"Arts, Faculty of"@en . "Philosophy, Department of"@en . "DSpace"@en . "UBCV"@en . "Whiteley, Diane Elizabeth"@en . "2009-06-25T18:31:33Z"@en . "1998"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "In this study I tackle the problem of justifying criminal punishment. Although I\r\ntake heed of a traditional line of theorizing which says that punishment is an\r\nexpressive and communicative endeavour, my theory breaks away from\r\ntraditional approaches. This break is motivated by a recognition that theorists\r\nworking in the traditional framework have failed to resolve the tension between\r\nretributivist and consequentialist reasons for punishment. I argue that\r\npunishment is justified as a type of communication from those affected by the\r\ncrime to distinct and varied audiences.\r\nMy naturalistic theory is structured around two fundamental themes, naturalism\r\nand pluralism about aims. The naturalism consists in the fact that the theory\r\ntakes an empirically informed descriptive approach to the problem of justifying\r\npunishment. This foundation provides the resources for developing a balanced\r\nview of the moral agent which takes into account not only cognitive but also\r\nemotional capacities. This broader, deeper view of agency permits, indeed calls\r\nfor, an analysis of the moral psychologies of those involved in the social practice.\r\nThat analysis leads to the explanation that punishment is a type of\r\ncommunication of, among other things, strong but justified moral sentiments.\r\nFurther development of this view suggests that punishment's various messages\r\nare intended for a variety of audiences - not just the wrongdoer but also the victim\r\nand community.\r\nThat explication supports my other fundamental theme, pluralism about aims.\r\nThe social institution of punishment is a complex one involving stakeholders who\r\nhave differing motives and needs. Consequently, we should reject strategies\r\nwhich claim that punishment's justification can be reduced to one reason such as,\r\nfor example, that the criminal deserves it. I argue that punishment's justification\r\nis multifaceted and complex.\r\nThe arguments I put forward to justify punishment also bring to light aspects of\r\nthe existing social institution that need reform. In general, they point to the need\r\nto design penal measures that promote communication among wrongdoer, victim\r\nand community. But I also call for a specific reform. I argue that the victim,\r\nwhose concerns have traditionally been disregarded, should also be given a voice\r\nwithin the social institution."@en . "https://circle.library.ubc.ca/rest/handle/2429/9610?expand=metadata"@en . "10834777 bytes"@en . "application/pdf"@en . "A N A T U R A L I S T I C J U S T I F I C A T I O N F O R C R I M I N A L P U N I S H M E N T by D I A N E E L I Z A B E T H W H I T E L E Y B. A . , Northeastern University, 1969 M . A . , Simon Fraser University, 1992 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y i n T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Philosophy We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A August 1998 \u00C2\u00A9Diane Elizabeth Whiteley, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada Date DE-6 (2/88) Abstract In this study I tackle the problem of justifying criminal punishment. Although I take heed of a traditional line of theorizing which says that punishment is an expressive and communicative endeavour, my theory breaks away from traditional approaches. This break is motivated by a recognition that theorists working in the traditional framework have failed to resolve the tension between retributivist and consequentialist reasons for punishment. I argue that punishment is justified as a type of communication from those affected by the crime to distinct and varied audiences. My naturalistic theory is structured around two fundamental themes, naturalism and pluralism about aims. The naturalism consists in the fact that the theory takes an empirically informed descriptive approach to the problem of justifying punishment. This foundation provides the resources for developing a balanced view of the moral agent which takes into account not only cognitive but also emotional capacities. This broader, deeper view of agency permits, indeed calls for, an analysis of the moral psychologies of those involved in the social practice. That analysis leads to the explanation that punishment is a type of communication of, among other things, strong but justified moral sentiments. Further development of this view suggests that punishment's various messages are intended for a variety of audiences - not just the wrongdoer but also the victim and community. That explication supports my other fundamental theme, pluralism about aims. The social institution of punishment is a complex one involving stakeholders who ii have differing motives and needs. Consequently, we should reject strategies which claim that punishment's justification can be reduced to one reason such as, for example, that the criminal deserves it. I argue that punishment's justification is multifaceted and complex. The arguments I put forward to justify punishment also bring to light aspects of the existing social institution that need reform. In general, they point to the need to design penal measures that promote communication among wrongdoer, victim and community. But I also call for a specific reform. I argue that the victim, whose concerns have traditionally been disregarded, should also be given a voice within the social institution. iii Contents Abstract ii List of Tables iv Dedication v Acknowledgements vi Introduction 1 Parti. The Theory Chapter 1 Why Develop a Naturalistic Justification for Punishment? 10 Chapter 2 Punishment: Communication with Various Audiences 41 Part II. Clarifications Chapter 3 A Comparison with R. A. Duffs Expressivist View 75 Chapter 4 Punishment and Respect for the Wrongdoer's Dignity 102 Chapter 5 Wrongful Intent, Harm Done and the Sentence 136 Part III. An Application Chapter 6 The Role of the Victim 168 Conclusion 202 Bibliography 209 List of Tables THE AIMS OF PUNISHMENT The Uiufying Theme: Communication of the Moral Demand from Various Stakeholders to a Variety of Audiences iv For my mother Anna Kenney Sinewski who taught me the value of an education v Acknowledgments I would especially like to thank Paul Russell, senior supervisor of my graduate studies at the University of British Columbia, who has provided philosophical guidance, insight, support and a sense of humour when needed. I would also like to thank to the other members of my supervisory committee, Jim Dybikowski and Alan Richardson who have provided invaluable philosophical guidance. I am grateful to my son Garth Whiteley. His circumstances in life roused my interest in a variety of ethical problems. Above all, I am grateful to Don Whiteley who has provided encouragement, intellectual stimulation, editorial comment and enduring personal support. vi Introduction Criminal punishment is the deliberate infliction of physical and/or psychological suffering on a person by the state for a crime.1 It raises a question. Is it ever justifiable for citizens of a civilized and humane society to punish a fellow human being?2 This problem of punishment has attracted significant philosophical attention because it is a pressing practical concern as well as a hard philosophical problem. The various traditional theories put forward to justify the practice include a variety of important reasons for punishment. Nonetheless, the seemingly irreconcilable conflicts among them suggest that none of them has the resources to explain fully or to justify the practice. In this study I put forward a theory that breaks away from traditional approaches. This naturalistic theory explains and justifies the social institution of criminal punishment by taking into account not only human cognitive but also human emotional capacities. Naturalism, of course, encompasses a wide variety of views. In this study I use the term narrowly and specifically as the basis for an empirically informed descriptive approach to the problem of justifying punishment. Traditional theories can be categorized broadly in four ways - retributivist, consequentialist, mixed theories and expressivist. Retributive theories claim that 1 I interpret punishment, also commonly called \"hard treatment,\" broadly to encompass a variety of penal measures such as incarceration, mandatory supervision, electronic monitoring, diversion to community service, etc. 2 I make the idealizing assumption that the state's laws are mostly just and that, when they are not, there are formal methods which permit citizens to bring about changes in the laws. I think it is clear, for example, that a state which punishes political reformers without a fair trial is acting unjustly and that the resultant punishment cannot be justified. 1 punishment is justified because, and only because, the wrongdoer deserves it for having done wrong.3 These theories are classified as backward-looking because they 'look back' to the wrong action in order to explain why the person who performed it deserves punishment. They explain why we punish only wrongdoers and not those who are innocent of wrongdoing. They are rightly criticized, however, for providing a necessary but not a sufficient condition for punishment. In particular, they do not explain the benefit of punishment for the community at large. Consequentialist theories justify punishment on the basis of some future benefit for society.4 They are classified as forward-looking because they take the future benefit of society to be a necessary condition for punishment. Criticisms of consequentialist approaches centre on worries that social utility will take precedence over wrongdoers' rights or even lead to the punishment of innocents.5 Mixed or teleological retributivist theories have been put forward in response to some of the inadequacies of the other theories just mentioned.6 Proponents 6 Kant is the most renowned defender of retributivism. Jeffrie G. Murphy is a well-known contemporary retributivist who bases his defence of a retributive theory on Kantian insights. Jeffrie G. Murphy, Retribution, Justice, and Therapy (Boston: D. Reidel Publishing Company, 1979). 4 Jeremy Bentham, a prominent utilitarian and social reformer of the nineteenth century, presented and defended a consequentialist theory of punishment. John Rawls presents a contemporary defence of the consequentialist approach. John Rawls, \"Two Concepts of Rules,\" The Philosophical Review, 44 (1955), 3-13. 5 Many theorists have discussed the various criticisms of the consequentialist position. See, for example, H . J . McCloskey, \"A Non-Utilitarian Approach to Punishment,\" Philosophical Perspectives on Punishment, ed. Gertrude Ezorsky (Albany, New York: State University of New York Press, 1972), pp. 119-34. 6 H . L. A. Hart is a perhaps the most well-known contemporary defender of teleological retributivism. H . L. A. Hart, \"Prolegomenon to the Principles of Punishment,\" Punishment and Responsibility: Essays in the Philosophy of Law (New York: Oxford University Press, 1968), pp. 1-27. 2 correctly point out that punishment is a complex social institution which has a plurality of aims. Their theories are the result of attempts to reflect that complexity by integrating both backward and forward-looking reasons into the justification. But in attempting merely to combine salient aspects of consequentialist and retributive accounts, these theories miss the mark. They leave out reasons for punishment that are important to its justification. In particular, like the consequentialist and retributive accounts from which they arise, they fail to explain adequately how the victim's concerns relate to justification of the practice. Another significant line of theorizing centres on the theme that punishment is an expressive and communicative endeavour. Contemporary theorists such as Duff, Feinberg, Hampton and Primoratz develop this line in various ways.7 They do not agree, however, on whether punishment as expression is backward-looking, forward-looking or both. Invariably, depending on which it is deemed to be, the defence for the particular expressivist view falls back on traditional retributivist, consequentialist or \"mixed theory\" arguments. In my approach I take heed of the notion that punishment is an expressive and communicative practice, but I develop the approach in a way that contrasts sharply with the expressivist accounts just mentioned. In my view, those 1 R. A. Duff, Trials and Punishments (New York: Cambridge University Press, 1986). Joel Feinberg, Chapter 5, \"The Expressive Function of Punishment,\" Doing and Deserving (New Jersey: Princeton University Press, 1970), pp. 95-118. Jean Hampton, \" A New Theory of Retribution,\" Liability and Responsibility, ed. R. G. Frey and Christopher W. Morris (New York: Cambridge University Press, 1991), pp. 377-414. Igor Primoratz, \"Punishment as Language,\" Philosophy, 64 (1989), 187-205. Reprinted in Philosophy of Law, ed. Joel Feinberg and Hyman Gross (Belmont, Ca.: Wadsworth Publishing Co., 1995), pp. 602-12. 3 contemporary theorists buy into a traditional theoretical framework which served a useful purpose but must now be judged inadequate. While important reasons for punishment such as desert, social benefit and expression were brought to light in that framework, those reasons have neither been fully explained nor successfully integrated into one theory. The problem is that the traditional framework draws the battle lines for justification along retributive/consequentialist lines - or, more simply, in terms of backward and forward-looking reasons. Theorists have focused on resolving the apparently - intractable tension between the two types of reasons while other reasons important to the justification of the practice have been downplayed or ignored. I argue that punishment is best explained and justified in the context of a naturalistic theory of moral responsibility. When the problem of punishment is placed in this framework, two key elements, which traditional theorists have overlooked, are readily identified and analyzed. The first is that punishment is, among other things, an expression and attempted communication of specific emotional attitudes, called reactive attitudes by P. F. Strawson8 and moral sentiments (the term I adopt) by Adam Smith.9 The second key element is that punishment's plurality of aims is best unified and explained not in terms of backward and forward-looking reasons but as a type of communication from various players to a variety of audiences - specifically, the wrongdoer, the community and the victim. 8 Peter Strawson, \"Freedom and Resentment,\" Free Will, ed. Gary Watson (New York: Oxford University Press, 1982), pp. 59-80. 9 I use Adam Smith's term \"moral sentiments\" in order to acknowledge the philosophical tradition. Adam Smith, The Theory of the Moral Sentiments, ed. D. D. Raphael and A. L. Macfie (Indianapolis, Indiana: Liberty Fund Inc., 1982). 