by criminal law. Tort law recognises a right not
to have emotional distress inflicted on one, but criminal law does not recognise such a
25
right (even though we can be blameworthy for inflicting emotional distress). I believe
that there is no overarching principle that can explain and justify the different account
of rights and duties recognised by the two domains. I take it that one of the jobs of legal
philosophers (along with other scholars) is to determine, for each individual right that
is protected by one domain, what its status should be in the other domain.
However, the two accounts of responsibility that are in play in the two domains
can help to explain certain crucial differences between tort and criminal law. Moreover,
the two-pronged account of rights and responsibility can help shed some light on why,
on the one hand, some people are tempted to think that there is a principled way of
solving the Demarcation Problem, even though many are sceptical of such a solution
given the seemingly haphazard evolution of the law.
26
Chapter 2: Ripstein’s Answer
Violating Rights and Choosing to Violate Rights
1 Introduction
In Responsibility, Equality, and the Law, Arthur Ripstein argues that whether one commits
a crime or a tort depends on whether one makes a certain sort of choice (1999).
27
According to Ripstein, both tort law and criminal law specify the boundaries of
acceptable behaviour explicated in terms of the rights that people have. The difference
between the two lies in the fact that criminal law deals with a smaller class of behaviour
that crosses these boundaries, namely, behaviour that exhibits the choice to deny the
boundaries. In particular, he claims that violating criminal law involves choosing to
violate or risk violating rights (22).
Here, he is referring not to all rights, but those that
are protected by criminal and tort law. In contrast, a tort, for Ripstein, can be
committed without the defendant choosing to violate or risk violating rights, even
though committing a tort always involves taking a risk that someone’s rights will be
violated where ‘taking’ does not require ‘choosing’ the risk.
Ripstein’s answer to the Demarcation Question differs from the two most
prominent answers to it that were examined in the previous chapter. We have seen the
problems faced by both the Corrective-Retributive Answer and the Public-Private
Answer. They are unable to explain the fact that on item of conduct can constitute both
a tort and a crime and what the differences are between, say, the tort of wrongful death
and homicide, or the tort of conversion and theft. As indicated in the last chapter, I
think that progress can be made by examining these differences. In the next chapter, I
focus on concepts that appear as elements in both torts and crimes and examine their
differences to see if they can provide some principled way of explicating the distinction
between tort law and criminal law.
In this chapter, however, I focus on Ripstein’s answer partly because his account
is informed by the elements of torts and crimes. Moreover, his answer is sufficiently
plausible to merit close examination. It appeals to the claim that to commit a crime,
one must choose to violate a right (or choose to risk violating a right) and this can
explain a general doctrine of criminal law is that there is no culpability without a guilty
mind.
28
Arguably, one’s ‘mind’ is guilty when one chooses to violate a right. Also, this,
in turn, can explain why different consequences attach to criminals and tortfeasors.
However, as I shall show, the claim that choice is required for criminal law (but
not for tort law) is incompatible with another claim made by Ripstein: that both
criminal law and tort law are objective. I show, by providing examples, how the agent
27
All parenthetical page references to Ripstein are to this book.
28
This is usually expressed in the Latin phrase “actus non facit reum nisi mens sit rea” which translates to “the
act is not culpable unless the mind is guilty”.
27
whom Ripstein takes to have committed a crime need not choose to commit acts that
exceed the boundaries of acceptable behaviour set by criminal law (or, equivalently for
Ripstein, need not choose to violate the rights of others). This exploration of the tension
reveals challenges faced by any account of criminal law that tries to both (i) make
intelligible certain doctrines of criminal law that use the ‘reasonableness’ or objective
standard; and (ii) make intelligible why punishment is an appropriate response to those
found guilty of crimes.
Moreover, since Ripstein is correct to assert that criminal law and tort law
sometimes use the objective standard, as a matter of fact, the arguments of this chapter
can establish that any attempt to distinguish between tort law and criminal law by
appealing to the choice of the defendants in the two domains of law faces the objections
I raise.
2 What the Defendant Chooses in Tort and Criminal Law
As noted, committing a crime, for Ripstein, requires not merely violating (or risking the
violation of) a right that is protected by criminal law, but also choosing to violate, or at
least, choosing to risk violating, that right. Of this choice, which is meant to distinguish
criminal from tort law, Ripstein writes: “The requisite notion of choice is explicated in
terms of the distinction between the rational and the reasonable” (134).
29
The rational
person “does what seems best from her situation given her ends” whereas the
reasonable person “takes appropriate regard for the interests of others” (7). And to find
out what counts as the relevant interests (and hence what counts as rights protected by
tort and criminal law) requires specifying the “fair terms of interaction”, that is, the
basic terms of interaction in a just society (9). Accordingly, the standard of private
rationality “justifies outcomes by their beneficial consequences for the decision maker”
(59) whereas the public standard of reasonableness sets the fair terms of interaction and
hence the rights that are protected by law.
What, then, is the relationship between the choice that one makes when
committing a crime, on the one hand, and the contrast between (public) reasonableness
and (private) rationality, on the other? Ripstein writes: “Criminal acts are those acts in
which one person seeks to substitute private rationality for public standards of
reasonableness” (134). To get a better sense of what counts as the kind of substitution
that Ripstein has in mind, consider the following case:
29
There is some affinity between Ripstein’s view of the criminal law and that developed by Larry
Alexander and Kim Ferzan (2009a). For them, what matters for an account of culpability in the criminal
law are the reasons that an agent has for acting in a certain way. There is some similarity between one’s
reasons, as this notion is developed by them, and the kind of choice Ripstein invokes. In particular and
as we will see, for Ripstein, choice is connected to the idea of substitution of private rationality for public
standards of reasonableness. This apparent affinity between Ripstein and Alexander and Ferzan is
interesting, but will not be pursued here.
28
Iris is deciding whether to kill her cousin, Jacob. If Jacob dies and she
is not caught, then she will receive a larger inheritance from their
deceased grandparents. She deliberates about what would be best for
her and she decides not to kill Jacob.
It seems that she could be described as substituting private rationality for public
standards of reasonableness. However, she does not commit a crime. So perhaps we
should interpret Ripstein as claiming that substituting private rationality for public
standards of reasonableness is a necessary condition, but not a sufficient one, for an act
to be criminal.
30
However, there is an alternative interpretation of Ripstein available. Substituting
private rationality for public standards of reasonableness is sufficient, but perhaps
Ripstein could claim that, in this case, there is no substitution since Iris does not take a
course of action that is different from the one mandated by the public standards of
reasonableness.
31
This move is available to Ripstein, since it is actions, according to
him, that count as substitutions.
32
Or perhaps, he could respond by claiming that the
kind of substitution he has in mind is substitution that is accompanied by a risk
imposition and that Iris does not impose any risks. However, this requires a further
clarification about what exactly is entitled by risk imposition. It is true that given that
Iris decides not to kill Jacob, Jacob will not be killed. Hence, Iris does not risk violating
Jacob’s right to life. But suppose Iris decides to kill Jacob and attempts to do so, but
Jacob will not be killed because there is a major flaw in Iris’s plan. Iris does substitute
private rationality for public standards of reasonableness, but is this accompanied by
risk imposition? There may be an account of risk imposition that delivers the correct
results in these two cases and others, but it suffices to note that Ripstein’s notion of
substitution requires further unpacking.
The examples discussed in this section raised questions about whether substitution
of private rationality for public standards of reasonableness is necessary or sufficient for
choosing to violate or risk violating a right, and what counts as substituting private
rationality for public standards of reasonableness. It is not entirely clear what the
answers to these questions are. But, thankfully, my objections to his view as an account
30
Here is another example that supports this interpretation. Suppose I publish an offensive book that
violates the public standards of reasonableness, solely for my benefit. There is a sense in which I substitute
private rationality for public standards of reasonableness, but I do not commit a crime.
31
Regarding the example in footnote 4, Ripstein could say that that there is no substitution because
publishing an offensive book does not, in fact, violate public standards of reasonableness (or the fair terms
of interaction). Hence, if we have the correct understanding of what counts as a public standard of
reasonableness, then this example is not a counterexample to substitution being a sufficient condition for
an act to be a crime.
32
He writes: “Intent or recklessness is an essential element of core areas of criminality because a person
must be aware that the rights of others are in jeopardy if his action is to count as such a substitution”
(134; my italics). Of course, whether or not the claim that such awareness is necessary for the kind of
substitution Ripstein has in mind is plausible depends on, inter alia, what counts as private rationality
and there being a public standard of reasonableness.
29
of the distinction between tort law and criminal law do not rely on answering these
questions (though Ripstein’s answers to these questions might be illuminating).
For an application of Ripstein’s account to criminal law, consider an example: I
take the book that is on the table. I am aware that if I take it, I will violate your right
because I know that the book belongs to you and that you do not consent to my taking
it. But I take the book because it is much more convenient for me to do so than to check
out the copy of that book from the library. Here, I commit a crime and Ripstein’s
account can explain this: I choose to take the book, knowing that I am violating your
right that is protected by the criminal law. Hence, I can be described as choosing to
violate your right. Equivalently, according to Ripstein, I can be properly described as
substituting my private rationality for public standards of reasonableness. I certainly
meet Ripstein’s stated requirement for such a substitution: I am aware that I am putting
your rights in jeopardy.
My awareness that the book belongs to you is crucial in this case because it is this
awareness that justifies describing me as choosing to violate your right. Suppose that this
awareness is missing: I am under the misapprehension that the copy of the book on the
table is the one that I just bought. In this case, I do not commit a crime and, again,
Ripstein can explain this, as I do not choose to violate your rights. (Equivalently, I do
not seek to substitute my private rationality for public standards of reasonableness.)
However, even though I do not commit a crime (in this second example),
Ripstein’s account correctly takes me to be liable for the tort of conversion. By taking
your book, I do, in fact, violate your right, even though I do not choose to violate the
right.
33
This is in line with conversion since the rule for liability for conversion does not
require that I intend to take possession of something that I know does not belong to me.
Indeed, it is sufficient that I intend to take the book from the table and that the book
turns out to be a property of yours and I, in fact, do not have permission to do so.
34
Given that liability for conversion does not require that I am aware that the property
that I take does not belong to me, it is understandable why Ripstein thinks that while
the choice to violate the rights of others is required for there to be a crime, such a choice
is not required for there to be a tort.
Ripstein’s account of the contrast between criminal law and tort law can be
sharpened by thinking about the tort of negligence
35
, which, for Ripstein, is the
33
Some might want to resist the claim that I violate your right when I commit the tort of conversion and
instead claim that I infringe your right. What is important is that your right is infringed by my action
and so I choose to perform an action that constitutes an infringement of your right (although I do not
choose to infringe your right).
34
Note that it is important that, even for liability in tort, the taking of the book is an act that can be
attributed to me. If someone pushes my arm to swipe the book in to my bag, then plausibly, it is not I
who violated your right, but the person who forcibly moved my arm who violated your right. I come
back to this point in Section 4.
35
to be distinguished from strict liability torts like conversion as well as intentional torts. (Indeed,
although it is common to regard conversion as a strict liability tort, the intention element of conversion
is similar to the intention element of intentional torts, such as battery. To commit battery, the defendant
must intend to perform to make physical contact with the plaintiff (even though she need not think that
30
“primary basis of tort liability in Anglo-American legal systems” (48).
36
It is not difficult
to see why he thinks that the choice to take the risk is present in criminal law, but not
in the tort of negligence. Suppose that I am working at a warehouse and it is my job to
store all the cardboard boxes that come to the warehouse. What I should do is either
secure the boxes so that the chance of them tumbling down is low or flatten the boxes
and store them that way. However, I do not care about doing a good job and so I do
not give much thought to the method that I should use before I start. I idly stack the
boxes on top of each other without bothering to secure them and without realising that
since the boxes are not properly secured, they could tumble down. Suppose that the
boxes fall and they damage some fragile items nearby. It is true that I caused the
damage to the fragile items and that I intentionally engaged in an activity that risked
damage to them. Here, unlike the case where I choose to take your book, knowing that
it belongs to you, I did not choose to damage the item or even choose to risk damaging
the item. I merely performed an action that was risky and hence, one that I should not
have performed. This makes me liable for the tort of negligence and liable for the
damages thus caused. But there is no crime here.
Ripstein’s notion of choice is meant to distinguish between tort law and criminal
law. Moreover, Ripstein argues that the fact that committing a crime entails that you
choose to violate or risk violating the rights of others and the fact that committing a tort
does not entail such a choice can make intelligible the different legal consequences that
are faced by tortfeasors and criminals. In tort law, according to Ripstein, “[t]he
question is not whether I am being careful by the standards of what I am doing, but
whether I am being appropriately careful in light of my neighbor’s interests in security
and mine in liberty” (58). That is, tort law strikes the balance between everyone’s
interests in security and liberty and determines what rights we all have. If I, by engaging
in some activity, violate your right, then I am responsible for that violation. “The idea
that people should bear the costs of their choices requires that the defendant bear the
costs” (58). Hence, Ripstein thinks that compensation (usually in the form of monetary
damages) is a legitimate response to tortfeasors.
In contrast, he argues that the appropriate response to a crime is punishment that
involves “the intentional infliction of hard treatment” (140). This is because, according
to Ripstein, only hard treatment “negates” the choice to violate the rights of others (and
accordingly, the substitution of private rationality for public standards of
reasonableness). When someone commits a crime, she chooses to violate the rights of
others—she is acting to her own advantage, disregarding the rights of others (which are
set by the public standards of reasonableness).
37
Ripstein argues that the appropriate
the contact is consensual). Similarly, to commit conversion, the defendant must intend to take possession
of the property (even though she need not think that she lacks permission to take it).
36
In contrast, he thinks that negligence, as a basis of culpability in criminal law, is an anomaly.
37
It was suggested to me that perhaps Ripstein ought to have said “choose to violate another’s right, or
choose to behave as if the person has forfeited or waived the right in question”. However, this additional
clause is problematic. If the defendants think that the other has forfeited or waived the right, then they
31
response to the criminal wrongdoer must be “addressed to the private perspective from
which the wrongdoer acted” and must “seek to cancel the crime by canceling its
apparent advantage from the point of view from which the criminal acted” (140).
Ripstein’s account of punishment that “combines retribution and deterrence” is a
complex account and a thorough examination of that account will not be undertaken
here. However, I think that it is important to bear in mind that Ripstein’s notion of
choice is meant not only to distinguish between the two domains of law, but also make
intelligible the different legal sanctions that are imposed.
3 Objectivity of Criminal Law
As I outlined in the last section, it seemed that when Ripstein first introduces the idea
that criminals seek to substitute their private rationality for public standards of
reasonableness, such substitution requires them to be aware of the rights of others that
are put into jeopardy by their actions. This would explain why intention and
recklessness are core mens rea elements of crimes.
38
However, in this section, I turn to
Ripstein’s claim that criminal law is objective. There have been various understandings
of the objectivism of tort law and criminal law, at least from the time of Oliver Wendell
Holmes.
39
But the version of objectivism that is of interest to us states that the standard
of reasonableness that delineates the class of acceptable behaviour from criminal law’s
point of view departs from the standard that seems reasonable from the point of view
of the defendant (given the information available to her). Hence, “what a person
thought he was doing is often not sufficient to exculpate him” (135). This section is
devoted to explaining Ripstein’s understanding of objectivism. In the next section, I
argue that this objectivism is in tension with Ripstein’s claim that committing a crime
requires the defendant to choose to violate or risk violating a right.
According to the version of subjectivism that Ripstein argues against, committing
a crime requires that the agent acted voluntarily where “[a]n act is only voluntary with
respect to some circumstance or consequence if that circumstance or consequence
came before the agent’s mind” (174). Ripstein argues against subjectivism by discussing
mistakes. This is because he thinks that his objectivism can explain why the law makes
a distinction between mistakes of fact and mistakes of law. More precisely, he shows
how appealing to reasonableness can explain why punishment may be mitigated or
could not have thought that they were violating the right since it is forfeited or waived. But the choice to
violate a right is supposed to be the distinguishing feature of a criminal act. So this additional disjunct
seems not to fit with Ripstein’s account.
38
I argue in Section 5 that there is a plausible understanding of recklessness that is not compatible with
Ripstein’s notion of choice (which involves awareness that the rights of others are in jeopardy).
39
Holmes sometimes spoke of objectivism interchangeably with externalism (though not always). He
writes: “The tests of liability are external, and independent of the degree of evil in the particular person’s
motives or intentions” (1881: 50).
32
even excluded when one makes a mistake of fact, but not when one makes a mistake of
law. Ripstein’s point is that a subjectivist understanding of criminal law cannot make
sense of this distinction. Ripstein, echoing Holmes, objects that given this
understanding of voluntary action, subjectivists must treat both mistakes of fact and
mistakes of law as exculpating. If I take the book on the table without realising that the
book belongs to you (a mistake of fact), then my act is involuntary with respect to theft.
Similarly, “[i]f I act without realizing that murder or theft is wrong, give no thought to
the moral or legal status of my act, believe myself to be exempted from some law, my
act is also involuntary with respect to that wrong” (179-180).
Of course, the subjectivist might think that she can point to some independent
reason why subjectivism does not entail that both mistakes of fact and mistakes of law
exculpate. For all that has been said here, there may be some distinction that can be
drawn between fact and law of which the subjectivist can make use. Alternatively, she
could agree with Ripstein that subjectivism entails that both mistakes of fact and
mistakes of law exculpate according to subjectivism, but argue that this is a virtue, not
a cost. Another response is to say that the kind of pure subjectivism that Ripstein has
in mind is a strawman, or at least, that there is a version of subjectivism that is more
plausible. Perhaps a subjectivist is someone who thinks that the standards of
reasonableness are subjective in the sense that they are relative to the defendant’s
knowledge base. According to this broader subjectivism, the fact that one should have
known could be a fact that is relative to one’s knowledge base. After all, it is a fact about
what is actually available to one, not a fact about the abstract reasonable person.
40
But
since the kind of subjectivism Ripstein has in mind says that the standards of
reasonableness are relative to the defendant’s occurrent beliefs (that is, the beliefs that the
defendant has in mind), I set aside this broader (perhaps, more plausible) version of
subjectivism.
It turns out that Ripstein, in arguing against subjectivism, makes claims about
elements of various crimes that are incompatible with his contention that the criminal
chooses to violate the rights of others. To see this, consider the two doctrines in criminal
law that are concerned with mistakes: self-defence and consent (both of which are
discussed by Ripstein).
Consider this (hypothetical) self-defence case where mistakes are salient:
Caleb is walking home when a stranger, Elliot, knocks Caleb to the
ground and threatens to kill him. Elliot walks away. Caleb falsely
believes that Elliot has a gun and he falsely believes that Elliot is about
to shoot him. So Caleb pulls out his gun and shoots Elliot.
A (purely) subjectivist view would predict that since Caleb genuinely believed that his
life was threatened by Elliot, the plea of self-defence should be successful. If we take the
world as Caleb believed it to be, then he would be justified in killing Elliot and hence,
40
The kind of subjectivism I have in mind has a certain affinity to Derek Parfit’s notion of evidence-
relative wrong (in contrast to belief-relative wrong (2011).
33
would not be guilty of homicide. In contrast, objectivism states that not all genuine
beliefs count: the relevant genuine belief must be reasonable, in the relevant legal sense.
Hence, we must find out whether Caleb’s beliefs were reasonable in that sense.
As a matter of law, in order to plead self-defence successfully, it is not sufficient for
Caleb to prove (by the relevant evidentiary standard) that he genuinely believed his life
to be in danger. If Elliot dies as a result of being shot by Caleb, Caleb may be found
guilty of homicide if it was unreasonable for Caleb to believe that his life was threatened
by Elliot. That is, his belief that the danger to his life was imminent must be reasonable
for a successful plea of self-defence.
41
Hence, Ripstein claims, objectivism can vindicate
(while subjectivism cannot) this legal doctrine entrenched in most jurisdictions (if not
all).
42
Turning from self-defence to consent, Ripstein argues that objectivism can also
vindicate the doctrine that a defendant may be found guilty of rape even when he
genuinely, but unreasonably, believed that the victim consented to having sex. He
discusses the (rather disturbing) case of DPP v. Morgan [1976]. This is how Ripstein
describes the case:
An R.A.F. officer invited a number of men under his command home
to have sexual intercourse with his wife. Morgan told them that his wife
would heighten her own pleasure by pretending to resist. When she
resisted, they proceeded anyway, and at their trial claimed that they
believed that she had consented. (176)
The issue in question is whether or not the mens rea element of rape is intent to have
nonconsensual sexual intercourse, to be contrasted with intent to have sexual
intercourse (where it is unreasonable to think that the sexual intercourse is consensual).
If the former intention is required for rape, then belief that the victim consented seems
to negate that intention (although the defendant intended to have sex which, as a matter
of fact, was nonconsensual).
43
Ripstein argues that this should not be the law. This is
because, as he sees it, the law should protect people equally and requiring an intention
to have nonconsensual sex which can be negated by a genuine, but unreasonable belief
that the victim consented, favours the defendant.
Of course, as Ripstein claims, from a subjectivist point of view it makes sense why
41
Self-defence also requires that the defendant reasonably believe that the harm he inflicts on the victim
is proportionate to the amount of harm that the victim would have inflicted on him (requirement of
proportionality) and that the defendant reasonably believe that no morally acceptable alternative to using
this much force against the victim would have adverted the attack (requirement of necessity).
42
It is true that subjectivism gets the law wrong as it stands. But it is an open question whether we should
have this reasonableness requirement. I do not take up this issue here.
43
Actually, this is a controversial issue in the philosophy of action and philosophy of rationality. Does
intending to φ entail believing that you will φ? The view that answers this question in the affirmative is
called “Cognitivism”. (See, for instance, Gilbert Harman (1997) who defends this view.) However, many
reject Cognitivism. (See, for instance, Michael Bratman (2009).) I assume, for the purposes of this
chapter, that when the defendant believes the victim to be consenting, he does not have the intention to
have nonconsensual sex. But I want to note that it is not obvious that the intention to have nonconsensual
sex is always negated by lacking the belief that victim does not consent.
34
genuine, albeit unreasonable, belief negates the intent element of rape. According to
subjectivists, having a guilty mind is a required mens rea element of a crime and a
defendant who makes this kind of mistake lacks a guilty mind. Arguably, the defendants
in Morgan who had sex with Morgan’s wife should have realised that she was not
consenting. But given that they believed that she was consenting, they were not doing
anything that they thought was wrong.
However, as a matter of law, a defendant who thought that he was having
consensual sex and genuinely believed that the victim consented can still be convicted
of rape as long as his belief that the victim consented is an unreasonable one.
Accordingly, not all mistakes negate the mens rea element of the relevant crimes.
44
Indeed, this is how he distinguishes mistakes of fact from mistakes of law. Mistakes of
fact that are unreasonable are, for Ripstein, mistakes of law—you are making a mistake
about what the rights of others are and mistakes about what rights people have are
mistakes about the law. (The defendants in Morgan who had sex with Morgan’s wife, for
instance, make a mistake about the kinds of things that count as consenting to sexual
intercourse).
45
This is an idiosyncratic understanding of the distinction between
mistakes of fact and mistakes of law. But, for now, it suffices to note that objectivism
can explain why it is the case (as a matter of law currently) that the defendants in Morgan
are guilty of rape even though they genuinely believed that they had the victim’s
consent. After all, the relevant legal doctrine is that a defendant who thought that he
was having consensual sex and genuinely believed that the victim consented can still be
convicted of rape as long as his belief that the victim consented is an unreasonable one.
4 Tension Between Choice and Objectivity
So how does the discussion in the preceding section show that there is a tension in
Ripstein’s account overall? Recall that, for Ripstein, a criminal can be distinguished
from a tortfeasor by the fact that only the former is aware that her action puts a right
of another in jeopardy, but nonetheless chooses to perform that action. However, is it
accurate to describe Caleb in my hypothetical self-defence case and the defendants in
Morgan as choosing to violate the rights of others? After all, for this to be the case, those
44
It is slightly difficult to identify what the mens rea element of rape is, given this objectivism. Intention to
have nonconsensual sex is too narrow because the defendant who unreasonably believes that the victim
consents does not intend to have nonconsensual sex. Intention to have sex which turns out to be
nonconsensual is too broad because the defendant who reasonably, but falsely, believes that the victim
consents has this intention, but does not commit rape. I think the most promising way of encapsulating
the mens rea element is to explain it in terms of the rights of the victim. She has a right not to be subject
to nonconsensual sex that a reasonable person would believe is nonconsensual. It seems to follow from
this that we do not have a right to be free from nonconsensual sex simpliciter, at least, not one that is
protected by the laws against the crime of rape, as a matter of contingent fact.
45
Similarly, according to Ripstein, ignorance that the book belongs to another is a kind of mistake of
law, if such ignorance is unreasonable.
35
defendants would have to be aware that the rights of others are put in jeopardy by their
intended actions.
Take the defendants in Morgan. They genuinely (although unreasonably) believed
that Morgan’s wife consented to have sex. If this is true, then plausibly, they did not
think that they were in danger of violating her right to bodily integrity. Hence, we
cannot assert, even on Ripstein’s own terms, that they were choosing to violate the
victim’s right (or substituting private rationality for public standards of reasonableness)
even though each of the defendants is guilty of committing a crime.
Perhaps one could respond on behalf of Ripstein by claiming that even though
they did not choose to violate the victim’s right, they chose to risk violating her rights
because they chose to act in a way that did not take sufficient regard of her rights. Recall
Ripstein’s claim that even if the defendants in Morgan genuinely believed that the victim
consented, their belief was unreasonable because it is unreasonable to treat the victim’s
husband saying that she consents as an indication that she consented.
46
If this is right,
then the defendants made a mistake about the status of the victim’s right to bodily
integrity (and, more precisely, under what conditions she consents to physical contact)
and so they made a mistake about the law.
47
This is significant, the rejoinder could
continue, because the fact that the defendants in Morgan make a mistake of law, not a
mistake of fact, is relevant to whether or not they substitute their own private rationality
for public standards of reasonableness and hence, relevant to whether or not they chose
to violate (or risk violating) a right as Ripstein understands choice.
48
Perhaps when you
make a mistake of law, you necessarily substitute your own private rationality for public
standards of reasonableness. After all, what rights one has is a public matter and so
when the defendants in Morgan (wrongly and unreasonably) act on their belief that the
husband’s say-so is sufficient evidence of the victim’s consent, they are substituting their
private rationality for public standards of reasonableness. And since choosing to violate
or risk violating a right, according to Ripstein, is equivalent to substituting private
46
Actually, this is how Ripstein chooses to interpret the case. However, in fact, the court found that the
defendants did not genuinely believe that the victim consented. They reasoned as follows: since the belief
that the victim consented was so manifestly unreasonable, the defendants could not have genuinely
believed that the victim consented. This treats reasonableness as an evidentiary indicator of what the
defendants genuinely believed. Of course, Ripstein wants the reasonableness to delineate the boundaries
of acceptable behaviour rather than merely being used as evidence. In discussing this case, I follow
Ripstein and grant that the defendants genuinely, but unreasonably, believed that the victim consented.
47
Ripstein, in personal communication, claimed that Morgan reveals that the very thing that makes an
unreasonable mistake about consent unreasonable is that it is a mistake about what counts as consent.
Since what counts as consent is a legal and therefore a public issue for Ripstein, a mistake about what
counts as consent is regarded by him as a mistake of law, not a mistake of fact.
48
Ripstein, again in personal communication, acknowledged that there is a tension between a focus on
choice in the characterisation of the criminal defendant and the role of objective standards in what counts
as a criminal act, but added that his view has the resources to deal with the tension, namely, his
(idiosyncratic) way of understanding what counts as a mistake of law.
36
rationality for public standards of reasonableness, the defendants in Morgan choose to
violate or risk violating a right and hence, do not pose a problem for Ripstein’s view.
49
I think this rejoinder faces two objections. First, even if the case can be made that
the unreasonable mistake that the defendants in Morgan make can be construed as a
mistake of law, it is not at all clear that all unreasonable mistakes can be so construed.
Recall the hypothetical self-defence case: Elliot, the aggressor, is walking away and is
no longer threatening Caleb. Let us grant that it is unreasonable for Caleb to think that
Elliot has a gun and is about to shoot him. As we said of the defendants in Morgan, we
could say that even though Caleb does not choose to violate Elliot’s rights, he chooses to
risk violating Elliot’s rights.
50
After all, he chooses to defend himself, thereby risking
getting it wrong, and given the particular circumstances of the case, Caleb acts in a way
that assumes that Elliot does not have the right to life or, more plausibly, that Elliot has
forfeited the right to life.
When discussing the defendants in Morgan, we said that their choice to risk
violating the right of another revealed that they made a mistake about the law (by
making a mistake about the status of the victim’s rights). However, it is not obvious that
Caleb made a mistake about the law by making a mistake about Elliot’s rights. It seems
that Caleb simply made a factual error about what Elliot was doing. It is a stretch to
think that Caleb was mistaken about Elliot’s rights (even though he acted in a way that
violated Elliot’s rights). Caleb genuinely (although unreasonably) believed that his life
was under imminent threat and that the only way to survive was to kill the attacker,
Elliot. Plausibly, Caleb did not think that he was in danger of violating Elliot’s right to
life—after all, in Caleb’s mind, Elliot was the aggressor and he himself was the victim.
To clarify the objection, contrast this self-defence case with the Morgan case. If we
look at the world as the defendants believed it to be, we see that (i) the victim consented
to sexual intercourse with the defendants; and (ii) the victim’s husband saying that she
consents is sufficient evidence of consent. Of course, both claims are false, but only (ii)
is a claim about the status of the right of the victim. Hence, the defendants in Morgan
make a mistake of law, as Ripstein urges. But let us now look at the world as Caleb
believed it to be: (a) Elliot pulled out his gun and threatened Caleb’s life; and (b) the
only way to neutralise this threat is to shoot Elliot. Again, both claims are false, but
neither (a) nor (b) are claims about the status of the right of the victim. Hence, Caleb
does not make a mistake of law.
49
Another way of putting this idea is in terms of respect. By thinking that the husband’s say-so is sufficient
evidence of consent by the wife, the defendants in Morgan do not show sufficient respect for the wife as
a bearer of rights. Perhaps this idea of respecting another as a bearer of rights is what is important for
Ripstein and choosing to violate a right is simply a paradigmatic case of this kind of disrespect. If this is
right, then perhaps this kind of respect would be something that is required for committing a crime,
though it would not be required for committing a tort (even though committing a tort can manifest such
disrespect).
50
This point was raised by Jonathan Quong. See his (2012) for his account of the conditions for liability
to defensive harm.
37
Another way of pressing this objection further is to contrast the case of Caleb with
a kind of case where the defendant (call her Katie) attempts to use force to defend
herself, but the use of force is justified (because the belief that she is under imminent
threat from the victim is reasonable). We said that Caleb acts in a way that assumes
that the victim has forfeited the right to life and because he uses force on this
assumption, he risks getting it wrong. This allowed us to pursue the possibility that
Caleb chose to risk violating Elliot’s right and hence, made a mistake about the law.
However, Katie also acts in a way that assumes that the victim has forfeited the right
to life and she also risks getting it wrong. Of course, there is a significant difference
between the two cases: Caleb, in unreasonably judging Elliot to have forfeited the right
to life, ought not have acted in the way he did whereas Katie’s belief is not unreasonable
and hence she was justified in using force. But the rejoinder that I am considering on
behalf of Ripstein is that Caleb made a mistake of law because he chose to risk violating
a right (since he was risking getting the facts wrong about whether or not the victim has
indeed forfeited the right to life). I have just shown that anyone who attempts to use
force to defend herself risks getting it wrong and hence chooses to risk violating a right.
Hence, risking getting it wrong cannot be used to distinguish between mistakes of law
and mistakes of fact. That is, there is pressure on someone who wants to pursue this
rejoinder to come up with a reason for thinking that Caleb is making a mistake of law
that does not also apply to those who justifiably use force against aggressors.
51
Furthermore, even if one could argue that Caleb did make a mistake of law and
that all unreasonable mistakes are mistakes of law in the Ripsteinian way (and are
mistakes about someone’s rights), the rejoinder faces another, perhaps more pressing,
objection. The rejoinder relies on the claim that one always substitutes one’s own
private rationality for public standards of reasonableness when one makes a mistake of
law. However, this is in tension with the characterisation of substitution given by
Ripstein. Recall that there is a tight connection between the kind of substitution in
which Ripstein is interested and the awareness that the rights of others are in jeopardy.
52
This awareness was what seemed essential to an agent choosing to violate a right rather
than merely choosing to perform an act that happened to violate the right. However,
Caleb is not aware that he is doing anything unreasonable, let alone that he is making an
unreasonable mistake about the status of Elliot’s right to life.
53
51
In footnote 23, I suggested that we could construe the kind of rejoinder I have in mind in terms of
disrespect of an agent as a bearer of rights. The objection that I have been discussing in the main text
can also be raised against this gloss. After all, it is not clear that the kind of disrespect I described is
required for all crimes. Just as Caleb does not seem to be making a mistake of law, he does not seem to
be disrespecting Elliot as a bearer of rights. Hence, it does not seem that disrespect can do the work of
demarcating between criminal law and tort law.
52
Recall the previously quoted claim that “a person must be aware that the rights of others are in
jeopardy if his action is to count as such a substitution” (134).
53
We might even question whether Caleb has any beliefs about Elliot’s right to life. At any rate, plausibly,
he lacks any occurrent belief that he is risking a rights violation.
38
Hence, the response according to which Caleb chooses to risk violating Elliot’s
right is trapped in a dilemma. If, on the one hand, we hold fixed the idea that choosing
to violate a right and seeking to substitute private rationality for public standards of
reasonableness require awareness that a right is in jeopardy, then Caleb does not choose
to violate a right or seek such a substitution. But if we weaken the notion of choice and
substitution and do not require that they involve awareness that another’s right is in
jeopardy (in direct contradiction to what Ripstein says), it is not clear how these weaker
notions of choice and substitution can be used to distinguish between tort law and
criminal law. For this to work, we have to argue that these weaker notions are necessary
for every criminal act and that they are not necessary for any tortious act. I explore this
horn of the dilemma in Section 6.
This dilemma can also be run with the defendants in Morgan in mind. If, on the
one hand, we hold fixed the idea that choosing to violate a right and seeking to
substitute private rationality for public standards of reasonableness requires awareness
that a right is in jeopardy, then the defendants in Morgan do not choose to violate a right
or seek such a substitution, even if it is the case that they made mistakes of law. If, on
the other hand, we weaken the notions of choice and substitution and do not require
that they involve awareness that another’s right is in jeopardy (so that when making a
mistake of law, you count as choosing to risk violating a right and substituting private
rationality for public standards of reasonableness) then, it is not clear how these weaker
notions of choice and substitution can distinguish between tort law and criminal law.
If they genuinely (although unreasonably) believed that the victim consented, they
are probably not thinking that they were violating the victim’s right. This shows that
one can be rightly convicted of a crime, according to Ripstein, even though one does
not choose to violate a right. So it seems that Ripstein cannot maintain that the crucial
difference between criminal law and tort law is that the former requires the choice to
violate the rights of others while, at the same time, holding onto objectivism unless
‘choice’ is understood, counterintuitively, as having nothing to do with how agents
understand what they are doing. As noted above, I explore this horn of the dilemma in
Section 6.
5 Recklessness: Objective or Subjective?
However, before exploring the weaker notions of choice and substitution, I want to
examine what Ripstein has to say about recklessness. When Ripstein discusses the
distinguishing mark of criminal law as choice, he talks about both intent and
recklessness.
54
So far I have avoided the issue about recklessness by focusing on crimes
54
Recall: “Intent or recklessness is an essential element of core areas of criminality because a person
must be aware that the rights of others are in jeopardy if his action is to count as such a substitution”
(134).
39
that require intention to satisfy the mens rea element. In this section, I argue that
Ripstein’s objectivism might be used to argue that we should understand recklessness
objectively rather than subjectively. If this is right, then this puts further pressure on
Ripstein’s contention that violating criminal law involves choosing to violate the rights of
others.
Ripstein thinks that “[b]oth intentional and reckless wrongdoing are sufficient
levels of awareness. The intentional killer adverts to the risk, for he intends to realize it.
The reckless killer adverts to it and proceeds despite awareness of it. Both consider the
rights of their victims, and act in the face of them” (233-234).
55
Here, Ripstein seems
to be claiming that a defendant who satisfies the recklessness element is aware that her
action puts the rights of others in jeopardy and that she counts as substituting her
private rationality for public standards of reasonableness. That recklessness involves
this awareness may seem like a plausible claim about the criminal law. After all, the
paradigmatic case of recklessness is a case like the following:
SUBJECTIVE RECKLESSNESS: Emmet is shooting his gun around and
one of the bullets hits Enrique and he is injured. Emmet does not
intend to injure Enrique—in fact, he does not intend to injure
anyone. However, he thinks that it is quite likely that one of his
bullets (if not more) will hit and injure someone. But he does not
care about that likelihood and chooses to shoot his gun towards
the ceiling in the lobby of the bank.
Emmet is conscious of the fact that the rights of others are put in jeopardy by his action,
but he chooses to put these rights in jeopardy. When he acts on that choice, he can be
described by Ripstein as substituting his private rationally for public standards of
reasonableness. Presumably, this is the kind of case that Ripstein had in mind when he
claimed that recklessness is a component of the mens rea element of a crime which shows
that the defendant chooses to violate the rights of others where the choice to violate the
right requires awareness that the right was put in jeopardy.
However, one might wonder whether someone who lacks the conscious awareness
that her action puts the rights of others in jeopardy can also be described as satisfying
the recklessness component of mens rea. Here is the contrasting example:
OBJECTIVE RECKLESSNESS: Emma is shooting her gun around and one
of the bullets hits Enrique and he is injured. Emma also does not
intend to injure anyone; she is not even thinking that the people
in the bank could be hurt. But she is enraged and she lets off steam
by shooting her gun towards the ceiling in the lobby of the bank.
The question is whether or not Emma satisfies the mens rea of recklessness. Emma lacks
the occurrent belief that shooting her gun in the bank risks violation of rights. However,
55
There is also the other category of ‘knowing’ killers, those who do not intend to kill, but know that
their actions will result in deaths but proceeds with their actions. This group is different from reckless
killers who take the risk that they will kill, but do not know that they will kill.
40
someone who was moved by Ripstein’s objectivism might argue that this lack of a belief
is not sufficient to deny that she committed a crime. After all, Emma’s failure to believe
that shooting her gun risks violation of rights is unreasonable—a reasonable person in
Emma’s position would have realised that her firing her gun in a bank would risk injury.
So, there is some reason for thinking that objective recklessness satisfies the mens rea of
recklessness. If so, plausibly, recklessness also does not involve the kind of choice (which
requires awareness) that Ripstein has in mind.
One might deny that Ripstein ought to be interpreted as requiring conscious
awareness (understood in terms of occurrent beliefs, perhaps) for choice. After all, he
does not explicitly say that his notion of choice requires conscious awareness. What he
says is that a criminal act, which necessarily involves choosing to violate a right, is
performed when the agent seeks to substitute private rationality for public standards of
reasonableness and that awareness that the rights of others are in jeopardy is required
for such a substitution. There is no claim that this awareness must be “conscious”
awareness. However, here I offer some textual evidence for thinking that the requisite
notion of choice does require conscious awareness that a right is put in jeopardy.
56
It is telling, I think, that Ripstein discusses both intentional and reckless
wrongdoing, but not negligent wrongdoing when discussing criminal law (even though
the tort of negligence is unsurprisingly given extensive treatment). He writes that for
criminality, “both intentional and reckless wrongdoing are sufficient levels of
awareness. The intentional killer adverts to the risk [of a rights violation], for he intends
to realize it. The reckless killer adverts to it and proceeds despite awareness of it. Both
consider the rights of their victims, and act in the face of them” (233-4). Of course,
“awareness” could be interpreted as not requiring conscious awareness or occurrent
belief of the risk of a rights violation. However, if this is so, it is odd that negligent
wrongdoing is not mentioned as satisfying the awareness condition. Since one common
way of distinguishing between negligence and recklessness is that the latter requires
conscious awareness of the risk of a rights violation whereas the former does not, I think
this provides some reason for thinking that the requisite notion of choice requires
conscious awareness. This is not particularly conclusive, however. Hence, as promised
in the previous section, I will consider next whether the sort of choice required for
criminal law could be one that does not require awareness. The next section considers
two such ‘thin’ notions of choice. I then argue that neither notion can be used to
distinguish criminal law from tort law, as both notions of choice are also required for
tort law.
56
One might think that there is also a problem for Ripstein if he does not think that the awareness must
be conscious. This is because if he does not think this, he would be using ‘choice’ in a stipulative way and
he would need to say a lot more about what kind of thing he means by ‘choice’. Thanks to Gary Watson
for pointing this out.
41
6 Thinner Notions of Choice
We have just encountered one way for Ripstein to resolve the tension between the
reliance on choice and objectivity of criminal law. This response is to deny that
awareness is required for the requisite notions of choice and substitution. In this section,
I explore what notion of choice he could use instead (so that he can maintain his
objectivism and the claim that the distinction between criminal law and tort is the
notion of choice involved in criminal law). In particular, I consider one notion of choice
and ultimately argue that this notion cannot do the job. I then briefly consider another
notion of choice, but that notion is subject to the same criticisms as Ripstein’s notion.
It is not clear to me that any other notion can do any better than the ones that I
consider.
Since the requisite notion of choice cannot require awareness (while being
compatible with objectivism), one might think that what Ripstein needs is the choice to
perform the act that, as a matter of fact, violates rights. This notion of choice, call it the
“thin notion” of choice, is present in the cases we have discussed already although the
notion of choice that required awareness is not present. After all, no one denies that the
defendants in Morgan, Caleb (who kills Elliot in what he thinks is self-defence), or Emmet
and Emma (who recklessly injure Enrique in the bank) chose to perform the relevant
acts. The defendants in Morgan chose to have sex even if they were not aware that the
victim’s right was in jeopardy. Moreover, they did violate the relevant right (as a
reasonable person would believe that the sex was nonconsensual). Similarly, Caleb
chooses to kill Elliot (even though he was not aware that he was putting Elliot’s right to
life in jeopardy). And Emmet and Emma choose to fire the gun.
This thin notion of choice cannot be used by Ripstein to distinguish criminal law
from tort law as it is also required for committing many sorts of torts. This can be easily
forgotten, especially when one focuses on negligence as the primary basis for tort
liability, as Ripstein does.
57
However, there are intentional torts that illustrate that this
thin notion of choice is important to tort law. In general, one need not intend to violate
the rights of the plaintiff in order to be found liable for an intentional tort. A case
involving medical battery illustrates this clearly.
Omar needs an operation on his left ear and gives consent to Peter to
perform the operation. Peter performs the operation on Omar’s right
ear, thinking that he had Omar’s consent.
Peter is liable for the tort of battery even though he did not intend to violate Omar’s
right. That is, what is sufficient for the intent element of tort of battery is the intent to
perform an act which, as a matter of fact, results in the violation of rights.
58
Three
57
In fact, even in negligence, this thin notion of choice must be present for liability although this is
complicated by the case of omissions.
58
This is not true in all jurisdictions. In Colorado, for instance, what is required for the intention element
for battery is the intent to cause the contact as well as the intent to cause harm or intent to make non-
42
elements must be satisfied for Peter to be liable for battery: (1) Peter’s intention to
perform the operation on Omar’s right ear; (2) Peter’s acting on that intention; and (3)
the absence of consent for that operation by Omar. Notice that (1) specifies only the
thin notion of choice that I introduced. Intentional torts involve the thin notion of
choice, and hence, Ripstein cannot appeal to this notion of choice to distinguish tort
law from criminal law.
59
Ripstein could argue that intentional torts are outliers or at least that it suffices for
him if negligence can be distinguished from criminal law. This response relies on a
contentious claim that negligence is the primary basis of tort liability.
60
However, even
if I concede that negligence is the primary basis of tort liability and that it is not a
problem if there are outliers, I can show that negligence can also involve the thin notion
of choice. This discussion about negligence and the requisite notion of choice is
complicated by the fact that one can be liable for negligence for not performing any
action. This is because what could ground your liability for negligence is not your
choice to do something that resulted in an injury but your omission and the absence of
your choice to perform the action that you should have performed. However, we can
focus on cases where there is liability for negligence on the basis that the agent performs
an action that fails to meet the standard of reasonable care.
Consider the following hypothetical case:
Arturo is trying to get on the train, carrying a very heavy package. The
conductor pushes Arturo onto the train and causes him to let go of the
package. The package is dropped on the foot of a nearby customer,
Esther, who is thereby injured.
61
The conductor (along with the railroad company) is liable for negligence.
62
He should
not have pushed Arturo—it was reasonably foreseeable that pushing a customer onto
the train may cause injury to those nearby.
63
Since he chose to perform an act that he
should not have performed, he is liable for negligence.
Moreover, when Ripstein talks about what is an adequate response to tortfeasors,
he says that they should bear the costs of their choices (58). That is, if you choose to
consensual contact. This approach is known as the “dual intent” approach and is criticized by Kenneth
Simons (2006). Nancy Moore (2012), in contrast, advocates for the “dual intent” approach. But, of
course, this stronger notion does not help Ripstein as it blurs the boundaries even more.
59
Some strict liability torts also involve the thin notion of choice. Suppose I am blasting a tunnel (for
which I have permission). But unfortunately, I also blast the nearby garage (for which I do not have
permission) even though I took all reasonable precaution in ensuring that the blasting is contained. I am
liable for the property damage that I caused since I chose to engage in an activity that imposes risks (even
though I neither chose to violate anyone’s rights nor acted negligently). See Spano v. Perini Corp. (1969).
60
Recall that in the last chapter, I criticised Jules Coleman for claiming, without argument, that
intentional torts are paradigmatic.
61
This is a variation of the famous case, Palsgraf v. Long Island Railroad Co. (1928).
62
The conductor may also be liable for the intentional tort of battery as, arguably, Arturo did not consent
to being pushed. But the injury to Esther is the harm that was negligently caused by the conductor (even
though the conduct that is the basis of the tort of negligence (for the harm caused to Esther) is the same
as the conduct that is the basis of the tort of battery (for the harm caused to Arturo).
63
There is also no issue of proximate causation here, unlike the actual case.
43
engage in a particular activity, then you bear the costs of that activity (since the act of
engaging in that activity belong to you). Of course if you do not choose to engage in
that activity because, say, you are sleeping or you have been drugged, or are otherwise
unable to make choices, then you do not perform actions that can properly be described
as belonging to you and hence, you do not commit any torts. But the same is true in
criminal law. Hence, the thin notion of choice that involves choosing to perform an act
that, as a matter of fact, violates a right is required for a tort, and hence is not distinctive
of criminal law.
One might respond to my line of reasoning by claiming that there is an alternative
notion of choice that is thinner than the notion of choice that Ripstein started with but
not as thin as the ‘thin notion’ that I introduced here. Recall the example of me taking
the book from the table. Suppose that I think that the book is mine and that is why I
take the book. In this case, I do not commit a crime—I lack the mens rea element
required for the crime of theft. However, I do choose to do something that, as a matter
of fact, violates the rights of others. I choose to take the book and the book belongs to
you and you do not consent to my taking it. My thin notion of choice is present in this
case even though there is no crime. Hence, one might object, I have chosen too thin a
notion for Ripstein and so it does not matter that this notion is also implicated in tort.
Moreover, the objector may claim that the kind of cases that motivated Ripstein to be
an objectivist were the cases involving mistakes. His claim was that only reasonable
mistakes negate or mitigate guilt or punishment. So an alternative to my thin choice
that is worth examining is the choice to do something that the chooser takes to be
unreasonable.
64
After all, the defendants in Morgan did choose to do something that is
unreasonable and they counted as having committed a crime according to Ripstein.
However, this thicker notion fails to provide Ripstein with what he needs. For
Ripstein, what is reasonable and what rights we have are intimately connected. More
specifically, what legal rights we have (that are protected by tort and criminal law) are
constrained by what counts as reasonable. Moreover, if the person is not aware that
what they are doing is unreasonable, then they cannot be choosing to do what they take
to be unreasonable. That is, this notion, unsurprisingly, is subject to the same objection
to which Ripstein’s original notion was subject. This point can be illustrated by thinking
about the defendants in Morgan. Did they choose to do something that they took to be
unreasonable? On Ripstein’s story about what that would involve, the answer is ‘no’:
the defendants were not aware that what they were doing was unreasonable. So
Ripstein would say that they do not choose to do what they take to be unreasonable.
But the defendants in Morgan did commit crimes. (Notice that the defendants do make
the thin kind of choice: they chose to do something that, as a matter of fact, was
unreasonable. Hence, the only way to make it the case that the defendants in Morgan
64
The alternative must be the choice to do something that the chooser takes to be unreasonable rather
than something that is unreasonable defined objectively (even though the latter coheres better with
Ripstein’s objectivism). This is because the latter just is the thin notion of choice that I discussed.
44
chose to do something in a way that is useful to Ripstein is to use the thin notion of
choice.)
In this section, I argued that the notions of choice that seem to be available to
Ripstein are problematic. The thin notion of choice (choosing to perform acts that
violate the rights of others) is also present in tort law and hence cannot be used by
Ripstein to distinguish criminal law from tort law. The thicker notion of choice
(choosing to perform acts that the chooser takes to be unreasonable) suffers from the
same problem as Ripstein’s notion of choice, namely, it is incompatible with his
objectivist understanding of criminal law (and tort law, as a matter of fact).
7 Taking Stock
We started with Ripstein’s general idea that choice is what mattered for a clear
distinction between tort law and criminal law. This choice seemed to require awareness
on the part of the agent—awareness that her action will, or is likely to, result in injury
or damage. However, I have shown that this cannot be sustained if Ripstein is to hold
on to his objectivist perspective of both tort law and criminal law. What counts on that
view as violating the rights of others is measured by some objective standard of
reasonableness and it seems that one can be guilty of a crime (and liable for a tort) even
though one is unaware that one’s action puts a right in jeopardy. So what counts as
substituting one’s private rationality for public standards of reasonableness is infected
by some objective standard of reasonableness. Accordingly, the notion of choice cannot
be closely connected to the subjective mental states of defendants.
Then I argued that if we have a thinner notion of choice in mind, choice, in this
thin sense, must be present in intentional torts as well as the tort of negligence. Notice
that if this is right, in failing to find a plausible criterion of demarcation between tort
and criminal law, we have yet to find a basis for the distinctive practice of punishment
in the criminal law while the appropriate treatment of tortfeasors is monetary damages.
After all, the difference between criminal and tort law explicated in terms of choice
mattered for Ripstein because it explained different responses. But take cases of
intentional torts where the person intends to injure, assault, falsely imprison, or inflict
emotional distress. And take cases of criminal recklessness, understood objectively, in
which the person simply failed to meet the objective standard of reasonableness. Why
is hard treatment appropriate for the latter, but not for the former? It is true that the
former is also guilty of homicide and hence if punishment is justified for criminals, but
not tortfeasors, we can explain the difference. However, we can ask why punishment is
justified for criminals. A natural and powerful thought is that a person who chooses to
violate the rights of others deserves punishment.
65
65
This is not to imply that Ripstein himself is a desert-based retributivist even though he is partly a
retributivist. Ripstein seems to appeal to the idea that punishment is required to cancel out the apparent
45
Nevertheless, as we have seen, this choice to violate the rights of others is not
required for committing a crime. Hence, we can question why hard treatment is
deserved for an agent who commits a crime because she has made a mistake (albeit an
unreasonable one). From the point of view of desert (or blameworthiness, which is tied
in with desert), there may be no real difference between the negligent person and the
(objectively) reckless person or the person who falsely and unreasonably thinks that he
needs to kill the aggressor in order to survive.
Someone who was convinced by the arguments presented in this chapter could
give up on trying to understand the difference between criminal law and tort law in
terms of choice. They could instead try and find a different way of distinguishing
between the two. But a serious challenge remains: any such approach must both explain
the objectivity of tort and criminal law and account for the fact that different kinds of
legal sanctions are imposed in the two domains of the law.
advantage of the crime and he could hold that it is important wrongdoers do not benefit from their
crimes without thereby thinking that they deserve to suffer as a result of their wrongdoing.
46
Chapter 3: Mental States, Defences and
Different Accounts of Responsibility
1 Introduction
The previous chapter argued that Arthur Ripstein’s way of distinguishing between tort
law and criminal law fails. But it was also noted that I endorse his approach to
answering the Demarcation Question which involves examining the elements of torts
and crimes and identifying what it takes to commit them. In this chapter, I outline some
differences that emerge when looking at the elements of torts and crimes. These
differences, I argue, indicate that tort law and criminal law implicate different accounts
of responsibility and this will be the guiding principle for the rest of the dissertation. In
particular, the claim that I want to defend (as a part of my answer to the Demarcation
Question) is that the criminal law regards those who commit crimes as blameworthy.
That is, when one commits a crime, one is blameworthy or culpable for committing the
crime. This can be contrasted with the claim that one need not be blameworthy or
culpable to commit a tort. (One still needs to be responsible to commit a tort and the
account of responsibility is implicated in tort law will be unpacked in Chapter 5.)
One reason for endorsing the claim that when one commits a crime, one is
blameworthy or culpable is that criminal law explicitly appeals to the notion of
culpability. This is evidenced by the fact that one element that is required for crimes is
called “mens rea” which means “guilty mind”.
66
Moreover, this claim can explain
criminal law’s punitive response. My claim is neither that punishment is only justified
when the person being punished is blameworthy for committing the crime nor that
blameworthiness is a necessary condition for punishment to be justified. I will be
arguing for the claim is that criminal law requires blameworthiness on the part of the
guilty defendants by examining the mental states that are required for crimes and those
conditions that count as excusing conditions in criminal law.
With this in mind, this chapter has two main sections. Section 2 discusses the kinds
of mental states that are required by tort and criminal law. In particular, I examine the
intention element of intentional torts. Intention torts are often described by legal scholars
as fault-based torts which means that they have the best chance of requiring
blameworthiness. I argue that the mental states that are required to satisfy the intention
element in tort law do not entail that those who commit intentional torts are
blameworthy. I also examine what is required to satisfy the criminal law’s intention or
purpose element. In addition, I outline other mental states that are required for
66
See, for instance, the Model Penal Code (MPC), Section 2.02, which outlines the different mental
states that count as mens rea. The section is called “General Requirements of Culpability”. See also New
Zealand’s Crimes Act 1961, Section 160, which outlines what count as “culpable homicide” and claims
in subsection 4 that “Homicide that is not culpable is not an offence” (subsection 4).
47
different torts and crimes to argue for the claim that criminal law implicates
blameworthiness whereas tort law does not. Section 3 is devoted to exploring defenses
in criminal and tort law. In particular, I argue that different conditions count as excuses
in the two domains. These differences suggest that there is a difference in the kind of
responsibility that is attributed to tortfeasors and criminals. The fact that conditions
that count as excuses in criminal law do not count as excuses in tort law suggest that
the notion of responsibility that criminal law implicates is stronger than the one that is
implicated in tort law. Moreover, the conditions that count as excuses in the criminal
law indicate that criminal law implicates blameworthiness.
2 Mental States
One main kind of difference between criminal law and tort law is that different mental
states are required for the verdict that one is tortiously or criminally liable. One of the
objections to the Corrective-Retributive and the Public-Private Answers was that either
they do not explain the fact that, say, killing another human being could make one
liable for the tort of wrongful death and guilty of homicide or if they can, then they do
not explain the differences between, say, the tort of wrongful death and the crime of
homicide. One difference between torts and crimes that can be committed by the same
behaviour is that different mental states are required. In this section, I outline the
different mental states that are required for different torts and crimes and show how
they suggest that committing a crime implicates blameworthiness whereas committing
a tort implicates responsibility that is weaker than blameworthiness.
2.1 Intention
Both tort law and criminal law employ the notion of intention. Since what constitutes
torts and crimes are actions, an intention is required for one to commit a tort or a
crime.
67
However, different torts and crimes require different intentions. Of course,
different mental states are required for different torts and the same is true of different
crimes. I noted in previous chapters that there is a tripartite taxonomy of torts:
intentional torts, negligence and strict liability. These can be distinguished by the
different mental states are required for them. Similarly, there are four different mental
states that can satisfy the mens rea element of a crime namely, purpose
68
, knowledge,
recklessness
69
and negligence. I unpack what mental states are required to satisfy these
67
I am taking it as given that action is intentional behaviour.
68
The mens rea of purpose is sometimes referred to as ‘intention’.
69
The mental state that can satisfy the mens rea of recklessness is believing that the act is likely to result
in the harm in question (but disregarding it).
48
different intention or mens rea elements to show that criminal law implicates
blameworthiness whereas tort law does not.
2.1.1 Intention in Tort Law
The intention element of intentional torts is usually thought to be the key feature that
distinguishes intentional torts from the tort of negligence as well as strict liability torts.
The tort of negligence places an obligation to take reasonable care which is understood as
an obligation to conform one’s conduct to the dictates of whatever it is that due care
demands in the circumstances at hand. You are liable for negligence when you fail to
exercise such due care and your failure causes some harm.
70
It is not required that you
intend to cause or risk causing harm.
71
Torts that impose strict liability also do not
require that the defendant intend to cause harm. One paradigmatic strict liability tort
is one that imposes liability for causing (physical) harm in the course of engaging in
what is considered an abnormally dangerous activity.
72
A defendant who is liable for a
strict liability tort intends to engage in the abnormally dangerous activity but need not
intend to cause harm. In contrast, intentional torts, as the name might suggest, require
the defendant to intend to do more than just perform the actions that constitutes the
torts. Intentional torts include battery, assault, and false imprisonment. If intentional
torts are committed when one intends to make nonconsensual contact, to frighten, or
to imprison someone, then we may think that intentional torts are fault-based or that
they involve deliberate wrongdoing. For instance, Robert Keeton, Lewis Sargentich and
Greg Keating claim that “[t]he unity of intentional tort liability ... does not lie in the
unity of the interests that it protects – they are enormously varied – but in the fact that
it redresses forms of deliberate wrongdoing” (2004: 30). In this section, I explore exactly
what the intention elements are that are required for intentional torts to see if
intentional torts do indeed involve deliberate wrongdoing. After all, if intentional torts
are deliberate wrongs, then a defendant who is liable for an intentional tort may be
blameworthy for the wrong. However, I argue that the intention elements that are
required for intentional torts vindicate neither the claim that intentional torts are
deliberate wrongs nor the claim that defendants liable for intentional torts are
blameworthy.
70
Note that the tort of negligence does not protect from all types of harm. Due care arguably demands
that a teacher grades student assessment within a reasonable period of time and failing to exercise such
due care might cause harm to certain students. But the teacher does not commit a tort of negligence.
71
The foundational case of negligence is the case of Donoghue v. Stevenson. Donoghue consumed a ginger
beer float that was bought for her by a friend. The bottle of ginger beer, manufactured by Stevenson,
contained a decomposed snail. Donoghue was required to consult a doctor after suffering from
abdominal pain and was subsequently diagnosed with severe gastroenteritis. The manufacturer,
responsible for ensuring that the products are safe did not intentionally harm Donoghue, but was found
liable for negligence.
72
The paradigmatic example of an abnormally dangerous activity is that of blasting (used as a method
in construction).
49
To illustrate the role that intention plays in intentional torts, let us consider battery,
an intentional tort. In order for the defendant to be found liable for a tort of battery,
the plaintiff must prove (on the balance of probabilities) each of the following elements:
1. The defendant performed an act that resulted in making physical
contact with the plaintiff;
2. The plaintiff did not consent to the resulting physical contact;
3. The resulting contact is harmful or otherwise offensive
73
; and
4. The defendant intended to perform the act that results in the
nonconsensual contact that is either harmful or offensive.
74
The key element for our purposes is the intention element specified in (4). We can see
that there are four readings of (4) that are available. The weakest reading is where the
defendant must intend to perform the act in question, but need not have any mental
states with respect to the resulting contact. That is, he need not even think that contact
will result from her action. The second weakest reading is where the defendant must
intend to make contact, though she need not have any mental states with respect to the
nature of the contact. The third weakest reading is where the defendant must intend to
make nonconsensual contact and hence believe the contact to be nonconsensual, but
does not believe the contact to be harmful or offensive. The strongest reading is where
the defendant must intend to make harmful or offensive, nonconsensual contact.
If the tort of battery required the strongest reading, it seems that the defendant
who is liable for battery would be blameworthy for causing harmful or offensive contact.
And perhaps the tortfeasor would be blameworthy if battery required the third reading
since the defendant intends to make nonconsensual contact. However, she would not
be blameworthy if the second reading was required since she is not required to believe
that the contact that she intends is nonconsensual.
Here is a case that rules out the strongest reading which requires the defendant to
intend to cause harmful or offensive contact. In White v. University of Idaho (1990),
73
Harmful contact is unauthorised or nonconsensual contact that causes physical harm whereas offensive
contact is unauthorised or nonconsensual contact that is an affront to the dignity of the person. (For a
discussion of this understanding of this in the United States, see the Restatement (Second) of Torts, §18,
Comment c, 31.) We should also note that what counts as offensive is what a “reasonable person” would
find offensive. This understanding of what counts as offensive is described as using the “objective”
standard. The idea is that there is a standard of what counts as offensive that is set by the law
independently of what the particular plaintiff thinks. So not all instances of nonconsensual contact count
as offensive for the purposes of the tort of battery. For instance, being brushed by a passerby in a crowded
street does not count as offensive. (For commentary on this in the United States, see again the Restatement
(Second) of Torts, §19, Comment a, Illustration 2.)
74
Note that typical strict liability is a kind of intentional tort in the thin sense: Blaster performed an act
that caused an explosion. The explosion was harmful to plaintiff. Blaster intended to perform the act that
results in harmful consequences. However, there is one difference between battery and strict liability.
For battery, the defendant must not only intend to perform the act that results in harmful or offensive
contact, she must intend to make contact. In contrast, no such intention is required for strict liability –
the intention to engage in the activity (that happens to be abnormally dangerous) is sufficient.
Nevertheless, it is clear that neither those who are liable for battery nor those who commit strict liability
torts are blameworthy. Strict liability torts are discussed further in Section 2.2.1.
50
“[u]nanticipated by Mrs. White, Professor Neher walked up behind her and touched
her back with both of his hands in a movement later described as one a pianist would
make in striking and lifting the fingers from a keyboard” (109). It was held that he did
not intend to harm or cause offence and that he did not believe the contact to cause
injury or offence.
75
Unfortunately, as a result of the contact, “Mrs. White suffered
thoracic outlet syndrome on the right side of her body, requiring the removal of the
first rib on the right side. She also experienced scarring of the brachial plexus nerve
which necessitated the severing of the scalenus anterior muscles” (109). The court held,
following Rajspic v. National Mut. Ins. Co. (Idaho 1986), that the “intent necessary to
commit a battery was intent to commit the act, not the intent to cause harm” (110).
Quoting Rajspic, the Supreme Court of Idaho generalised the rule for battery: “The
intent element of battery … is satisfied if the actor’s affirmative act causes an intended
contact which is unpermitted and which is harmful or offensive” (111; 828 in Rajspic).
The defendant need not intend to cause harmful or offensive contact or believe that the
contact is harmful or offensive.