4 In explaining the first key element, I take up Strawson's claim that our practice of punishment is essentially connected to moral sentiments that are part of our emotional make-up as human beings, a 'given' of human nature. The idea is that when a person violates the moral demand for some degree of goodwill or regard by acting with contempt, indifference or malevolence towards another, we react with resentment or moral indignation. My analysis of these moral sentiments shows that they underlie and, to a great extent, explain the social practice of punishment. In other words, when the problem of punishment is placed in a naturalistic framework, its justification can admit analysis of the moral psychologies of those involved - the wrongdoer, the members of the community and, significantly, the victim. It becomes clear that punishment is a type of communication which takes place not only at a cognitive level but also at an emotional level. Through punishment, the victim and community express and convey their moral sentiments using both verbal and nonverbal means \u00E2\u0080\u0094 the nonverbal mode being particularly suited to the communication of emotions. Acknowledgment that punishment involves wrongdoer, community and victim brings us to the second key element. Punishment's plurality of aims can be unified and explained as a type of communication from various players to various audiences. In other words, the sentence is not aimed merely at the criminal. It is intended to impart various messages to various audiences. In short, the approach assumes not only that everyone affected by the crime has a stake in the punishment of the criminal but also that their perspectives on the 5 punishment and their aims in having it inflicted - or, in the case of the criminal, suffering it - differ. From the victim's perspective, the punishment expresses his resentment and associated condemnation of the wrongdoer. It communicates in a strong, nonverbal way his demand for a minimum degree of respect from the wrongdoer and others. From the community's perspective, punishment conveys a variety of messages - moral indignation towards the wrongdoer, authoritative disavowal of her act, affirmation of the community's values, and specific and general deterrence. And from the wrongdoer's perspective, in those cases where she repents, certain types of sentences may signify to the victim and community her guilt and remorse. This justification, then, is pluralist. Punishment's plurality of aims is unified by the idea that it is a type of communication. The communication sends various messages from the stakeholders in the social institution to its audiences. Pluralism about aims entails that no one of them on its own can be considered the necessary one that justifies the social institution of punishment, although, in a particular instance of punishment, communication of some or one of the aims may be sufficient to justify that instance of punishment. As the discussion to this point indicates, the theory is structured around two fundamental themes. The first is naturalism. The naturalistic foundation provides the resources for developing a balanced view of the human moral agent which takes into account not only cognitive but also emotional capacities. This 6 broader, deeper view of agency permits, indeed calls for, an analysis of the moral psychologies of those involved in the social practice of punishment. That analysis leads, in turn, to the explanation that punishment is a communication of, among other things, strong but justified moral sentiments. Further development of the view that punishment is a strong form of communication suggests that its various messages are intended for a variety of audiences \u00E2\u0080\u0094 not merely wrongdoer but also victim and community at large. In short, the explication of punishment as a form of communication with a variety of audiences is well supported by the naturalistic context in which it is defended. Moreover, that explication supports my other fundamental theme. The second fundamental theme of this study is pluralism about punishment's aims. The social institution of punishment is a complex one involving stakeholders who have differing motives and needs. When we keep this fact in mind, it becomes clear that we should reject strategies which claim that punishment's justification can be reduced to one simple reason such as, for example, that the criminal deserves it. I argue that punishment's justification is multifaceted and complex. These two fundamental themes are crucial to the justification and, as a result, are prominent threads in the chapters that form the fabric of this thesis which, for clarity, is divided into three parts. In Part I, which comprises the first two chapters, I explicate this naturalistic theory of punishment. I first set out an expressive view of the practice in the context of naturalism. The naturalistic foundation provides the resources - missing from previous expressivist theories -7 needed to explain and justify punishment. In Chapter 2 I analyze punishment as a form of communication from various stakeholders to a variety of audiences. Part II, which includes Chapters 3, 4 and 5, focuses on clarifications of the theory. In Chapter 3 I compare the theory to R. A. Duffs expressivist and rationalistic account. The comparison highlights the advantages of the naturalistic theory which, in contrast with Duffs narrow approach, can accommodate a plurality of aims. In Chapter 4 the question of how punishment can be said to respect the wrongdoer's dignity is addressed. In Chapter 5 I tackle the difficult problem of how to relate wrongful intent and harm done to the severity of the sentence. The naturalistic framework provides a fresh perspective which clears up some questions that are problematic for traditional approaches. In Part III, which consists of Chapter 6,1 apply the theory by examining various proposed and actual reforms from the naturalistic perspective. My primary purpose in this thesis is to justify punishment as a type of communication from those affected by the crime to distinct and varied audiences. The arguments I put forward in support of this view, however, also bring to light aspects of the existing social institution that need reform. In general terms, these arguments point to the need to design penal measures that promote communication among wrongdoer, victim and community. But I also call for a specific reform. The existing justice system has concerned itself almost exclusively with satisfying the needs of the public and protecting the rights of the 8 wrongdoer. The victim's concerns have been repeatedly disregarded. My arguments in support of the naturalistic justification show that the victim should also be given a voice in the social institution of punishment. 9 Part I. The Theory Chapter 1 Why Develop a Naturalistic Justification for Punishment? In this chapter I explain why we should develop a naturalistic justification for punishment. I begin by pointing out some problems with two promising traditional approaches, H. L. A. Hart's teleological retributivist account and Igor Primoratz's expressivist theory. The problems centre on the fact that these theories discount the emotions of those involved in the social practice of punishment. I submit that P. F. Strawson provides the key to an alternative approach. Taking up Strawson's insight that we demand some degree of goodwill from those involved in social interactions with us, I argue that the victim and community inevitably experience negative moral sentiments towards one who does harm. Traditional theories of punishment neither acknowledge nor explain this need to express retributive sentiments. Yet these sentiments are, in fact, an important, aspect of the social practice of punishment and a justifying reason for it. I contend that a theory based on a naturalistic foundation has the resources to explain them. The Inadequacies of Two Promising Attempts to Justify Punishment Traditional theories of punishment attempt to justify the practice by focusing on one of punishment's aims to the exclusion of others. Consequentialists focus on forward-looking social goals. Retributivists concentrate on the abstract need to balance the scales of justice by giving the wrongdoer what she deserves. Both 10 types of theories bring to light significant reasons that one may want to include in the justification of punishment, but neither captures the complete picture. In this section I consider two theories each of which attempts, in its own way, to deal with the failings of these approaches. H. L. A. Hart endeavours to combine consequentialist and retributivist reasons in one theory. Taking a different tack, Igor Primoratz follows the expressivist tradition claiming that punishment is a form of communication. I argue that these accounts also fail to include all the justifying reasons for punishment, and I explain why. In \"Prolegomenon to the Principles of Punishment\" H. L. A. Hart attempts to reconcile the conflicting values of the consequentialist and retributivist by setting out a framework for combining teleological and retributive principles.1 According to Hart, the justification of punishment cannot be both retributive in the positive sense and utilitarian.2 The positive retributivist argues that the wrongdoer deserves to suffer even if that suffering increases the total amount of suffering in society. This principle conflicts with the utilitarian commitment to the maximization of utility. To avoid the conflict, Hart dissects the complex question of justification into interrelated but separate questions of Definition, General Justifying Aim and Distribution. Consider briefly his analysis of the General Justifying Aim and Distribution. 1 Hart, \"Prolegomenon to the Principles of Punishment,\" pp. 1-27. 2 Positive retributivism is the principle that one who is guilty ought to be punished. Negative retributivism is the principle that one who is not guilty must not be punished. Hart does not use these terms which I take from Mackie. Nonetheless, i t seems obvious that Hart's \"Retribution in General Justifying A i m \" is positive retributivism and his \"Retribution in Distribution\" is negative retributivism. J . L. Mackie, Chapter X V , \"Morality and the Retributive Emotions,\" Persons and Values (Oxford: Oxford University Press, 1985), p 207. 11 Hart claims that \"much confusing shadow fighting\" between retributivists and utilitarians can be avoided once we recognize that ...it is perfectly consistent to assert both that the General Justifying Aim of the practice of punishment is its beneficial consequences and that the pursuit of this General Aim should be qualified or restricted out of deference to principles of Distribution which require that punishment should be only of an offender for an offence.3 According to Hart, then, the inadequacies of theories that rely on a single principle can be met by one that incorporates different values by separating the distinct questions involved in justifying the whole social institution. The assumption underlying this strategy is that a social institution such as punishment has \"a plurality of features which can only be understood as a compromise between partly discrepant principles.\"4 Although retribution in General Justifying Aim would conflict with the utilitarian value of social benefit, retribution in Distribution merely places a restriction on the unqualified pursuit of that value. Limiting the aim in this way avoids the conflict. It also satisfies our intuition that punishment should be applied only to the guilty. While it is undoubtedly right to limit punishment to wrongdoers, the question arises as to whether relegating the retributivist principle to a mere formula for distribution really does satisfy our retributivist feelings. The common reaction, for example, upon hearing of Paul Bernardo's torture and murder of Leslie Mahaffy and the subsequent mutilation of her body, is one of reprobation, outrage and even revulsion. As John Mackie points out, we feel that wrong actions call for 3 Hart, \"Prolegomenon to the Principles of Punishment/' p. 9. 4 Ibid., p. 10. 12 a hostile response. The positive retributivist principle has \"immediate, underived moral appeal.\"5 According to Mackie, if we did not feel the pull of positive retributivism, we would not be persuaded that arguments justifying punishment on the basis of some social benefit \"would make it morally right to inflict suffering or deprivation on the criminal.\"6 These observations indicate that feelings of reprobation towards the wrongdoer should not and, I shall argue, cannot be ignored. They need to be expressed. It is through punishment that the community expresses and attempts to communicate reprobation. Punishment has an expressive/communicative function. On Hart's view, however, denunciation of the crime is not a justification for punishment. He says the suggestion that it is should be treated as \"...a blurred statement of the truth that the aim not of punishment, but of criminal legislation is indeed to denounce certain types of conduct as something not to be practiced.\"7 It is true that denunciation of certain types of actions is one of the purposes of the criminal law. It does not follow, however, that denunciation of a particular wrong action and reprobation towards the wrongdoer are not reasons for punishment \u00E2\u0080\u0094 reasons which are independent of utilitarian aims and which are, in fact, pertinent to the justification of punishment. Hart fails to recognize the reprobative nature of punishment because he views a crime merely as a violation of a legal rule which states that a given type of action b Mackie, \"Morality and the Retributive Emotions,\" p. 208. 6 Ibid. 7 Hart, \"Prolegomenon to the Principles of Punishment,\" p. 8. 13 is forbidden.8 Criminal punishment for Hart is a method of social regulation which \"...defers action until harm is done; its primary operation consists simply in announcing certain standards of behavior and attaching penalties for deviation....\"9 But, as Joel Feinberg points out, characterizations such as this one fail to distinguish punishment from mere penalties such as the fines levied for infractions of traffic laws. While both penalties and punishment are authoritative deprivations for deviations from a legal rule, punishment has the additional function of expressing moral condemnation. In other words, there is a significant difference of kind and degree of purpose between a penalty and punishment. Feinberg says, ...punishment is a conventional device for the expression of attitudes of resentment and indignation, and of judgments of disapproval and reprobation, on the part of the punishing authority himself or of those 'in whose name' the punishment is inflicted. Punishment, in short, has a symbolic significance...10 Punishment, then, expresses the very real retributive attitudes felt by victim and community in the aftermath of a crime. \u00C2\u00B0 Hart is renowned for giving legal positivism its most powerful articulation and defence. See H . L. A. Hart, The Concept of Law (Oxford: Oxford University Press, 1961). His legal positivism calls for a sharp distinction between law and morality. It takes a rule to be the central explanatory concept of the law. Legal obligation has its seat in rules which Hart characterizes as primary or secondary. Primary rules tell all individuals in the social group how they must act. Secondary rules are about the primary rules. They specify how they may be \"ascertained, introduced, eliminated, varied, and the fact of their violation conclusively determined.\" Ibid., p. 92. It should be clear, in light of his legal positivism, why Hart believes denunciation of the criminal has no place in the justification of legal punishment. In my view, Ronald Dworkin convincingly refutes Hart's legal positivism arguing that Hart's model of judicial decision in hard cases is defective because it focuses too exclusively on rules. The upshot is that it cannot explain how such decisions essentially and necessarily involve appeals to moral considerations -considerations that form a part, either explicitly or latently, of the moral traditions of the system in question. 9 Hart, \"Prolegomenon to the Principles of Punishment,\" p. 23. 10 Feinberg, \"The Expressive Function of Punishment,\" p. 98. 14 Let us examine more closely Hart's claim that denunciation is the purpose of criminal legislation and not of punishment. There can be no doubt that criminal legislation defines and denounces certain types of actions in a formal, impersonal way. The legislation, which is educative, is directed at the entire community. It communicates to the members of society which types of actions are forbidden. It has the socially beneficial aim of deterring crime. Criminal legislation is a form of impersonal communication with the entire community. When a crime is committed, however, the victim and community react with feelings of extreme disapproval which they want to express and communicate to the criminal. I shall argue that it is through punishment that victim and community express their outrage at the victim's being devalued. In other words, punishment is, among other things, a form of personal communication with the wrongdoer. It is significant that Hart never even considers the question of denunciation of the individual criminal. His account cannot, in fact, accommodate the notion that condemnation of the criminal is an aim of punishment. Why? He is committed to the view that the General Justifying Aim of punishment is one of social benefit.11 On his hybrid theory, other values of importance to society may be included as part of a \"morally acceptable account\"12 of the whole social institution but not as the General Justifying Aim of punishment itself.13 In other words, there is a gap H Hart, \"Prolegomenon to the Principles of Punishment,\" p. 9. Hart indicates that the General Justifying Aim is the utilitarian one of social benefit, but does not argue for that view here. 12 Ibid., p. 3. 