What has been said so far in White v. University of Idaho does not rule out any of the
other readings, including the weakest reading which required the defendant to intend
to perform the act which happens to be nonconsensual and harmful or offensive.
However, many cases can rule out this reading. Villanueva v. Comparetto (N.Y. 1992) held
that “[a]lthough the injury may be unintended, accidental, or unforeseen a plaintiff
seeking to establish a civil battery need only prove that the defendant intentionally touched
his person without his or her consent” (629; my emphasis). This seems to require, at
least, the second weakest reading that requires the defendant to intend to make contact.
Moreover, the “medical battery” cases also show that what is required to satisfy the
intention element of battery is the intent to make contact with the plaintiff.
76
Bettel v.
Yim (1978) held that in Ontario, if the physical contact itself is intended, the fact that its
magnitude exceeded all reasonable and intended expectations makes no difference.
This again shows that you do not need to be aware of the extent of harm that results
from the physical contact so long as you intend to make the contact. That is, the weakest
and the strongest readings are ruled out.
77
75
Professor Neher explained that “his purpose was to demonstrate the sensation of this particular
movement by a pianist, not to cause any harm” and that “he has occasionally used this contact method
in teaching his piano students” (109).
76
See, for instance, Meyers v. Epstein (N.Y. 2002), Mohr v. Williams (Minn. 1905), and Mink v. University of
Chicago (Ill. 1978).
77
This is not the law in Colorado as decided in White v. Muniz (2000): “the law of Colorado requires the
jury to conclude that the defendant both intended the contact and intended it to be harmful or offensive”
(818). This ruling by the Supreme Court of Colorado seems to have been motivated by the wording of
the Restatement (Second) of Torts §13(1)(a) which requires that the defendant “acts intending to cause a
harmful or offensive contact”. The court interpreted this as requiring both (i) the intention to cause
contact; and (ii) the intention that the contact is either harmful or offensive. Accordingly, it distinguished
this interpretation with the requirement that the defendant intentionally contacts another which, as a
matter of fact, results in a harmful or offensive contact. The intention element that is required in
Colorado is called “dual-intent”: the intention to make contact and the intention for the contact to be
either harmful or offensive. See Nancy Moore (2012) who argues for the dual-intent approach. However,
51
The two readings that remain are the second weakest reading and the second
strongest reading. That is, we now know that the defendant must intend to make
contact, but we do not know whether battery requires intention to make nonconsensual
contact or whether it requires intention to make contact that happens to be
nonconsensual. If battery merely requires intention understood in the latter way then
tortfeasors who are liable for intentional torts cannot be deemed blameworthy for
committing the tort since they need not intend the contact to be nonconsensual (let
alone harmful or offensive). Indeed, most jurisdictions do require intention understood
in the latter way. (That is, there is no requirement that one intends to make
nonconsensual contact, understood de dicto.
78
)
White v. University of Idaho is again an example. There, the court held that one
satisfies the intention element of battery if one intends to perform the act of making physical
contact. That is, one must intend that one’s act results in physical contact, but that one
need not intend that the physical contact be nonconsensual. Indeed, the defendant
there made contact with the plaintiff as a teaching method and did not think the contact
was nonconsensual. Moreover, Wagner v. State (Utah 2005) explicitly held that the
defendant does not need to believe that the contact is nonconsensual: “it is not an
element of the tort that the actor appreciate that the contact is unwanted” (610).
What we have seen is that the intentional tort of battery merely requires the
intention to make physical contact to be liable for battery and does not require the
defendant to believe the contact to be nonconsensual or harmful or offensive. This
means that battery need not involve deliberate wrongdoing, contra those who have
claimed that battery, along with other intentional torts, is a fault-based tort because it
involves deliberate wrongdoing. Hence, I have undermined one argument for the
conclusion that those who commit intentional torts are blameworthy because those who
commit intentional torts engage in deliberate wrongdoing. In Sections 2.2 and 2.3, I
examine the mental states that are required to commit the tort of negligence and strict
liability torts to argue that tortfeasors need not be blameworthy to be liable.
2.1.2 Intention in Criminal Law
we should note that if the intention element of an intentional tort requires an intention to φ, then it can
be satisfied by knowing that one’s conduct will result in one’s φ-ing. See Garratt v. Dailey (Wash. 1995)
that held that knowing, with substantial certainty, that one’s action would result in contact is sufficient
to satisfy the intention element of the tort of battery.
78
Note that the debate on the intention element of battery is framed as a debate about whether or not
tort law requires (or should require) single-intent or dual-intent where the single-intent is understood as
requiring the intention to make contact whereas the dual-intent is understood as requiring the intention
to make harmful or offensive contact. My analysis shows that there is another option which is to require
the intention to make nonconsensual contact, understood de dicto. However, it seems that all
jurisdictions either require single-intent or the dual-intent. See Kenneth Simons (2006) who frames the
debates this way (in the section entitled “Dual Intent or Single Intent?). He argues that the single-intent
approach is “the only plausible interpretation of the [relevant] case law” (1067).
52
We are now in a position to contrast the notion of intention that is at work in tort law
with the notion of intention that is working in the criminal law.
79
I illustrate the thicker
notion of intention that criminal law has in mind by considering the crime of burglary.
The intention element of burglary is the intention to commit a felony on the premises
(including automobiles) that the defendant has entered unlawfully.
80
Here is a
paradigmatic example:
Tina wants to take some jewellery and other valuable items she finds in
a house that is not hers. So, she breaks a window to unlock the door of
the house and enters the house.
Tina satisfies the intention element of intending to commit a felony in the house
because she formed the intention to take jewellery before she entered the house.
81
Moreover, she intends to steal and believes that the jewellery does not belong to her and
that she does not have permission to take it. Contrast this case with the following case
where the intention element of burglary is not satisfied:
Eugene wants to take some jewellery that is in the house. He thinks that
the house and the jewels belong to his friend Vera, who has given
permission for Eugene to take to sell. But, unfortunately, Eugene has
the wrong address and the house belongs to a stranger, Wynona.
Eugene needs the jewellery urgently. Unable to reach Vera, he breaks
a window to unlock the door and enters the house.
Eugene intends to perform an action that is a felony since the jewels belong to Wynona
who has not consented to Eugene taking them. However, he does not satisfy the
intention element of burglary because he believes that he does have permission to take
the jewels that he intends to take. That is, the intention to commit a felony therein is interpreted
as de dicto rather than de re. That is, it is not sufficient that the agent intends to perform
an act that happens to be an act of committing a felony.
82
This is different from the intentional element of battery. The intention to make
nonconsensual contact was interpreted de re, that is, as requiring the intention to make
physical contact that happens to be nonconsensual (and happens to be either harmful
79
Recall that intention or purpose is one of four different ways of satisfying the mens rea element of a
crime. Intention or purpose is regarded as being the most culpable and negligence as being the least
culpable. Many treat intention or purpose as more culpable than knowledge and knowledge as more
culpable than recklessness. However, I think that there is an argument for treating the hierarchy of
culpability as a weak ordering, allowing some instances of recklessness to be as culpable as intention.
80
The common law burglary was defined as the breaking and entering of a house in the night time.
Some jurisdictions have dropped the requirement that the act occurs at night. (California, for instance,
dropped this requirement in 1982 and the Model Penal Code does not require that the breaking and
entering happened at night.) In line with this, in some jurisdictions, a breaking and entering committed
during the day time (with the intention to commit a felony therein) is technically not ‘burglary’, but
‘housebreaking’.
81
This is not sufficient to make taking the jewels a felony but the intention to take them is sufficient to
satisfy the intention element of the felony.
82
Actually, the defendant need not believe the act to constitute a felony. It is sufficient that the defendant
believes it to be illegal or that she believes certain conditions to obtain (say, that she lacks permission to
take the jewels) which are the very conditions that make the act a felony.
53
or offensive). The difference between the intention element in tort law and the intention
element in criminal law can be cashed out by noticing that criminal law requires the
relevant mens rea with respect to each material element of the crime. Since it is an element
of the crime that Eugene does not have consent from the owner to take the jewellery,
Eugene must have the relevant mens rea with respect to his lacking consent to take the
jewellery.
83
However, there is no such general requirement to be liable for an
intentional tort. That is, it is not the case, as a general rule, that a defendant must have
the relevant mental state with respect to every element of the tort. Hence, satisfying the
intention element in tort law cannot suffice to satisfying the intention element in
criminal law since criminal law requires the defendant to satisfy the intention element
with respect to more elements.
I started the section on intentions by mentioning that the role that intention plays
is fairly truncated in tort law given the existence of the tort of negligence as well as strict
liability torts. Of course, this by itself does not establish that intentions play a bigger
role in criminal law than in tort law. As noted, intention is only one of four different
mental states that can satisfy the mens rea element of a crime. However, when we look
at the elements of torts and crimes, we see that the intention element required in tort
law is thinner than the intention element required in criminal law.
To make the contrast vivid, we can compare a tort and a crime that seem to protect
the same right, or at least, seem to govern similar situations. Recall Eugene who takes
Wynona’s jewellery without Wynona’s consent. Even though Eugene took items that,
in fact, he does not have permission to take, he is not guilty of the crime of theft. This
is because although Eugene intends to take the jewellery (that, in fact, he does not have
permission to take), he does not intend to steal it. To intend to steal something, plausibly,
you must believe that you lack permission to take it. Since Eugene believes he does have
permission to take it, he does not intend to steal it. We cashed out the content of the
intention that is required for the crime of theft by taking the de dicto reading of the
following intention: the intention to take items that he does not have permission to take.
Since Eugene lacks this intention, he does not satisfy the intention element of theft and
hence is not guilty of theft.
However, Eugene would be liable for the tort of conversion. This is because
conversion requires the defendant to intend to take the items that she does not, in fact,
have permission to take, but does not require the defendant to intend to steal the items.
So Eugene satisfies the intention element of conversion: the de re intention to take items
that he does not have permission to take. The distinction between de re and de dicto is
helpful in showing the difference between the intention element that is required by
criminal and tort law. Since the de dicto intention is stronger than the de re intention, the
intention element of criminal law is stronger than the intention element of tort law.
83
For the crime of burglary, the relevant mens rea element can be satisfied either by intention to steal
the jewellery or the knowledge that the one lacks permission to take the jewellery.
54
What does this show about the accounts of responsibility that are implicated in
criminal law and tort law? Tina has the de dicto intention to take items that she does not
have permission to take and hence satisfies the intention element of theft.
84
In contrast,
Eugene lacks this de dicto intention, but has the de re intention to take items that he
happens not to have permission to take. They each take jewellery that they do not have
permission to take and hence they each do what they should not have done.
85
However,
even if Tina is blameworthy for committing the crime of theft (and burglary), arguably,
Eugene is not blameworthy for committing the tort of conversion.
86
Eugene is not liable
to blame or resentment or condemnation. This again suggests that criminal law
implicates blameworthiness while tort law does not. Moreover, even though Eugene is
not blameworthy for taking Wynona’s jewellery, he is still responsible for it. This can help
explain why he should apologise to Wynona
87
and why he should return the jewellery
to Wynona or compensate him for if he has already sold the jewellery.
88
I think this
suggests that tort law implicates some notion of responsibility, albeit a notion that is
weaker than blameworthiness.
To hammer home the contrast, let us compare the intentional tort of battery and
the crime of assault.
89
For one to be liable for a tort of harmful battery, one must intend
to make nonconsensual contact. But again, what is required to satisfy the intention
element is the intention to make nonconsensual contact understood de re. In contrast,
84
She also satisfies the intention element for burglary since she intends to break and enter and she intends
to commit felony once she enters. However, for the contrast, we should focus on the crime of theft only
and the intention to take items that she does not have permission to take.
85
Some believe that intentions are relevant to determining permissibility and may claim that Eugene
acted permissibly (even though Tina acted impermissibly since she intended to steal). Hence, some might
argue that intentions are relevant to determining criminal permissibility although not tortious
permissibility. Michael Moore (2011) is someone who accepts that intentions are relevant to determining
both moral permissibility as well as legal permissibility. (When he talks about legal permissibility, he only
seems to have in mind the criminal law, and not tort law.) But this is a controversial thesis. Moreover,
even those who deny this thesis are likely to accept the claim that intentions are relevant to responsibility
and since this less controversial thesis can explain the difference between tort law and criminal law, I
focus on this claim. Jonathan Bennett (1980) is an example of someone who denies that intentions are
relevant to (moral) permissibility though he accepts that they are relevant to questions of responsibility.
86
Tina may not be blameworthy even if she has this de dicto intention since other conditions required for
blameworthiness must be met. Suppose, for instance, that she stole under duress. In that case, she would
not be blameworthy. My point, however, is that even if Tina is blameworthy because she satisfies the
other conditions, Eugene, who also satisfies these conditions, would not be blameworthy.
87
Of course, there are cases where someone responsible for performing some wrong act ought not
apologise to the victim (say, when the victim does not want to be contacted by the wrongdoer), but even
in those cases, the wrongdoer should feel regret for violating the right of the victim.
88
One might think that the order of explanation goes the other way: The fact that Eugene should
apologise or feel remorse and the fact that he should try to right the wrong (by returning the jewellery or
compensating Wynona for the loss of the jewellery) explain that he is responsible for taking Wynona’s
jewellery. I do not take a stand on which order of explanation is correct. What is important is that the
normative facts about what Eugene should do cohere with the judgement that he is responsible.
89
The tortious analogy of the crime of assault is the tort of battery (not the intentional tort of assault
which protects the right to be free from apprehension of imminent physical threat). Moreover, there is
no criminal analogue to the intentional tort of offensive battery where the nonconsensual contact des not
result in injury, but is an affront to one’s dignity. Hence, I focus on the contrast between the crime of
assault and the tort of harmful battery.
55
the crime of assault requires the intention to make harmful nonconsensual contact
understood de dicto. That is, the defendant must believe that the contact she is making
is nonconsensual and that the contact she is making is harmful. Moreover, someone
who has a de dicto intention to make harmful nonconsensual contact is blameworthy for
the harm caused although someone who merely has the relevant de re intention is not.
90
Hence, criminal law implicates blameworthiness while tort law does not.
This section attempted to show that even when intention is required to be liable
for a tort, the kind of intention that is required is thinner than the kind of intention that
is required to be liable for a crime. This was cashed out by appealing to the distinction
between de re and de dicto intentions. Tort law requires de re intentions whereas criminal
law requires de dicto intentions. Moreover, it is plausible to think that this difference
maps onto a difference in two notions of responsibility, namely blameworthiness and
something weaker than blameworthiness.
2.2 Strict Liability
Another reason for thinking that tort law does not implicate blameworthiness is the
existence of strict liability torts. In this section, I describe strict liability tort in an attempt
to show that we are not blameworthy when we are strictly liable for causing harm in
tort law. Even if that is right, one might think that the existence of strict liability crimes
shows that criminal law does not implicate blameworthiness either. Hence, I describe
strict liability crimes and identify some differences between strict liability crime and
strict liability torts and explain how we can be blameworthy even when we are strictly
liable for a crime.
2.2.1 Strict Liability Torts
Strict liability torts are usually defined negatively as neither requiring intent nor
negligence (or at least, not requiring the proof of intent or negligence). Just as the unity
of intentional torts does not come from the unity of rights that are protected, the unity
of strict liability (if there is one) does not come from the unity of rights that are
protected.
91
90
This is not strictly true. This is because even one who has a de dicto intention to make harmful
nonconsensual contact is not blameworthy if one is justified in making the contact, say, in self-defence
or defence of others (and the contact is a proportionate response). But this is compatible with my view
since self-defence is a recognised defence in criminal law and one who successfully pleads self-defence
will not be found guilty.
91
This point is explicitly raised in the Restatement (Third) of Torts: “There is ... no general rule of strict
liability in tort ... Instead, there are a number of particular rules that impose strict liability in certain
circumstances. Each of these rules has its own elements, which the plaintiff must prove in order to render
the rule operational” (228). I do think that there is some unity to the strict liability torts which has to do
with the question of who should bear the cost of engaging in an activity that has some inherent risks of
harm.
56
The paradigmatic strict liability tort imposes liability for causing (physical) harm
as a result of engaging in an activity that is considered to be abnormally dangerous. The
reason why this tort is characterised as imposing strict liability is because the defendant
is liable for harm caused in the course of her conducting the abnormally dangerous
activity even if she has exercised all reasonable care not to cause harm. That is, the
plaintiff need not prove that there was some reasonable precaution that the defendant
failed to take and that the harm resulted from the lack of such precaution. One
paradigmatic example of an abnormally dangerous activity is that of blasting. This is
because engaging in blasting satisfies two conditions:
(i) Dangerous: Blasting creates a foreseeable, significant risk of physical
harm even when reasonable care is exercised by the agents
involved in the activity.
92
(ii) Abnormal: Blasting is not a common activity.
93
94
To illustrate, let us consider the case of Spano v. Perini (1969).
95
The defendants
were constructing a tunnel pursuant to a contract with the City of New York and for
that work, set off dynamite at the construction site. Two garages in the vicinity were
damaged and the owners of the garages sued for damages. There was no attempt made
to establish that the defendants were negligent; that they had failed to exercise
reasonable care when blasting. However, it was held that the defendants could be held
liable for the damage to the property.
96
The court was not deciding “whether it was
lawful or proper to engage in blasting but who should bear the cost of any resulting
damage—the person who engaged in the dangerous activity or the innocent neighbor
injured thereby” (17). By finding the defendants liable, the court imposed strict liability
92
The foreseeability condition does not require the defendant to actually foresee the riskiness of the
activity. A defendant who does not in fact possess such knowledge can be strictly liable if she has “reason
to know or should know of the riskiness of its activity” (Restatement (Third) of Torts, Comment i, 238).
The case where the foreseeability condition is not met is when the dangers of the activity is not known
(say, when we do not know the harmful effects of a certain chemical).
93
Driving is often cited as an example of a dangerous activity that is not considered abnormally
dangerous.
94
Accordingly, the Restatement (Third) of Torts, §20(b) states that “An activity is abnormally dangerous
if: (1) the activity creates a foreseeable and highly significant risk of physical harm even when reasonable
care is exercised by all actors; and (2) the activity is not one of common usage” (229).
95
The foundational case of this kind is Rylands v Fletcher [1868] where the House of Lords upheld the
decision of the Court of Exchequer Chamber that “the person who for his own purpose brings on his
lands and collects and keeps there anything likely to do mischief, if it escapes, must keep it in at his peril,
and if he does not do so, is prima facie answerable for all the damage which is the natural consequence
of its escape” (Blackburn J., LR 1 Ex 265). The House of Lords added another limitation on liability that
the land from which the escape occurs must have been modified in a way deemed “non-natural” (Lord
Cairns, LR 3 HL 330). Some later courts have interpreted this non-natural condition as requiring the
activity to be abnormal or not common. I choose to use the case of Spano v. Perini as a case in point for
strict liability for abnormally dangerous activities since this case does not restrict liability to situations in
which the defendant brings onto her land some dangerous or mischievous object.
96
This case overturns Booth v. Rome (1893) that rejected the rule in Rylands v Fletcher and held that proof
of negligence was required unless the blasting was accompanied by an actual physical invasion of the
damaged property. That is, Spano extended strict liability to harm caused by debris from the blasting to
harm caused by vibrations generated by blasting.
57
for harm caused in the course of conduct an abnormally dangerous activity even though
the harm was caused despite the fact that “the utmost care was exercised to prevent the
harm”.
97
It is clear that the defendants in Spano are not blameworthy for the damage caused.
The defendants were authorised by the city to engage in blasting and so arguably, the
defendants were not doing anything that they should not have been doing. Moreover,
there is no argument that the defendants were reckless or careless: the defendants took
all reasonable precautions. Hence, I take the existence of this kind of strict liability torts
as evidence for the claim that tort law does not implicate blameworthiness.
98
Of course,
this does not mean that the defendants are not responsible in any sense. After all, they
engaged in an activity that is inherently dangerous (in the sense that there is a risk of
significant harm even when all reasonable care is exercised). As was held in Spano
“[s]ince blasting involves a substantial risk of harm no matter the degree of care
exercised, we perceive no reason for ever permitting a person who engages in such an
activity to impose this risk upon nearby persons or property without assuming
responsibility therefore” (18). Hence, although tort law does not implicate
blameworthiness, arguably, it implicates some notion of responsibility that is weaker
than blameworthiness. I take up the challenge of determining which notion of
responsibility is implicated in tort law in Chapter 4.
2.2.2 Strict Liability Crimes
As mentioned, there are strict liability crimes as well as strict liability torts. The issue to
resolve is whether the existence of strict liability crimes undermines the claim that
criminal law implicates blameworthiness, contrary to my thesis. Strict liability crimes
can be defined as crimes that do not require a mens rea element for at least one of the
non-mens rea elements.
99
We can illustrate this by considering a non-strict liability crime,
97
The chapter on strict liability in the Restatement (Third) of Torts also provides specific sections on
strict liability imposed for harm caused by non-domestic animals (under three sections: §21. Intrusion by
Livestock or other Animals; §22. Wild Animals; and §23. Abnormally Dangerous Animals). A small
number of jurisdictions do not hold the wild animal owner strictly liable unless some negligence is found
in keeping the animal. For instance, Connecticut (see Blanchard v. Bridgeport (1983)) and Montana (see
Hanson v. Brogan (1965)). What seems to unify all these torts is that liability is imposed for harm caused by
voluntarily engaging in an activity that has inherent risks of harm.
98
One could respond by arguing that strict liability torts do not form part of the core of tort law and can
be set aside. Indeed, as we saw in Chapter 1, Jules Coleman regards strict liability torts as lying outside
of the core of tort law, or at least, non-paradigmatic of tort law. However, as I noted, this is a controversial
claim. Moreover, I argued in Section 2.1 of this chapter that those who commit intentional torts are not
blameworthy and as I shall argue in Section 2.3, those who commit the tort of negligence are also not
blameworthy. Hence, arguing that strict liability torts should be set aside is not a compelling response to
my claim that tort law does not implicate blameworthiness.
99
Recall my claim that one difference between crimes that require intention as the mens rea element and
intentional torts is that the former requires intention with respect to each (non-mens rea) element whereas
the latter does not.
58
such as the crime of theft. To be guilty of theft, one must intend to steal: that is, one
must intend to take the item and one must intend to take the item which one lacks
permission to take (or at least believe that one lacks permission). By contrast, one
common example of a strict liability crime is possession of illicit drugs. For one to be
guilty, one must possess the drugs, but one need not know that the items that one has
in possession are illicit drugs. That is, there is one element of the crime (namely, that
the item is illicit) toward which the defendant need not bear a propositional attitude.
Another common example of strict liability crime is statutory rape since the defendant
need not have a mental state with respect to one element of the crime, namely whether
the victim is below the statutory age.
100
To see whether the existence of strict liability crimes undermines the claim that
criminal law implicates blameworthiness, let us consider a kind of case where the
defendant would not be blameworthy for possessing illicit drugs:
Adrian has a mailbox at the local post shop and he is expecting a parcel
of some socks and shirts from his mum. When he goes to the post shop
to pick up his parcel, the person behind the counter makes a mistake
and gives him a parcel that was not intended for Adrian, but for
someone else who has a mailbox at the same post shop. The package
that he receives is of a similar weight and size to the one he was
expecting and he takes the package home and puts it down on the table.
The moment after he does this, some police officers enter the house,
open the package and find that the parcel contains heroin.
Plausibly, we would not regard Adrian as being blameworthy for possessing illicit drugs.
After all, he had no reason to suspect that the contents of the package were illicit and
he was genuinely mistaken as to the actual contents of the package and did not think
that the contents of the package were illicit. Also, he had no reasonable opportunity to
open the package and report that he accidentally came into possession of illicit drugs.
So, if Adrian were to be found guilty of a crime, then this would be counter-evidence
for the claim that criminal law implicates blameworthiness. However, Adrian would
not be found guilty.
This is because there has been a recent trend for courts to require the prosecution
to prove that the defendant knew that she was involve with drugs. That is, the relevant
provisions that prohibit the possession of controlled substances have been interpreted,
either by statute or by judicial decision, so that knowledge is an element of the crime of
possession in the majority of U.S. jurisdictions. Hence, in those jurisdictions, at least,
Adrian would not be found guilty because the crime of possession of illicit drugs is not
a strict liability crime.
101
Indeed, more generally, the U.S. Supreme Court has
100
Both of these crimes impose strict liability with respect to the circumstance element of the crime
(namely, the illicit nature of the drugs and the age of the victim) rather than the result element (which is
an element that the defendant can bring about).
101
The only two jurisdictions in the U.S. where this is not the case are North Dakota and Washington,
but as we shall see, both jurisdictions allow for the defence of “unwitting possession”.
59
interpreted statutes dealing with activities that have typically imposed strict liability
(including gun registration, and transportation of pornographic films involving minors)
to require mens rea as to each element of the offence.
102
Hence, the area of criminal law
that imposes strict liability has been getting smaller.
Moreover, even in jurisdictions where the crime of possession of illicit drugs is a
strict liability crime in the sense that there is one non-mens rea element for which mens
rea is not required, there is a defence available such that Adrian would not be found
guilty. That is, he would not be found guilty even in North Dakota and Washington,
the two states according to which knowledge is not an element of the crime of possession.
This is because they recognise a defence of unwitting possession.
103
Since Adrian can
prove that he took all reasonable care and exercised due diligence to avoid the
commission of the relevant offence, he would not be found guilty. Moreover these very
conditions that Adrian needs to prove for the defence are the same conditions that
meant that he was not blameworthy for possession. This is not just in the case of the
two U.S. states, but other jurisdictions such as England and Canada.
104
Moreover, this
kind of defence is available not just for the crime of possession of illicit drugs, but other
crimes. In relation to the crime of statutory rape, if a defendant can prove that the
reasonable person in his position would have also thought that the victim was of
consenting age, then the defendant would not be found guilty of statutory rape. Again,
the thought that it was reasonable for the defendant to think that the victim was of
consenting age is what makes the defendant not blameworthy for committing statutory
rape.
105
The upshot of this section is that in many jurisdictions, crimes that have imposed
strict liability are no longer strict liability crimes. Moreover, even in those jurisdictions
where strict liability is imposed for some crimes (in the sense that there is at least one
non-mens rea element for which no mens rea is required), a defence is available such that
102
See, for instance, United States v. X-Citement Video (1994) in which the Supreme Court interpreted a
statute that made illegal the knowing transportation of a film involving the use of minors in sexually
explicit conduct as requiring not only that the defendant knew that he was transporting a film and that
the film showed sexual conduct, but that he also knew that the film involved a minor (in a sexually explicit
conduct). Hence, the transportation of pornographic films involving minors is no longer a strict liability
crime.
103
See State v. Michlitsch (N.D. 1989) and State v. Cleppe (Wash. 1981).
104
See, for instance, the English legislation of the Misuse of Drugs Act 1971, §28: Proof of lack of
knowledge etc. to be a defence in proceedings for certain offences. It provides that the accused (b) “shall
be acquitted thereof: (i) if he proves that he neither believed nor suspected nor had reason to suspect that
the substance or product in question was a controlled drug; or (ii) if he proves that he believed the
substance or product in question to be a controlled drug, or a controlled drug of a description, such that,
if it had in fact been that controlled drug or a controlled drug of that description, he would not at the
material time have been committing any offence to which this section applies”. See also R. v City of Sault
Ste-Marie (1978) in which the Supreme Court of Canada held that the offences of strict liability would
grant the accused a defence of due diligence.
105
For an example of another strict liability crime for which a defence available, see England’s Food
Safety Act 1990, §21: a defendant has a defence if he proved that he took all reasonable precautions and
exercised all due diligence to avoid the commission of the offence by himself or a person under his
control.
60
a defendant who is not blameworthy (such as Adrian) would not be found guilty. Hence,
the existence of strict liability crimes does not defeat the thesis that criminal law
implicates blameworthiness.
2.3 Negligence
One might also note the fact that the tort of negligence is a paradigmatic tort and think
that this gives us yet another reason for thinking that tort law does not implicate
blameworthiness. For this reasoning to go through, it must be the case that we are not
blameworthy for negligence. But this is controversial. For one thing, everyone accepts
that we can sometimes be blameworthy for negligence especially when the negligent
conduct or the negligent causing of some harm can be traced back to a choice or an act
for which one is blameworthy. However, some theorists have argued that we can never
be blameworthy for negligence in those cases where we cannot trace back to some
blameworthy act
106
that is causally related in an appropriate way to the negligent
conduct. That is, they have argued that we are never non-derivatively blameworthy for
negligent conduct. The next chapter explores arguments for this claim as well as the
arguments for the claim that we are sometimes non-derivatively blameworthy for
negligence. Hence, whether or not the fact that there is a tort of negligence is evidence
for the claim that tort law does not implicate blameworthiness is a complicated issue
which merits further exploration.
Moreover, negligence also plays a role in criminal law since one of the mental
states that can satisfy a mens rea element is that of negligence (along with recklessness,
knowledge, and intention/purpose). Hence, we may be tempted to think that
negligence is not a dimension along which tort law and criminal law differ with respect
to the notion of responsibility each domain implicates. However, although each domain
employs the concept of negligence, the tort of negligence is paradigmatic whereas the role
that negligence plays in criminal law is marginal. After all, there are only two crimes
whose mens rea element can be satisfied by negligence, namely homicide and the neglect
of children. In contrast, the tort of negligence is a paradigmatic tort. If one is tempted
to think that we are not blameworthy for negligence, then the fact that the role that
negligence plays in criminal law is marginal whereas the role that negligence plays in
tort law is vast may be one reason for thinking that tort law does not implicate
blameworthiness. I set aside this issue for now, but the next chapter is devoted to
exploring arguments for and against the claim that we are sometimes non-derivatively
blameworthy for negligent conduct.
2.4 Upshot
106
More precisely, the act for which we are blameworthy
61
Although the previous subsection ended on a hesitant note, I think that the reasons put
forward in this section on mental states show that criminal law and tort law implicate
different accounts of responsibility. Moreover, criminal law implicates blameworthiness
or culpability whereas tort law does not.
One piece of evidence for these claims was that criminal law requires the defendant
to have the relevant mental state (that satisfies the mens rea element for the particular
crime) with respect to each non-mens rea element of the crime. Hence, criminal law
requires the defendant to have a de dicto mental state with respect to the proposition that
states the non-mens rea elements of the crime. In contrast, tort law merely requires the
defendant to have a de re mental state. Moreover, plausibly, one is not blameworthy
when one merely has the relevant de re intention but one is blameworthy when one has
the relevant de dicto intention. Hence, we have good reason to think that criminal law
implicates blameworthiness whereas tort law does not.
The other piece of evidence had to do with strict liability. The existence of strict
liability torts was taken as evidence for the claim that tort law does not implicate
blameworthiness. Of course, there are strict liability crimes and hence we wondered
whether this means that we can no longer think that criminal law implicates
blameworthiness. But we saw that criminal law allows the defence of due diligence for
strict liability crimes whereas tort law does not. It seemed that given the availability of
the defence of due diligence, one could be blameworthy for committing a strict liability
crime, even though one is not necessarily blameworthy for committing a strict liability
tort. However, not all jurisdictions allow for such a defence. Fortunately for my
argument, we saw that in many, if not all, of those jurisdictions, the courts and
legislatures have moved to regard what were usually considered to be strict liability
crimes as requiring knowledge of each element of the crime. That is, many crimes that
used to be strict liability crimes are no longer strict liability crimes. Hence, the existence
of strict liability crimes seems not to pose a threat to the claim that criminal law
implicates blameworthiness.
3 Defences
Another issue that is treated differently by the two domains of the law is that of defences.
There are more conditions that count as defenses in criminal law than in tort law. In
particular, there are no excuses in tort law, if excuses are to be understood as non-
justificatory affirmative defences.