1^ Honderich points out that it is not at all clear why Hart, who thinks a compromise or two-value justification of punishment is needed, should also think that the practice is justified by only one of them. He suggests that Hart conflates the question of aim with that of justification. The two may be, but need not be, the same. Ted Honderich, Chapter 5, \"Compromises,\" Punishment: the supposed justifications (London: Hutchinson and Company Ltd., 1969), pp. 138-43. 15 in Hart's account. He fails to appreciate that denunciation of the individual wrongdoer might well be one of punishment's justifying aims. The commonsense view that there is an ineliminable reprobative aspect to punishment, which is directed at the criminal, is a strong one. In the traditional framework for discussion of punishment, the aim of condemnation of the criminal is considered to be \"backward-looking.\" Characterizing it in this way, however, immediately sets it in opposition to another value that seems important, the fact that punishment should benefit the community. This aim is traditionally characterized as \"forward-looking.\" The upshot is that theorists working in the traditional framework set themselves an apparently intractable problem. Both the backward and forward-looking reasons seem important but, when taken as aims of punishment, they appear to be inherently opposed. This problem, which arises out of the traditional framework, can be avoided by recognizing that punishment is a form of communication having a variety of audiences. Once this fact is appreciated, the way is opened to explaining how punishment can have a plurality of aims including the need to express condemnation towards the criminal. To appreciate why interpreting punishment as communication will help to resolve the problem set by the traditional framework, it will be useful briefly to consider the notion of communication using terms typically used by traditional theorists. Suppose, for example, that a child is caught defacing the walls of one of the school's washrooms. The school authorities deal with the problem by giving him 16 and the rest of his class a lecture on the value of school property and end it saying \"don't ever do that deplorable action again.\" The lecture is a communication directed at two different audiences, the child who did the wrong and the rest of the class. For the child who defaced the walls, the message is one of denunciation. Traditional theorists would say it 'looks back' to the wrong action denouncing it and the child who did it. Again using traditional terms, however, the message can also be said to 'look forward' to future interactions by letting that child know what is expected. Moreover, for the rest of the class, the message is forward-looking. It tells them both the standards of conduct and the consequences to expect if the standard is not met. The example shows that a particular communication can and often does incorporate both backward and forward-looking aims. It also shows that a communication may be aimed at more than one audience sending different messages to each. To sum up then, Hart fails to recognize that punishment is a form of communication which, among other things, denounces the individual wrongdoer for his wrong action. Punishment is, in part, an attempt to communicate the victim's and community's emotions to the wrongdoer - how they feel as a result of the crime. While there is hope that the communication will motivate the wrongdoer to change his conduct in future, such a change is not a necessary condition for punishment to be justified. The communication, in and of itself, fulfills a need to respond to the wrongdoer. The idea that punishment is a form of communication is not new. As noted, some theorists working in the traditional framework have developed expressivist 17 theories. Feinberg says that punishment has an expressive function. And in \"Punishment as Language\" Primoratz takes up where Feinberg left off. He considers various attempts to justify punishment as the expression of condemnation. He argues, however, that the various attempts made are forms of extrinsic expressionism which fail to justify punishment because they view it \"as a means only, and as social in the sense that its immediate effects on society are what count most.\"14 In other words, they justify punishment on the basis of the good it accomplishes for society. Primoratz considers these theories to be, at bottom, consequentialist and vulnerable to the familiar criticisms of consequentialist justifications. In particular, he claims they do not rule out punishment of the innocent or mere apparent punishment of the guilty. He proposes, instead, intrinsic expressionism. On his view, punishment is an expression of condemnation which \"indicates the law has been broken, reaffirms the right which has been violated, and demonstrates that the misdeed was indeed a crime.\"15 These stated purposes \"are inherent to punishment: they are not things distinguished from punishment and achieved by means of it, but rather tasks accomplished in punishing.\"16 This understanding and justification is \"entirely backward-looking.\"17 It entails that the state has a duty to punish criminals, \"...its failure to do so would be incompatible with the law it promulgates and the rights it proclaims.\"18 1 4 Igor Primoratz, \"Punishment as Language,\" p. 605. 1 5 Ibid., p. 607. 1 6 Ibid., p. 608. 1 7 Ibid. 1 8 Ibid. 18 Primoratz's defence of intrinsic expressionism falls back on traditional retributivist arguments. First, in the tradition of positive retributivism, he argues that punishment of the criminal is required. Society has an obligation to demonstrate that its laws and the related rights they establish must be respected. And second, he contends that punishment is the only way to accomplish this backward-looking goal. Consequently, the state \"has a duty to punish\" even if \"the condemnation expressed through punishment proves inefficient in preventing future crimes.\"19 As a version of positive retributivism, intrinsic expressionism is susceptible to the usual criticisms of exclusively retributive theories. In particular, it is not at all clear why the aim of social benefit should be subordinated to punishment's backward-looking goals. Furthermore, the claim that punishment is the only way to accomplish society's backward-looking goals is open to challenge. The upshot is that, although Primoratz's theory is significant because it highlights the expression of condemnation as central to the justification of punishment, he offers no new insight as to why we should accept a purely retributive theory -even if it is an expressivist theory. His theory is simply another way of saying that the wrongdoer deserves punishment. Like both consequentialists and retributivists before him, Primoratz fails to acknowledge that the expression of condemnation towards the criminal has its seat in the emotions of those affected by the crime. This is the insight that Primoratz misses. 1 9 Ibid. 19 What is still needed, then, is a theory that justifies punishment by incorporating our commonsense judgments about the victim's and community's need to express condemnation and those about the need to achieve some social benefit from punishment. In the sections that follow I argue that a theory which can do just that is one based on the insight that the need to express condemnation originates in emotions that are experienced when harm is done. Strawson's Insight In administering punishment our justice system attempts to take into account a variety of aims such as desert, deterrence, education, and reform. We require a theoretical approach that both reconciles these aims and accommodates the needs of victim and community to communicate their emotions to the wrongdoer. Peter Strawson, in his seminal account of moral responsibility and the related practices, gives us the key to just such an approach.20 Strawson describes competing views of moral responsibility in a general way as those of the optimist and pessimist. He points out that something vital is left out of the traditional debate between them regarding the practices of praising, blaming, rewarding and, in particular for our purposes, punishing. The optimist's exclusively forward-looking \"social regulation\" account fails to capture the essentially reprobative character of blame and punishment. The pessimist's account also fails, however, because it needlessly goes beyond the facts as we Strawson, \"Freedom and Resentment,\" pp. 59-80. 20 know them claiming that blame and punishment are justified only if responsibility is grounded in true freedom - the Kantian transcendental freedom of the rational self. Optimists and pessimists can be reconciled when both acknowledge the common belief that it matters a great deal to us whether \"the actions of other people...reflect attitudes towards us of goodwill, affection, or esteem on the one hand or contempt, indifference, or malevolence on the other.\"21 As human beings we \"...demand some degree of goodwill or regard on the part of those who stand in ...relationships to us.\"22 (my emphasis) Strawson's argument makes it clear, against the \"social regulation\" view, that the practices of blame and punishment really are expressions of our moral attitudes and not just devices we employ for regulative purposes. \"Our practices do not merely exploit our natures, they express them.\"23 (my emphasis) Strawson's project centres on the notion that the human susceptibility to moral sentiments is a necessary condition of moral responsibility. Our practices of praising, blaming, rewarding and punishing are essentially connected to reactive attitudes that are part of our emotional make-up as human beings, a 'given' of human nature. When a person violates the moral demand for some degree of goodwill or regard by acting with contempt, indifference or malevolence towards another, we react with resentment if personally involved in the interaction or 2 1 Ibid., p. 63. 22 ibid. 2 3 Ibid., p. 80. 21 moral indignation if indirectly involved. When the violation is criminal - that is, when the wrongdoer fails to show even a minimal regard for the moral demand -these moral sentiments are expressed through the social practice of punishment. Strawson, then, provides us with a new way to look at punishment! He brings to light the significance of human moral psychology for the practices associated with moral responsibility - practices such as punishment. A Naturalistic Theory: A Context for the Justification of Punishment In the remainder of this chapter and the chapters that follow, I make use of Strawson's insight to develop a naturalistic theory of punishment. In this section I set out an account of moral agency that centres on natural human capacities. That account provides the resources to explain the moral sentiments and the role they play in interpersonal relationships. In the next section I begin my discussion of the retributive moral sentiments focusing on some significant features that have relevance for the social practice of punishment. A foundation in naturalism typically raises questions. Accordingly, before discussing my naturalistic concept of moral agency, it will be worthwhile to discuss briefly my naturalistic methodology. To begin with, it will be helpful to consider in general how an empirically informed, naturalistic methodology relates to the normative aspects of moral theorizing. Bernard Williams has commented with insight on the general relationship of normative to naturalistic explanation. He begins by reminding us that any attempt to reduce morality to sociobiology is 22 \"naturalistic fallacy territory.\"24 In other words, is does not imply ought. He suggests, nonetheless, that We all ought to remember at this point a companion to 'is does not imply ought,' and which came out of the same stable, is that 'ought does imply can.' Presumably one of the items that sociobiologists might like to deliver to us is a proposition about what we cannot do, what society cannot do, on what sorts of norms it cannot run. Although that will not give us ought, it can give us not ought: if 'ought implies can,' then 'cannot implies not ought'; that is, you had better forget that one as a norm.25 A sociobiological explanation of human nature, then, does not tell us what human beings ought to do - what our normative claims, in effect, should be. Rather, it tells us what human beings are and are not capable of doing.26 This explanation of capabilities serves a twofold purpose in a naturalistic theory. First, it reveals what capacities humans have, thereby providing the factual basis for claims about moral agency. Second, it exposes the limitations of those capacities. It tells us what human beings cannot do without terrific costs. For as Williams notes \"there is a Spielraum, an area in which it is possible for human beings individually - or even for a time societally - to do things of a certain kind, but it is so against the grain that some things are just...too much to ask.\"27 In exposing the limitations of human capacities, the description of human nature / 4 B. A. O. Williams, \"Conclusion,\" Morality as a Biological Phenomenon: The Presuppositions of Sociobiological Research, ed. Gunther S. Stent (Berkeley and Los Angeles, California: University of California Press Ltd., 1980), p. 281. 2 5 Ibid. 2 ^ Williams is careful to point out that when sociobiologists say a human being 'cannot' do something, they do not mean 'absolutely cannot' because in that strongest case of 'cannot' the question of whether we 'could' would never arise. Rather, they mean that human beings in general 'cannot' do such and such without terrific costs. Ibid., p. 282. 2 7 Ibid. 23 indicates what the constraints on a justificatory theory might be, thereby informing the development of the theory. This naturalistic approach to punishment, then, brings an empirically informed description of human nature to bear on the problem of punishment. That description reveals that the experiencing of retributive emotions and the need to express them underlie and, to a great extent, explain the practice of punishment. The recognition that an explanation of the practice must make reference to human moral psychology points to the type of justification required. That justification must take account of both the different perspectives of all those involved and the need for the victim and community to express their retributive sentiments. In short, the justification must take into account not only human cognitive but also human emotional capacities. With the descriptive foundation laid, the theory goes on to consider the actual practice of punishment. That practice is then criticized and evaluated on the basis of reasons and within the context of accepted moral and social norms. Given this general characterization of methodology, we can now develop a naturalistic account of moral agency. In the ongoing debate about moral responsibility and the related practices, contemporary compatibilists have made headway by avoiding attempts to ground moral concepts in metaphysical assumptions. Instead, compatibilists such as Frankfurt, Watson and Wolf attempt to discern and understand the natural human capacities necessary for 24 moral agency.28 Strawson takes a different but complementary tack focusing on the actual practices involved in holding people responsible. Combining these insights proves fruitful for our understanding of moral agency. First consider what compatibilists concerned with the structure of a person's will have to say about the capacities needed for responsible agency. On the contemporary view, a moral agent has two key capacities, moral understanding and self-evaluation. His distinctly rational capacity of moral understanding allows him to grasp and apply moral reasons both to practical problems regarding his actions and to evaluation of his own character. The related capacity of self-evaluation allows him critically to reflect on and evaluate both the beliefs, desires and feelings that constitute his character and the ends he is inclined to pursue. Moreover, the moral agent is able to bring the judgments of his critical reflection to bear on his character. He has the capacity to modify or at least attempt to modify his character in the light of criticism. He may endorse some desires and reject others. He may accept some of his feelings as appropriate and reject others as inappropriate. Implicit in this description of the moral agent is the notion that he is an intentional agent. If asked to explain his motivations or actions, he can give an intentional explanation. And, significantly, he recognizes that he is one among other similar intentional agents who make up the society in which he lives. With 2 8 These contemporary compatibilists present hierarchical accounts which focus on an agent's capacity to distinguish mere desires from the reflective desires that he endorses and wants as his wil l . See Harry G. Frankfurt, \"Freedom of the Wil l and the Concept of a Person,\" Free Will, ed. Gary Watson (New York: Oxford University Press, 1982), pp. 81-95; Gary Watson, \"Free 25 that recognition comes the expectation of due regard from other similar agents. In other words, every intentional agent recognizes that he has the capacity to set standards to live by and be judged by and he expects others to do the same. As Strawson points out, the moral demand for goodwill is the expectation of due regard that such agents call for - the minimal expectation as it were. Underlying this minimal expectation is the picture of an agent having the essential capacities of moral understanding and self-evaluation. These capacities distinguish normal adult human beings from animals, children and the mentally ill.2 9 And, according to contemporary compatibilists, it is in virtue of an agent's possessing these capacities that we view him as having dignity.30 There is more to the story, however. As one among many who demand at least a minimal degree of goodwill, the agent does not do critical self-evaluation in isolation. Most often, it is others in society who supply him with the raw material. Herein lies the connection between the compatibilist picture that focuses on the agent's capacity to have the will he wants to have and the Strawsonian picture that depicts the agent as one who is deeply affected by the attitudes and feelings of others. Agency,\" Free Will, ed. Gary Watson (New York: Oxford University Press, 1982), pp. 96-110; Susan Wolf, Freedom Within Reason (New York: Oxford University Press, 1990.) 2 9 We view children as potential moral agents who are developing the necessary capacities. We consider the mentally i l l to be potential moral agents who have impaired capacities or perhaps even lack them altogether. 30 My discussion here relies a great deal on Paul Russell's brief but insightful analysis of moral agency in Freedom and Moral Sentiment: Hume's Way of Naturalizing Responsibility (New York: Oxford University Press, 1995), pp. 91-2. 26 In keeping with the spirit of contemporary compatibilism then, I assume that moral agency has its seat in the natural capacities of moral understanding and self-evaluation. I submit, however, that these capacities are best explicated in a framework which highlights another significant natural capacity, that of empathy. As Strawson indicates, moral agency can really be understood only in a social context which takes into account both the agent's interaction with others and the values and goals of society as a whole. Human beings are motivated to evaluate their feelings, desires and goals because they have the capacity for empathy and, therefore, are affected by others' expressions of approval or disapproval. The capacity for empathy is not mentioned by theorists concerned with the structure of the agent's will. Nonetheless, if we look at the broad picture, the everyday picture of humans reacting and responding to one another in a social setting, it is clear that they are affected by and often motivated by others' expressions of approval or disapproval. A complete picture of the moral agent, then, must include this crucial capacity of empathy. Human beings are moral agents and appropriate objects of moral sentiment because they possess the three capacities of moral understanding, self-evaluation and empathy. It is in virtue of the agent possessing these three capacities that we view him as having dignity. In other words, we can formulate a norm of human dignity based, not on a Kantian metaphysical conception of the individual as having strong free will, but on this naturalistic description of the agent. I elaborate on this notion of dignity in Chapter 4. 27 At this juncture I want to discuss the capacity for empathy. For it is through viewing the moral agent in the context of interpersonal relationships which call upon the capacity for empathy that we can begin to appreciate the relevance and importance of moral sentiments to our social practices. The assumption that humans are social and mutually caring is not new. It has historical roots. For example, in The Theory of the Moral Sentiments, Smith notes at the outset, How selfish soever man may be supposed, there is evidently some principles in his nature, which interest him in the fortunes of others, and render their happiness necessary to him, though he derives nothing from it except the pleasure of seeing it. Of this kind is pity or compassion, the emotion which we feel for the misery of others, when we either see it, or are made to conceive it in a very lively manner. That we often derive sorrow from the sorrow of others is a matter of fact too obvious to require instances to prove it; for this sentiment, like all the other original passions of human nature, is by no means confined to the virtuous and humane, though they perhaps may feel it with the most exquisite sensibility. The greatest ruffian, the most hardened violator of the laws of society, is not altogether without it.3 1 With the dominance of Kantian thought in philosophy, however, the human capacity for empathy fell into disrepute and was relegated to a minor role. It is only in recent times with the rise of the relatively new science of sociobiology and the formulation of theories of moral development in psychology that the capacity for empathy has again gained prominence in moral theorizing. Naturalistic approaches share similar assumptions regarding morality with these contemporary scientific approaches. They all consider human beings to be social 31 Smith, The Theory of the Moral Sentiments, I.i.1.3, p. 9. 28 animals whose moral behavior evolved out of strategies for optimizing human welfare.32 Those strategies are social. Strawson's insight comes in pointing out the significance of our social interactions for morality - not merely interactions with family and friends but also those with chance parties in an enormous range of encounters. He marks the fact that the attitudes of others towards us and our reactions towards them are essential components of moral responsibility and the related practices. To be sure, those attitudes and reactions call upon our capacities of moral understanding and self-evaluation because they are backed up by reasons. But they also call upon our capacity for empathy because they are experienced and expressed through emotions. Strawson's commonsense observations regarding the emotional basis of interpersonal relationships are supported by empirical evidence. In defining emotions from the evolutionary point of view, Robert Plutchik says, \"emotions are phylogenetically ancient complex types of reactions that communicate information from one individual to another, that regulate the reactions among individuals, and that contribute to the likelihood of survival.\"33 (my emphasis) And even more to the point, in Emotional Intelligence: Why it can matter more than IQ, Daniel Goleman notes that animals that have the amygdala, the part of the limbic brain that is the seat of the emotions, removed or severed \"lack fear ^2 Gunther S. Stent, \"Introduction,\" Morality as a Biological Phenomenon: The Presuppositions of Sociobiological Research, ed. Gunther S. Stent (Berkeley and Los Angeles, California: University of California Press Ltd., 1980), p. 1. 33 Robert Plutchik, The Psychology and Biology of Emotion (New York: Harper Collins College Publishers, 1994), p. 364. 29 and rage, lose the urge to compete or cooperate, and no longer have any sense of their place in the social order; emotion is absent or blunted.\"34 (my emphasis) This evidence indicates that emotions are essential to the interpersonal interactions which are basic to human society. How is the capacity for empathy related to the other key capacities of the moral agent? As Strawson points out, the practices associated with moral responsibility arise out of a mutual expectation of a certain degree of goodwill. In an interaction, when one person expresses an attitude towards another through word or action, that expectation is either met or not met. If it is met, the corresponding emotional reaction of the recipient in the encounter is a positive moral sentiment. If it is not met, the recipient experiences a negative moral sentiment. The moral sentiment may be expressed either verbally or nonverbally. A positive reaction communicates approval, a negative one, disapproval. The person who is the object of the moral sentiment expressed, in virtue of his capacity for moral understanding, can understand the other person's reasons for reacting with approval or disapproval. If it is disapproval that is communicated, the object of the moral sentiment, in virtue of his capacity for empathy, may well be motivated to take up self-evaluation. To sum up then, moral agents are capable of understanding and responding to others' expressions of approval and disapproval in virtue of these three natural capacities. 0 ( 4 Daniel Goleman, Emotional Intelligence: Why it can matter more than IQ (New York: Bantam Books, 1995), p. 15. 30 The naturalistic theory claims that punishment is, among other things, an expression of strong disapproval. In other words, it communicates the victim's and community's retributive moral sentiments. In the next section I discuss those moral sentiments. The Retributive Moral Sentiments: Resentment, A Paradigm Case In this section I examine the retributive moral sentiments which are, I shall argue, inevitably experienced when one is harmed by another person. What are they? Why do they arise? What distinguishes them as moral sentiments? Do we need to express these negative emotions? I begin by looking at Strawson's and Mackie's accounts. Strawson sets out a distinction between the retributive emotions of resentment and moral indignation which I shall use. And Mackie offers an illuminating hypothetical explanation of the spontaneous, reprobative and persistent character of these responses to morally wrong actions. Consider first the distinction between resentment and moral indignation. The victim of a crime naturally reacts with intense personal resentment. This moral sentiment is a negative response to the wrongdoer because of his wrong action. On a Strawsonian interpretation, it is a natural psychological reaction which arises because the wrongdoer failed to meet the moral demand for a certain degree of goodwill which underlies human interpersonal relationships. It is important to understand that the naturalistic theory incorporates a cognitive theory of emotions. The victim's resentment is an emotion that arises naturally, 31 but it has cognitive content. Consequently, it can be evaluated as appropriate or not. The psychological reactions of the members of the community to a crime are closely related to, but different from, that of the victim. According to Strawson, the reactions of others to the plight of the person directly involved \"...rest on, and reflect, exactly the same expectation or demand [of a certain degree of goodwill] in a generalized form.\"35 The idea is that, when one reacts to a situation in which one is not personally involved, one views the situation with a degree of disinterest that is often absent from personal interactions. The experience of moral indignation reflects sympathy for the victim but it tends to be less intense than the personal reaction of resentment. Nonetheless, like the victim's reaction, the community's impersonal reaction rests on and reflects the moral demand for a certain degree of goodwill. Moreover, it too has cognitive content and can be evaluated as appropriate or not. A key claim in the naturalistic account is that it is part of our emotional make-up as human beings to experience retributive sentiments when a harm is done. In other words, in the absence of some pathology, experiencing these emotions is inevitable. It is compatible with this inevitability thesis that some individuals may intentionally decide to give up experiencing these attitudes, for example, for religious reasons. For, as Williams notes, human psychology is such that it is possible for human beings individually or even for a time societally to do certain Strawson, \"Freedom and Resentment,\" p. 71. 32 things even though doing so goes against the human psychological 'grain.' Strawson suggests, nonetheless, that any wholesale suspension of these sentiments by human beings is \"practically inconceivable.\"36 He offers only a sketchy explanation as to why we could not give them up, though. Mackie, however, sets out a plausible hypothetical account that explains their inevitability. Mackie considers the question of why human beings have an \"ingrained tendency\"37 to respond with hostility to wrong actions. He suggests that the tendency to experience resentment or moral indignation has a biological explanation in the theory of natural selection. It is probable that \"non-moral retributive emotions\"38 developed as an accompaniment to defensive retaliatory tendencies which evolved to ward off future injury by aggressors. The further development of resentment and moral indignation out of their non-moral counterparts is explained through the evolution of social cooperation. In the social context these moral sentiments play a strategic role in interaction. Members of the community influence one another to comply with morality by reacting with retributive emotions when the moral demand for goodwill is disregarded. According to Mackie, these retributive emotions are spontaneous and \"essentially connected with previous harmful...actions.\"39 3 6 Ibid., p. 68. 3 7 Mackie, \"Morality and the Retributive Emotions,\" p. 215. 3 8 Ibid., p. 216. 3 9 Ibid., p. 219. 33 Mackie's hypothetical account is useful because it explains the spontaneity and inevitability of our experiencing these emotions in evolutionary terms. But it over-emphasizes their instinctive component. Although Mackie concedes that resentment of certain kinds of behavior and the social interaction that supports it in society \"presupposes fairly advanced intellectual powers associated with the retributive emotions,\"40 he does not distinguish the retributive sentiments themselves as having cognitive content and, therefore, being susceptible to rational justification. Nor does he make a distinction between experiencing them and expressing them. What, then, is a moral sentiment? To begin with, it is worth pointing out that a moral sentiment may judged justified or not. I offer some examples of unjustified moral sentiments presently. A justified moral sentiment is an \"intelligent\" emotion in the sense that it conveys a \"considered\" emotional reaction - one which is based on reasons and takes account of the circumstances. It is not an uncontrolled passion such as vengefulness which is characterized by inappropriate intensity or a failure duly to consider facts and circumstances. A moral sentiment can be evaluated, judged appropriate or not and controlled. It is not a mere feeling - a pleasant or unpleasant sensation such as a pounding heart or queasy stomach. While there is no denying that a moral sentiment may have a physiological component, it is, in part, a cognitive state of belief. It is about something. It is a way of seeing and engaging with the world. In short, a moral sentiment is a complex emotion that involves a belief, an evaluation and a relation to action. 