107
This difference is manifested in various ways. I
explicate the difference by looking at particular affirmative defences of criminal law,
107
The claim that a very few conditions count as excuses in tort law has been defended elsewhere. See,
for instance, James Goudkamp (2013). Similarly, the claim that “tort law has little patience for excuses”
is a starting point in John Goldberg (2015).
62
namely insanity and duress. Arguably, insanity and duress undermine or mitigate
blameworthiness. However, they do not undermine or mitigate some weaker notion of
responsibility. Since insanity and duress are affirmative defences in criminal law, but
not in tort law, we have yet another reason for thinking that criminal law implicates
blameworthiness whereas tort law does not.
3.1. Insanity
The insanity defence is available as an affirmative defence in criminal law whereas it is
not recognised as such in tort law.
108
In criminal law, insanity is an affirmative defence
because a successful plea of the insanity defence results in a verdict that the defendant
is not guilty of the crime even if each element of the crime is proven (beyond a
reasonable doubt, which is the standard proof required in the criminal law). However,
if all the elements of a tort are proven (on the balance of probabilities or on the
preponderance of evidence, which is the standard of proof required in tort law), the
verdict will be that the defendant is liable for the tort. Of course, if the defendant is able
to show that one of the elements is absent because of her insanity, then the verdict will
be that she is not liable. But, of course the same is true in criminal law. That is, if the
defendant is able to show that one of the elements is absent because of her insanity, she
will be acquitted of the crime in question.
109
Hence, insanity plays a different (and
larger) role in criminal law than in tort law.
3.1.1 Insanity and Tort Law
In this section, I describe some tort cases to show that insanity is not an affirmative
defence in tort law. Moreover, this is true across many jurisdictions. Adamson v Motor
Vehicle Insurance Trust (1957) held that in Australia, one can be liable for negligence
despite one’s insanity. The defendant, due to insanity, believed that his work colleagues
planned to murder him. He fled from them in a motor vehicle (which he stole because
he thought that his workmates had sabotaged his own car) and disobeyed a stop sign in
order to get away. While disobeying a traffic cop, he ran down the plaintiff pedestrian.
He satisfied each element of the tort of negligence, but he pleaded insanity. The
Supreme Court of Western Australia held that the defendant is liable in negligence.
There are also many American cases that show that an insane defendant can be liable
for negligence. Williams v. Hays (1894) is the case in point where the captain of a ship
108
Goudkamp (2011) argues that insanity should be an affirmative defence in tort law as well. However,
his arguments are not relevant to the issue at hand since, as a matter of law, tort law does not allow
insanity as an affirmative defence. That is, even if Goudkamp is right that tort law should allow the
affirmative defence of insanity, and so perhaps should implicate blameworthiness, this does not affect my
arguments for the claim that tort law, as it stands, does not implicate blameworthiness.
109
Indeed, given that the standard of proof in criminal law beyond reasonable doubt, if the defendant
can show that there is some reasonable doubt that one of the elements is absent because of her insanity,
she will be acquitted. In tort law, the defendant must show that the one of the elements is absence because
of her insanity on the balance of probabilities.
63
had become delirious and refused to accept that the ship was in trouble and rejected
offers of assistance. The Court of Appeals of New York found in favour of the plaintiff
and held that “an insane person is just as responsible for his torts as a sane person”
(446).
Insane defendants are liable for intentional torts as well as negligence. The
defendant in Williams v. Kearbey (1989) was found liable for battery. The defendant,
Kearbey shot several people at a school. The jury found that the defendant was insane
at the time of the shooting. However, the Kansas Supreme Court imposed liability on
the defendant. The defendant in Krom v. Schoonmaker (NY 1848) was found liable for the
intentional tort of false imprisonment even though he was mentally ill at the time of
issuing a warrant. The defendant in Morse v. Crawford (1845) was a bailor for the
plaintiff’s oxen. (That is, the defendant had possession of plaintiff’s oxen for
safekeeping.) The defendant strangled an ox with a rope. Despite the finding that the
defendant was mentally disordered while strangling the ox, the plaintiff recovered
damages in conversion. The defendant in Tindale v Tindale [1950] attacked her daughter
with an axe while suffering under insane delusions. The Supreme Court of British
Columbia held that the fact that the defendant was “acting under the influence of
delusions and that her mind was in a severe state of perturbation” at the time of the
attack was no defence to an action in battery. The court allowed that if the defendant
could prove that she did not know what she was doing, then there would not be liability.
We find a similar ruling in Morriss v Marsden [1952]. The defendant was a catatonic
schizophrenic who attacked the plaintiff. It was found that the defendant knew “the
nature and quality of the act” since the defendant was not in a condition of automatism
or trance at the time of the attack.
110
It was also accepted that being a schizophrenic,
the “incapacity of reason arising from the disease of his mind was of so grave a character
that he did not know that what he was doing was wrong”. But that this was held to be
insufficient to undermine tortious liability.
111
James Goudkamp (2011) mentions some cases that seem to show that insanity is a
defence in tort law. One such case is Buckley and Toronto Transportation Commission v. Smith
Transport Ltd [1946]. The defendant insanely believed that his vehicle ‘was under some
sort of remote electrical control’ that rendered him unable to steer it. The plaintiff was
injured in the ensuing accident and sued the defendant in negligence. The Court of
Appeal for Ontario held that the plaintiff could not recover damages since the
motorist’s mind was ravaged by disease. However, this case does not show that inanity
is an affirmative defence in tort law. After all, given that the defendant believed that he
110
Stable J. writes: “On the whole, I accept the view that an intention – i.e. a voluntary act, the mind
prompting and directing the act which is relied on, as in this case, as the tortious act – must be averred
and proved. For example, I think that, if a person in a condition of complete automatism inflicted
grievous injury, that would not be actionable. In the same way, if a sleepwalker inadvertently, without
intention or without carelessness, broke a valuable vase, that would not be actionable.”
111
Stable J. writes: “I have come to the conclusion that knowledge of wrongdoing is an immaterial
averment, and that, where there is the capacity to know the nature and quality of the act, that is sufficient
although the mind directing the hand that did the wrong was diseased.”
64
was unable to steer the car, arguably, his delusions could have prevented him from
acting voluntarily when he injured the plaintiff.
112113
3.1.2 Comparing Tort and Criminal Law
So far, we have seen that insanity is not an affirmative defence in tort law. I have not
yet shown that insanity is an affirmative defence in criminal law, but this is a well-
established doctrine of the criminal law. Instead of outlining the affirmative defence of
insanity in criminal law, in this section, I compare some pairs of cases. Each pair of
cases arises from the same situation and has the same defendant, but one is a tort case
and the other is a criminal case. There is a plea of insanity in both cases, but it is only
successful in the criminal case. Since, as I shall argue, the defendant is blameworthy in
committing the crime whereas the defendant is not blameworthy for committing the
tort, the contrast between these pairs of cases illustrate that criminal law implicates
blameworthiness whereas tort law does not.
The first pair of cases involves Norman Russ, who suffered from “a severe case of
paranoid schizophrenia that involved delusions of persecution, grandeur, influence,
and reference, and also involved auditory hallucinations” (Polmatier v. Russ (1988): 231).
While experiencing delusions of persecution and auditory hallucinations, he shot and
killed his father-in-law, Arthur Polmatier. During an interview with a psychologist,
Russ claimed that “he believed that his father-in-law was a spy for the red Chinese and
that he believed his father-in-law was not only going to kill him, but going to harm his
infant child”. Russ cited these as reasons for his shooting the plaintiff.
112
Goudkamp discusses this case: “The court did not explain the precise avenue via which it released
the defendant from liability. Presumably, however, it absolved the defendant from liability on the basis
of an affirmative defence. This is because no trace can be found in the Court’s reasons of any suggestion
that the defendant’s insanity should be attributed to the reasonable person” (2011: 733, footnote 26). It
is not clear to me why the insanity defence is supposed to be regarded as an affirmative defence if insanity
cannot be attributed to the reasonable person. Perhaps the thought is that if insanity can be attributed
to the reasonable person, then insanity would be a defene that negates one of the elements. However,
there is another reason for thinking that insanity grounds a defence that negates one of the elements of
a tort, namely, that insanity causes the defendant to be not acting voluntarily. If the court held that the
defendant cannot have acted voluntarily while driving, then the defendant would not be liable in virtue
of his insanity, but this would not mean that insanity is an affirmative defence.
113
Goudkamp cites two other cases as cases that show that insanity is an affirmative defence in tort law.
One is Hutchings v. Nevin (1992) which follows Buckley v Smith [1946] and holds that liability for negligence
can be undermined if the defendant is able to establish that she did not understand and appreciate the
duty upon them to take care or if she did understand and appreciate that duty, she was prevented by the
particular disability from discharging it. It is not clear how to interpret this doctrine. In one way of
understanding it, it is inconsistent with the decision in Vaughan v. Menlove (1837) since Menlove did not
understand that he should not have stacked the hay in the way he did. But in another understanding of
the doctrine, it is consistent with Vaughan v. Menlove since Menlove understood and appreciated the duty
not to cause fire so close to the neighbouring property, but he did not realise that the very act of stacking
the hay in the way he did was in breach of that duty. And since he was not prevented from discharging
the duty by some mental illness, he would be liable under Buckley v Smith and Hutchings v. Nevin, understood
in this second way. Another case Goudkmap cites is White v. Pile (1950) in which the defendant committed
battery against the plaintiff under the delusion that the plaintiff was his wife (even though the defendant
was, in fact, unmarried). It was held that since the defendant did not know that his action was wrong, he
was not liable. But this case was overturned by Morriss v Marsden [1952].
65
This set of facts gave rise to two cases: a criminal charge of murder and a tort case
of wrongful death. As noted in the judgment of the tort case, he was found “not guilty
by reason of insanity pursuant to General Statutes §53a-13” (231). This section on ‘Lack
of capacity due to mental disease or defect as affirmative defense’ states:
(a) In any prosecution for an offense, it shall be an affirmative defense
that the defendant, at the time he committed the proscribed act or
acts, lacked substantial capacity, as a result of mental disease or defect,
either to appreciate the wrongfulness of his conduct or to control his
conduct within the requirements of the law.
The court in the tort case took as a matter of fact that Russ satisfied the condition set
out in the above section. However, although he was found not guilty of murder, he was
found liable for the tort of wrongful death. This was because there was “no evidence
indicating that the defendant’s acts were reflexive, convulsive or epileptic” (237).
Moreover, the fact that the defendant provided several reasons for why he killed the
plaintiff was taken as evidence that he intended to kill the plaintiff (even though he
believed that he was justified in killing him). Furthermore, the court rejected the
argument put forward on behalf of the defendant that he neither acted for the purpose
of causing nor with a desire to cause the injury in question. Glass J held that “[t]his
argument is more persuasive in its application to proof of the elements of crimes than
in its relation to civil liability” (239).
The fact that Russ lacked the capacity to form a rational choice (because he lacked
the capacity to “appreciate the wrongfulness of his conduct or to control his conduct
within the requirements of the law”) sufficed for a successful plea of insanity as an
affirmative defence for the criminal charge of murder. Plausibly, this is because the
defendant cannot be regarded as blameworthy since he lacked this capacity. However,
the fact that the defendant could make a choice (albeit “a schizophrenic or crazy
choice”) sufficed to satisfy the relevant intention element and hence was held liable for
a tort. This pair of cases illustrates how the existence of insanity as an affirmative
defence in criminal law (but not in tort law) suggests that criminal law implicates
blameworthiness whereas tort law does not.
The second pair of cases involves John Hinckley who attempted to assassinate
President Reagan. The shots that he fired struck the President as well as several
bystanders. Hinckley was charged with three federal and five District of Columbia
offences, including the attempted assassination of the President of the United States,
assault on a federal officer, assault with intent to kill and assault with a dangerous
weapon. In the criminal case, United States v. John W Hinckley (1982), Hinckley was
acquitted on all criminal charges of attempt by reason of insanity. He was under a
delusion that killing the President would cause the actress Jodie Foster to fall in love
with him and his obsession was taken as evidence for the claim that he failed the
Substantial Incapacity Test which was the law of insanity at the time for federal crimes:
A person is not responsible for criminal conduct if at the time of such
conduct as a result of mental disease or defect he lacks substantial
66
capacity either to appreciate the wrongfulness of his conduct or to
conform his conduct to the requirements of law (MPC, Section 4.01.
Mental Disease or Defect Excluding Responsibility).
He was found not guilty by reason of insanity under this test.
114
However, insanity did not undermine his civil liability. One civil proceeding that
was brought was for battery for the shooting of Thomas Delahanty, a bystander. In
Delahanty v. Hinckley (DDC 1992), it was argued on behalf of Hinckley that since he was
insane at the time of the shooting (which was proved beyond a reasonable doubt in the
criminal trial), that the civil proceeding should be summarily dismissed. However, the
court maintained that insane persons can be liable for their torts.
Hinckley’s delusions excused his criminal liability although they did not negate his
tortious liability. This again shows that the account of responsibility that criminal law
has in mind is more robust than the account of responsibility that tort law has in mind.
Moreover, since insanity is seen as a paradigmatic example of something that
undermines one’s blameworthiness, this pair of cases shows that criminal law implicates
blameworthiness whereas tort law does not. Furthermore, the affirmative defence of
duress that is available in criminal law, but unavailable in tort law is evidence for
thinking that criminal law implicates blameworthiness whereas tort law does not.
Imagine Beecham who has been coerced to shoot Cera by Dedre who threatens to kill
Beecham’s family if he does not shoot Cera. Beecham is less blameworthy than he
would have been if he had shot Cera without being coerced by anyone. And criminal
law reflects this by reducing Beecham’s sentence so that it is lower than it would have
been had Beecham shot Cera without being coerced by anyone. In contrast, duress
does not have such a mitigating factor in tort law. Beecham would be tortiously liable
for wrongful death and the compensatory damages he is liable to pay would be the
same as what he would have had to pay had he shot Cera without being coerced.
The upshot of this discussion is that one plausible way of explaining the general
fact about criminal law, namely that insanity and duress are affirmative defences, and
the general fact about tort law, namely that insanity and duress are not affirmative
defences, is to appeal to the idea that criminal law and tort law implicate different
notions of responsibility. In particular, criminal law implicates blameworthiness,
whereas tort law implicates a weaker notion of responsibility.
3.2 Minors
114
This led to a public outcry and the federal law on insanity reverted to the previous law, M’Naghten
Rule. Many states that had adopted the Substantial Incapacity Test followed suit. Hence it is an open
question whether or not Hinckley’s state of mind would suffice for a successful plea of insanity in all
jurisdictions. This, in particular, depends on whether or not Hinckley could appreciate the wrongfulness
of his conduct (since he satisfies the other prong of the M’Naghten Rule as he knew the nature and the
quality of the act that he was performing).
67
Yet another condition that counts as an excusing condition in the criminal law, but not
in tort law is that of being a minor. This is complicated as there is a juvenile criminal
system. However, it is a generally accepted doctrine that being a minor is not a feature
that undermines or mitigates tortious liability. What is important for tortious liability is
whether or not the defendant, a minor or not, understood the nature or the character
of her conduct (although not whether or not the defendant understood that her conduct
was wrong (either morally or legally). Garratt v. Dailey (1956) established that the five-
year-old defendant could be liable for an intentional tort of battery because he had the
capacity to form intentions (since he intended to play a joke on the plaintiff).
In contrast, however, the criminal law does care explicitly about the age of the
defendant. MPC s4.10(1) states: “A person shall not be tried for or convicted of an
offense if: (a) at the time of the conduct charged to constitute the offense he was less
than sixteen years of age, in which case the Juvenile Court shall have exclusive
jurisdiction”. The different treatment of minors is related to the difference between
what is required to commit a crime and what is required to commit a tort. Although a
14-year-old defendant has the capacity to form the requisite intention in tort law, she
may not have the capacity such that she can be blameworthy for committing a crime.
Hence, I take the different treatment of minors as another piece of evidence for thinking
that different notions of responsibility are implicated in the two domains of law.
4 Objection: Objective Standard
The preceding sections may have established that criminal law implicates an account
of responsibility that is stronger than one that is implicated by tort law. However, one
might still challenge my claim that criminal law implicates blameworthiness. The
objector may find support from the fact that criminal law takes the objective standard
to some of the elements of crimes in the sense explained in Chapter 2. There I argued
that Ripstein’s distinction between tort and criminal law fails because his distinction
fails to capture the fact that criminal law employs the objective standard. The objection
to the claim that criminal law implicates blameworthiness is that given the objective
standard, not all who are found guilty of a crime are blameworthy.
To explain the objection, it is helpful to have a case where the objective standard
is relevant. Recall DPP v Morgan case (1976): Morgan, “[a]n R.A.F. officer invited a
number of men under his command home to have sexual intercourse with his wife.
Morgan told them that his wife would heighten her own pleasure by pretending to
resist. When she resisted, they proceeded anyway, and at their trial claimed that they
believed that she had consented” (Ripstein 1999: 176). In the previous chapter, I
suggested that the most promising way of encapsulating the mens rea element is to
explain it in terms of the rights of the victim, namely, the right not to be subject to
nonconsensual sex that a reasonable person would believe is nonconsensual. (This right
68
was contrasted with the right to be free from nonconsensual sex simpliciter.) We might
think, given this kind of objectivism, the defendants in Morgan are not blameworthy (even
though they did breach the right). After all, they genuinely believed that the victim was
consenting.
My response is to argue that the defendants in Morgan are blameworthy for what
they did. Even though they genuinely believed that the victim was consenting, they are
blameworthy for failing to make sure that the victim was consenting. Or perhaps that
they are blameworthy for taking the victim’s husband’s say-so as sufficient evidence for
thinking that the victim was consenting. Hence, the defendants in Morgan are
blameworthy and hence the Morgan case and the objectivism it exhibits do not
undermine the claim that criminal law implicates blameworthiness.
115
This chapter argued for the claim that criminal law implicates blameworthiness. I
also argued for a negative claim that tort law does not implicate blameworthiness. I
indicated that it seems to implicate some notion of responsibility that is weaker than
blameworthiness although I have not tried to identify that notion here. In the next
chapter, I tackle head on an argument against the claim that criminal law implicates
blameworthiness which goes via the claim that we are not blameworthy for certain kind
of cases of negligence, namely, the kind of cases where the tracing strategy fails. I
examine arguments for the claim that we are not blameworthy in non-tracing cases of
negligence and find them to be wanting.
Chapter 4: Blameworthiness for (Criminal) Negligence
1 Introduction
FAVOURITE SONG: Eli’s favourite song comes on the radio while he is
driving to work. He takes his eyes off the road to change the
volume. He does not see the car moving into his lane in front of
him and hits it, injuring Jeanie, the driver of the other car.
Many will think that in at least some versions of this story, Eli is blameworthy for the
harm he causes Jeanie. It might depend on whether or not he would have avoided
hitting Jeanie’s car by not taking his eyes off the road and whether or not Jeanie was at
fault for changing lanes the way she did. However, some theorists have treated with
115
This raises the question about the necessary and sufficient conditions of blameworthiness. There are
competing views, some of which will be canvassed in the next chapter.
69
suspicion the claim that Eli is blameworthy for harming Jeanie because they think that
we are not blameworthy for inadvertently causing harm.
116
Some of them claim that there
are some necessary conditions that must be satisfied before an agent can be found
blameworthy and that these necessary conditions are missing in certain kinds of
inadvertence cases even when there is a strong pre-theoretical tendency for people to
attribute blame in such cases.
117
Of course, there are some who are sceptical about
blameworthiness altogether. But in this chapter, I am interested in those who think that
it is sometimes possible to meet the conditions for blameworthiness, but do not think
that those conditions are met when one inadvertently causes harm.
118
We are exploring this topic here because one objection to my claim that criminal
law implicates blameworthiness is that negligence is criminalised and we are not
blameworthy for negligence. Call this the Negligence Objection.
P1 We are never blameworthy for inadvertently causing harm.
P2 If we are never blameworthy for inadvertently causing harm, then
(criminal) negligence can be committed by an agent without being
blameworthy.
P3 If (criminal) negligence can be committed by an agent without
being blameworthy, then criminal law does not implicate
blameworthiness.
C1 Therefore, criminal law does not implicate blameworthiness.
Since C1 is one of the claims in my answer to the Demarcation Question, I must have
a good response to this objection.
If I could establish, contra P1, that we are sometimes blameworthy for negligence,
then we can adequately respond to this objection. Hence, in Section 3, I canvass two
accounts of blameworthiness that have been put forward with the goal of vindicating
the claim that we are sometimes blameworthy for inadvertently causing harm.
However, both of these accounts of blameworthiness are problematic and so I am not
able to deny P1.
Nevertheless, this does not mean that the argument is good. This is because even
if there is no positive argument for the claim that we are sometimes blameworthy for
inadvertently causing harm, I can put pressure on the claim by objecting to the
arguments for the claim. So in Section 4, I examine the sceptical arguments according
to which we are never blameworthy for negligence. I explore Larry Alexander and Kim
Ferzan’s argument that we do not have the requisite ability to choose otherwise in
negligence cases and find this argument wanting. I, then, turn to an argument that
116
See, for instance, Larry Alexander and Kim Ferzan (2009a) and (2009b), Matt King (2009), Gideon
Rosen (2003), and Michael Zimmerman (1997).
117
I describe the kind of inadvertence case for which this issue arises most acutely at the end of this
section.
118
Inadvertently causing harm can be distinguished from other cases where there is some awareness on
the part of the agent. Call the latter cases advertence cases since an agent in these cases adverts to, or is
aware of, the risk of harm that is increased by their conduct.
70
appeals to the claim that ignorance exculpates and that in negligence cases the agents
are ignorant of the relevant facts. I think that this is the most promising line of argument
for scepticism about blameworthy for negligence. However, this argument is
controversial and hence, I conclude that P1 of the Negligence Objection is
controversial.
Moreover, I can respond to the Negligence Objection by denying P3. In Section
5, I argue that even if we are never blameworthy for negligence, this does not
undermine the explanatory power of the claim that criminal law implicates
blameworthiness. I grant that my conclusions about our blameworthiness for
negligence are not very satisfying. However, I believe that this chapter does provide a
good response to the Negligence Objection.
2 Negligence and Inadvertence: Setting the Scene
Inadvertently causing harm can be distinguished from other cases where there is some
awareness on the part of the agent. Call the latter cases advertence cases since an agent
in these cases adverts to, or is aware of, the risk of harm that is increased by their
conduct. There are three types of advertence cases of causing harm: (i) purposefully or
deliberately causing harm (where an agent intends to cause harm), (ii) knowingly
causing harm (where an agent does not intend to cause harm and hence does not
purposefully cause harm, but causes harm intentionally and knowingly), and (iii)
willingly taking a significant risk of harm and thereby causing harm.
119
In the next
section, I examine accounts of blameworthiness according to which we are
blameworthy in both advertence and inadvertence cases. Those who argue that we can
be blameworthy for negligence provide examples where it is urged that the agents are
blameworthy for inadvertently causing harm. So consider the following four scenarios:
HOT DOG Alessandra, a soccer mom, has gone to pick up her children
at their elementary school. As usual, Alessandra is accompanied
by the family’s border collie … Although it is very hot, the pick-
up has never taken long, so Alessandra leaves Sheba [the border
collie] in the van while she goes to gather her children. This time,
however, Alessandra is greeted by a tangled tale of misbehavior,
ill-considered punishment, and administrative bungling which
requires several hours of indignant sorting out. During that time,
Sheba languishes, forgotten, in the locked car. (George Sher 2009:
24)
119
The three mental states that are involved with the three cases of advertently causing harm correspond
to the three of the four mens rea elements of criminal law: (i) purpose or intention; (ii) knowledge; and (iii)
recklessness (knowing or being certain that a particular activity has substantive risks of harm).
71
FORGOTTEN BIRTHDAY Angela forgot a close friend’s birthday. A few
days after the fact, she realized that this important date had come
and gone without her so much as sending a card or giving her a
call. (H. Smith 2011: 116)
120
FORGOTTEN BATH Joe, an otherwise loving parent, is busy giving a
bath to his 1-year-old son. The phone rings; he quickly leaves the
bathroom … to answer it, and becomes completely engrossed in
the ensuing conversation. He forgets about the bath, and his son
has drowned by the time he finally remembers. (Doug Husak
2011: 201)
121
COLICKY BABY Scout, a young woman of twenty-three, has been left
in charge of her sister’s baby. The infant is experiencing digestive
pains and has cried steadily for hours. Scout has made various
attempts to ease discomfort, but nothing has worked. Finally, to
make the child sleep, she mixes vodka with fruit juice. The child
is rushed to the hospital with alcohol poisoning. (Sher 2009: 26)
In each case, the agent inadvertently does something that should not have been done
(just as Eli should have refrained from changing the volume while driving). Alessandra
should not have left Sheba in the car for that long; Joe should not have let his baby
drown; and Scout should not have given the child alcohol poisoning.
122
But, just as Eli
does not choose to injure Jeanie, Alessandra does not choose to leave Sheba in the car
for as long as she did; Joe does not choose to let his baby drown; and Scout does not
choose to give the infant alcohol poisoning.
123
However, despite the fact that each agent
does not choose to harm another, many have the intuition that the agents in these
scenarios are blameworthy. It is not clear how widespread this intuition is, but we
should explain this intuition.
One way to explain this intuition is to provide an account of blameworthiness that
delivers the results that Alessandra, Angela, Joe and/or Scout are blameworthy. In the
next section, I examine this way of explaining the intuition. However, there is another
way to explain the intuition by paying attention to the thought that each agent may be
blameworthy for inadvertently causing harm because each agent is blameworthy for
some previous choice which is connected to the harm. Indeed, many sceptics employ
the so-called Tracing Strategy to argue that an agent can be blameworthy for
120
This example originally comes from A. Smith (2005: 236).
121
Alexander and Ferzan (2009a: 77), who have argued against criminal liability for negligence on the
ground that we are never blameworthy for inadvertently causing harm unless the tracing strategy is
available, accept that this kind of example (where the agent knows about the peril she has created, but
subsequently forgets about it) is the strongest example against their position.
122
It is true that Scout intentionally gives vodka, but she does not willingly give the baby alcohol
poisoning.
123
One might find some cases to be more plausible than others. I discuss each example in what follows
so it should suffice if you find at least one case to be a convincing case of blameworthiness.
72
inadvertently causing harm.
124
According to them, an agent is blameworthy for
inadvertently causing harm h if:
(1) the agent is blameworthy for some earlier choice, and
(2) this choice plays the right kind of causal role in h occurring.
125
Recall Eli in FAVOURITE SONG. We can see how this strategy might show that Eli
is blameworthy for injuring Jeanie. After all, Eli chose to take his eyes off the road and
change the volume and this lack of attention does play the right kind of causal role in
bringing about Jeanie’s injury. In a situation like this where a tracing explanation is
available, one of the standard stories about blameworthiness, we might say, following
Joseph Raz (2011), that Eli is derivatively blameworthy for causing harm and that he is
non-derivatively blameworthy for changing the volume and taking his eyes off the road.
Indeed, those who are sceptical about blameworthiness for many cases of
inadvertently causing harm accept the viability of the tracing strategy. Larry Alexander
and Kim Ferzan, who are sceptical about blameworthiness as well as criminal
culpability for inadvertently causing harm, employ the tracing strategy. In discussing
People v. Decina (1956) in which the defendant had an epileptic seizure while driving,
causing an accident and killing four people, they claim that that “the defendant
consciously disregarded the risk that he might suffer a seizure” since the defendant
knew he was prone to epilepsy. Hence, even though there is no choice to harm at the
moment of the seizure, according to Alexander and Ferzan, the defendant is
blameworthy for the choice to drive and this choice can be the basis of criminal liability
(2009a: 80). Michael Zimmerman discusses a similar case and, like Alexander and
Ferzan, insists that only tracing can explain blameworthiness for inadvertently causing
harm. He writes: “a drunk driver may be responsible for the accident that he [or she]
has caused, even though he was completely out of control at the time. But if he is, this
124
See, inter alia, John Martin Fischer and Neal Tognazzi (2009), King (2009), George Sher (2009: 34-
39), Holly Smith (1983) and (2011), Manuel Vargas (2005), and Zimmerman (1997).
125
My conditions are similar to those King (2009) proposes for tracing: “(1) the agent is responsible for
[the prior] action; and that action caused the agent to fail to satisfy the conditions on responsibility for
the later action” (580). His second condition is meant to capture a case where the “initial choice creates
a condition of impairment that later clearly contributes to some harm” (580). In particular, it captures
the following variant of the case:
Eva suspects that her partner Effie has been sleeping with someone else. Being distraught, she
decides to stop on her way home at a bar. She has a few drinks. She drives home and being
severely intoxicated, she cannot react quickly enough to brake for the pedestrian, Jeanie, who
is hit by Eva’s car.
Since Eva’s earlier choice to drink impairs her ability to drive, King’s condition can capture this case.
However, it is not clear that King’s conditions can easily capture the following case:
Ethan notices that the brake warning light is on. But being distracted, he heads on home
without stopping by at the mechanic’s. All of a sudden, the brake fails to respond and he swerves
to avoid the cars in front of him, which he manages to do. Unfortunately, he crashes into a
parked car and injures Jeanie.
Unlike Eva (whose choice to drink impairs her), arguably, Ethan’s decision not to go to the mechanic’s
does not impair him. My second condition is more general than King’s and it can capture Ethan’s case
as well as Eva’s case.
73
is because he is responsible for being out of control, and this requires that, somewhere
back down the line, he had been free to do other than he did” (1997: 411).
126
Hence, those sceptics who employ the Tracing Strategy may argue that the
intuition that Alessandra, Angela, Joe and Scout are blameworthy in the scenarios
above can be explained by the fact that Alessandra, Angela, Joe and Scout are derivatively
blameworthy.
However, George Sher and Holly Smith, who argue that these agents
are blameworthy, claim that even when we describe the cases so that the tracing
strategy is not available, we deem these agents to be blameworthy. That is, when we
imagine that Alessandra, in HOT DOG, simply forgets some relevant information
(namely, that Sheba is in the car), it is hard to think that there is some prior choice for
which she is blameworthy. Of course, as Holly Smith notes, we could claim that
Alessandra is blameworthy for bringing the dog with her on such a hot day, knowing
that she usually leaves the dog in the car (2011: 117).
127
But, we can imagine a case
where she is not blameworthy for doing so: the pick-up process has never taken very
long in the past and it might not be appropriate to leave the dog alone at home.
Moreover, even if we can understand this as a case where the tracing strategy is
available, we may think that her blameworthiness for forgetting about Sheba is more
severe than her blameworthiness for bringing her. Hence, we may think that “the full
extent of her blameworthiness for forgetting the dog can only be fully explained by
ascribing some non-tracing [blameworthiness] to her” (Smith 2011: 117).
128
In FORGOTTEN BIRTHDAY, we may think that Angela’s blameworthiness can be
traced back to her blameworthiness for failing to do something to jog her memory about
her friend’s birthday.
129
However, we can imagine a case where there is no such failure
and hence no blameworthiness for a prior choice. Similarly, in FORGOTTEN BATH, we
may think that Joe is blameworthy for answering the phone which explains his being
blameworthy for the resulting harm. However, I think that we can imagine scenarios
in which he is not blameworthy for leaving the bath. Answering the phone is not
126
In a similar vein, Gideon Rosen (2003) who is a sceptic about culpability for ignorance allows that
you can be culpable for acting from ignorance, but only if you are culpable for the ignorance from which
you act. If you are culpable for not knowing some fact, then you are culpable for acting from ignorance.
127
Sher (2009: 83) argues that understanding HOT DOG as a tracing case merely pushes back the
question. We need to ask whether Alessandra was blameworthy for not considering whether to bring the
dog with her when she left the house. The worry is that if we keep pushing the question back, there is no
conduct for which she is blameworthy that can explain her blameworthiness for forgetting about the dog.