4 0 Ibid., p. 218. 34 Consider a paradigm case - the victim's resentment.41 Resentment is a negative attitude directed at the wrongdoer because of his wrong action. It does not arise on injury by a natural disaster or an animal - at least not appropriately. It is experienced on being harmed by a person. The victim feels devalued. Her resentment arises naturally and inevitably. In other words, it is the natural reaction towards one who does not fulfill the moral demand for goodwill. In the case of resentment, the cognitive content consists in a belief that the wrongdoer did the crime. But the resentment also involves the victim's corresponding evaluation of the wrongdoer. It is an attitude towards him that stems from viewing him in the light of the belief that he did the harm and related beliefs about his moral qualities. The wrongdoer is evaluated as uncaring of the victim's value. The victim infers that he is capable of respecting her dignity but failed to do so. As Jeffrie Murphy points out, the reason we deeply resent moral injuries done to us \"...is not simply that they hurt us in some tangible or sensible way; it is because such injuries are also messages - symbolic communications. They are ways a wrongdoer has of saying to us, T count but you do not,' T can use you for my purposes,'...Intentional wrongdoing insults us and attempts (sometimes successfully) to degrade us....\"42 4 1 The concept of emotion is obscure and in need of a sustained philosophical analysis which is beyond the scope of this study. In this analysis of resentment I draw on Paul Russell's helpful discussion of moral sentiments. Russell, Freedom and Moral Sentiment, pp. 88-90. 4 2 Jeffrie Murphy and Jean Hampton, Forgiveness and Mercy (New York: Cambridge University Press, 1988), p. 25. 35 Resentment has, in addition, an affective component. It arouses in the victim desires or feelings directed at the wrongdoer. Those desires motivate the victim to act in ways characteristic of the emotion. The attitude that arises in the resenter is one of defiance in which she denies to herself and others the presumption, fostered by the wrongdoer's action, that she is low in value.43 This psychological reaction is analogous to one's physical reaction of striking out in self-defence at an assault. It motivates the victim to get back at the wrongdoer by expressing her resentment. Failure to express it, may signify acceptance of the wrongdoer's evaluation. And acceptance would indicate that she suffers some psychological pathology such as a severe lack of self-esteem.44 For example, a woman who suffers from the \"battered woman syndrome\" is a victim who does not react with resentment when she is beaten by another person. Instead, she accepts the batterer's assessment that she deserves to be beaten. She should resent the beatings. The fact that we judge her lack of resentment to be pathological indicates that experiencing resentment is an appropriate reaction to the deliberate harm of another.45 4 3 Hampton indicates that resentment is an act of defiance. It is important to recognize that resentment is an attitude, not an act. Experiencing the moral sentiment can be distinguished from expressing it. Hampton also points out that, in some cases, resentment involves the victim's fear that the wrongdoer's treatment of her is appropriate because she actually is low in value. I think Hampton is correct. Her analysis may point to the reason why some individuals are repeatedly victimized. Ibid., p. 56. 4 4 Murphy argues that resentment functions as a defence of certain values of the self. He says that the primary value defended is self-respect. On Murphy's view, the person who does not resent moral injuries done to him \"is almost necessarily a person lacking in self-respect.\" Ibid., p. 17. 4 5 For those cases i n which the victim of a crime has such low self-esteem that she fails to resent the wrongdoer's devaluing treatment it wil l be up to the community to stand up for the victim and express moral indignation (through punishment) on her behalf. 36 As a moral sentiment involving a belief, evaluation and affective component, the victim's resentment can be judged as justified or not. It is unjustified if the belief about the wrongdoer is false. For example, resentment towards the accused thief, Boggs, would be unjustified if Boggs did not, in fact, snatch the victim's purse but was mistakenly apprehended because he was in the vicinity and fit the description of the real thief. The resentment is unjustified if the victim's evaluation of the wrongdoer is incorrect. If Jones snatched the purse because he was coerced into doing so by a gang that was threatening his life, the victim's evaluation of Jones as uncaring of her value would be incorrect and unjustified. Finally, the affective component of the resentment may be inappropriate. If, on having her purse snatched, the victim's resentment is so intense that she would express it by cutting the thief s hands off so he could never steal again, we would judge her resentment unjustified because it is too intense. Moral sentiments, then, are natural but their instinctive component is not nearly as significant as Mackie implies. They can be justified. I also pointed out that Mackie fails to distinguish experiencing these retributive emotions from expressing them. The upshot is that his claim that experiencing these sentiments is inevitable can be distinguished from the question of whether expressing them is also inevitable. Do the retributive sentiments need to be expressed? I want to put forward the hypothesis that expressing these emotions is inevitable. I want to emphasize, however, that we have a choice about how to express them.46 In other words, the fact that we inevitably experience justified 4 t > For a discussion of this point in relation to Hume's teleological retributivist theory of punishment, see Russell, Freedom and Moral Sentiment, pp. 148-9. 37 moral sentiments does not entail that their expression through punishment is also inevitable and justified. Consider, for example, how the victim of a date rape might deal with the resentment she feels. On the one hand, there are a number of ways that she might express her resentment. She might confide to a friend about the rape. She might verbally express anger and resentment to the wrongdoer himself in the hope that communicating her feelings will influence his character and future actions. Or she might report the crime so that the rapist will be put on trial, publicly denounced and punished. On the other hand, she might feel so degraded by the devaluing experience that she suppresses her resentment never expressing it to anyone. It is true, then, that the expression of a retributive emotion may be repressed by a particular person in a particular circumstance. But a particular instance of non-expression does not indicate a commitment to never expressing one's moral sentiments. In the case of the date rape the victim, who feels too degraded to say anything to anyone, has what she considers to be a good reason for not expressing her emotion. But if a person never expressed her emotions, we would consider her failure to do so to be pathological. It might be objected that we actually admire some persons for their commitment to the non-expression of retributive emotions - persons such as Mahatma Gandhi, for example. I suggest, however, that Ghandi's advocacy of passive resistance and hunger strikes is not an example of a commitment to non-expression but rather an illustration of the fact that we have a choice about how to express our moral sentiments. His public demonstrations of passive 38 resistance in Africa, for instance, were intended both to express moral indignation over racism and to communicate to the community the need for reform. While I concede that it is possible for some few individuals to commit to the non-expression of moral sentiments, the wholesale repression of their expression by all human beings is neither possible nor desirable. It is not possible because our psychological make-up seems to require that we express our emotions. It is not desirable because living in society involves communicating with others. Expressing our moral responses towards others is a form of communication that has importance not only for our personal relationships but also for our institutionalized social practices. Although there is no possibility of suspending or abandoning the expression of retributive sentiments altogether, the naturalistic justification does not claim, as Primoratz's expressive theory does, that punishment is the only way to express condemnation of the criminal. The community does have a choice about how to express its retributive sentiments. I argue in the chapters that follow, however, that punishment is most often the appropriate way to express them given the gravity of the harm done.47 In summary then, in this chapter I claimed that traditional attempts to justify punishment fail because they downplay or ignore the emotions of those involved in the practice. Drawing on Strawson's insight that our social practices are essentially connected to moral sentiments that are part of our emotional make-4 7 In some circumstances other responses, such as treating the criminal with mercy (a reduced sentence) or even forgiving him (a pardon), might be appropriate. In Forgiveness and Mercy Hampton and Murphy offer a comprehensive study of these topics. 39 up, I pointed out that a justification for punishment must take those moral sentiments into account. I then took the initial steps towards explicating the naturalistic justification, setting out my naturalistic concept of moral agency and explaining what moral sentiments are. I argued that, when harm is done, retributive moral sentiments are inevitably experienced and expressed but that moral agents have a choice about how to express them. In the next chapter I explicate the social institution of punishment as a form of communication which aims to communicate retributive sentiments as well as various other messages from the stakeholders in the social institution to various audiences. 40 Part I. The Theory Chapter 2 Punishment: Communication with Various Audiences In the previous chapter I pointed out that the victim's and community's retributive moral sentiments underlie and, to a great extent, explain the social practice of punishment. I argued that, in contrast with traditional theories, a naturalistic theory has the resources to incorporate this key explanatory element into the justification. In this chapter I am concerned with a second key explanatory element which comes to light in the naturalistic context. I advance the hypothesis that punishment is a type of communication from various players to a variety of audiences. In developing the hypothesis I analyze the different perspectives of each of the stakeholders \u00E2\u0080\u0094 wrongdoer, victim and members of the community \u00E2\u0080\u0094 with respect to punishment and its aims. In the course of the explication it becomes clear that a theory that accommodates these various perspectives must incorporate a plurality of aims. The Moral Demand: The Unifying Theme In Chapter 11 noted that there have been a number of attempts to justify punishment along expressivist lines. Most expressivists have emphasized the notion that punishment is an expression of the community's condemnation though they sometimes incorporate related themes such as the idea that it is a form of moral education. Primoratz talks of punishment both as expression and as communication apparently using the terms interchangeably. Here I want to 41 shift the emphasis slightly, focusing on the notion of communication. Interpreting punishment as a form of communication has two important advantages. First, it suggests an alternative framework for explicating punishment's plurality of aims. This approach brings to the fore the idea of communicating from various stakeholders to various audiences. The victim, community and wrongdoer all have a crucial interest in the social institution and are also audiences for the communication. In the role of stakeholder, each wants the social institution to communicate various messages on his behalf to a variety of audiences in the community. The victim, for example, wants to communicate condemnation to the wrongdoer and a rejection of devaluation and assertion of self-respect to all. In other words, each of the stakeholders has a different perspective on the crime and different aims in inflicting punishment. It is worth mentioning that, when punishment's aims are set out in this alternative framework, there is no obvious overlap with the traditional interpretation of punishment's purposes as either backward or forward-looking. As I pointed out in the previous chapter, a particular communication can, at the same time, be both backward and forward-looking. And in fact, it will become clear that each of the stakeholders in punishment has both backward and forward-looking messages to convey to his audiences. A second advantage of interpreting punishment as communication is that it provides the opportunity to emphasize that the expression through punishment of condemnation, among other things, is intentional. The punishing authority 42 intends to convey to the wrongdoer, victim and community both that the punishment is being imposed because the wrongdoer did the crime and that condemnation, among other things, is being expressed to him for that reason. Moreover, it is intended that both wrongdoer and the public recognize that the punishment is being imposed for that reason and that they realize they are intended to recognize this fact.1 The shift in emphasis from expression to communication is particularly important for this naturalistic approach because of the crucial role that retributive sentiments play in explaining punishment. In day-to-day interpersonal relations individuals may express various emotions having no particular intention in mind. In other words, an expression of emotion may be done for its own sake, as a way of venting the emotion. In the case of the social institution of punishment, however, the punishment is intended to do more than merely express retributive emotions on behalf of the victim and community. The communication through punishment is intended to have meaning and that meaning is intended to be apparent to the victim, the community and the wrongdoer. I consider how the communication of punishment's various aims can be done most effectively initially in Chapter 4 and in greater detail in Chapter 6 where I consider sentencing options. 1 In his discussion of retributive punishment, Robert Nozick notes the importance of the intentional aspect of the message communicated through punishment. He also points out that this intentional view of communication - \"wherein something intentionally is produced in another with the intention that he realize why it was produced and that he realize he was intended to realize all this\" - fits H . P. Grice's account of meaning. Robert Nozick, \"Retributive Punishment,\" Philosophical Explanations, (Cambridge, Ma.: The Belknap Press of Harvard University, 1981), p. 370. 43 At first glance it may appear that interpreting punishment as communication of a variety of messages to various audiences is simply a facile way to include in the justification all the reasons we might ever want to include. Let there be no doubt that it is a major theme of this thesis that the social institution of punishment has a plurality of aims. Nonetheless, those aims do have a unifying theme -communication of the \"moral demand.\" Strawson articulates the moral demand in the following way: \"In general, we demand some degree of goodwill or regard on the part of those who stand in...relationships to us, though the forms we require it to take vary widely in different connections.\"2 In the case of criminal punishment, we are concerned with appropriate responses to failures to meet even the minimal demand for goodwill in the many social relationships required for the functioning of society. On the face of it, the moral demand may appear to be the Kantian demand that we treat each individual as an end worthy of our moral respect and never merely as a means. It is not. The Kantian and naturalistic demands to respect dignity are grounded in differing concepts of moral agency. For the Kantian, the practices of praising, blaming, rewarding and punishing are based on metaphysical assumptions about moral agency.3 On the naturalistic view, by contrast, moral agency is constituted in natural human capacities. One's dignity 2 Strawson, \"Freedom and Resentment,\" p. 63. 3 Various theorists concerned with the practices related to moral responsibility have questioned the appeal to metaphysical assumptions to explain and justify these practices. For example, in \"Freedom and Resentment\" Strawson rejects the \"panicky metaphysics\" of libertarianism. Ibid., pp. 60-2, 79-80. In Responsibility and the Moral Sentiments R. Jay Wallace rejects metaphysical interpretations which postulate facts about responsibility \"prior to and independent of our practice of treating people as responsible moral agents.\" R. Jay Wallace, Responsibility and the Moral Sentiments (Cambridge, Ma: Harvard University Press, 1994), p. 87. 44 has its seat in the natural capacities of moral understanding, self-evaluation and empathy. And it is one's natural expectation of goodwill from other people that justifies the practices associated with moral responsibility.4 When the wrongdoer commits a crime, however, she fails to meet even that minimal requirement. The victim and members of the community naturally react with retributive emotions. At first glance, it may seem that the expression of hostile emotions towards one who has offended does not treat him with dignity. Strawson suggests, however, that it does. He says, The partial withdrawal of goodwill which these attitudes entail, the modification they entail of the general demand that another should, if possible, be spared suffering, is, rather, the consequence of continuing to view him as a member of the moral community; only as one who has offended against its demands. So the preparedness to acquiesce in that infliction of suffering on the offender which is an essential part of punishment is all of a piece with this whole range of attitudes....5 In fact, the worry that the social institution of punishment disregards the wrongdoer's dignity is a pressing problem for all theories of punishment, not just this one. I address this worry in depth in Chapter 4. It is worth noting here that the first part of the argument in Chapter 4 is especially relevant to the hypothesis set out in this chapter. There I continue with the explication of punishment as a type of communication, focusing on the type of communication that it is \u00E2\u0080\u0094 in part at least, nonverbal communication. 4 In a justification for punishment clearly the focus is on negative practices that are the human response to a failure to meet the natural expectation of goodwill. It is worth mentioning that the associated naturalistic account of moral responsibility is symmetric. In other words, when the expectation of goodwill is met, the natural response is a positive one. When it is surpassed, human beings respond with praise and, in some cases, rewards. 5 Strawson, \"Freedom and Resentment,\" p. 77. 45 In the next three sections I consider crime and the ensuing punishment from the differing perspectives of each of the stakeholders. What does each of them want to communicate and who is his target audience? In other words, I examine the moral psychologies of wrongdoer, victim and community in order to understand their relationships to one another and their needs in the face of harm done. The claim that punishment has a plurality of aims introduces a certain unavoidable complexity into the task of justification. In the table that follows (on page 47) I organize the notions involved by categorizing the communicative aims of punishment from the differing perspectives of wrongdoer, victim and community.6 For each aim, the table indicates the emotion or attitude expressed, its relationship to the moral demand, its characteristics and the target audience of the communication. It may be helpful to refer to this table in the course of the discussion in this and upcoming chapters. The Wrongdoer's Disregard of the Moral Demand The problem of punishment begins with the crime. Let us first examine, then, the wrongdoer's perspective. In committing the crime he fails to meet even the minimal moral demand. Why? What is his moral psychology? While there may be no answer that applies in all cases, empirical research reveals some attitudinal trends. Ezzat A. Fattah points out that, prior to victimization, the wrongdoer undergoes one or more mental processes that can be roughly b The actual punishing authority, the state, is not mentioned on the table because ideally the state should reflect the various perspectives and related aims of all the stakeholders and communicate them through the social institution of punishment. Giving each of the stakeholders - community, wrongdoer and victim - a distinct role within the social institution is one way of ensuring that each perspective is represented. 46 \u00E2\u0080\u00A2 CO to (0 > \u00C2\u00A9 *1 CD O c O \"D 3 \"D o E o s 8 s E I s | .1 g o a\" \u00E2\u0080\u00A2s I S '\u00C2\u00AB 8 \u00C2\u00BB nj o 2 F T ! ' = s \u00E2\u0080\u00A2 3 o \"> E5 .s 8 i ! 2 c g i C 2> Q. 11\u00C2\u00AB= 01 E J! \u00E2\u0080\u0094 co s< SZ S O => = to *- 7: W a. 5 ^ CD co CD ci) \u00C2\u00AB ( / ) ( / ) 1 \u00C2\u00B0 8 S i \u00C2\u00A7 E o ~ Q) T J CC T3 ^ 5 CD ra 8 S > \u00C2\u00A7* S 2 CD i - o \u00C2\u00A3 \u00C2\u00AB\u00C2\u00A3 \"> \u00C2\u00B0 O \u00C2\u00A3 \u00C2\u00A9 c c \u00E2\u0080\u009E <= .2 o CO \u00E2\u0080\u0094 I S , , S 1 s g O \u00C2\u00AB Q- O (j \"D CO CO E > O -o \u00E2\u0080\u0094 CD \u00E2\u0080\u009E_ I I I CO CO I f , CO CO . E CD * \"o b. ca \u00C2\u00B0> \u00E2\u0080\u009E E l CD c o ? CO _ E - S c o co ,0 s.f I E s _ \u00C2\u00A3 .0 CO CO CO = I S ? o o C TJ I ~ g. E J a> \u00C2\u00BB cc Q o CO CD \u00C2\u00A3 E CO CD 2 E a I TI I f ? o a o CD lo-c i -1\u00C2\u00B0 TT C C CD o E CO T3 CD co CD U (fl CO ai 2 . _ ; CD SZ = o 1 o 8 >\u00E2\u0080\u00A2 o>-\u00C2\u00A3 1 5 3 CD E \u00C2\u00A3 4 7 distinguished as neutralization, redefinition or desensitization.7 These processes, which take place before victimization occurs, work towards breaking down the mechanisms of social control and avoiding guilt. Neutralization is the label given to a group of techniques used to justify breaking the bonds of social control. These techniques include denial of responsibility, condemnation of the condemners and appeal to higher loyalties. Delinquents, for example, commonly utter rationalizations such as \"He did it first,\" \"He is a no-good so-and-so himself,\" \"He had it coming to him.\"8 In redefinition the wrongdoer re-describes the criminal behavior to make it acceptable. Fattah recounts, for example, the actions of a church steward who sent boys to steal coal from railway cars. The steward did not perceive his action as theft but as a way for the poor church to get its coal supply for the harsh winter. On redefinition, the wrongdoer is able \"to engage in the victimizing behavior while avoiding guilt, damage to self-image, self-indignation, and the condemnation of all those who share the same redefinitions.\"9 (my emphasis) Desensitization refers to another group of techniques which permit the wrongdoer to inflict on a fellow human being the pain and suffering the crime will cause. Fattah notes that \"unless the victimizer becomes desensitized in advance, the victimization is bound to create moral tension and to elicit feelings of guilt, shame, 7 Ezzat A. Fattah, Understanding Criminal Victimization: An Introduction to Theoretical Victimology (Scarborough, Ontario: Prentice-Hall Canada Inc., 1991), p. 136. 8 Ibid., pp. 137-8 9 Ibid., p. 138. 48 remorse and reproach in the perpetrator.\"10 The various techniques include denial of the victim, objectifying her, denying any injury, blaming the victim and devaluing her. In rape, for example, the woman is negated as an equal human being. She is both objectified and devalued.11 Guilt and the related emotions mentioned are distinguished by Strawson as self-reactive attitudes, those \"associated with demands on oneself for others...such phenomena as feeling bound or obliged (the 'sense of obligation'); feeling compunction; feeling guilty or remorseful....\"12 The wrongdoer avoids experiencing such attitudes through the processes mentioned. From the victim's and community's perspectives, one of the aims of punishment is to attempt to tap the wrongdoer's capacity for empathy and motivate him to take up self-evaluation. The hope is that he will eventually experience self-reactive attitudes such as guilt which he has, in all likelihood, avoided. The wrongdoer, then, is obviously one of the audiences for punishment as communication. But he is also one of the stakeholders in the social institution. He cannot avoid the coercion of punishment if he is convicted. Nonetheless, he wants his dignity to be respected. The social institution must be designed to ensure that it is. Moreover, if the wrongdoer does acknowledge his guilt \u00E2\u0080\u0094 perhaps at the trial or during the course of his sentence, he may want to make reparation 1 0 Ibid., p. 139. 1 1 Ibid., pp. 139-40. 1 2 Strawson, \"Freedom and Resentment,\" pp. 71-2. 49 to the victim or community. Sentencing should be designed with options that allow for this possibility \u00E2\u0080\u0094 options such as victim/offender mediation, for example. The mention of options for sentencing raises an obvious prior question about punishment as communication. That is, why should society resort to punishment instead of using some other, more benign form of communication? We take the first steps towards answering that question by considering the crime from the perspectives of the victim and then of the community. The Perspective of the Victim We saw in the previous chapter that the victim's resentment is an essentially personal, hostile response that tends to be intense. The naturalistic justification claims that the need to express this emotion is a justifying reason for punishment. Consequently, this account must address some questions related to the victim's moral psychology. Is the victim's need to express this emotion justified? If it is, should the victim's personal response to the wrongdoer be reflected in the institutionalized response? If, as I shall argue, it should be incorporated in some way, the question arises as to how the victim's needs can be met. I consider that question in Chapter 6. One other question relates to the fact that the naturalistic theory allocates a distinct role to the victim yet many crimes are called \"victimless.\" I address that question in the next section. 50 The justice system grew out of the need to restrain the personal, overly intense reaction of vengefulness, so-called \"wild justice.\"13 It is, perhaps, for this reason that the retributive emotions in general and resentment in particular have been disparaged.14 In Chapter 11 argued, however, that resentment is a moral sentiment which can be evaluated by the community and judged justified or not. Justified resentment is an appropriate reaction to being intentionally harmed by another. In fact, we judge those who fail to resent deliberate harm - victims of the \"battered woman\" syndrome for example - to have a pathological condition. I also contended that retributive emotions are not only inevitably experienced but also inevitably expressed. So, although some victims in some circumstances may decide not to express their resentment, most want and need to express their retributive emotions. The victim has been harmed and devalued. She needs to defy openly and publicly the wrongdoer's attack on her value as a person. From the victim's perspective, then, an appropriate, public expression of her justified resentment through the justice system serves a number of purposes. First and foremost, it expresses how the victim feels about the crime. It is a form of catharsis - about which I have more to say below. It is, at the same time, more than a mere expression of emotion. The institutionalized expression of resentment on behalf of the victim is intentional. It is an attempt to communicate with the wrongdoer - to move him to feel guilt about the crime and 13 The full quotation from Bacon's \"Of Revenge\" is \"Revenge is a kind of Wilde Justice; which the more the Mans nature runs to, the more ought Law to weed it out.\" Francis Bacon, \"Of Revenge,\" The Essayes or Counsels, Civill and Morall, ed. Michael Kiernan (Toronto, Oxford University Press, 1985), p. 16. 1 4 Nozick discusses the distinction between retribution and revenge in detail. Nozick, \"Retributive Punishment,\" pp. 366-74. 51 acknowledge the victim's value. Fattah points out that the majority of wrongdoers are not unfeeling psychopaths. They have the capacity for empathy but undergo redefinition and desensitization in order to avoid self-reactive attitudes such as guilt. The communication of condemnation is intended to tap the wrongdoer's capacity for empathy thereby motivating him to take up self-evaluation about his crime and its effect oh others. The expression of resentment is also intended to assert to all the victim's value as a person. It serves as a declaration of self-respect telling the world that neither the wrongdoer nor others can get away with devaluing her. It demonstrates that the victim cares about herself and expects others to care about her too. Here it is important to recognize that self-respect is not only personally important. It has a social side. We care about what others think of us. Moreover, what others think may affect what one thinks of oneself. This social aspect of self-respect is commonly exhibited in concern for one's reputation, the public symbol of one's value. I mentioned above that the expression of resentment is a form of catharsis. The idea here is that the victim will be under psychological pressure to express it. This notion of catharsis is not based on a crude Freudian \"hydraulic model\"15 of emotions in which the human psyche is viewed as \"...a cauldron of pressures demanding their release in expression and action.\"16 But neither is it based on a !5 This psychological model originated with Thomas Hobbes. In this model, the emotions, which are not under the agent's control, are explicated as pressures from the unconscious that enter consciousness demanding to be discharged. Robert C. Solomon, The Passions (Garden City, New York: Anchor Press/Doubleday, 1977), pp. 140-44, 221-27. 1 6 Ibid., p. 142. 52 traditional cognitive theory such as Solomon's in which the need to express emotions is tied solely to cognitive purposes. Recall that the moral sentiment of resentment involves not just a belief but also an evaluation and an affective component. On Solomon's theory, an emotion such as resentment is related to its expression through an entirely cognitive process - through the logic of a practical syllogism such as \"'Offenders ought to be punished; Socrates is offensive; Socrates can best be punished by writing a sarcastic play about him....'\"17 This cognitive theory over-intellectualizes the moral sentiment by de-emphasizing its affective and evaluative components.18 An account of emotions that, in my view, better captures the intuitions underlying both the hydraulic and traditional cognitive views is the analysis presented by Patricia Greenspan.19 Greenspan accepts the basic assumptions of cognitive theorists \u00E2\u0080\u0094 that the judgment on which an emotion is based is partly factual and partly evaluative -but proposes a broader evaluative view. Although she does not discuss resentment in detail, she indicates that indignation and resentment are \"moral or quasi-moral\"20 forms of anger. She says emotions such as anger are \"pressuring.\"21 Anger, for instance, has an action requirement to get back at the object of one's anger.22 If the action requirement is not fulfilled, the agent 1 7 Ibid., p. 227. 1 8 The attempt to link evaluations with affect of some kind is controversial. 1 9 Patricia Greenspan, Emotions and Reasons: An Inquiry into Emotional Justification (New York: Routledge, Chapman and Hal l Inc., 1988), pp. 3-14, 48-55, 159-67. 2 0 Ibid., p. 53. 2 1 Ibid. 2 2 It is useful here to recall Mackie's analysis, discussed in Chapter 1, in which he says that underlying the idea of moral wrongness in general and positive retributivism in particular is the notion that a wrong action calls for a hostile response. 