128
This point was made earlier by Robert Adams (1985). Indeed, this puts pressure on the viability of
the tracing strategy. If one thinks that blameworthiness comes in degrees, then one would need to decide
whether one is blameworthy to the same degree for the resulting harm as one is for the relevant earlier
decision. Or if one thinks that there is a significant difference between being blameworthy for a decision
not to do something and being blameworthy for seriously injuring someone, then one might wonder
whether blameworthiness for causing serious harm can straightforwardly be attributed to someone just
because she is blameworthy for the relevant earlier decision. I set aside this issue for the purposes of this
chapter.
129
This possibility is noted by H. Smith (2011: 117).
74
normally an activity that need take very long and if you know that you are running a
bath, you would probably assume that you will cut it short.
130
Scout’s case is perhaps more difficult. It seems plausible to think that if Scout is
blameworthy it is because she is blameworthy for failing to investigate whether or not
it is safe to give vodka to an infant. Had she investigated, she might have found out that
she should not do so and she would have become aware that she was risking harm.
However, we can imagine that Scout was often in situations where information about
the effects of vodka on various people, including children, were being discussed, but
that somehow she either failed to take note of the discussions, or subsequently forgot
the content of them. Then, as with the other cases, there would be no prior action or
decision, as such, for which Scout is blameworthy. Or imagine that she is a very
irresponsible person as a matter of character who has a tendency not to think carefully
about consequences. So perhaps she knows that it is not safe to give alcohol to an infant,
but she just does not think.
131
We may deplore her being irresponsible, but there may
not be a prior choice for which she is blameworthy that can explain her
blameworthiness for giving the baby alcohol poisoning.
In discussing these four scenarios, I assume that the tracing strategy is not
available. Hence, in the next section, when I examine the accounts of blameworthiness
proffered by Holly Smith and George Sher, I take them provide accounts of
blameworthiness that deliver the result that Alessandra, Angela, Joe and Scout are non-
derivatively blameworthy for inadvertently causing harm. If either of these accounts is
adequate, then we will have shown that it is false that we are never blameworthy for
inadvertently causing harm.
3 Two Accounts of Blameworthiness
The two accounts of blameworthiness that are examined in this section can be seen as
further explicating the Humean idea that idea that blameworthiness for acts is tied to
our character.
132
If character flaws are what ground attributions of blameworthiness,
130
Even if you think that there are no circumstances in which a parent, running a bath for a young
infant, should answer the phone, there may be other circumstances in which the parent should leave the
bath. Maybe there is a burning smell that you notice and you just run to the kitchen to check that your
stove is turned off.
131
Indeed, this is the description of the case that Sher has in mind. He writes that her failure to realise
that she should not give the vodka to the baby can be attributed to “some combination of Scout’s
irresponsibility-related dispositions – her impulsiveness, perhaps, or her tendency not to consider the
consequences of what she does” (2009: 91).
132
Hume thought that we cannot blame an agent, however badly she acts, unless the bad act proceeds
from his character. He writes: “Actions are by their very nature temporary and perishing; and where
they proceed not from some cause in the characters and disposition of the person, who perform’d them,
they infix not themselves upon him, and can neither redound to his honour, if good, nor infamy, if evil.
… [T]he person is not responsible for it; and as it proceeded from nothing in him, that is durable or
constant, and leaves nothing of that nature behind it, tis impossible he can, upon its account, become
75
then to show that we are sometimes blameworthy for inadvertently causing harm, we
need to show that our inadvertence can be caused by or betray our character flaws.
Both Holly Smith and George Sher develop the thought that blameworthiness and
one’s character are connected, but rejects Hume’s specific claim that one is
blameworthy for precisely those acts that manifest some characteristic flaw.
133
That is,
both of these accounts can be seen as two different ways of developing a character-
based view without requiring that blameworthiness be connected to core, or
characteristic traits of agents. And this is a promising line of thought since, plausibly,
inadvertence can reveal one’s character and a wide variety of attitudes that are morally
significant.
134
As George Fletcher notes that “insensitivity and egocentricity are moral
flaws and both of these manifest themselves in incidents of negligent risk-taking” (1971:
417).
135
3.1 Smith’s Account
Holly Smith’s account of blameworthiness for inadvertence in non-tracing cases is an
extension of a view called Attributionism which “holds that we are as blameworthy for
our non-voluntary emotional reactions, spontaneous attitudes, and the ensuing patterns
of awareness as we are for our voluntary actions” (2011: 115). Attributionists focus on
the set of non-voluntary mental states and activities that reflect morally significant
‘‘evaluative judgments’’ held by the agent: bigoted or tolerant attitudes, valuing or
disvaluing of friends’ well-being, concern or lack of concern over individuals’ needs or
discomfort, amusement at ethnically insulting jokes, pleasure or pain in the success of
others, discounting the minor moral flaws of others, admiration or resentment of the
good qualities of others, and so forth. According to Attributionists, these evaluative
judgments form our ‘‘moral personality’’ and therefore can be held to our credit or
the object of punishment or vengeance.” (Book II, Part III, Section II, 411.) See also George Fletcher
who puts this thought thus: “To assess a man’s just desert, we must fathom the kind of man he is … The
choice to do harm manifests character flaws … and these flaws reveal the offender to be a man deserving
of punishment.” (1971: 417)
133
One might be tempted to reject Hume’s claim if one thought that one can be blameworthy for
conduct that did not reveal some stable character flaw and even if the wrong conduct was ‘out of
character’. Moreover, if Hume’s reason for thinking that we are only blameworthy for acts that reveal
our vices is because what grounds judgements of blameworthiness are our vices, then Hume must provide
a reason for thinking why we are blameworthy for conduct (that reveal our vices), and not simply for the
vices themselves (including those that do not lead to any conduct). Furthermore, one may object that
Hume’s claim (wrongly) implies that there is a match between the seriousness of the conduct (for which
one is alleged to be blameworthy) and the seriousness of the vice (from which one acted). After all,
arguably, one’s trivial flaw could lead an agent to perform especially bad acts. The latter two objections
are pressed by Sher and he ultimately rejects Hume’s claim (2006: Chapter 2).
134
Angela Smith (2005), for instance, has argued that a wide variety of attitudes that are morally
significant do not arise from conscious choice or decision
135
See also Sher’s discussion of BAD JOKE in which Ryland tells a story that is offensive to her audience
(2009: 28). She does not do this out of malice, but her telling the story is causally explained by her being
oblivious to the impact that her behavior will have on others. This is another plausible example of a non-
tracing case of inadvertence that manifests a character flaw, namely being self-absorbed or insensitive.
76
discredit. The non-voluntary responses that reflect these evaluative judgments are ones
over which we typically have little control. Precisely for this reason, they often provide
a better window into our souls than many actions, which can be strategically performed
in order to disguise our real feelings.
136
On the Attributionist view, if a person is blameworthy for an evaluative judgment,
then the person is also blameworthy for any non- voluntary reaction or response that
arises from this evaluative judgment. Then there is clearly an Attributionist argument
that an agent can be blameworthy for her non-tracing non-voluntary failure to notice
something morally significant. However, Smith notes that Attributionists rely mainly
on examples to elicit intuitions supporting their position. That is, the power of
Attributionism comes from the strength of our intuitions about the various cases.
Smith wants to provide a more principled argument for Attribution which relies
on psychological facts. She argues for deep similarities between two kinds of
psychological process: An (i) “Intention → Act” process where the agent’s mental state
(perhaps a combination of beliefs and desires, or perhaps the formation of an intention)
causes an act; and (ii) An “Attitude → Response” process where the agent’s evaluative
attitude causes either another mental state (for example, a failure to notice or a rush of
dismay), or a non-voluntary bodily state (for example, blushing or frowning).
137
She first develops an account of blameworthiness for choice (and by extension, act)
by paying attention to various psychological features of human agents. She argues that
if an agent is blameworthy for a choice, her psychology must in some way reflect a
response to the overall moral character of the act.
S is blameworthy for her choice to perform act A just in case either:
A. S believes that A would be all-things-considered wrong and
chooses A; or
B. Although S lacks any belief about A’s overall deontic status,
1. S believes A would have features F, G, and H;
2. S’s choice to perform A arises solely from her desires or
aversions (or lack thereof) with respect to A’s having F, G,
and H; and
3. If A occurred and had F, G, and H (and no other morally
relevant features), then A would be all-things-considered
136
Smith states that her account of blameworthiness is guided by Angela Smith’s version of
Attributionism. However, Angela Smith’s view is a view about responsibility which comes apart from
blameworthiness. She writes: “To say that a person is responsible for something [in the sense in which she
is interested] is only to say that she is open to moral appraisal on account of it (where nothing is implied
about what that appraisal, if any, should be). … [A]n assessment of blameworthiness … presupposes
responsibility in the more basic sense of attributability” (2005: 238). This explains my choice to examine
Holly Smith’s account despite the connection between her account and Angela Smith’s account.
137
Smith examines various arguments for thinking that the parallel between these two processes is not
strong enough to support the conclusion that individuals are responsible for non-voluntary cognitive and
emotional responses as well as for their acts. She concludes that these arguments fail. See her 2011: 123-
124.
77
wrong. (2011: 132-133)
But Smith argues that this is not sufficient and describes the following case:
BAD HAIR DAY Clara dislikes her classmate Bonnie, and has always
thought Bonnie’s hair style to be really unattractive. She has often
been tempted to wound Bonnie by saying something nasty about
her hair. However, because Clara doesn’t want to get a reputation
for being mean, and wants to impress her new boyfriend with
what a nice person she is, she has always resisted the urge to make
a nasty comment about Bonnie’s hair. During an experiment
Clara’s psychology teacher hypnotizes Clara. Part of the
experiment (unbeknownst to Clara) involves inhibiting, for
twenty-four hours, certain of Clara’s desires: her concern to
maintain a good reputation, and her desire to impress her
boyfriend with her good character. That evening, while cruising
social networking sites, Clara comes across a picture of Bonnie on
her Facebook page. Clara’s desire to wound Bonnie is aroused.
Since her other pertinent motives are ‘‘frozen’’ as a result of the
hypnosis, Clara decides to post a cutting attack on Bonnie’s
appearance. She knows the attack will be read by all their mutual
acquaintances, and deeply humiliate Bonnie. (2011: 133-134)
Smith suggests that reflection leads us to regard Clara as not blameworthy because
Clara has other motives that would normally have contributed to her making a
decision, and that these desires did not play their usual role because she was hypnotised.
That is, although her choice did arise from her own desire, and does reflect her entire
evaluative response at that moment to her options, “the choice did not arise from
anything like a reasonably full configuration of the motives that she actually has and
that would normally bear on such a decision” (2011: 134).
138
Hence, she adds the
following, additional, condition for B:
4. S’s desires and aversions (or lack thereof) with respect to F, G, and H
represent a sufficiently complete set of her desires and aversions that
are relevant to act A. (2011: 140)
That is, in order for an agent to be blameworthy for a choice, the relevant motivational
states must represent “a sufficiently complete set” of her motivational states that are
138
Smith anticipates an objection that states that hypnotism cases are idiosyncratic and cites
psychological experiments that illustrate processes that prevent significant components of an agent’s
motivational structure from playing their normal role. For example, facial expression as well as memory
performance has been shown to be affected by being presented with selected words (Choi et al. 2005, p.
320). Subliminally activating the African-American stereotype (which, according to the experimenter,
includes hostility) through subliminal exposure to faces of young Black men causes young white research
subjects to react with greater hostility to an experimenter’s request and some who were primed with
‘‘hostility’’ gave more intense electric shocks in Milgram-type experimental situations (Dijksterhuis and
Bargh 2001).
78
relevant to the choice at issue (2011: 140).
139
Now we have the complete account of
blameworthiness for choice, we can arrive at the analogous account of blameworthiness
for non-voluntary response:
S is blameworthy for her non-voluntary response R to situation X just in
case:
A. S believes X has F, G, and H;
B. S’s response R arises solely from her desires or aversions (or lack
thereof) with respect to X’s having F, G, and H;
C. If X had F, G, and H (and no other morally relevant features),
then response R to X would be all-things-considered bad; and
D. S’s desires and aversions (or lack thereof) with respect to F, G,
and H represent a sufficiently complete set of her desires and
aversions which are relevant to situation X. (2011: 140)
Let us see what this account suggests about the blameworthiness of Scout in COLICKY
BABY. Scout is blameworthy for giving the baby alcohol poisoning because she is
blameworthy for failing to realise that giving alcohol to a baby is not safe, and she is
blameworthy for failing to realise this because her failure arises from morally
objectionable evaluative attitudes on her part – her lack of sufficient concern for the
consequences of her action on others. Moreover, it is plausible that her failure to realise
that she should not give vodka arises, not just from an isolated objectionable attitude
triggered automatically by some stimulus, but rather from a full enough set of her
evaluative attitudes at the time that they adequately represent her moral personality.
140
Hence, Smith’s account entails that Scout is blameworthy. Thus it appears there may
be some non-tracing cases in which the agent is blameworthy for her wrongful act
because, ultimately, she is blameworthy for the evaluative attitudes that underlie her
failure to realise some relevant fact.
Of course, the fact that you can tell a story based on these evaluative attitudes does
not, all by itself, mean that there is not another story that also vindicates the judgement
of blameworthiness. We may think that when the case is described in more detail, we
can trace the blameworthiness back to blameworthiness for some prior choice of Scout.
Perhaps, she chose not to concentrate on what somebody once said about giving alcohol
to children. So we need to think of a version of the case where the tracing strategy is
almost certainly not available and ask if we would judge her to be blameworthy then.
If we do, then only then, does Smith’s account have an advantage. Moreover, there
will be many other more obviously non-tracing cases in which it is much less plausible
that the agent’s failure to notice arises from some objectionable attitude (which, in turn,
would explain her blameworthiness for the act). In HOT DOG, for example, it is hard
139
Smith acknowledges that this account must be supplemented by more detail about what counts as
‘‘sufficiently complete set” of the agent’s motivational states as well as what counts as motivational states
that are relevant to an act (2011: 140 and footnote 58).
140
This idea of ‘moral personality’ which Smith borrows from Hieronymi (2008) seems to be doing a lot
of work and hence more needs to be said to fill out this concept.
79
to see what is problematic about any of Alessandra’s evaluative attitudes, since she cares
greatly about the family dog.
141
How about Joe in FORGOTTEN BATH? Let us suppose
that the conversation that diverts Joe’s attention from his baby is not an urgent
conversation – he merely gets engrossed in some gossip. We want to say that he should
have remembered that his baby is in the bath and we may condemn the evaluative
attitude that gave rise to the failure to notice. (He shouldn’t have been that engrossed
by gossip.) But the evaluative attitude that gave rise to the failure to notice does not
reflect anything like what would be the agent’s full evaluation of the situation. Joe is not, all
things considered, willing to risk his baby drowning to listen to some gossip. Rather, an
environmental stimulus (that is, the engrossing nature of the gossip) compels Joe’s
attention in such a way that his other evaluative attitude remains inactive. Hence we
cannot judge that the incident reveals Joe’s full ‘moral personality’ to be objectionable
and so, on Smith’s account, Joe is not blameworthy for his wrongful act.
Perhaps Smith’s account accords with your intuition that Alessandra and Joe are
not blameworthy. But there is a dialectical worry with this result. This worry is not
directed at Smith’s account itself but is directed at those who want to utilise Smith’s
account to establish that the agents that we have been discussing are blameworthy.
After all, only Scout in COLICKY BABY is blameworthy according to Smith’s account.
This means that their intuitions about other cases of inadvertence are not vindicated
and if they are to be vindicated, they must point to some other theory of
blameworthiness.
A second worry has to do with whether she is really giving an account of
blameworthiness. For Smith, if the relevant evaluative attitudes that give rise to the choice
or the non-voluntary response are morally objectionable, then you are blameworthy
for the choice or the non-voluntary response. That is, the moral character of the agent’s
overall configuration of motivational states renders her choice or non-voluntary
response blameworthy. But what supports the latter claim of blameworthiness? One
could argue that the moral flaw in the evaluative attitudes is what renders the choice or
non-voluntary response morally flawed. And since the choice or non-voluntary
response arises out of the objectionable evaluative attitudes of the agent, the morally
sub-par choice and non-voluntary response may be attributable to the agent, in the
sense that the evaluation of the choice or non-voluntary response is something that can
be made of the agent.
However, it is a leap to think that the agent is judged to be blameworthy for the
choice or response. If the objectionable attitude is not necessarily something for which
the agent can be blamed, it is not obvious that there is any blame to transmit to the
non-voluntary response to which the objectionable attitude gives rise. Indeed, Angela
Smith’s Attributionism that Holly Smith develops is only concerned with whether an
agent is open to moral appraisal on account of her non-voluntary responses. Angela
Smith explicitly states that “nothing is implied about what that appraisal, if any, should
141
Sher Concedes this (2009: 131).
80
be” (A. Smith 2005: 237). I think that (Holly) Smith owes us a reason for concluding
that we do not only “own” our non-voluntary responses that arise out of our evaluative
attitudes, but that we are blameworthy for our non-voluntary responses that arise out
of our (morally objectionable) evaluative attitudes.
3.2 Sher’s Account
As noted, Sher’s account is also a character-based account. He argues that if the agent’s
bad act is in some sense rooted in his character, then he is blameworthy. He attempts to
find the connection that explains what it is about the act-agent relation that makes it
reasonable to extend our condemnation from a bad act to the agent who performed it.
Sher argues that what is necessary for an agent to be blameworthy for some
conduct is the connection between wrongdoing and “the interaction of a complex
subset of the desires, beliefs, and fine-grained dispositions that together make their
possessor the person he is” (2006: 49). That is, character is to be construed broadly:
character is not some core, stable character traits of the agent, but includes the attitudes,
dispositions and traits, and the interaction between them. If an agent’s inadvertence is
caused by her constitutive attitudes, dispositions, and traits, then it is her inadvertence
and she can be blameworthy for inadvertently causing harm.
Sher’s main aim is to flesh out the necessary epistemic condition on
blameworthiness. The complete, albeit unwieldy, condition is the following:
When someone performs an act in a way that satisfies the voluntariness
condition, and when he also satisfies any other conditions for
responsibility
142
that are independent of the epistemic condition, he is
responsible for his act’s morally or prudentially relevant feature if, but
only if, he either
(1) is consciously aware that the act has that feature (i.e., is wrong
or foolish or right or prudent) when he performs it; or else
(2) is unaware that the act is wrong or foolish despite having
evidence for its wrongness or foolishness his failure to recognize
which
a. falls below some applicable standard, and
b. is caused by the interaction of some combination of his
constitutive attitudes, dispositions, and traits; or else
(3) is unaware that the act is right or prudent despite having made
enough cognitive contact with the evidence for its rightness or
prudence to enable him to perform the act on that basis. (2009:
143)
142
Although Sher talks about responsibility to remain neutral between acts that are wrong or right, if the
act is wrong, then the relevant judgement about the agent’s responsibility is that the agent is blameworthy
for performing the act.
81
The relevant condition for us is (2). Alessandra, Joe and Scout should have realised that
they were acting wrongly; hence they satisfy (2a). The more interesting question is
whether they satisfy (2b). That is, can the epistemic shortcomings of Alessandra, Joe
and Scout “be traced to the psychological features that make them the individuals they
are?” (2009: 90). In other words, we need to know whether the failure to realise the
wrongness of their conduct is caused by some combination of their constitutive traits,
attitudes, and dispositions. Let us take each agent in turn.
Although Alessandra is not consciously aware that Sheba is in the hot car, Sher
claims, she does dispositionally believe that Sheba is in the hot car. After all, “although
the proposition is not before her mind, she would sincerely assent to it if prompted”
(2009: 91). Moreover, her failure to recall that proposition and conclude that she should
hurry to let Sheba out of the car “is explained by some further combination of her
attitudes and traits” (2009: 92). This is not to condemn her attitudes and traits, but
merely to note that if they were different in certain ways, she would not have forgotten
about Sheba. Similarly, Joe’s epistemic shortcoming (namely, the failure to bring before
his mind that his baby is in the bath) is presumably caused by some combination of his
constitutive traits, attitudes, and dispositions. And again, Scout’s epistemic failure (the
failure to realise that she should not give the vodka to the baby) can also be traced to
some psychological facts about Scout. Sher, in discussing Scout, thinks that, arguably,
it is “some combination of Scout’s irresponsibility-related dispositions – her
impulsiveness, perhaps, or her tendency not to consider the consequences of what she
does – that prevents the idea that she may be harming the baby from even entering her
mind” (2009:91).
My worry with Sher’s account is that it proves too much. After all, arguably, the
kind of causal explanation that Sher has in mind is going to be available in most cases
of failing to believe, advert to, or recall, some proposition. Consider the following case:
Ivan finds out that his partner Jamie has been diagnosed with a terrible
condition. After an emotional conversation, trying to deal with this
terrible news, Ivan fails to remember to set the alarm for the next
morning. Ivan, despite normally being vigilant about being on time, is
late to work.
When we ask what causes Ivan to forget to set the alarm, presumably, it is some
combination of his attitudes, dispositions and traits. However, we may think that given
the gruelling nature of the talk, Ivan is excused from forgetting to set the alarm and that
he is not blameworthy for coming late to work.
143
Perhaps his failure to recognise that he should set the alarm does not fall below
some applicable standard. That is, even if Ivan satisfies condition (2b), he fails to satisfy
(2a). After all, it is not unreasonable for someone in Ivan’s position to forget to set the
alarm. However, although the applicable standards vary depending on the situation,
143
Of course, Ivan did something wrong and others, including his employer, can legitimately feel
annoyed about it. But this does not imply that Ivan is blameworthy for being late.
82
Sher does not allow the agent’s “beliefs, desires, and traits that prevent an agent from
realizing that he is acting wrongly or foolishly [to be] themselves part of his situation”
(2009: 98). That is, according to Sher, “the relevant standards cannot classify too many
of the determinants of an agent’s cognitive failure as aspects of his situation” (2009: 21).
Hence, unless there is some external manipulation that causes our failure to be aware
of some relevant fact, we are going to be blameworthy for such failures. So I think it is
going to be very difficult to find any cases where we are not blameworthy for our
epistemic shortcomings that lead to inadvertently causing harm.
144
One may reply on behalf of Sher that it is open to him to tighten what counts as
attitudes, dispositions and traits that are constitutive of the agent. Perhaps we can see
whether the attitudes, dispositions and traits that caused the failure are some stable and
core features of the agent. However, there are two problems with this response. First,
this reply is not available to Sher since he argues, contra Hume, that you are not only
blameworthy for (bad) acts that reflect some stable character flaws. That is, Sher
explicitly allows that you can be blameworthy for an act even if that act does not
manifest any characteristic flaw. Relatedly, this reply will not get right results that Sher
wants. The reason why he wanted to move away from Hume’s claim was because he
thought that it would not deliver the right results in cases like HOT DOG. Recall that
there is nothing stable and characteristic about Alessandra that causes her to forget to
get back to the car for Sheba. Hence, I think that Sher’s account does not succeed in
providing an adequate account that allows us to be blameworthy, not only for
intentionally or knowingly causing harm, but also for inadvertently causing harm.
To summarise: Neither Smith nor Sher offers an account of responsibility that
explains why we should attribute blameworthiness for inadvertent acts in precisely the
cases where we feel most inclined to do this. Of course, I have not examined all accounts
of blameworthiness that attempt to vindicate that we are sometimes blameworthy for
inadvertently causing harm. However, I hope to have shown that it is likely that an
account of blameworthiness will either prove too much (so that we are blameworthy
for any conduct that can reasonably be attributed to us) or prove too little (so that we
are only blameworthy for inadvertently causing harm in cases which reveals the kind
of people that we are). This means that I have not yet established the falsity of P1 which
states that we are never blameworthy for inadvertently causing harm.
4 Sceptical Arguments
In this section, I examine two different arguments for P1. By putting pressure on these
arguments, I hope to show that P1 is controversial and that the burden to support P1
144
Jonathan McKeown-Green (2006), who discusses moral responsibility in general, starts from the idea
that the boundaries of the self are unclear, and possibly too unclear for it to be useful for thinking about
moral responsibility. If those arguments are persuasive, then projects like Sher’s seem destined to fail.
83
is squarely on those who wish to use it as a premise in the Negligence Objection against
my claim that criminal law implicates blameworthiness.
4.1 Ability to Choose to Advert to the Risk
Many have argued that the agent who causes harm must choose to harm in order to be
blameworthy for the harm.
145
Certainly, in the paradigmatic cases of blameworthiness,
where an agent knowingly and intentionally causes harm, it seems natural to explain
the blameworthiness in terms of the kind of choice made by the agent. In non-tracing
inadvertence cases, there is not an obvious choice on which to focus in this way, but
perhaps the blameworthiness could be explained, if we knew why choice seemed so
important. If we knew this, maybe we would see that the factors that make choice
important can also be factors in non-tracing cases; this would enable us to explain why
there seems to be blameworthiness here. On the other hand, of course, a sceptic about
blameworthiness in these cases might be able to show that every factor that might
explain why choice seems to matter for blameworthiness is absent in the non-tracing
cases.
One reason for thinking that choice matters is that when one chooses harm, one
is aware of the harm and only when one has awareness of harm, can one be
blameworthy.
146
Hence, Alexander and Ferzan claim that “[w]e are not morally
culpable for taking risks of which we are unaware” (2009a: 71).
147
In general, it is very
plausible that when one intentionally φs, one adverts to the risk that one φs.
148
So we
can say that choosing to harm and thereby intentionally harming bring with them the
awareness that one will or is likely to cause harm. According to this line of reasoning,
choosing to harm is required for judgements of blameworthiness because awareness of
risking harm is required for judgements of blameworthiness.
149
But we can ask why awareness or adverting to the risk of harm is required for
blameworthiness. One natural answer is that only when one has the awareness, does
one have sufficient control over the result of one’s actions.
150
After all, many think that
145
Jerome Hall (1963) is an example. He, in fact, talked about culpability, but I do not think that the
difference in terminology amounts to any significant conceptual difference for our purposes. Another
theorist who claims that culpability depends on choice is Michael Moore (1997).
146
J.W.C. Turner has argued that awareness or what he calls ‘foresight of consequences’ is necessary for
criminal responsibility. See his 1945 and the discussion of his argument by H.L.A. Hart (2008).
147
See also Alexander (1990).
148
In what follows, I assume that adverting entails consciously noting and hence, being aware. This is
what those sceptics who discuss awareness seem to assume. See Husak (2011) who claims that there has
not been enough attention paid in the literature to what the relevant notion of awareness should be in
this context.
149
George Sher describes (and rejects) a view of responsibility according to which an agent’s
responsibility extends only as far as her awareness of what she is doing. He calls this the ‘Searchlight
View’ (2009: 4).
150
Sher (2009) discusses the idea that the Searchlight View is supported by the thought that conscious
awareness is a necessary component of control.
84
the fact that something was out of one’s control can exculpate one.
151
That is, many
take the view that blameworthiness requires control.
152
One way of understanding an
agent’s control over her action that has been discussed in the literature is in terms of
her ability to choose otherwise.
153
This account of control is something that can be applied
to cases of inadvertence, as H.L.A. Hart (2008) noted. Just as we have the capacity to
refrain from harming (even when we do not exercise the capacity), we have the capacity
to advert to, be consciously aware of, or perceive, our conduct and its risks. Hence, the
pertinent question about control in inadvertence cases is: did the agent have the
capacity to perceive the risk (even though the agent did not, as a matter of fact, perceive
the risk)?
154
If the answer is ‘yes’, then the agent could have chosen to do otherwise than
cause the harm. Recall Joe in FORGOTTEN BATH. Those who react to this case by
judging Joe to be blameworthy presumably take Joe to have the capacity to remember
that his baby is in the bath before the baby drowned. After all, they might think, had
he remembered, he would have chosen to rush to the bath and hence would not have
let the baby drown.
155
Hence, he had control on this understanding of control and
therefore satisfies the necessary control condition for blameworthiness.
Larry Alexander rejects this Hartian claim that the capacity to avert to, or
perceive, the risk is the capacity that suffices for control. Instead, he argues that the
agent who has this capacity (to advert to the risk) still lacks the requisite control for
blameworthiness. This is because, according to Alexander, “it is simply false that the
[negligent agent] ‘could have’ chosen differently in any sense that has normative bite”
151
Not everyone agrees that control is required for blameworthiness. See, for instance, Nagel (1976).
152
See, for instance, Rosen (2004). Some have explicitly claimed that control is required for criminal
culpability. See Husak (1987: Ch. 4) as well as Alexander and Ferzan (2009a). Relatedly, many also
accept that having free will is equivalent to or at least requires having a certain kind of control. See, for
instance, Randolph Clarke and Justin Capes (2013).
153
Harry Frankfurt (1969) has argued that one can be responsible for an item of conduct even if one
could not have done otherwise so long as one could have chosen otherwise. [Gary: On his analysis, this is
not necessary either. The counterfactual intervener (who would prevent him from choosing otherwise if
one were to begin to do so) would make this false.] Although I am presently interested in an analysis of
control, and not responsibility, I examine the weaker claim that control requires the ability to choose
otherwise.
154
Since control (understood as the ability to choose otherwise) is only taken to be a necessary
requirement for blameworthiness, I am exploring whether the capacity to perceive the risk is also
necessary for blameworthiness. Another question we may need to ask to establish that someone is
blameworthy for inadvertently causing harm is whether or not she would have engaged in the relevant
behaviour even if she had been aware of the risk. This is the test that Kenneth Simons (1994) suggests
for determining whether someone can be culpable for inadvertently causing harm.
155
I am not offering a counterfactual analysis of capacity here. My point is that we assume that Joe is a
fairly ordinary individual and it seems natural to presume that Joe has this capacity to remember and
that since he is a loving parent, he would have chosen to rush to the bath if he had remembered. Of
course, if we are told that Joe does not have the capacity to remember his baby even though he had just
put him there, most people would retract their judgment of blameworthiness. So, I shall assume that Joe
has the capacity to remember. (I do not think this assumption is problematic. After all, Alexander and
Ferzan, whose objection to Hart we are about to consider and who consider a case like Joe’s, do not
argue that Joe is not blameworthy because he does not have the capacity to remember. In what follows,
I discuss why they take Joe to be blameless.
85
(1990: 99).
156
He, later joined by Ferzan, claims that for the ability to choose otherwise
to have normative bite, it must be such that the agent had reason to choose differently
given what she believed. That is, the agent must have been aware of some reason to
choose differently. But in the non-tracing inadvertence cases, they claim, the agent has
no reason that she is aware of to choose differently from the way she chose. Their
thought is that if we are going to find an agent blameworthy, we must be able to point
to some feature of the agent to which the agent had access such that that feature is a
reason for the agent to choose to do differently than she did.
157
So the pertinent issue,
according to Alexander and Ferzan, is: whether the agent had “any internal reason to
choose differently from the way he chose” (2009a: 83). In cases of inadvertence, the
relevant act that the agent must choose differently, according to this line of reasoning,
is the act of adverting to the risk of harm. The question is then: Is adverting to the risk
the kind of thing that you can choose to do?
Suppose that you just are not conscious of some background belief that would
make you choose to advert to the risk. Alessandra, for example, does not advert to the
risk of harm to Sheba because she is not conscious of the relevant background fact that
Sheba is in the car. So, in order to claim that Alessandra could have chosen to advert
to the relevant risk of harm, we have to say that Alessandra could have chosen to be
aware of the fact that Sheba is in the car (and hence had the right kind of control over
believing that Sheba is in the car). Alexander and Ferzan do not say any more on this
issue. They simply claim that one would then have to show why it was a “culpable
decision not to form that belief [about the relevant background fact]” (2009a: 83).