53 experiences emotional discomfort. On Greenspan's view, the discomfort itself supplies a self-interested and pressing reason for the agent to act. She says, \"To the extent that I am uncomfortable about not yet acting on it [anger at X], the thought that I ought to get back at X is not something I can simply drop or ignore. ...The threat at continuing discomfort, in short, supplies a self-interested and pressing reason for action from the agent's standpoint....\"23 If Greenspan's analysis is correct - and I think it is, the resentment itself gives the victim a pressing reason for expressing it. If it is not expressed, he experiences psychological tension. Hence, from the victim's perspective, there is a non-intellectual, affective reason - called an \"extrajudgmental\" reason by Greenspan \u00E2\u0080\u0094 to express the emotion. This suggestion that the emotion motivates the agent to act - to express it in some way \u00E2\u0080\u0094 helps to explain why we view the retributive emotions with such suspicion. The \"extrajudgmental\" motivation to retaliate for a harm could be dangerous indeed if it were to override the victim's reflective evaluation motivating him to act without sufficient reflection or on his own rather than through the social institution of punishment. I want to emphasize, however, that the moral sentiments are considered emotional reactions which can be adjusted to suit the circumstances. As I pointed out in Chapter 1, they are susceptible to justification. The victim is a member of society who, presumably, has been socialized to respect moral and social norms. Even as he experiences intense resentment in the aftermath of the crime, he will have some sense of what is appropriate to the circumstances. Moreover, as a moral agent who is affected by 2 3 Greenspan, Emotions and Reasons, p. 159. 54 others' criticism or endorsement of his resentment, he can be influenced to adjust it to be appropriate to the circumstances. Giving the victim a distinct voice within the social institution of punishment would be one way of both guarding against him taking matters into his own hands and influencing him to moderate his resentment to an appropriate level. The victim, then, has a strong and justifiable emotional stake in what happens to the criminal. Given that this is so, one wonders why victims have not been given a more prominent role to play in the justice system. I have already indicated one reason - the mistaken suspicion that the victim's resentment is mere vengefulness. It should be clear by now that it need not be. Nonetheless, there is reason to be concerned about the personal, intense nature of resentment and the corresponding variability of victims' reactions. We do not want wrongdoers who commit similar crimes to be punished differently merely because their victims have vastly different, perhaps idiosyncratic, reactions to the harm done. Underlying this worry is our strong commitment to fairness. That commitment is incorporated in our justice system through the principle of proportionality. More serious crimes are punished more severely and, to some extent, similar crimes receive similar punishments. Is our commitment to fairness incompatible with taking the victim's resentment into account? I contend that it is not. While the justice system must preclude overly intense or otherwise idiosyncratic reactions, it should not set aside victims' reactions. It should control how they are expressed by giving the victim an institutionalized role. Yet in our justice system the victim does not have a 55 distinct, institutionalized role. If there is a trial, she may be called as a witness or she may not. That decision is not up to her; it is up to the prosecution. If the wrongdoer pleads guilty and there is no trial, the victim may have no role. At this point, it will be helpful to take a look at factors that affect the intensity of the victim's resentment. A better understanding of those factors will indicate how the community can judge whether the resentment is appropriate or not. The discussion will bring to light two good reasons for giving the victim a role in the social institution of punishment. The first is that the victim's resentment and the reasons for it provide important information to both the community and the wrongdoer about the crime. The second reason is that giving the victim a role in the social institution provides society with the opportunity to influence the victim to moderate his resentment to be appropriate. Consider first how the intensity of the resentment provides information about the crime. To begin with, it is important to appreciate that the removal of goodwill by wrongdoers is not an all or nothing thing. It should be viewed as on a continuum from less to more severe depending on the wrongdoer's motives. The wrongdoer may have committed the crime out of indifference, negligence, selfishness, contempt or outright malevolence. If the victim is cognizant of the wrongdoer's motives - and if she knows him she may be - the intensity of her resentment will reflect that awareness and, perhaps other aspects of their relationship.24 If, for 2 4 A substantial number of crimes, particularly violent crimes, are inflicted on persons known to the wrongdoers. Fattah says, \"a large number of...violent victimizations...occur between persons well known to each other, between individuals bound by interpersonal bonds, and mostly, between family, friends, or those bound by affective ties.\" Fattah, Understanding Criminal Victimization, p. 158. For example, a study of homicide in Canada from 1976 to 56 example, she is sure he beat her because he enjoys seeing her suffer, she knows he acted out of malevolence and her resentment will be extreme. Giving the victim an opportunity to explain why she feels the way she does may reveal important information about the wrongdoer's intent. It is worth noting here that the relationship between wrongful intent and harm done is not at all straightforward. Wrongful intent may be similar, for example, in an attempted robbery and in one that is completed. The harm done, however, is greater in the completed crime. And, in pur justice system, the punishment inflicted is more severe. I shall put off until Chapter 5 the complex question of how to relate the severity of punishment to intent and harm done. There I argue that determining the sentence should involve consideration of both wrongful intent and the gravity of harm done. I submit that, in contrast with traditional theories, the naturalistic account has the resources to explain why. In many cases, the victim will be unaware of the wrongdoer's motives and, as a result, will react to the harm done on the assumption that it is an accurate reflection of intent. Even in these cases the victim's resentment and her reasons for it can tell us something about the crime. The victim has information about the harm done that no one else can give us. To ignore her perspective, to leave it out of the assessment of the crime, is to omit information that may well be pertinent to assessing the gravity of the crime. 1985 found that 76.9 percent of all solved homicides involved victims and suspects who were known to each other. The types of relationships were almost equally divided between domestic (38.4 percent) and social or business (38.5 percent.) Ibid., p. 161. 57 Doubtless, her resentment will reflect not only her assessment of the harm done but also personal factors related to her natural disposition, character and the extent of her support system. Does the fact that resentment is personal and varies in intensity mean that it should be set aside by the justice system? Surely it does not. Solomon suggests that the idea that it should stems from the fact that our concept of justice has been overly influenced by Kant. He notes that \"[W]ith Kant, philosophical thinking about values became singularly obsessed with 'reason' and rational principles and abusive of the passions.\"25 This unbalanced view stems from the Kantian concept of moral agency which fails to acknowledge the significance of the natural human capacity for empathy. It fails to recognize that human beings are social creatures who are motivated by the personal reactions of others as well as social standards set by the community.26 Our justice system has incorporated these Kantian notions. It has ruled out the victim's emotions on the assumption that they are uncontrolled and irrational. The upshot is that the system has, to a great extent, insulated wrongdoers from the human impact of their crimes. Mark Chupp notes, Rarely have offenders had to face their victims and the human impact of their crimes. Offenders perceive their victims to be deserving of the crime and able quickly to recover losses. Victims have not only been overlooked by their offenders but many times by the entire criminal justice system.27 ^ 5 Robert C. Solomon, Chapter 5, \"Justice and the Moral Sentiments,\" A Passion for Justice: Emotions and the Origins of the Social Contract (Don Mills , Ontario: Addison-Wesley Publishing Company, Inc., 1990), p. 204. 26 The relevance of the victim's role is also tied to the question of whether the impact of the crime (harm done) should affect the degree of punishment. As noted above, I shall argue that the practice of taking harm done into account in sentencing is justified and that the naturalistic justification has the resources to explain why. Jeffrie Murphy also presents a provocative discussion suggesting that the victim's personal response to the wrongdoer has relevance for sentencing. Jeffrie G. Murphy, \"Getting Even: The Role of the Victim,\" Philosophy of Law, ed. Joel Feinberg and Hyman Gross (Belmont, Ca: Wadsworth Publishing Company, 1995), pp. 692-3 and footnote 29. 2 7 Mark Chupp, a proponent of face-to-face victim/offender mediation, was the director of the Victim-Offender Reconciliation Program in Elkhart, Indiana. Mark Chupp, Chapter 4, \"Reconciliation Procedures and Rationale,\" Mediation and Criminal Justice: Victims, Offenders 58 Chupp's observation highlights how the existing system may frustrate the aim of motivating the wrongdoer to take up self-evaluation and through it, perhaps eventually, to feel guilt about the crime. In failing to give the victim a distinct role within the social institution, the justice system may actually prevent the wrongdoer from gaining a full understanding of the harm he has done. For some wrongdoers communicating to them about the personal suffering of the victim may well be an essential step in achieving that purpose. A second reason to give the victim a role in the social institution is that doing so would provide the community with the opportunity to influence the victim to moderate his resentment. As I have already pointed out, the victim's resentment must be expressed. Given that victims have no distinct role in the existing system, it should be no surprise that some have turned to the media and to public protests to make their case. These ways of communicating retributive emotions are fueled by excess rather than moderation. The media thrive on sensationalism, and public protests glean more support when the victims involved play upon the fears of the public by emphasizing the horrors of crime. If victims were given a role within the system, there would be an alternative to the sensationalistic methods of communication that some victims have adopted. More importantly, however, the community would have the opportunity to examine critically the victim's resentment and his reasons for it in a formal proceeding having suitable procedural safeguards. I will have more to say on this topic in the section on the mutual influence of the community and the victim. and Community, ed. Martin Wright and Burt Galaway, (Newbury Park, Ca: SAGE Publications 59 We should not, then, set aside the victim's justified resentment keeping it outside our institutionalized response to the wrongdoer. The victim, through no fault of his own, is inextricably involved in the crime. He is, in the majority of cases, the one who suffered most. He has good reasons to want, and certainly deserves, vindication. He should be given the opportunity to express his resentment - to make his case, so to speak, and not just as a witness who is included only if it suits the prosecution's purposes. Giving the victim a distinct role would give him the opportunity to communicate his perspective on the crime to society and to his offender. It would acknowledge his rights as a stakeholder - his personal interest in achieving justice. Victimless Crimes The naturalistic justification calls attention to the importance of the victim's perspective on the crime. What, then, can be said about victimless crimes? It will be helpful to consider this question in the light of Feinberg's definition of a crime. According to Feinberg, crime is conduct against persons or property which produces serious harm or unreasonable risk of harm to others or conduct which causes harm to the public in general.28 So-called victimless crimes are those characterized in the definition as harm to the public in general. Typical crimes in this category are, in Feinberg's words, \"...counterfeiting, smuggling, income tax evasion, contempt of court, and violation of zoning and antipollution Ltd., 1989), p. 56. 2 8 Joel Feinberg, Harm to Others: The Moral Limits of the Criminal Law (New York: Oxford University Press, 1984), pp. 10-11. 60 ordinances.\"29 They are called victimless because there is no identifiable person who is the victim.30 Rather, the harm is to the public which is, of course, composed of private individuals who stand in complex social and legal relations to one another. In some cases of public harm such as the poisoning of a city's water supply or the undermining of a country's currency there are, in fact, many individual victims who suffer the harm directly. In other cases such as one person's income tax evasion the harm to any particular individual is unappreciable. In the latter types of cases, the harm results from the weakening of the public institutions which are essential to the functioning of society. The entire community, then, has a stake in punishing wrongdoers who commit such crimes. There is unlikely to be an individual victim's perspective on these crimes. Rather, the community has a perspective which takes into account the interests of all the individuals who comprise the society. In the next section I discuss the community's perspective on crimes. Needless to say, the community has its own set of aims for punishment both when there are identifiable victims and when the crime is victimless and it is the public interest alone that has been harmed. 2 9 Ibid., p. l l . 30 Some would argue that crimes such as possession of marijuana are 'victimless.' The attempt to categorize such crimes in this way is, in reality, an attempt to argue for decriminalization. Calling the crime 'victimless' is simply another way of saying that no harm is done and that, therefore, the conduct in question should not be a crime. Those who believe the conduct should be a crime, however, argue that such conduct is harmful or is a serious risk of harm to individuals or to the public in general. The fact that possession of marijuana, for example, is a crime indicates that there is consensus in society, at this time, that the conduct causes harm or serious risk of harm to individuals or the public. 