It seems that their argument goes thus: Holding fixed the attitudes that the agent
had at the time of, or just prior to, the relevant action, there is nothing that we could
point to such that it would give a reason for the agent to choose to do differently than
she did. So, if we are to hold that agent blameworthy, we must be able to say that the
agent could have chosen to advert to some fact that would give her a reason to choose
to do differently. That is, to hold Alessandra blameworthy for forgetting about the dog,
we have to say that she could have chosen to remember that Sheba is in the car. But,
we cannot choose to remember something.
158
Alexander and Ferzan neither explicitly
claim this nor provide any reasons for it, but we might think that this is independently
plausible.
159
So, perhaps this argument succeeds (at least in so far as it applies to
Alessandra).
156
See also Alexander and Ferzan (2009a: 83).
157
The features that Alexander and Ferzan take to be salient are: (1) the belief that the agent has about
the risk that he is imposing on the interests of others; and (2) the reasons he has for performing the act
(and for imposing the risk). They further claim that the agent “must be aware of the reasons”, even
though she “need not be motivated by them” (2009a: 24).
158
One might claim that we lack the ability to decide which mental states are occurrent (or consciously
entertained) at any given time. If this is right, then it follows that one lacks the ability to decide to
remember something at any given time.
159
Although in order to make this argument, Alexander and Ferzan only need to claim that we cannot
choose to remember, elsewhere, they make a broader (and, more controversial) claim that “we cannot
86
However, it is not clear why the relevant ability that one must possess in order to
have control that has ‘normative bite’ is the ability to choose to advert to the risk, rather
than the ability to choose to refrain from performing the risky conduct. The latter
ability, which Hart thinks is what is necessary for a negligent agent to be blameworthy,
is deemed inadequate by Alexander and Ferzan, but they do not argue for this view.
So even if they are right that we cannot choose to remember or choose to advert to
risks of which we are currently unconscious, they need to provide support for the claim
that for an agent to be blameworthy for creating some risk, they have to, not only have
the ability to choose to refrain from creating the risk, but also have the ability to choose
to advert to the risk. Absent this support, Alexander and Ferzan’s view seems under-
motivated.
Moreover, even if Alexander and Ferzan could motivate their claim about what
kind of ability is needed, one could respond by claiming that their argument only
succeeds against cases of inadvertence where the agent fails to advert to the risk because
she forgets some relevant background fact. Had she remembered that fact, she would
have adverted to the risk, but since she cannot choose to remember it, she is not
blameworthy. However, it may be urged, there are other kinds of inadvertence cases
where the agent fails to advert to the risk, but not because she forgets some relevant
fact. Consider the following case:
HOME FOR THE HOLIDAYS Joliet, who is afraid of burglars, is alone
in the house. Panicked by sounds of movement in her kitchen, she
grabs her husband’s gun, tiptoes down the stairs, and shoots the
intruder. It is her son, who has come home early for the holidays.
(Sher 2009: 26)
Let us further suppose that Joliet did not think that there was any chance that her son
might be home: when she first hears some noise, she may have considered and rejected
that possibility; but at the moment when she becomes aware that there are sounds of
movement and grabs her gun, that possibility is not before her mind at all. Now contrast
that with a variant of HOME FOR THE HOLIDAYS where Juliet thinks, at the time of
grabbing the gun that there is a very slim chance that her son is home, but she thinks
that it is so unlikely that she dismisses it as a real possibility and sets aside this small risk.
In this case, given her awareness of the slim chance, there was an “internal reason to
choose differently”. Hence, on Alexander and Ferzan’s view, Juliet counts as having
the requisite control whereas Joliet does not.
160
One worry with this analysis of the examples is that the difference between Joliet
and Juliet is so minor that it is odd if one is not blameworthy at all and one is
blameworthy. Hence, one might conclude that the requirement that there be a mental
control how we perceive, what we infer and what we believe” (2009b: 273). They argue that this means
that we should not be blamed for how we perceive, what we infer and what we believe. Cf. Pamela
Hieronymi (2008).
160
Since shooting the intruder is unjustified, on Alexander and Ferzan’s view, Juliet is blameworthy.
87
state that would have given a reason to choose differently may not be the right kind of
requirement for attributions of blameworthiness.
161
This puts pressure on the idea that
Alexander and Ferzan could provide support for the claim that the ability to choose to
advert to the risks is the ability that we are after when thinking about control and
blameworthiness.
Alexander and Ferzan may respond by pointing out that they are not, at this
juncture, trying to argue that people are not blameworthy. All they are doing is trying
to show that a particular way of attributing blameworthiness fails, namely that Joliet
could have chosen to advert to the possibility that it is her son that is making the noise.
They claim that there is no ‘internal reason’ why Joliet should have thought about the
possibility her son might be there. Hence Joliet could not have chosen to advert to the
relevant risk (in a way that has normative bite), and so this criterion gives us no mandate
for thinking that Joliet is blameworthy. Moreover, we might think that this view of
Alexander and Ferzan’s delivers the right result in recklessness cases. Recall Ida, the
bank robber, who consciously adverts to the risk that her firing the gun might injure
someone. Since she is aware of that risk, we can point to a mental state that the agent
had that would give Ida a reason to choose to refrain from firing the gun.
However, it is not clear to me that the mere fact that she was aware of the relevant
risk means that she could have chosen differently (than firing the gun) in a way that has
normative bite. After all, we are meant to hold fixed all the mental states that the agent
had at the time of, or just prior to, the relevant action. That is, we hold fixed that Ida
wants to show everyone that she is serious about her threat of violence; that she wants
to rob the bank successfully; that she believes that firing the gun would show that she is
serious; and that she believes that showing that she is serious is crucial to robbing the
bank successfully. It seems ad hoc to simply point to the belief about risk and say that
this would give her a reason to choose differently when the whole set of mental states
would not give her a reason to choose differently.
Alexander and Ferzan have argued that to be non-derivatively blameworthy for
inadvertently causing harm, one must not merely have the ability to advert to the risk,
but the ability to choose to advert to the risk. I think that this claim is under-motivated.
They neither provide any arguments for why Hartian ability to choose to refrain from
engaging in risky conduct is inadequate nor provide arguments for why their preferred
ability (to choose to advert to the risk) has normative bite. Moreover, I have also tried
to show their preferred ability gives some unsatisfying results in some of the cases.
Perhaps what I have said is not conclusive. However, I hope to have shown that sceptics
like Alexander and Ferzan must take on the task of providing support for the claim that
161
Of course, it is open to Alexander and Ferzan to say that Juliet is not blameworthy for some other
reason. Indeed their discussion elsewhere suggests other conditions for being blameworthy. Suppose we
can describe her as “sincerely believing” that the person she shoots is an intruder and not her son. If that
belief justifies her shooting the intruder, then she would not be blameworthy. On the other hand, if it
turns out that even if it is an intruder, she was not justified in shooting the intruder, then she would be
blameworthy for shooting the intruder. (See their 2009a: Chapter 2.)
88
the ability to choose to advert to the risk is the relevant ability that is needed for control (rather
than the ability to choose to perform the act that minimises the risk).
At this point, one might wonder whether the Hartian ability can ground
attributions of blameworthiness and so we have another positive account of
blameworthiness that can accommodate the intuition that we are sometimes
blameworthy for inadvertently causing harm. After all, Alessandra had the ability to
choose to return to the car earlier. Joe had the ability to choose to ignore the phone or
end the conversation earlier; Scout had the ability to choose not to give the child vodka.
However, whether the ability to choose otherwise is a necessary and sufficient condition
for blameworthiness is either controversial or implausible.
Harry Frankfurt has tried to show that ability to choose otherwise is not necessary
for blameworthiness.
162
However, this has been a hugely controversial issue. As
Michael McKenna and Justin Coates note, “an enormous (and intricate) literature has
emerged around the success of Frankfurt's argument and [t] he debate is very much
alive, and no clear victor has emerged” (2015).
163
But perhaps this is not a controversy
with which we must contend. After all, we wanted to see whether or not we could
establish that we are sometimes blameworthy for inadvertently causing harm. So even
if the ability to choose otherwise is not necessary for blameworthiness, so long as it is
sufficient for blameworthiness, we have been able to make our case. However, there
are reasons to doubt that the ability to choose otherwise is sufficient to ground
blameworthiness. This is because even in cases where agents do not seem blameworthy,
the agents could have chosen to do otherwise. Consider agent-regret cases discussed by
Bernard Williams (1981). He discusses a lorry driver who hits a child through no fault
of his own (because the child runs out onto the street out of nowhere). The driver causes
the injury and (rightfully rationally) feels agent-regret, says Williams, but he is not
blameworthy (even if it grounds a moral obligation to apologise or feel regret (hence
agent-regret).
164
But the driver did have the ability to choose not to drive down
that road (or the ability to choose to drive way below the speed limit, giving him plenty
of time to brake for the child). Moreover the driver has the ability to choose otherwise
in the same way that Alessandra had the ability to choose return to the car earlier.
Hence, this account of blameworthiness is inadequate because it is too strong and
overgeneralises. Moreover, this account of blameworthiness is something that is too
weak to be an account of blameworthiness that is implicated only in criminal law, but
not in tort law. Therefore, it is not the right kind of account of blameworthiness that is
in play in P1 of the Negligence Objection.
162
Harry Frankfurt (1969) initially argued against the necessity of the ability to do otherwise for
responsibility. Moreover, ability to choose that one can be responsible for an item of conduct even if one
could not have done otherwise so long as one could have chosen otherwise.
163
They cite numerous critics of Frankfurt’s argument as well as its advocates.
164
We might wonder whether we are even responsible for the harm in agent-regret cases. The next chapter
on the account of responsibility implicated in tort law discusses these agent-regret cases.
89
4.2 Ignorance Exculpates
Perhaps we are never blameworthy for inadvertently causing harm because in an
inadvertence case, there is some fact of which the agent is ignorant and this ignorance
grounds the claim that the agent is not blameworthy.
165
This argument relies on the
claim that ignorance exculpates. That is, there is some relevant fact of which you must
be aware for you to be blameworthy. If we can show that there is some fact of which
an agent is ignorant in each non-tracing case of ignorance and awareness of this fact is
necessary for blameworthiness, then we have another argument that concludes that we
are not blameworthy in non-tracing cases of inadvertence.
Alessandra is not aware that her dog is languishing in the car; Joe is not aware that
his child is in danger of drowning; and Scout is not aware that giving vodka to a baby
is dangerous. It is true that had they been aware of these facts, they would have acted
differently: Alessandra and Joe would have hurried back to the car and bathroom,
respectively; and Scout would not have given vodka to the baby. But the pertinent issue
is whether the lack of awareness of these facts makes it the case that these agents are
not blameworthy. Perhaps they would be blameworthy if they were also blameworthy
for their ignorance. This suggests the following sceptical argument:
P1 Agents in non-tracing cases of inadvertence are ignorant of some
salient information.
P2 Ignorance of that salient information exculpates unless the agents
are blameworthy for their ignorance.
P3 They are not blameworthy for their ignorance.
C Therefore, agents in non-tracing cases of inadvertence are not
blameworthy.
Accordingly, if can show that Alessandra, Joe and Scout are ignorant of some fact, but
are not blameworthy for this ignorance, then we could show that they are not
blameworthy for causing harm. I think P1 and P3 can be challenged.
4.2.1 Ignorance: Challenging P1
In order for the above argument to succeed, we have to show that P1 is true. That is,
we have to show that Alessandra, Joe and Scout are indeed ignorant of some salient
fact. It is true that, at the relevant time, they are not consciously aware of the salient
fact. But this is not a typical case of ignorance. To see this, contrast Joe in FORGOTTEN
BATH with Moe, his partner. Moe comes home from work, sees Joe on the phone, and
starts making dinner, thinking that the baby must be in bed already. Plausibly, Moe’s
not blameworthy for letting the baby drown. Moe’s ignorance of the fact that baby is
alone in the bathtub is different in quality from Joe’s ignorance. One natural way of
165
This is an argument that somebody sceptical about blameworthiness for inadvertently causing harm
might consider, though, as far as I know, nobody has. Thanks to Steve Bero for encouraging me to
consider this kind of argument.
90
cashing out this difference is to invoke the distinction between occurrent and dispositional
beliefs. It is true that, at the relevant time, both Joe and Moe lack the occurrent belief that
the baby is in the bathtub alone. But Joe, unlike Moe, has a dispositional belief that the
baby is in the bathtub, at the relevant time. One way to support the claim that Joe has
this dispositional belief is to see what Joe would do if he was asked where the baby is.
Suppose Moe interrupts Joe’s phone call to ask where the baby is. Chances are that Joe
would remember that the baby is in the bath and hurry to the bathroom. This might
incline one to think that Joe is not ignorant of the salient information since he believes
the salient information (albeit dispositionally).
Here is a reason for thinking that having the relevant dispositional belief means
that the agent is not ignorant. Suppose Caitlin believes that her dog is a Labrador and
we are interested in whether she knows that her dog is a Labrador. We must ask some
questions: (i) Is Caitlin’s dog indeed a Labrador?; (ii) Is Caitlin justified in believing that
her dog is a Labrador?; and (iii) Is she in some kind of Gettier-style scenario that defeats
knowledge? If we answer ‘yes’ to the first two questions and ‘no’ to the third one, we
can say that Caitlin knows that her dog is a Labrador.
166
But we did not ask whether her
belief is occurrent or merely dispositional. This suggests when one has a dispositional
belief that p and one satisfies the other conditions of knowledge, one knows that p. This
means that Alessandra knows that her dog is languishing in the hot car and that Joe
knows that he is putting his baby at risk by leaving the baby alone in the bathtub. If this
is right, then plausibly, Alessandra and Joe are not ignorant of the salient information.
It is open to the sceptic, at this point, to argue that the slogan behind P1 that
‘ignorance exculpates’ should be understood as ‘lack of conscious awareness
exculpates’. Modifying the rest of the argument, we get an argument one of whose
premises is: Not being consciously aware of the salient fact exculpates unless the agents
are blameworthy for their unawareness. But this is the very premise for which the
sceptic must provide justification.
4.2.2 Blameworthiness for Ignorance: Challenging P3
Let us now turn to P3 according to which the agents in non-tracing cases of
inadvertence are not blameworthy for their ignorance. Alexander and Ferzan think
that we are never blameworthy for our ignorance because we are never blameworthy
for “how we perceive, what we infer and what we believe” (2009b: 273). But this
requires justification. Moreover, other sceptics are open to the possibility that we can
be blameworthy for our ignorance. Zimmerman (1997), for instance, claims that
166
There may be another condition that must be satisfied for knowledge. Suppose I dispositionally
believe that the nearest branch of my bank is open on Saturdays and that this belief is true and justified
and I am not in a Gettier-style scenario. If the stakes are sufficiently high, I might not count as knowing
that the bank is open on Saturdays. For discussions of stakes and attributions of knowledge, see Keith
DeRose (1992), Jeremy Fantl and Matthew McGrath (2002), John Hawthorne (2004), and Jason Stanley
(2005). For criticism of the claim that stakes affect knowledge attribution, see Buckwalter and Schaffer
(2015) and Charity Anderson and John Hawthorne (forthcoming).
91
ignorance can exculpate if we do not have sufficient control over our ignorance. This
suggests that if one has sufficient control over one’s ignorance, one could be
blameworthy for such ignorance. In addition, Gideon Rosen argues that only blameless
ignorance exculpates. This means that we need to determine whether Alessandra, Joe,
and Scout are blameworthy for their ignorance. To do so, we need to have an adequate
account of epistemic blameworthiness. This is a largely neglected issue.
167
I do not take
up this project here, but one upshot of the discussion of the sceptical argument explored
in this section is that examining the conditions of doxastic blameworthiness may not
only be an important project in its own right, but that it may be important to our
attempt to answer questions about moral (non-epistemic) blameworthiness in non-
tracing cases of inadvertence.
168
Another issue we should explore is whether all ignorance exculpates or whether
there is a useful distinction to be made between blameworthiness of factual ignorance
and blameworthiness for moral ignorance. The kind of sceptic in whom I am interested
would want to appeal to this distinction between moral and nonmoral ignorance
because she must be able to say that we are blameworthy in advertence cases where we
intentionally, knowingly, or recklessly cause harm. But when we judged someone to be
blameworthy for intentionally harming someone, we do not inquire whether or not she
knew what she was doing was wrong. If moral ignorance can also exculpate, then we
seem to lose the distinction between the paradigmatic cases of blameworthiness, on the
one hand, and inadvertently causing harm, on the other.
Claiming that only nonmoral ignorance exculpates helps my sceptic to claim that
Alessandra in HOT DOG, Joe in FORGOTTEN BATH and Scout in COLICKY BABY are
not blameworthy. They were ignorant of the risks of harm that they were creating (at
the time they were creating the risk) and if we hold fixed their nonmoral beliefs, then
what they do was justified (since, according to them, they were not creating any risks of
harm).
169
The upshot of this view is that, in non-tracing inadvertence cases, when we
hold fixed their nonmoral beliefs, what they do is always justified. Hence, agents who
inadvertently cause harm are never blameworthy unless the tracing strategy is available.
So if it is true that nonmoral ignorance exculpates, then we can conclude that we are
never blameworthy in non-tracing cases of inadvertently causing harm.
But since the argument requires that there is this distinction between moral and
nonmoral ignorance such that only the former exculpates, then one could challenge the
argument by objecting to this premise. Indeed, Zimmerman (1997) and Rosen (2003)
have argued that both moral and nonmoral ignorance can exculpate if we do not have
167
One exception is Rik Peels (2013) who argues that doxastic blameworthiness entails the ability to
believe otherwise.
168
See Lloyd Fields (1994) who argues that although sometimes factual or nonmoral ignorance excuses,
we can be blameworthy for moral ignorance.
169
This view allows for the tracing strategy. A drunk driver can be blameworthy for causing harm if she
was aware of the risks when she started drinking so much that her capacities would be impaired.
92
sufficient control over our ignorance.
170
To be sure, when a person has some gruesome
normative beliefs (it’s okay to kill you if I feel like it, say), we have a very negative
reactive attitude to that person and maybe what we think of her as an agent is much
worse than someone who does the wrong thing despite having the right normative
attitudes (I knew it was wrong to kill you, but I was overcome by my anger, say). And
perhaps, we can be blameworthy for our moral ignorance because such ignorance
displays insufficient concern or wanton disregard. But this is a hugely controversial issue
and so the claim that we are never blameworthy for inadvertently causing harm should
not be used in a premise of an argument, including the Negligence Objection.
5 Objection Reconsidered
The reason why we have been considering blameworthiness for inadvertently causing
harm is because the claim that we are never blameworthy for inadvertently causing
harm is used in a premise of an argument that is an objection to my claim that criminal
law implicates blameworthiness. We saw in the last two sections that it is very
controversial whether or not we can be blameworthy for negligence. However, in this
section, I want to evaluate the plausibility of another premise in the Negligence
Objection according to which if criminal negligence can be committed by an agent who
is not blameworthy, then criminal law does not implicate blameworthiness.
My objection to this premise concerns the role that negligence plays in the criminal
law. Although all jurisdictions criminalise certain kinds of inadvertently causing death,
most crimes require that you intentionally, knowingly or recklessly cause the prohibited
result. Consider arson, for instance. Forgetting to turn off the stove and burning down
the house is not a crime even if it is a crime to purposefully refrain from turning off the
stove in order to burn down the house. That is, not all cases of inadvertently causing
harm are crimes. Unlike the other three mens rea, negligence does not satisfy the mens rea
element of crimes against property. Negligence does satisfy the mens rea element of
homicide.
171
However, negligence only satisfies the mens rea element of assault in a few
jurisdictions. Moreover, not all cases of inadvertently causing harm to a person are
crimes. For one thing, although some jurisdictions requires a failure to perceive “a
170
Zimmerman writes: “there are some who claim that it is easier to excuse moral ignorance that is due
to nonmoral error than moral ignorance that is due to moral error. I see no reason to think that this is
so. The argument [about ignorance] applies in both cases” (422-3). Rosen argues that “blameless moral
ignorance is a possibility” and that “insofar as [an agent] acts from blameless ignorance, it would be a
mistake for us to blame [her]” (66). Although Zimmerman and Rosen are skeptical of blameworthiness,
they are global sceptics, and not the kind of sceptics who are sceptical about blameworthiness for
negligence, but not sceptical about blameworthiness for intentional, knowing, or reckless wrongdoing.
171
This crime goes by the name ‘criminally neg1igent homicide’ in the United States and Canada, but
has other names in other jurisdictions. For example, in England and Wales, it is called ‘gross negligence
manslaughter’.
93
substantial and unjustifiable risk” that the harm will occur
172
, other jurisdictions require
wanton or reckless disregard
173
which may not be attributable to Alessandra, Joe, or
Scout.
In addition, although we have been treating cases of Alessandra, Joe and Scout as
non-tracing cases to make it plausible that we are never blameworthy for inadvertently
causing harm, we could regard these cases as tracing cases. And even if we should not
understand these particular cases as tracing cases, many cases of inadvertence that have
been deemed criminal negligence in the courts are, arguably, cases for which the
tracing strategy is available. Furthermore, the inconsistency in the treatment of what
appears to be non-tracing cases of negligence may simply indicate the controversial
nature of blameworthiness, rather than undermining the claim that criminal law
implicates blameworthiness.
174
Hence, showing that criminal negligence can be committed by someone who is
merely derivatively blameworthy does not undermine the claim that criminal law does
not implicate blameworthiness. It is true that there are defendants who have been found
guilty of criminal negligence even in non-tracing cases. However, the scope of this is so
small as to not really undermine the explanatory power of the claim that criminal law
implicates blameworthiness. In the next chapter, I consider accounts of responsibility
that are weaker than blameworthiness and hence candidates for the kind of
responsibility that are implicated in tort law.
172
See New York’s Penal Code, S15.05 (4) “Criminal negligence.” A person acts with criminal
negligence with respect to a result or to a circumstance described by a statute defining an offense when
he fails to perceive a substantial and unjustifiable risk that such result will occur or that such circumstance
exists. The risk must be of such nature and degree that the failure to perceive it constitutes a gross
deviation from the standard of care that a reasonable person would observe in the situation. See also,
MPC, 2.02(d).
173
See the Canadian Criminal Code, s219(1) which states that “Everyone is criminally negligent who (a)
in doing anything, or (b) in omitting to do anything that it is his duty to do, shows wanton and reckless disregard
for the lives or safety of other persons” (my emphasis).
174
Gene Weingarten (2009) discusses ‘death by hyperthermia’ cases which is similar to the case of
Alessandra in HOT DOG, except that the victim is a child. He notes the inconsistency among different
courts: “40 percent of cases authorities examine the evidence, determine that the child’s death was a
terrible accident -- a mistake of memory that delivers a lifelong sentence of guilt far greater than any a
judge or jury could mete out -- and file no charges. In the other 60 percent of the cases, parsing essentially
identical facts and applying them to essentially identical laws, authorities decide that the negligence was
so great and the injury so grievous that it must be called a felony, and it must be aggressively pursued.
94
Chapter 5: Responsibility and Tort Law
1 Introduction
The previous chapter examined whether or not one can be blameworthy for inadvertently
causing harm to another and what effect this has on the claim that criminal law
implicates blameworthiness. I hope to have established that our blameworthiness for
negligence (or lack thereof) does not undermine that claim, partly because the
interesting discussions about blameworthiness for negligence focus on non-tracing cases
of negligence and partly because negligence forms a small part of criminal law. In
contrast to criminal, the tort of negligence is regarded by many legal theorists as a
paradigmatic tort.
175
Moreover, in contrast to the controversial nature of the debate
about blameworthiness for negligence, it is not controversial whether we are responsible
(in some weaker sense) for our negligence. This supports the claim that tort law, unlike
criminal law, implicates an account of responsibility that is weaker than
blameworthiness. Moreover, when one is found liable under tort law for inadvertently
causing harm, one is liable to pay compensatory damages to the person who was
harmed. I take this as suggesting that we are imputing some kind of responsibility to
the tortfeasor. In this chapter, I outline some desiderata for the account of responsibility
that is implicated in tort law in the attempt to find the best account of responsibility
that is implicated in tort law.
2 Two Desiderata
As mentioned earlier, negligence is a paradigmatic tort. So to vindicate the claim that
tort law implicates responsibility (that is weaker than blameworthiness), it must be the
case that we are responsible for inadvertently causing harm. This is the first
desideratum for an account of responsibility that is implicated in tort law. The second
desideratum concerns strict liability torts. If the best explanation of tort liability involves
appealing to the claim that tortfeasors are responsible (just as the best explanation of
criminal liability involves appealing to the claim that criminals are blameworthy), then
those defendants who commit strict liability torts must be responsible. In the subsection
below (2.1), I take time to describe these cases because they are the cases where
somebody without experience of the law is perhaps least likely to attribute
responsibility.
One obvious account that satisfies both of these desiderata is an account of causal
responsibility. An example of this account of responsibility is proffered by Richard
175
See, for instance, Jules Coleman (2001) and Arthur Ripstein (1999, especially at 48).
95
Epstein (1973). He proposes that an agent can be responsible for, and is liable to pay
compensatory damages for, harm caused inadvertently because it is a harm that she
caused. Moreover, Epstein understands causation as but-for causation.
176
It is true that
every tortfeasor is responsible in this very minimal sense. But my aim in this chapter is
to identify the strongest account of responsibility that is implicated in tort law. After all,
an account of responsibility that satisfies the desiderata above will also be implicated in
criminal law since a defendant who is guilty of committing a crime will be responsible for
committing a crime. This is because blameworthiness entails responsibility.
177
However, responsibility does not entail blameworthiness. So if blameworthiness is
implicated by criminal law, but not by tort law, we can explain certain features of
criminal law (such as the mental states that are required for crimes and what count as
excuses in the criminal law). Moreover, the attribution of blameworthiness to an agent
for some harmful or otherwise wrongful outcome could ground the judgement that she
is liable to punishment. In contrast, the claim that one is merely responsible (in the
relevant sense that satisfies our two desiderata) for some harmful outcome, arguably,
does not ground the judgement that one is liable to punishment, even though it could
ground the judgement that one is liable to pay compensatory damages. When we have
these two different notions where one (namely, blameworthiness) is implicated in
criminal law and the other (namely, responsibility) is implicated in tort law, then this
can provide an answer to the Demarcation Question.
2.1 Strict Liability Torts
Before I examine some accounts of responsibility that are stronger that but-for
causation, let us consider strict liability torts. Recall Spano v. Perini (1969). There was no
attempt made to establish that the defendants were negligent; that they had failed to
exercise reasonable care when blasting. But the defendants could be held liable for the
damage to the property. That is, strict liability was imposed for “abnormally dangerous
activity” even though the harm was caused despite the fact that “the utmost care is
exercise to prevent the harm”. In all of these cases where strict liability is imposed, the
defendant’s agency is involved and so the defendant satisfies a minimal condition of
causal responsibility. After all, the agent who is engaging in abnormally dangerous
activity is the one who caused the damage. But in the next section, I discuss an account
of responsibility that is stronger than this minimal causal responsibility account.
176
But-for causation is a counterfactual theory of causation. To determine whether or not A caused B,
the theory urges us to determine whether if A had not occurred, B would not have occurred. For a fairly
comprehensive discussion of counterfactual theories of causation and some of its major challenges, see,
for instance, John Collins, Ned Hall and L.A. Paul (2004).
177
After all, one might think that being responsible is a necessary condition for being blameworthy.
96
3 Outcome-Responsibility
In the previous chapter, we encountered Hart’s analysis of ability to choose otherwise.
According to Hart, criminal punishment is justified only when the defendant, at the
time she acted, had “the normal capacities, physical and mental for doing what the law
requires and abstaining from what it forbids, and a fair opportunity to exercise those
capacities” (1968: 152). Stephen Perry proposes an account of responsibility, which he
calls Outcome-Responsibility, which is meant to echo Hart’s account.
178
According to
Perry, “[a]n agent is outcome-responsible for an outcome if and only if he causally
contributed to it, possessed the capacity to foresee it, and had the ability and
opportunity to take steps, on the basis of what could have been foreseen, to avoid it”
(81).
179
Perry claims that the “two criteria [that Hart outlines] -- the capacity to foresee
the harm and the ability, on the basis of such foresight, to avoid it” are “necessary and
sufficient conditions for responsibility for harmful outcomes” (2001: 88). However,
Perry overstates the analogy with Hart here, since he has another condition that is also
necessary for outcome-responsibility. This is the condition that is present even in the
minimal account of responsibility, namely, the condition that the agent is a cause of the
harmful outcome in question (81).
180
In the rest of this section, I discuss whether
outcome-responsibility satisfies the two desiderata outlined earlier.
Since outcome-responsibility does not require awareness on the part of the
defendant, it is a promising candidate for an account of responsibility that satisfies the
first desideratum. To see this, recall (from the previous chapter) the following
hypothetical:
FAVOURITE SONG: Eli’s favourite song comes on the radio while he is
driving to work. He takes his eyes off the road to change the
volume. He does not see the car moving into his lane in front of
him and hits it, injuring Jeanie, the driver of the other car.
Now we want to know if Eli is outcome-responsible for injuring Jeanie. We can go
through the three conditions that are necessary and jointly sufficient for outcome
responsibility.
1. Causation: This first condition is obviously met since Eli causally
contributed to Jeanie’s injury. After all, he hit Jeanie with his
car.
181
2. Foreseeability: Plausibly, Eli possessed the capacity to foresee the
injury. It seems plausible, absent further information to the
contrary, that Eli, at the time he changed the volume, had the
capacity to realise that fiddling with the radio and not paying
attention to the road could result in him causing an accident.
178
The term ‘outcome responsibility’ originally comes from Tony Honoré (1988).
179
Unless otherwise specified, all page references to Perry are to his (2001).
180
As we shall see, Perry’s understanding of the foreseeability condition differs from Hart’s.
181
Note that the notion of causation invoked here is counterfactual or but-for causation.
97
3. Avoidability: Eli not only could have foreseen the risk of harm, but
he also had the opportunity to avoid creating that risk. After all, he
could have refrained from fiddling with the radio. For all we know,
he was not under some compulsion to turn the volume up.
Hence, Eli satisfies the three conditions for outcome-responsibility and so he is
outcome-responsible for Jeanie’s injury.
182
But note that this account of responsibility does not entail that one is always
responsible when one inadvertently causes harm. To see this, let us consider the famous
case of Palsgraf v. Long Island Rail Road (1928) where the defendant (even though he was
acting negligently) was deemed not tortiously liable for a harmful outcome that he
caused. There, it was decided that the employee of the Long Island Rail Road (and,
given vicarious liability, the Long Island Rail Road itself) was not liable for the injury
to Palsgraf, the plaintiff, even though the employee was acting negligently (by pushing
a passenger into the train from the platform which he should not have done) and even
though he causally contributed to the injury to Palsgraf.
183
Cardozo, writing for the
majority, established that a necessary condition for liability for negligence was
foreseeability of the harm caused. But “there was nothing in the situation to suggest to the
most cautious mind that the parcel wrapped in newspaper would spread wreckage
through the station” (345). That is, the defendant in Palsgraf does not satisfy Perry’s
second condition of outcome-responsibility, namely, foreseeability. Since the kind of
harm that was caused to Palsgraf was not the kind of harm that was foreseeable as a
result of pushing the passenger, the defendant is not outcome-responsible for the injury
to Palsgraf and hence is not liable for that injury in tort law.
Perry distinguishes between circumstances in which the defendant should have
avoided creating the risk of harm and those in which it is not the case that the defendant
should have avoided creating the risk of harm. In the former cases, the defendant is
liable in tort law if she is also outcome-responsible for the harm. This class of
circumstances is meant to cover the tort of negligence (93).
184
The latter class of cases
covers strict liability torts: it is not that the defendant failed to do something that she
should have done, but rather that she is nonetheless outcome-responsible for the harm.
182
Note that even Alexander and Ferzan (our ‘sceptics’ from the last chapter) would say that Eli is
responsible, and even blameworthy, for the injury to Jeanie. This is because they endorse the tracing
strategy. One difference between their explanation for Eli’s responsibility and Perry’s is that Perry
delivers the result that Eli is responsible without relying on the tracing strategy. For Perry, it is sufficient
that Eli could have foreseen the harm and could have avoided that harm given its foreseeability. That
Eli chose to fiddle with the radio and that he is responsible for that decision is not necessary for the
judgement that Eli is outcome-responsible for Jeanie’s injury.
183
After all, it was the defendant pushing the passenger that caused the passenger to drop the package
containing fireworks, causing the package to explode, which in turn, caused the scales at the other end
of the platform to be knocked down, which injured Palsgraf.
184
Perry says that this is a case of fault. (See also page 119.) But I find that term problematic. It suffices
to note that there is a normative claim that the agent should have avoided the harm. Moreover, although
Perry has torts of negligence in mind, the formulation “agent not only could have avoided the [harmful]
outcome, but should have done so” equally applies to intentional torts.
98
He also distinguishes between risks that are jointly created and risks that are unilaterally
imposed. This distinction relies on the notion of a normal or acceptable pattern of
interaction. Risks that are jointly created are created in interactions that are normal or
acceptable. An example that Perry gives is walking and driving. Suppose that Dryden
is driving as carefully as can be expected. If so, the risks that he creates are within the
normal or customary range. And suppose that Pedro is walking nearby and there is
nothing that he is doing that is abnormal or risky. Given that the risks created by
Dryden and Pedro are within the normal range of risks attendant when there is both
driving and walking in the same area, the risks are jointly created.
Suppose that, despite both Dryden and Pedro refraining from acting in a way that
increases the normal risks associated with their activities (that is, both are acting non-
negligently), the normal risk materialises and someone gets injured. Say that a stray dog
suddenly runs out onto the road and Dryden swerves to avoid the dog causing the car
to skid onto the footpath and hit Pedro.
185
Dryden would not be liable for the harm in
tort law and Perry’s account predicts this because the risk of the harm is within the
normal range. That is, there is nothing that Dryden did that increased the risk of driving
(unlike Eli who increases the risk of driving above a normal range by acting negligently).
Hence, there is nothing that Dryden did that he should not have done.
However, according to Perry, he is outcome-responsible for the harm. After all,
Dryden satisfies all three conditions of outcome-responsibility: (1) Dryden is a cause of
the injury to Pedro; (2) Dryden, at the time he decided to drive, knew that driving could
result in him causing the accident. That is, given the normal range of risks associated
with driving, it is foreseeable that Dryden would cause harm; and (3) Dryden could
have refrained from driving. After all, he was not being compelled to drive. Dryden
satisfies the foreseeability condition because he had the capacity to realise that driving
comes with risks of harm. He writes: “From the general perspective of the activity he
was engaged in, namely, driving, an accident of this kind was clearly foreseeable and it
also could have been easily avoided (by not driving)” (123: footnote 27). Moreover,
Dryden satisfies the avoidability condition, not because there was something he could
have done at the time or just before he swerves to avoid the dog, but because he could
have refrained from driving altogether. That is, the avoidability condition is now also
interpreted very weakly as being able to avoid the activity that has foreseeable risks of
harm.
But given how he understands the two conditions, Pedro also satisfies them and is
outcome-responsible for the harm to himself. He had the capacity to foresee that
walking where cars are around could result in injury and he could have refrained from
walking there. This reveals how weak the notion of outcome-responsibility is. We can
contrast this with an alternative, stronger, notion whose foreseeability condition is
stronger according to which neither Dryden nor Pedro had the capacity to foresee the
185
This is an agent-regret case. A case like this involving a lorry driver was first introduced in this
dissertation in the last chapter, in Section 4.1.
99
injury. After all, there is a difference between being able to foresee the general dangers
of driving and walking near cars and being able to foresee that a stray dog will run out
onto the road. Furthermore, we could have a stronger notion of outcome-responsibility
whose avoidability condition is stronger also. We can ask whether Dryden, at the time
or just before he swerves to avoid the dog, could have avoided hitting Pedro by driving
differently (rather than asking whether Dryden could have avoided hitting Pedro by
refraining from driving altogether).
The example of Dryden and Pedro was an example where risks created by their
actions are risks that fall within the normal range of risks within a normal and
acceptable pattern of interaction. But some risks that arise even within a normal and
acceptable pattern of interaction like driving and walking can be unilaterally imposed if
“the defendant acted so as to increase the risk substantially” (199-120; my emphasis).
Recall Eli who fiddles with the radio when he should have refrained from doing so. He
does something that he should not have done and by doing so, he substantially increases
the risks associated with driving and pushes it over the normal and acceptable range.
So the risk that Eli creates is unilaterally imposed and since that risk materialised and
Eli is outcome-responsible for the harm to Jeanie, Eli is liable for the harm in tort law.
Another case that Perry discusses where the defendant is liable because she fails to
do what she should have done is a case where the defendant does not substantially increase
the risk, but increases it nonetheless. If it is also the case that the defendant could have
reduced or eliminated the risk at a relatively low cost, then she should have done what
was required to reduce or eliminate the increased risk. Perry cites Wagon Mound [No. 2]
[1967] to illustrate. The defendant owned a freighter ship named the Wagon Mound
that leaked some furnace oil into the harbour. There were some welders working on a
different ship and the sparks from the welders caused the leaked oil to ignite destroying
the plaintiff’s two ships that were moored nearby. The Privy Council found that a
reasonable person in the position of the ship’s engineer would have been aware that
allowing oil to be discharged imposed a risk of fire. Since the defendant could have
eliminated the risk of fire at a relatively low cost, the defendant was liable for the
damage to the plaintiff’s ships.
Perry’s analysis of this case goes thus: The ship’s engineer is outcome-responsible
for the harm caused to the plaintiff’s ships since he meets the three conditions:
(1) Since he was in charge of the Wagon Mound, he was a cause of
the leaking of the oil from Wagon Mound. And since the leaking
of the oil was a cause of the fire, he was a cause of the fire and the
damage to the plaintiff’s ships;
(2) At the time of the oil leak, he could have foreseen that allowing oil
to be discharged imposed a risk of fire; and
(3) He had the opportunity to avoid the risk. Even though the risk that
he increased was non-substantial, since he could have eliminated
the risk at a relatively low cost, he should have eliminated the risk.
Since he fails to do something that he should have done and he is
100
outcome-responsible for the harm, he is liable for the harm in tort
law.
So there are two kinds of cases where the defendant is liable for harm caused within a
normal pattern of interaction:
(i) The defendant substantially increases the risk associated with the
normal activity by doing something that he shouldn’t have done.
(Eli provided the example of this kind of a case.)
(ii) The defendant non-substantially increase the risk associated with
the normal activity, but she could have reduced or eliminated the
risk at a relatively low cost and hence she should reduce or
eliminate the risk. (Wagon Mound [No. 2] provided the example of
this kind of a case.)
This leaves us with the last class of circumstances where the defendant unilaterally imposes
risks and is liable for harm in tort law (if she is also outcome-responsible for the harm).
Here, the risks do not “originate within an established pattern of interaction at all” (75).
A paradigmatic example of this kind of risk is the kind of risk that is imposed when
engaging in an abnormally dangerous activity, such as blasting. Recall Spano which held
that the defendants engaging in blasting were liable for the damage even though no
negligence had been established. The defendants created risks that do not arise from a
normal activity since blasting does not fall within the pattern of normal interaction.
Hence, the defendants unilaterally created the risks of injury and property damage and
when the risks of property damage materialised, the defendants are liable for a tort
since they were outcome-responsible for the damage. (1) The defendants causally
contributed to the damage to the garages. (2) The defendants possessed the capacity to
foresee the injury since they knew the dangers involved with blasting. (3) The
defendants could have refrained from blasting and hence could have avoided the
damage. The defendants were outcome-responsible for the damage and the risk of
damage was unilaterally imposed since blasting is not a normal or acceptable pattern
of interaction.
Note how weak the foreseeability condition is. What sufficed for the defendants to
satisfy the foreseeability condition is not their capacity to foresee at the time of the
blasting that this particular act of blasting was likely to cause harm to the plaintiff’s
garages, but a more general capacity to foresee that blasting is dangerous and can
sometimes cause harm despite due diligence. Moreover, they satisfy the avoidability
condition, not because they could have engaged in the act of blasting in a different way
that would have avoided the harm, but because they could have refrained from
engaging in the activity of blasting.
However, these weak understandings of the foreseeability and avoidability
conditions make Perry’s conception of outcome-responsibility a fairly minimal account
of responsibility (although stronger than mere but-for causation). So, is there another
notion of responsibility that is thicker than the watered-down version of outcome-
101
responsibility that can satisfy both desiderata? In the next section, I examine Raz’s
account of responsibility and see if it can meet this challenge and if it can provide an
answer to the Demarcation Question.
4 Raz’s Account of Responsibility
Raz proposes and defends an account of responsibility – call it Responsibility R – according
to which we can be responsible for an action or an outcome in spite of lack of requisite
control. Raz calls this view the Rational Functioning Principle (RFP). According to it,
“[c]onduct for which we are (non-derivatively) responsibleR is conduct that is the result
of the function, successful or failed, of our powers of rational agency, provided those
powers were not suspended in a way affecting the action” (2011: 231).
186
According to
Raz, our being responsible and being held responsible is important because it informs
our “Being in the World” and the fact that “we are attached to our actions and their
consequences” (232).
187
Responsibility R is worth exploring as an account of responsibility that is implicated
in tort law for three reasons. First, Raz’s understanding of responsibility is not
equivalent to blameworthiness or liability to blame.
188
Hence, it is a promising
candidate for an account of responsibility that is implicated in tort law, but not in
criminal law. Moreover, Responsibility R is designed explicitly to deal with negligence,
so it satisfies our first desideratum. Indeed, as I show below, it provides a novel way of
delivering the result that we are non-derivatively responsible for our negligence. Third,
Responsibility R is stronger than Outcome-Responsibility. This is because although we
are outcome-responsible in agent-regret cases, we are not responsible R in agent-regret
cases. The accident of the kind in which Dryden, our careful driver, was involved
(namely, hitting Pedro as he swerves to miss the dog) is foreseeable and Dryden could
have avoided this foreseeable risk of harm by refraining from driving. Since Dryden
was clearly the cause of this harm, he is outcome-responsible for injuring Pedro. In
contrast, the injury to Pedro is not traceable to any malfunctioning of Dryden’s rational
capacities. Hence, he is not responsible R for the outcome. However, ultimately, I
conclude that Responsibility R cannot be an account of responsibility that is implicated
186
Raz actually writes that the defendant is ‘responsible 2’ for the harm and this is what I am calling
responsibility R. Raz distinguishes between three different “uses” of the term ‘responsible’ (227-8). The
first kind of responsibility (Responsibility 1) has to do with the general capacity for rational action (as in
‘he is not in his right mind and therefore not responsible for his actions’). The third kind (Responsibility 3)
had to do with duties (as in ‘it was your responsibility to secure the building’). I replace ‘responsible 2’
whenever it occurs in a quote with ‘responsible R’ to make it clear that I am talking about Raz’s account
of responsibility.
187
All parenthetical references to Raz are to his 2011.
188
Indeed, Raz laments the preoccupation with blame. He writes: “the preoccupation with praise and
blame, natural in our blame society, misses the central role of responsibility. We need to bear in mind
the full range of evaluation for which responsibility R is a condition” (265).
102
in tort law. This is because it fails to satisfy the second desideratum that we are
responsible for committing strict liability torts.
To see that we are responsible R for negligence, let us contrast RFP with what Raz
calls the Guidance Principle for a judgement of responsibility. According to the Guidance
Principle, “we are responsible R for φ-ing if and only if our φ-ing was guided and
controlled by our powers of rational agency” (229).
189
What is important about this
principle, according to Raz, is that it “makes responsibility R turn on successful guidance”
(230; my emphasis). But, although Raz thinks that this principle specifies a sufficient
condition for ResponsibilityR (231)
190
, he claims that we are also sometimes
responsible R for actions when the guidance fails; that is, when our actions are caused
not by successful function of our capacities of rational agency, but malfunctioning of those
capacities. Hence, the Guidance Principle is not necessary for responsibility R. That is, he
defends the view according to which we can be responsible for an action or an outcome
in spite of lack of requisite control.
RFP is meant to imply that we can be responsible R for inadvertently causing
harm.
191
Let me explicate this principle by applying it to Eli in FAVOURITE SONG who
fails to pay adequate attention and hits and injures Jeanie. Plausibly, the act of hitting
and injuring Jeanie is neither guided nor controlled by Eli. He did not intend to injure
or hit Jeanie and if he had his way, he would not have hit Jeanie. That is, he is not
responsible R under the Guidance Principle. But given RFP, he is still responsible R for
injuring Jeanie. This is because Eli’s hitting and injuring Jeanie is the result of the failure
of Eli to exercise his rational capacities since driving carefully and paying adequate
attention to the road is within Eli’s domain of secure competence. This notion of competence
is crucial to understanding and justifying RFP.
192
According to Raz, the domain of
secure competence is the domain in which I can not only succeed, but trust myself to
succeed. And what we do successfully in our domains of secure competence can be
189
I omit many of the details about how this principle is to be applied. In particular, I omit the two
conditions that Raz sets out that must be met for an action to be guided and controlled by our powers of
rational agency.
190
For some remarks on whether this is correct, see Gary Watson (2014: §2).
191
Of course, there are other (non-negligent) kinds of malfunctioning of our rational capacities for which
we are also responsible R. Watson outlines how Raz’s account can also show that we are responsible in
cases of weakness of will and (failed) attempts (2014: §10).
192
Raz justifies RFP by claiming that “we hold ourselves and others responsible R for actions within our
respective spheres of secure competence, and we do so even when actions within the domain fail” (245).
Moreover, he argues that this practice of holding ourselves responsible R in this way is justified by the role
that the practice plays in “maintaining our sense of who we are, and of our relations to the world” (245).
I find this justification of the RFP wanting. Here is an analogy to illustrate my problem. Most people
think that we can sometimes act out of altruism and our practice of gratitude perhaps reveals this. But
even though this practice plays an important role in maintaining our sense of who we are, and of our
relations to the world, that does not seem to be a good argument for the view that altruism is possible.
There is much more to be said about Raz’s justification of the RFP. For the purposes of my project,
however, I focus on whether responsibility R is an account of responsibility that is implicated by tort law.
103
contrasted with what we successfully do as a matter of luck.
193
Since Eli trusts himself
to drive successfully (without injuring anyone), driving is within his domain of secure
competence. And his failure to drive without injuring anyone is conduct for which he
is responsibleR.
Moreover, Eli’s powers of rational agency were not suspended in a way affecting
the action. Again, this idea is explicated by appealing to the notion of competence. Eli
is responsible R for injuring Jeanie only if there are no competence-defeating conditions. Raz
does not specify what these are though he says that they “suspend
194
one’s ability to
function as a competent rational agent” (249). He also provides various examples of
such conditions along the way. These include sleepwalking, hypnosis, paralysis,
earthquakes, seizures, drops in blood pressure, biased teachers, and frailty. Since there
are no competence-defeating conditions present for Eli, Raz delivers the result that Eli
is responsible R for his negligence.
195
However, as I mentioned above, Responsibility R cannot be an account of
responsibility that is implicated in tort law. This is because we are not responsible R for
committing strict liability torts.
196
Indeed, Raz sets aside strict liability torts and
distinguishes them from the tort of negligence precisely on the grounds that the duty to
compensate for harm under strict liability does “not depend on establishing that the
defendant is responsible R for the harm” (260). Instead, Raz regards strict liability in law
as a way of fairly distributing ‘‘risk of liability without attributing responsibility’’ (259).
To see why we are not responsible R in strict liability tort cases, recall the defendants in
the blasting case of Spano v. Perini (1969). Although the damage to property in the
vicinity of the construction site is foreseeable, this outcome is neither intended nor
foreseen. Moreover, this outcome does not result from malfunctioning of the
defendants’ capacities of agency. Since these defendants were tortiously held liable for
the outcome, tort liability cannot be explained by ResponsibilityR.
5 Concluding Remarks
In this chapter, I searched for the strongest account of responsibility that is implicated
in tort law. I argued that Hart-inspired account of outcome-responsibility is the
193
Suppose I am trying archery for the first time. The first couple of times I shoot, I manage to hit the
target. But I do not trust myself to hit the target the next time. I think that I got lucky. So, archery is not
within my domain of secure competence and I am not responsible R for hitting or failing to hit the target.
194
He also speaks of our rational capacities or competences being ‘blocked’. (See 228, 231, and 243.)
195
Of course, not all occurrences of seizures excuse the agent from responsibility. Recall People v. Decina
(1956) discussed in the last chapter where the defendant was found criminally liable for harm caused by
a seizure because he knew that he suffered from seizures. But, presumably, the defendant in that case is
derivatively responsible R even if, he is not non-derivatively responsible R.
196
This is not to claim that Raz’s account is otherwise without problems. Indeed, one major worry with
his account is the unclarity about how to understand competence-defeating conditions that play a crucial
role in explicating responsibility R. See Watson (2014: §15).
104
strongest account of responsibility that satisfies the two desiderata, namely that we are
responsible for negligence and that we are responsible when we commit strict liability
torts. I showed that Raz’s account satisfies the first, but not the second, desideratum.
My answer to the Demarcation Question, then, is that blameworthiness is
implicated in criminal law but that only outcome-responsibility is implicated in tort law.
In the next chapter, however, I argue that the distinction between blameworthiness and
outcome-responsibility cannot, by itself, provide an adequate answer to the
Demarcation Question. This is because blameworthiness is strictly stronger than
responsibility. That is, blameworthiness entails outcome-responsibility. Hence, if the
distinction between criminal and tort law is to be explicated solely by an appeal to the
distinction between blameworthiness and responsibility, then criminal liability would
entail tortious liability. But it is not true that whenever one is criminally liable, then one
is tortiously liable. This is because there are some rights that are protected by criminal
law that are not protected by tort law. Attempting to injure or kill someone could
constitute crimes even though one cannot be liable in tort law for attempts.
197
Moreover, not all cases of blameworthiness are criminally liable. One could be
blameworthy for intentionally inflicting emotional distress on someone, but one would
not be criminally liable since criminal law does not protect the right not to be
emotionally harmed (without also being physically harmed). Hence, in the next and
final chapter, I try and see if the responsibility-based distinction can serve to demarcate
between tort and criminal law if it is supplemented by a different distinction that pays
attention to different rights that are protected by the two domains.
197
Along with the inchoate offense of criminal attempts, another inchoate offense, namely that of
conspiracy is also an example where the agent can be blameworthy and hence responsible for the
outcome but the agent would only be criminally liable and not tortiously liable.
105
Chapter 6: Rights and Responsibility
My Answer to the Demarcation Question
1 Introduction
The driving thought in this dissertation has been that the Demarcation Question can
be answered by distinguishing between two different accounts of responsibility. In
particular, the answer defended claims that criminal liability implicates
blameworthiness while tortious liability implicates outcome-responsibility, an account
of responsibility that is weaker than blameworthiness.
In Chapter 4, I took seriously one obstacle to thinking that criminal liability
implicates blameworthiness, namely the fact that we can be criminally liability for
negligence. I criticised two different accounts of blameworthiness that deliver the result
that we are sometimes blameworthy for negligence. Moreover, I have attempted to
show the limited way which the existence of criminal negligence undermines the
explanatory power of the claim that criminal law implicates blameworthiness. First,
many criminal negligence cases are cases for which the tracing strategy is available and
hence defendants in those cases are derivatively blameworthy for negligence. Second,
most crimes require that you intentionally, knowingly or recklessly cause the prohibited
result and only death and very serious bodily harm caused inadvertently trigger
criminal liability in most jurisdictions.
In Chapter 5, I focused on tort law and explored different accounts of
responsibility that are weaker than blameworthiness. In the course of that discussion, I
put forward two desiderata that the account of responsibility that can explain tort
liability. First desideratum required that we are responsible for negligence, and the
second required that we are responsible when we commit strict liability torts. The main
aim of the previous chapter was to identify the strongest account of responsibility that
satisfies both desiderata. I argued that Raz’s account of responsibility cannot satisfy
both desiderata because we are not responsible R for committing strict liability torts. I
also considered different understandings of foreseeability and avoidability conditions
that are appealed to by Hart and Perry. If the avoidability condition is understood as
requiring that the agent can avoid causing harm by taking reasonable care, then the
defendants who are strictly liable will not count as being responsible. Hence, Perry
understands avoidability more weakly and his account of outcome-responsibility
delivers the result that a defendant who is strictly liable is responsible. This is because
the avoidability condition is satisfied if the defendant had the capacity to refrain from
engaging in the activity in question. I suggested that this is a weak account of
responsibility, but that this is the strongest account of responsibility that can satisfy both
desiderata and hence the strongest account of responsibility that can be implicated in
tort law.
106
However, I concluded the last chapter by suggesting that solely appealing to the
distinction between blameworthiness and outcome-responsibility cannot provide an
adequate answer to the Demarcation Question. In this chapter, I further explicate my
reasons for this claim and how this distinction can be supplemented to provide an
adequate answer. In particular, I argue that we can arrive at a good answer by
supplementing with an account of different rights that are protected by the two domains
or an account of the different standards of conduct that are required by the two
domains. Accordingly, my answer to the Demarcation Question – call it the
Responsibility-Rights Answer – is two-pronged. The first prong appeals to the
distinction between blameworthiness and outcome-responsibility. The second prong
appeals to the differences in the rights that are protected by criminal law and tort law.
I show how supplementing the responsibility-based distinction with a rights-based
distinction can provide an adequate answer to the Demarcation Question by
recognising that there are two different ways in which tort law and criminal law can be
distinguished.
2 Pitfalls of the Blameworthiness-Responsibility Answer
One main reason why solely appealing to the distinction between blameworthiness and
outcome-responsibility cannot provide an adequate answer to the Demarcation
Question is because blameworthiness entails outcome-responsibility. This is a problem
for this answer – call it the Blameworthiness-Responsibility Answer – because criminal
liability does not entail tortious liability. There is an obvious way in which criminal
liability does not entail tortious liability which has to do with legal procedure. Suppose
Caitlin is a bank robber who takes Elroy, a customer at the bank, as a hostage and locks
him in a room to compel the bank employee to unlock the safe. Further suppose that
Caitlin is arrested by the police and is charged with a crime and the prosecution is able
to prove beyond a reasonable doubt that she committed the crime.
198
Caitlin is criminal
liable for falsely imprisoning Elroy and taking him as a hostage. But suppose Elroy does
not want to sue Caitlin for false imprisonment because he wants to forget the traumatic
incident as soon as possible. In that case, Caitlin would not be found tortiously liable
since in order for that to happen, Elroy, as the plaintiff must bring forward the suit. So
Caitlin who is criminal liable is not tortiously liable. However, although this is a
counter-example to the claim that criminal liability entails tortious liability, it is not a
198
A relevant legislation may be the United States’ Hostage Taking Act which makes taking hostages a
federal crime. See Title 18 of the United States Code, §1203(a) which states that “whoever, whether
inside or outside the United States, seizes or detains and threatens to kill, to injure, or to continue to
detain another person in order to compel a third person or a governmental organization to do or abstain
from doing any act as an explicit or implicit condition for the release of the person detained, or attempts
or conspires to do so, shall be punished by imprisonment for any term of years or for life and, if the death
of any person results, shall be punished by death or life imprisonment”.
107
counter-example to the spirit of that claim since Caitlin would be tortiously liable for
false imprisonment had Elroy sued Caitlin. Moreover, there are more compelling
counter-examples to the claim that criminal liability entails tortious liability.
One such counter-example is that of criminal attempts.
199
Here is an example of
a criminal attempt.
ATTEMPTED ARSON: Ember intends to burn a building that houses a
company that fired her recently. She buys a lighter and the day’s
newspapers. She has these items in her possession and is on her
way towards the building with the goal of burning the building
down. But she is stopped just in front of the building as an alarm
has gone off in the building. The police search her as she is acting
suspiciously and Ember is arrested for attempted arson.
200
Ember would be criminally liable and she is blameworthy for choosing to take the
means to burning a building. However, she would not be tortiously liable because there
is no tort of attempt.
Another kind of counter-example to the claim that criminal liability entails tortious
liability comes from the so-called public-order crimes.
201
These include prostitution and
use, sale, and possession of certain prohibited drugs. Whether or not they should be
criminalised is a controversial question, but even when one is criminally liable for, say,
having in possession certain amount of prohibited drug, one would not be tortiously
liable for it. In addition, there are crimes against animals, but no tortious equivalent.
Cruelty to animals and wildlife smuggling are criminal, but there is no tortious liability
for harming an animal.
202
Another kind of reason why Blameworthiness-Responsibility Answer fails has to
do with the fact that not all acts or outcomes for which we are blameworthy trigger
criminal liability. Suppose Enya is blameworthy for causing emotional distress to Odo.
Enya would not be criminal liability because criminal law does not protect the right not
to be emotionally distressed.
Similarly, not all acts or outcomes for which we are outcome-responsible trigger
tortious liability. Suppose I publish an offensive book
203
and that this causes Pierre to
be offended and it is foreseeable to me that my publishing the book would offend him.
Suppose I knew that publishing the book would cause Pierre offence, but I do it, not in
spite of this, but because of it (as he has done to greatly annoy me and I want to offend
199
Attempt is an inchoate crime along with conspiracy, incitement and solicitation. There is a tort of
conspiracy although the paradigmatic example of civil conspiracy occurs in the context of business. Also,
see Steven Weingarten (1984) who argues against the current law that does not allow tort liability for
incitement.
200
This example is a modified version of an example of attempted arson given by Michael Moore (1993:
18).
201
These crimes are sometimes referred to as ‘victimless’ crimes.
202
There is tort liability for failing to take reasonable care to prevent one’s own animals from harming
other human beings. But there is no tortious liability for causing harm to animals.
203
This example was first introduced in Chapter 1, Section 2, footnote 4.
108
him). Since I could have avoided offending Pierre by refraining from publishing the
book, I outcome-responsible for causing offence to him, but I would not be tortiously
liable.
204
This is because mere offence is not the kind of harm that is protected by tort
law.
205
3 Responsibility-Rights Answer
The discussion in the previous section suggests that the Blameworthiness-Responsibility
Answer must be supplemented with an account of the different rights that are protected
by criminal law and tort law. Hence, in this section, I put forward the Responsibility-
Rights Answer to the Demarcation Question that pays attention to both the distinction
between blameworthiness and outcome-responsibility and the differences between the
rights that are protected by the two domains.
206
Chapters 3 to 5 have focused on the
responsibility prong. But in order to fill out the details of this answer, we need to figure
out what rights are recognised and protected by tort law and criminal law.
3.1 The ‘Rights’ Prong
Tort law protects a wide variety of rights including the right to not to be harmed, right
to exclusive possession of one’s property, right not to be defamed, and right not to be
falsely imprisoned. Greg Keating (2012) claims that a more unifying story can be told
about these different rights and that tort law is fundamentally concerned with two
different kinds of rights: the right to be free from harm (which grounds harm-based
torts) and the right to autonomy (which grounds what he calls ‘sovereignty-based’ torts).
I do not take issue with this claim. I do not ask whether all of the rights that are
protected by tort law fit one of these two general rights. Nor do I ask what
understanding of harm and sovereignty is required to vindicate Keating’s claim. This
is partly because an account of rights that forms a part of the Responsibility-Rights
Answer to the Demarcation Question must be more specific than this. After all,
criminal law is also concerned with the right to be free from harm and the right to
autonomy.
207
Hence, the approach I prefer is to examine the elements of different torts
204
As before, I am imagining the example, although this book causes offence, it is not libellous.
205
This shows that there is a problem that Ripstein has (other than the one raised in Chapter 2). He says
that when you take a risk, you are liable for a tort. But not all takings of (unreasonable) risks trigger tort
liability. (I can take an unreasonable risk that in saying something, I will hurt your feelings, but I do not
incur any tortious liability because tort law does not recognise or protect a general right not to have our
feelings hurt.)
206
Recall that legal rights (as I understand them) are correlative to legal duties. Accordingly, if I have a
right to φ, this means that you have a duty to refrain from preventing me from φ-ing.
207
It is true that criminal law may also be fundamentally concerned with the duties derived from justice
(as evidenced by crimes against justice such as obstruction of justice and perjury, among others). In
109
and crimes. This is also in line with my methodology for approaching the Demarcation
Question.
208
As mentioned, both criminal law and tort law recognise and protect a right to be
free from physical harm. Which fine-grained rights under the broad heading of ‘right
to be free from physical harm’ are protected by the two domains? We have already
collected some data to answer what rights are protected by tort law. There are many
torts that are concerned with physical harm: the intentional tort of battery and the tort of
negligence, as well as strict liability torts. However, tort law does not protect a right not
to be physically harmed simpliciter. We have seen that battery protects a right to be free
from nonconsensual contact and the right to be free from physical harm that results
from such nonconsensual contact. We have also seen that strict liability torts protect
the right not to be harmed by someone engaging in an abnormally dangerous
activity.
209
The tort of negligence also protects the right not to be physically harmed, but only
in cases where the person fails to take reasonable care. We should note that one can fail
to take reasonable care even though one acted to the best of one’s own. This is because
tort law endorses the Reasonable Person Standard. The seminal case that illustrates
this standard is Vaughn v. Menlove. Menlove, the defendant, built a hay rick in a way that
risked the haystack igniting spontaneously and this hay rick bordered the plaintiff’s
land. It was held that a reasonable person, a person of “ordinary prudence” would not
build the haystack interest eh way that Menlove did and that she would not place the
hayrick so close to the neighbouring property. That is, the tort of negligence required
the defendant to conform to a particular standard of conduct. Hence, the tort of
negligence protects a right not to be physically harmed by someone failing to conform
to the objective standard of conduct that is pertinent to the activity in question.
210
addition, criminal law is also concerned with duties to maintain public order (as evidenced by public-
order crimes against prostitution, public indecency, recreational drug use, among others).
208
This methodology is employed by Arthur Ripstein. See Chapter 2.
209
One might argue that strict liability recognise the right to be compensated for the harm caused by
someone engaging in an abnormal activity (although it does not recognise the right not to be harmed by
someone engaging in an abnormal activity). I have reasons for thinking that the reasons for imposing
liability in these cases cited by the courts support the right not to be harmed, but not much hangs on this
for the purposes of the dissertation. The important point is that tort law does not protect a right not to
be harmed simpliciter and hence we must work out which fine-grained rights are protected by tort law by
looking at the elements of various harm-based torts.
210
It was held that a claim of negligence cannot be defeated by showing that the defendant had acted to
the best of his own judgement; that he did not himself foresee that the haystack would ignite and cause
damage to the plaintiff’s property. This shows that we must understand the foreseeability condition
objectively. One satisfies the foreseeability condition if the harm or the risk of harm is reasonably
foreseeable. This makes the notion of outcome-responsibility even weaker. See Ripstein (1999: Chapter
4) who also argues for understanding Perry’s foreseeability condition objectively, as foresight of a
reasonable or ordinary person. The objectivity of foresight that is required for an adequate account of
responsibility that is implicated in tort law also shows why Raz’s account of responsibility is too strong.
To see this, note that the harm caused to Vaughn is not due to a malfunctioning of Menlove’s capacities
of rational agency.
110
One upshot of determining which fine-grained rights are protected by tort law is
that the Responsibility-Rights Answer can deliver the correct result that there is no tort
liability in the agent-regret case involving Dryden and Pedro described in the last
chapter. Recall that Dryden is driving as carefully as can be expected but causes Pedro
harm through no fault of his own. We saw that Dryden is outcome-responsible for the
harm. But, he is not tortiously liable for the harm and this is because since he is driving
as carefully as can be expected, he neither violates the right not be physically harmed
by someone failing to conform to the objective standard of conduct that is pertinent to
the activity of driving nor the right not to be physically harmed by someone engaging
in an abnormally dangerous activity (since driving is not an abnormally dangerous
activity).
211
We should also consider what other rights (other than the right to be free from
physical harm) are recognised by tort law and criminal law. Tort law recognises the
right to be free from emotional distress, as evidenced by the tort of intentional infliction
of emotional distress. However, criminal law does not recognise such a right. In
addition, criminal law recognises a duty not to attempt to cause harm whereas tort law
does not recognise such a right. The Rights prong of my Responsibility-Rights Answer
can explain these facts. Moreover, it can explain why one may be tortiously liable but
not criminally liable for breaching a right that is protected by both domains.
4 Objections to the Responsibility-Rights Answer
In this section, I examine two objections to my answer to the Demarcation Question.
According to the first objection, the mere laundry list of rights that are protected by the
two domains is unsatisfactory. According to the second, if we have sufficiently fine-
grained rights that are protected by the two domains, there is no need to appeal to the
distinction between blameworthiness and outcome-responsibility to answer the
Demarcation Question.
4.1. Triviality Objection to the Rights Prong
One objection to the Rights prong of my Responsibility-Rights Answer is that it fails to
provide an account of kinds of rights that are protected by tort law and criminal law. It
would be better, so the thought goes, if there was some overarching principle of why
tort and criminal law protect the rights that they do, rather than some other rights in
the neighbourhood. One might urge that there are more general principles that can
explain these facts, such as a particular conception of autonomy that tort law is
interested in or the notion of political community that is implicit in criminal law.
211
This also shows that in order to understand precisely which right is recognised by strict liability torts,
we have to determine which activities tort law deems to be abnormally dangerous.
111
I agree that it would be better if there is a more general principle that can explain
the fact that law protects the rights that it does and it would be great if there are two
(or more) distinct principles that can explain both the similarities and the differences in
the rights that are protected in the two domains. One such general characterisation of
the rights in the offing is that tort law protects private rights whereas criminal law
protects public rights. However, we have seen the objections to the various plausible
specifications of publicness and privateness.
But perhaps there are other characterisations of the difference in the kinds of rights
that are protected. One natural thought is to think that tort law is, at bottom, concerned
with harm and this can explain the absence of tortious liability for attempts. However,
criminal law is also concerned with harm. But perhaps there is a difference. Here is a
way of characterising the difference between tort law’s response to harm and criminal
law’s response to harm: Tort law is concerned exclusively with harm whereas criminal
law is only derivatively concerned with harm because it is concerned, in general, with
failures to treat others with the respect that they are owed.
However, to vindicate this thesis, we must get clearer about how to understand the
harm with which tort law is exclusively concerned. After all, tort law protects against a
variety of harms. Not only does tort law protect against physical harm, but it also
protects against emotional distress and harm to one’s reputation as well as financial
harm
212
. Moreover, arguably, tort law is also concerned with failure to treat others with
the respect that they are owed. Take, for instance, the so-called “dignitary torts” which
include defamation, invasion of privacy, breach of confidence and malicious
prosecution. In addition, the torts of battery and false imprisonment are concerned with
our right to bodily integrity and not just physical harm. One can be nominally liable
for battery and false imprisonment even though they do not cause any physical injury
to the plaintiff. Furthermore, tort law, like criminal law, is concerned with some harm
because it is the result of a failure to treat another with the respect that she is owed. The
harm caused by nonconsensual contact is recoverable in tort law because the harm is a
result of nonconsensual contact. One difference between criminal and tort law may be
that criminal law, unlike tort law, is concerned, in general, with culpable failures to treat
others with the respect that they are owed. I agree with this. However, this does not
undermine a need for more specific accounts of rights that are protected by the two
domains.
In addition, it is not sufficient to claim that tort law is exclusively concerned with
harm as there are some harms with which tort law is not concerned. We saw that
driving is not considered to be an abnormally dangerous activity by tort law and hence
there is no right not to be injured by someone driving so long as the driving conforms
to the relevant standards. It is conceivable, with the development of technology that
makes feasible self-driving cars, that one day, tort law will begin to regard driving
manually (that is, without the aid of the technology of self-driving cars) to be an
212
as evidenced by various economic torts.
112
abnormally dangerous activity. Also, although emotional harm is currently protected,
tort law did not always concern itself with emotional distress. My approach to filling
out the details of the Rights prong allows the flexibility to fill out the details and keep
up with the evolution of the law.
I think that the changeable nature of the rights, together with some theoretical
unity provided by the accounts of responsibility can help explain why the Demarcation
Question itself seems both pressing and intractable. That is, we can begin to appreciate
why some have given up on there being a principled answer to the Demarcation Question
(that doesn't merely provide a laundry list of some of the differences). And we can
appreciate why some (myself included) have been drawn to the thought that there is
some stable distinction between the two that can help make sense of certain features of
tort and criminal law.
But one might now wonder if any difference between tort and criminal law can be
cashed out in terms of the different rights that are protected under the two domains. If
we had a more precise understanding of which rights are protected by the two domains,
could we accommodate the fact that there would be tortious liability, but not criminal
liability, by simply appealing to the rights, without looking at different accounts of
responsibility? The next section is devoted to exploring this objection to the composite
account.
4.2 An Objection to the Responsibility Prong
According to this objection, Responsibility-Rights Answer is inadequate because the
Responsibility prong is redundant. After all, according to the Responsibility-Rights
Answer, we can explain tortious liability by determining (1) whether or not the
defendant breached a right that is protected by tort law; and (2) whether or not the
defendant was outcome-responsible for breaching the right. The objection is that that
once we determine that the defendant breached a tortious right (a right that is protected
by tort law), then we do not need to ascertain whether or not she was outcome-
responsible. This is because, the objection supposes, one does not count as having
breached a tortious right unless one was outcome-responsible. That is, in order to
breach a tortious right, one must be outcome-responsible. If this is correct, then all we
need is the Rights prong.
However, there are cases where one is not tortiously liable because one is not
outcome-responsible. Recall that in Palsgraf, the defendant was found not liable for the
injury to the plaintiff even though he violated a right protected by tort law because he
failed to discharge his duty to conform to the relevant standard of conduct by pushing
a passenger onto the train from the platform. However, the claim that tortious liability
implicates outcome-responsibility can explain why the defendant was not liable: the
defendant did not satisfy the foreseeability condition of outcome-responsibility. The
113
kind of harm that the plaintiff suffered was not the kind of harm that was reasonably
foreseeable as a result of pushing the passenger.
The objector may respond by claiming that there is an even more fine-grained
right that is protected by the tort of negligence and that the defendant in Palsgraf does
not breach that right. This right is the right not to have harm h inflicted on us by action
a where h belongs to a type of harm t and t is reasonably foreseeable when engaging in
a. The defendant in Palsgraf does not infringe this right since the harm that the plaintiff
suffers is not reasonably foreseeable.
However, this does not undermine the need for the Responsibility prong of my
answer to the Demarcation Question. This is because it is still worth distinguishing
between cases where there is no breach because we do not think the kind of act
performed is one that people have rights against and cases where there is no breach
because the responsibility condition is not met. That is, even if we can put the features
of responsibility into our descriptions of the rights, this seems a mere terminological
move rather than a substantive one. After all, there are two different kinds of
explanations for why there is no breach of a right in a particular situation. To make
this more vivid, imagine that Armando is offended by Estella’s use of a racial slur. Even
though her action is morally problematic, it does not breach a right that is protected by
tort law. In contrast, sometimes there is no breach because there is no one who is
responsible for the right-violating caused. Imagine that Estella used a racial slur but
that this was a symptom of her Tourette’s syndrome. Even if tort law was changed so
that it protected the right not to be offended by racial epithets, Estella would not be
liable given that she is not outcome-responsible. (Plausibly, Estella cannot satisfy the
avoidability condition.)
Moreover, we saw that insanity and duress count as excuses in criminal law but
not in tort law. To the extent that this can be explained by appealing to the distinction
between criminal law and tort law, this explanation will be available even if the fine-
grained descriptions of rights include these excusing conditions. That there are these
two kinds of explanations is particularly important because we can imagine the law
evolving to protect different kinds of rights. However, to change the law so that an
injury caused by someone under hypnosis, say, counts as a breach of a right protected
by tort law would be a fundamental change to tort law and it might be the kind that
means that the change has created a very different kind of institution.
Furthermore, appealing to different accounts of responsibility can help to explain
the radically different responses to the violation of a right given by tort law and criminal
law. One of the data to be explained in answering the Demarcation Question is that a
tortfeasor is, as a default rule, liable to pay compensatory damages whereas the criminal
is, as a default rule, liable to punishment. Even if the rights can described in a fine-
grained way, this difference is plausibly explained by the distinction between
blameworthiness and outcome-responsibility.
114
In tort law, the parties to a suit are the defendant and the plaintiff. If the defendant
is found liable for a tort, then the defendant is found to have violated a right of the
plaintiff. One important question to answer when faced with this rights-violation is to
think about how the wrong should be addressed. A natural thought is that the person
who violated the rights should try and undo the wrong. As argued for in Chapter 2, the
fact that tort law requires the tortfeasor to compensate for the loss suffered by the
plaintiff can be explained the defendant is responsible for violating the right (and the
plaintiff is not responsible for violating the right
213
). Hence, given that the defendant is
responsible for the rights-violation and the fact that someone must bear the cost of the
rights-infringement, we can make sense of the fact that defendant is required by tort
law to bear the costs of the rights-violation.
Moreover, the claim that criminal law implicates blameworthiness can explain the
punitive nature of the response to criminal wrongdoing. Given that punishment is
intentional infliction of pain, it makes sense that criminal law requires defendants to be
blameworthy for the crimes that they commit.
214
This goes hand in hand with the
thought that being coerced to commit a crime under duress mitigates one’s
blameworthiness and hence the severity of the punishment.
5 Conclusion
This chapter attempted to defend the Responsibility-Rights Answer to the
Demarcation Question. I do not claim that this answer can explain each and every
difference between tort law and criminal law. What I have hoped to show is that it has
enough resources to be able to highlight some of the main differences as well as be able
to pinpoint when tort and criminal law share some similarities (by protecting the same
right or a similar right).
My answer can also overcome some of the obstacles faced by the Corrective-
Retributive Answer examined in Chapter 1. One objection that I raised against it was
that the principles of corrective and retributive justice are second-order principles and
so merely appealing to the distinction between corrective and retributive justice cannot
explain the fact that different rights are protected by the two domains.
215
My answer,
in contrast, can explain this and hence it is not subject to an incompleteness objection
213
An exception might be the case of contributory negligence where, arguably, the plaintiff is outcome-
responsible for violating a right.
214
Of course, there are competing explanations of the punitive nature of the criminal law’s response. In
particular, one might think that it makes equal sense that criminal law requires that the infliction of pain
has some deterrent value. I agree with this if we are focusing on the punitive nature of the criminal law's
response. But as noted in Chapter 3, there are other aspects of the criminal law that indicate that criminal
law implicates blameworthiness. Moreover, the claim that criminal law implicates blameworthiness can
explain other non-punitive responses that are available in criminal law. I return to this point in Section
5.
215
See Chapter 1, Section 3.1.
115
that is faced by the Corrective-Retributive Answer. Furthermore, my answer is also not
subject to another kind of incompleteness objection faced by that answer. This is
because although the principle of corrective justice cannot make sense of injunctive
remedies, my outcome-responsibility does not rule out injunctive remedies.
216
Similarly, although the principle of retributive justice cannot make sense of non-
punitive response, the claim that a criminal is blameworthy for committing the crime
does not rule out the intelligibility of keeping a criminal record, or requiring that the
criminal undergo drug rehabilitation to prevent her from reoffending.
One dissatisfaction with my answer to the Demarcation Question may come from
the fact that it tried to accommodate two opposing viewpoints. Some have argued that
the differences between the two domains are arbitrary and products of historical
accidents. Hence, it may be thought that to provide a principled and coherent
framework that makes sense of the differences as well as the similarities is a fool’s errand.
But I have taken on this fool’s errand. In contrast, others have approached the
Demarcation Question by looking at the function of criminal law and tort law and have
argued that those general functions can give us all we need to answer the Demarcation
Question. But I have rejected this approach in favour of an approach that takes a closer
look at the details of substantive criminal and tort law.
However, I think that the Responsibility-Rights Answer, rather than being an
unstable position, has two advantages. First, it can accommodate the grain of truth in
the claim that the law sometimes evolves haphazardly. Why is it, for instance, that the
right to be free from emotional distress is recognised in tort law, but not in criminal
law? Given my approach to filling out the details of the Rights prong of the
Responsibility-Rights Answer, I do not attempt to answer that question. This is a
normative question and a job for moral and political philosophers to determine which
rights should be protected by the law.
This brings us to the second advantage. The Responsibility-Rights Answer is an
answer to the Demarcation Question understood as an explanatory question rather
than a justificatory question. One may think that this is an objection to my answer.
However, I think that my answer allows us to make careful and well-founded
distinctions between the two domains of law which can aid in asking this justificatory
question. More carefully, it shows that there are two different kinds of justificatory
questions to be asked. We can ask which rights should be protected by criminal and
tort law. We can also ask whether a defendant should be found criminally liable only if
she is blameworthy and whether a defendant should be found tortiously liability only if
she is responsible.
If criminal law and tort law, as currently practised, do indeed implicate different
notions of responsibility, as argued here, then we can ask whether this is justified.
216
It may rule punishment (perhaps, including punitive damages) if we think that blameworthiness is
required for punishment to be justified. It is unsurprising that the availability of punitive damages in tort
law is highly controversial.
116
Depending on how we answer this, there will be different upshots to how we should
reform our law. If we think that blameworthiness should be a necessary condition for
criminalisation, then what we think about blameworthiness for negligence, for instance,
will have a direct bearing on what should and should not be criminalised. Another
implication of my view is that we should re-examine whether we are blameworthy for
being ignorant of the law, and for committing a crime as a result of entrapment (public
and private). Moreover, we may be encouraged to think about different excusing
conditions that undermine blameworthiness, such as one’s social and economic
background and one’s upbringing. Turning briefly to tort law, if tort law does not and
should not implicate blameworthiness of the kind that is required for punishment to be
justified, then my account can shed some light on the debate about whether punitive
damages should be awarded in tort law. That is, by focusing on current doctrines, my
two-pronged approach has both explanatory and critical power and clears the field for
asking important justificatory questions about both criminal and tort law.
117
References
Robert Adams (1985): “Involuntary Sins”, Philosophical Review, 94(1): 3-31.
Larry Alexander (1987): “Causation and Corrective Justice: Does Tort Law Make
Sense?”, Law and Philosophy, 6(1): 1-23.
------ (1990): “Reconsidering the Relationship among Voluntary Acts, Strict Liability,
and Negligence in Criminal Law”, Social Philosophy and Policy, 7(2): 84-104.
Larry Alexander and Kim Ferzan (2009a): Crime and Culpability: A Theory of Criminal Law,
Cambridge University Press.
------ (2009b): “Against Negligence Liability” in Criminal Law Conversations, Paul
Robinson, Stephen Garvey and Kim Ferzan (eds.), Oxford; New York: Oxford
University Press, 273-281.
Jonathan Bennett (1980): “Morality and Consequences”, The Tanner lectures on Human
Values, delivered at Oxford University.
Michael Bratman (2009): “Intention, Belief, Practical, Theoretical” in Spheres of Reason,
Simon Robertson (ed.), Oxford University Press, 29-61.
Guido Calabresi (1970): The Costs of Accidents, New Haven: Yale University Press.
------ (1975): “Concerning Cause and the Law of Torts: An Essay for Harry Kalven,
Jr.”, University of Chicago Law Review, 43: 69-108.
Bruce Chapman (1995): “Wrongdoing, Welfare, and Damages: Recovery for Non-
Pecuniary Loss in Corrective Justice” in Philosophical Foundations of the Law of Torts,
David Owen (ed.), Oxford: Oxford University Press, 409–426.
Randolph Clark and Justin Capes (2013): “Incompatibilist (Nondeterministic) Theories
of Free Will”, The Stanford Encyclopedia of Philosophy (Spring 2013 Edition), Edward
N. Zalta (ed.).
Jules Coleman (1974): “On the Moral Argument for the Fault System”, Journal of
Philosophy, 71(14): 473-490.
------ (1982): “Moral Theories of Torts: Their Scope and Limits: Part One”, Law and
Philosophy, 1(3):371-390.
------ (1992): Risks and Wrongs, Cambridge University Press.
------ (2001): The Practice of Principle: In Defence of a Pragmatist Approach to Legal Theory,
Oxford; New York: Oxford University Press.
Jules Coleman and Arthur Ripstein (1995): “Mischief and Misfortune”, McGill Law
Journal, 41(1): 91-130.
John Cottingham (1979): “Varieties of Retribution”, Philosophical Quarterly, 29(116):
238–246.
Antony Duff (2007): Answering for Crime: Responsibility and Liability in the Criminal Law,
Oxford and Portland, Oregon: Hart Publishing.
Antony Duff and Sandra Marshall (1998): “Criminalization and Sharing Wrongs”,
Canadian Journal of Law and Jurisprudence, 11(1): 7-22.
118
------ (2010): “Public and Private Wrongs” in Essays in Criminal Law in Honour of Sir Gerald
Gordon, James Chalmers, Fiona Leverick and Lindsay Farmer (eds.), Edinburgh
University Press, 70-85.
Richard Epstein (1973): “A Theory of Strict Liability”, Journal of Legal Studies, 2(1): 151-
204.
Joel Feinberg (1987): “Voluntary Euthanasia and the Inalienable Right to Life”,
Philosophy and Public Affairs, 7(2): 93-123.
John Martin Fischer (1994): The Metaphysics of Free Will, Oxford: Blackwell Publishers.
John Martin Fischer and Neal Tognazzini (2009): “The Truth about Tracing”, Noûs,
43(3): 531-556.
George Fletcher (1971): “The Theory of Criminal Negligence: A Comparative
Analysis”, University of Pennsylvania Law Review, 119(3): 401-438.
Harry Frankfurt (1969): “Alternate Possibilities and Moral Responsibility”, Journal of
Philosophy, 66(3): 829-839. Also in The Importance of What We Care About, Cambridge
University Press, 1988: 1-10.
John Gardner (2011): “What is Tort Law For? Part 1: The Place of Corrective Justice”,
Law and Philosophy, 30(1): 1-50.
Mark Geistfeld (2014) “Compensation as a Tort Norm” in Philosophical Foundations of the
Law of Torts, John Oberdiek (ed.), Oxford: Oxford University Press, 65-85.
John Goldberg (2015): “Inexcusable Wrongs”, California Law Review, 103(3): 467-512.
Gerald Gordon (2000): The Criminal Law of Scotland, 3
rd
Edition, Vol. 1, Michael Christie
(ed.), Edinburgh: W. Green & Son Ltd / Scottish Universities Law Institute.
James Goudkamp (2011): “Insanity as a Tort Defence”, Oxford Journal of Legal Studies,
31(4): 727-754.
------ (2013): Tort Law Defences, Hart Publishing.
Jerome Hall (1963): “Negligence Should Be Excluded from Penal Liability”, Columbia
Law Review, 63: 632-644.
Gilbert Harman (1997): “Practical Reasoning” in The Philosophy of Action, Alfred Mele
(ed.), Oxford University Press, 149–177.
H.L.A. Hart (1968): Punishment and Responsibility: Essays on the Philosophy of Law, H.L.A.
Hart and John Gardner (eds.), Oxford University Press.
Pamela Hieronymi (2008): “Responsibility for Believing”, Synthese, 161(3): 357-373.
Oliver Wendell Holmes (1881): The Common Law, Boston: Little, Brown, and Company.
Tony Honoré (1988): “Responsibility and Luck: The Moral Basis of Strict Liability”,
Law Quarterly Review, 104: 530-553. Also in Responsibility and Fault, Oxford
University Press, 1999, Chapter 2.
David Hume (1960): A Treatise of Human Nature, L. A. Selby-Bigge (ed.), Oxford
University Press.
Douglas Husak (1987): Philosophy of Criminal Law, Rowman and Littlefield.
------ (2008): Overcriminalization: The Limits of the Criminal Law, Oxford University Press.
119
------ (2011): “Negligence, Belief, Blame and Criminal Liability: The Special Case of
Forgetting”, Criminal Law and Philosophy, 5(2): 199-218.
John Josephson and Susan Josephson (1994): Abductive Inference: Computation, Philosophy,
Technology. Cambridge; New York: Cambridge University Press.
Greg Keating (2012): “The Priority of Respect over Repair”, Legal Theory, 18(3): 293-
337.
Robert Keeton, Lewis Sargentich and Greg Keating (2004): Tort and Accidental Law:
Cases and Materials, Fourth Edition, American Casebook Series, Thomson West.
Matt King (2009): “The Problem with Negligence”, Social Theory and Practice, 35(4): 577-
595.
Grant Lamond (2007): “What is a Crime?”, Oxford Journal of Legal Studies, 27(4): 609-
632.
Ambrose Y. K. Lee (2015): “Public Wrongs and the Criminal Law”, Criminal Law and
Philosophy, 9(1): 155-170.
Michael McKenna and Justin Coates (2015): “Compatibilism” in The Stanford
Encyclopedia of Philosophy (Summer Edition), Edward N. Zalta (ed.).
Jonathan McKeown-Green (2006): “The Real Me” in Judging and Understanding: Essays
on Free Will, Narrative, Meaning and the Ethical Limits of Condemnation, Pedro Tabensky
(ed.), Ashgate, 200-218.
Michael Moore (1993): Act and Crime, Oxford: Clarendon Press.
------ (1997): Placing Blame, Oxford: Clarendon Press.
------ (2011): “Intention as a Marker of Moral Culpability and Legal Punishability” in
Philosophical Foundations of Criminal Law, R.A. Duff and Stuart Greens (eds.), Oxford
University Press, 179-205.
Nancy Moore (2012): “Intent and Consent in the Tort of Battery: Confusion and
Controversy”, American University Law Review, 61(6): 1585-1654.
Jeffrie Murphy (2007): “Legal Moralism and Retribution Revisited”, Criminal Law and
Philosophy, 1(1):5-20.
Thomas Nagel (1976): “Moral Luck”, Proceedings of the Aristotelian Society, 50: 137-51. Also
in Mortal Questions, Cambridge University Press, Chapter 3.
Robert Nozick (1974): Anarchy, State, and Utopia, Oxford: Basil Blackwell.
David Owen (1985): “Deterrence and Desert in Tort: A Comment”, California Law
Review, 73(3): 665-676.
Derek Parfit (2011): On What Matters, Oxford University Press.
Stephen Perry (1988): “The Impossibility of General Strict Liability”, Canadian Journal
of Law and Jurisprudence, 1(2): 147-171.
------ (1997): “Libertarianism, Entitlement, and Responsibility”, Philosophy and Public
Affairs, 26(4): 351-396.
------ (2001): “Responsibility for Outcomes, Risk, and the Law of Torts” in Philosophy
and the Law of Torts, Gerald Postema (ed.), Cambridge University Press, 72-130.
120
Jonathan Quong (2012): “Liability to Defensive Harm”, Philosophy and Public Affairs,
40(1): 45-77.
Joseph Raz (2011): From Normativity to Responsibility, Oxford University Press.
Arthur Ripstein (1999): Responsibility, Equality, and the Law, Cambridge University Press.
Gideon Rosen (2003): “Culpability and Ignorance”, Proceedings of the Aristotelian Society,
103(1): 61-84.
------ (2004): “Skepticism about Moral Responsibility,” Philosophical Perspectives, 18(1):
295-313.
David J. Seipp (1996): “The Distinction between Crime and Tort in the Early Common
Law”, Boston University Law Review, 76: 59-87.
Hanoch Sheinman (2003): “Tort Law and Corrective Justice”, Law and Philosophy, 22(1):
21-73.
George Sher (2006): In Praise of Blame, Oxford University Press.
George Sher (2009): Who Knew?: Responsibility without Awareness, Oxford University Press.
Kenneth Simons (1994): “Culpability and Retributive Theory: The Problem of
Criminal Negligence,” Journal of Contemporary Legal Issues, 5: 365-398.
------ (2006): “A Restatement (Third) of Intentional Torts?” in Symposium, 48 Arizona
Law Review 1061.
------ (2008): “The Crime/Tort Distinction: Legal Doctrine and Normative
Perspectives”, Widener Law Journal, 17: 719-732.
Angela Smith (2005): “Responsibility for Attitudes: Activity and Passivity in Mental
Life”, Ethics, 115(2): 236–271.
Holly Smith (1983): “Culpable Ignorance”, Philosophical Review, 92(4): 543-571.
Holly Smith (2011): “Non-Tracing Cases of Culpable Ignorance”, Criminal Law and
Philosophy, 5(2): 115-146
Victor Tadros (2005): Criminal Responsibility, Oxford University Press.
J.W.C. Turner (1945): The Modern Approach to Criminal Law, Leon Radzinowicz and
J.W.C. Turner (eds.), Macmillan.
Manuel Vargas (2005): “The Trouble with Tracing”, Midwest Studies in Philosophy, 29(1):
269-291.
Gary Watson (2014): “Raz on Responsibility”, Criminal Law and Philosophy, Published
online.
Gene Weingarten (2009): “Fatal Distraction: Forgetting a Child in the Backseat of a
Car Is a Horrifying Mistake. Is It a Crime?”, Washington Post, March 8.
Steven Weingarten (1984): “Tort Liability for Nonlibelous Negligent Statements: First
Amendment Considerations”, Yale Law Journal, 93(4): 744-762.
Ernest Weinrib (1995): The Idea of Private Law, Cambridge: Harvard University Press.
Bernard Williams (1981): Moral Luck: Philosophical Papers 1973-1980, Cambridge
University Press.
Gideon Yaffe (2012): “The Voluntary Act Requirement” in Routledge Companion to
Philosophy of Law, Andrei Marmor (ed.), Routledge, 174-190.
121
Michael Zimmerman (1997): “Moral Responsibility and Ignorance”, Ethics, 107(3):
410-426.
Cases:
Adamson v Motor Vehicle Insurance Trust (1957) 58 WALR 56 (SC).
Bettel v Yim (1978) 88 D.L.R. (3d) 543 (Ont. Co. Ct.).
Blanchard v. Bridgeport, 463 A.2d 553 (Conn. 1983)
Booth v. Rome, W. & OTR Co., 140 N.Y. 267 (1893)
Buckley and Toronto Transportation Commission v Smith Transport Ltd [1946] OR 798 (CA).
Hanson v. Brogan, 400 P.2d 265 (Mont. 1965)
Dawkins v. State, 313 Md. 638, 547 A.2d 1041 (1988).
Delahanty v Hinckley, 799 F. Supp. 184 (DDC 1992).
DPP v Morgan [1976] AC 1.
Garratt v. Dailey, 46 Wash. 2d 197, 279 P.2d 1091 (Wash. 1955)
Hutchings v Nevin (1992) 9 OR (3d) 776 (OCJ).
Krom v. Schoonmaker, 3 Barb 647 (NY 1848).
Meyers v. Epstein, F.Supp.2d 192 (N.Y. 2002).
Mohr v. Williams, 104 N.W. 12 (Minn. 1905).
Mink v. University of Chicago, 460 F. Supp. 713 (Ill. 1978).
Morriss v Marsden [1952] All ER 1 925.
Morse v Crawford, 17 Vt. 499 (1845).
Palsgraf v. Long Island Railroad Co., 248 N.Y. 339, 162 N.E. 99 (N.Y. 1928)
People v. Camp, 104 Cal.App.3d 244 (1980).
People v. Decina, 138 N.E.2d 799 (N.Y. 1956)
Polmatier v. Russ, 206 Conn. 229 (1988).
R. v. Aguiar, 2002 CarswellOnt 4271
Rajspic v. National Mut. Ins. Co., 718 P.2d 1167 (Idaho 1986).
Rylands v Fletcher [1868] UKHL 1, 3 HL 330, (1868) LR 3 HL 330
Spano v. Perini Corp. et al, 25 NY 2d 11 (1969)
State v. Burns, 457 S.W.2d 721 (Mo. 1970).
State v. Cleppe, 635 P.2d 435 (Wash. 1981).
State v. Flinchpaugh, 232 Kan. 831, 659 P.2d 208 (1983).
State v. Harvey, 281 N.C. 1, 187 S.E.2d 706 (1972).
State v. Michlitsch, 438 N.W.2d 175 (N.D.1989).
Tindale v Tindale [1950] 4 DLR 363 (BCSC).
United States v. John W Hinckley, Findings and Order, Criminal Number 81-306, August
10, 1982.
United States v. Weaver, 458 F.2d 825, 826 (D.C. Cir.1972).
United States v. X-Citement Video, 115 S.Ct, 464 (1994).
Vaughan v. Menlove, 3 Bing. N.C. 468, 132 Eng. Rep. 490 (1837).
122
Villanueva v. Comparetto, 180 A.D.2d 627 (N.Y. 1992).
Vincent v. Lake Erie Transportation Co., 109 Minn. 456, 124 N.W. 221 (1910)
Wagner v. State, 122 P.3d 599 (Utah 2005).
Wagon Mound (No. 2) (Also known as Overseas Tankship (UK) Ltd v The Miller Steamship Co.)
[1967] 1 AC 617
Walker v. State, 356 So.2d 674 (Ala Crim. App. 1977).
White v. Muniz, 999 P.2d 814 (Colo. 2000).
White v. Pile, 68 Wash. 176 (1950).
White v. University of Idaho, 797 P.2d 108 (Idaho 1990).
Williams v Hays, 143 NY 442 (1894).
Williams v Kearbey, 13 Kan. App.2d 565, 775 P.2d 670 (1989).
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Crime and punishment: examining proportionality issues within criminal law
PDF
Responsibility and the emotional structure of relationships
PDF
Responding to harm
PDF
Aggregating happiness: seeking and identifying a single plausible unifying theory
PDF
Aggregating complaints
PDF
A deontological explanation of accessibilism
PDF
The constitution of action
PDF
The minimal approval view of attributional-responsibility
PDF
Units of agency in ethics
PDF
Getting our act(s) together: the normativity of diachronic choice
PDF
The boundaries of the intentional
PDF
Defending Rawls's Law of Peoples against cosmopolitan critiques
PDF
On law enforcement officers’ liability to defensive harm
PDF
The scope and significance of George Berkeley's language model
PDF
Contract, from promise to commodity
PDF
Thomas Reid on singular thought
PDF
Vagueness, legal content, and legal interpretation
PDF
Describing authority
PDF
Distributive equality and labor burdens
PDF
Negotiating pluralism and tribalism in liberal democratic societies
Asset Metadata
Creator
Webster, Aness Gaeyoung Kim
(author)
Core Title
Responsibility and rights: a search for a principled distinction between criminal law and tort law
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Philosophy
Publication Date
07/08/2016
Defense Date
05/10/2016
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
criminal law,distinction between criminal and tort law,OAI-PMH Harvest,philosophy of law,responsibility,tort law
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Watson, Gary (
committee chair
), Keating, Greg (
committee member
), Quong, Jonathan (
committee member
), Yaffe, Gideon (
committee member
)
Creator Email
Aness.Webster@nottingham.ac.uk,anessweb@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-264066
Unique identifier
UC11281188
Identifier
etd-WebsterAne-4523.pdf (filename),usctheses-c40-264066 (legacy record id)
Legacy Identifier
etd-WebsterAne-4523.pdf
Dmrecord
264066
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Webster, Aness Gaeyoung Kim
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
distinction between criminal and tort law
philosophy of law
responsibility
tort law