61 The Perspective of the Community In this section I discuss the aims of punishment from the community's perspective. As was the case with the other stakeholders, it will be helpful to begin with consideration of the moral psychology of the members of the community. In Chapter 11 set out Strawson's distinction between resentment and moral indignation. We saw that, when members of the community react to a harm in which they are not personally involved, they view the situation with a certain disinterest. Their experience of moral indignation reflects sympathy for the victim but tends to be less intense than resentment. Like resentment, however, moral indignation rests on and reflects the moral demand for some degree of goodwill. Moral indignation is a complex emotion. It is similar to resentment in that it involves a belief, a related evaluation and an affective component. In addition however, it has a normative dimension. It is grounded in approval of the victim's resentment. It reflects sympathy for the victim from an impersonal standpoint. As such, it is not a mere feeling of sympathy arising out of sentimentality. Moral indignation involves a reflective evaluation of the victim's resentment which results in a judgment to endorse it. The cognitive content of the moral indignation is the belief that the wrongdoer did the crime. The related evaluation is that the wrongdoer is uncaring, to some degree, of the victim's value. The affective component, which is normally less intense than that of resentment, involves sympathy for the victim and a desire to get back at the wrongdoer on behalf of 62 the victim.31 The fact that moral indignation involves a judgment to endorse the victim's resentment entails that it may be modified or even withdrawn when the facts of the case are examined. Feinberg points out that at a criminal trial the absolution of the accuser often hangs as much in balance as the guilt of the accused.32 In the case of date rape, for example, the victim's motives and actions may be examined and questioned equally with those of the accused. The upshot is that, in our system in which the value of fairness is given a high priority, the community withholds its expression of moral indignation until the wrongdoer is pronounced guilty. Only then does the community express its moral indignation through punishment. Now, keeping in mind the moral psychology of the members of the community, consider the communicative aims of punishment from their perspective. In punishing, the community has the opportunity to express to all moral condemnation of the wrongdoer for her wrong action. The expression is intended to send a message. The community intends that the wrongdoer and public recognize that the punishment means condemnation, among other things. In a crime such as robbery, assault, rape or murder the wrongdoer fails to meet the 3 1 Similarly, Smith describes what I have called moral indignation - our sense of demerit regarding an action - as a compound of two sentiments. He says, \"...we cannot indeed enter into the resentment of the sufferer, unless our heart beforehand disapproves the motives of the agent, and renounces all fellow-feeling with them; so upon this account the sense of demerit...seems to be a compounded sentiment, and to be made up of two distinct emotions; a direct antipathy to the sentiments of the agent, and an indirect sympathy with the resentment of the sufferer.\" Smith, The Theory of the Moral Sentiments, II.i.5.4, p. 75. 3 2 Feinberg, \"The Expressive Function of Punishment,\" p. 105. 63 demand for goodwill to a lesser or greater degree. In extreme cases such as violent, brutal murder her actions express outright malevolence. But in all cases, a criminal act shows that the wrongdoer views the victim as having no or reduced moral value. If the community were to fail to express reprobation, it would be acquiescing in the wrongdoer's devaluation of the victim. Jean Hampton suggests that the community's response involves \"a kind of fear and defiance.\"33 On her view, the community fears that by not opposing the wrongdoer's challenge to its values, it invites further challenges. Therefore, it defies the challenge. I think Hampton's analysis is correct for our existing justice system. But if the system were changed to accommodate the victim's concerns by giving him a role - as I have suggested it should be, the community's response could reflect sympathy for him, the sympathy inherent in moral indignation. Punishment would be condemnation of the wrongdoer to be sure, but also a way of standing up for the person devalued by the crime.34 A communicative aim that is closely related to condemnation is called \"authoritative disavowal\"35 by Feinberg. He explains this purpose with the example of an airplane from one nation firing on that of another over international waters. Undoubtedly, the offending nation would be called upon by the other to punish the pilot who ordered the attack. Feinberg points out that 33 Murphy and Hampton, Forgiveness and Mercy, p. 59. 3 4 In With Justice for Some George P. Fletcher argues that the justice system is \"rent by structural defects\" which result in its failure to stand by the victims of crimes. He believes expressing solidarity with the victim should be an important purpose of punishment. He says, \" A primary function of punishment, then, is to express solidarity with the victim. It is a way of saying to the victim and his or her family: Y o u are not alone. We stand with you, against the criminal.'\" George P. Fletcher, With Justice for Some: Victims'Rights in Criminal Trials (Don Mills , Ontario: Addison-Wesley Publishing Company, 1995), p. 203. 3 5 Feinberg, \"The Expressive Function of Punishment,\" p. 101. 64 \"Punishing the pilot is an emphatic, dramatic, and well-understood way of condemning and thereby disavowing his act. It tells the world that...his government does not condone that sort of thing.\"36 Similarly, when the community punishes the wrongdoer, it disavows the criminal act. If it did not respond in this unmistakably clear way, it would be seen to be implicitly condoning the crime. It is through enforcement of its laws that the community confirms to all that particular types of actions are forbidden. Another of the community's aims is affirmation of its values. Communicating and upholding society's values are primary reasons for setting up social institutions. And it is through the social institution of punishment that the community affirms the value that underlies the prohibition stated in the law.37 In punishing a wrongdoer for assault, for instance, the community affirms the value of respect for the safety and security of all persons. Society cannot exist without social order. And maintaining the social order involves enforcing existing laws in order to deter the wrongdoer and others from criminal activity. A community that fails to enforce its laws and the associated values loses its credibility and, eventually, its ability to govern. As noted, the affirmation of the community's values is related to the aim of deterrence. I have discussed it as a distinct aim, however, in order to emphasize the importance the community attaches to its values. 3 6 Ibid., p. 102. 3 7 Jean Hampton characterizes moral indignation as an emotional protest against immoral treatment which is aimed at defending the value which the action violated. Murphy and Hampton, Forgiveness and Mercy, p. 59. 65 What can be said about deterrence? Proponents of consequentialist justifications invariably focus on it as a primary reason for punishment as do many professionals in the criminal justice system and a large segment of the public. Retributivists disagree claiming that the wrongdoer should be punished for his crime irrespective of any social benefit. The debate between them, fueled by inconclusive empirical evidence as to whether punishment deters, has become the locus of much philosophical wrangling. The naturalistic justification, which embraces a plurality of aims, has the resources to resolve this tension. The view is that punishment communicates not only the condemnation that flows from retributive sentiments but also society's message of deterrence. Traditional theories characterize these two messages as backward and forward-looking respectively and usually take them to be at odds. When punishment is viewed as communication, however, it becomes clear that these messages are consistent and complementary. Primoratz, in discussing his expressive account, makes this point saying, Punishment...aims at preventing crime. This is effected by its deterrent impact and by way of disabling the criminal from breaking the law again. But the expressive and deterrent functions of punishment are in no way inconsistent: the former reinforces the latter, disgrace deters.38 Nonetheless, it seems fairly obvious that some penal measures are better suited to achieving deterrence and others to communicating retributive sentiments. In Primoratz, \"Punishment as Language,\" p. 605. 66 other words, these aims may be consistent and complementary but different types of sentences will emphasize one aim more than the other. At this point it is worth reiterating that I have used the term \"punishment\" broadly to encompass a spectrum of penal measures \u00E2\u0080\u0094 incarceration, mandatory supervision, electronic monitoring, diversion to community service etc. Clearly, the justice system has a variety of ways of dealing with offenders. If punishment's aim is taken to be primarily prevention, for example, the implication seems to be that incarceration - the longer and more severe, the better - is the most appropriate response to the criminal. If it is taken to be primarily communication of retributive sentiments, other measures in addition to incarceration may be required. In the latter case, the wrongdoer is one of the target audiences and one of the aims of the communication is to bring about a change in his moral psychology by provoking self-evaluation. In this regard, programs that combine incarceration with educative measures may be most appropriate. In Chapter 6 I discuss innovations which are designed both to educate and reintegrate the wrongdoer and to provide reparation to the victim. Through punishment, then, the community communicates condemnation and deterrence to the wrongdoer. It communicates condemnation, deterrence, authoritative disavowal and affirmation of the community's values to the public. But what of the victim? I suggested in the previous section that the victim is a stakeholder in the social institution and, as such, should be given an . institutionalized role. I also pointed out earlier on that, in some cases, the wrongdoer who takes up self-evaluation may want to address the victim as 67 audience by apologizing or making reparation. The community, however, should also recognize the victim as one of its audiences. Punishment certainly has the potential to be an acknowledgment of the suffering of the victim, his family and friends. Traditionally however, the victim has no sanctioned opportunity to influence the members of the community.39 If he were given a role, he would have that opportunity. There would be a forum within the social institution for communication between victim and community. And the community, in giving him a voice, would open itself to being influenced. It could then include standing up for the victim among its aims. To accomplish that aim, the community would undoubtedly need to adopt innovative penal measures designed to make clear its endorsement of the victim's resentment and acknowledgment of his suffering. In the next section I reflect on the prospects for mutual influence between victim and community. The Mutual Influence of Community and Victim Recall that the psychological reactions of the members of the community to a crime are closely related to, but different from, that of the victim. Strawson suggests that moral indignation and resentment are \"connected humanly.\"40 But what does this mean? The idea is that the experience of moral indignation is 3 9 Reparation is one way to acknowledge the victim's suffering and has been, to some extent, incorporated in the justice system. The use of Victim Impact Statements is another example -albeit a controversial one - of how the victim may be given an expanded role. 4 0 Strawson, \"Freedom and Resentment,\" p. 71. 68 sympathetic but tends to be less intense because the vicarious victim is not personally involved. As moral agents, both victim and vicarious victim have the capacity for empathy. Both inevitably experience an emotional engagement which entails that they can sympathize with and be influenced by what one another feels. This mutual influence may affect the intensity of the retributive emotions. On the one hand, the victim who has a tendency to over-react may be influenced positively by the impersonal, less intense reactions of the members of the community. For example, if a victim of a minor robbery felt so devalued and disgusted by the crime that he insisted the wrongdoer should go to prison for ten years, the victim's friends would likely try to influence him to adjust his resentment to appropriate levels. If he were asked to testify at the trial and expressed his inappropriate reaction there, the jury would be liable to reject it as unreasonable. Furthermore, they might well be inclined to view the rest of his testimony with skepticism. On the other hand, the members of the community may be affected negatively by sympathy with the victim that is too strong. For example, if the crime and subsequent attempt to apprehend the wrongdoer is sensationalized in the media as was the O. J. Simpson case, it may be difficult to find citizens who remain impartial and can serve on the jury. Members of the community, then, must avoid actual emotional identification with the victim. 69 How might the community influence the victim? Moral indignation is grounded in approval of the victim's resentful reaction. Resentment, however, involves not only the victim but also the wrongdoer who is its target. It is, then, an open possibility that the community could come to sympathize with and support the wrongdoer rather than the victim. In The Theory of the Moral Sentiments Adam Smith relates this possibility to the propriety of the victim's resentment. He says, \"As they are both men, we are concerned for both, and our fear for what the one may suffer, damps our resentment for what the other has suffered.\"41 If the victim's resentment is too intense the community may sway towards sympathy for the wrongdoer. As Smith explains, \"...violent resentment instead of carrying us along with it, becomes itself the object of resentment and indignation. We enter into the opposite resentment of the person who is the object of this unjust emotion, and who is in danger of suffering from it.\"42 In other words, if the victim's resentment is judged too intense, the community may sympathize with the offender. In considering what it takes for the victim to enlist the sympathy of others, Smith reinforces the important point that the community considers resentment to be justified when it is a defence of one's self-respect and reputation. Moreover, the propriety of the resenter's reaction is a factor in securing the emotional support of others.43 The upshot is that the victim, in order to gain the support of 4 1 Smith, The Theory of the Moral Sentiments, I.ii.3.1, p. 34. 4 2 Ibid., II.i.5.8, p. 77. 4 3 Ibid., I.ii.3.8, p. 38. 70 others, must adjust the intensity of his expression to conform to social norms. Practically speaking, if the victim were given a distinct role in the justice system, he would have an opportunity to communicate his resentment. The community and even the offender could then judge whether it was appropriate. Moreover, the fact that the victim's perspective and reactions were included and publicly open to examination and criticism would probably have a moderating influence on his resentment.44 What can be said about the victim's effect on the members of the community? It is possible for the community's sympathy for the victim to be too strong. The vicarious victim does not normally feel resentment because he is not directly involved in the crime. But in the case of shocking and frightful acts, some members of the community may feel directly affected by the devaluing of the victim and experience emotional identification with him. The victim's family and close friends are obvious examples. But others may also feel a close kinship to the victim.45 There is a danger that, if the victims are not given an institutionalized role, they may express their resentment in ways that undermine justice 4 6 In our society victims increasingly have expressed their resentment in 4 4 In victim/offender mediation the intensity of the victim's resentment often decreases after a face-to-face meeting. Mark Chupp notes, \"The power of the personal face-to-face encounter breaks down many stereotypes and previous hostilities. Many victims will..offer some type of acceptance of...[an]apology or extend forgiveness to the offender, despite their earlier intentions.\" (my emphasis) Chupp, Mediation and Criminal Justice, p. 62. 4 ^ Jean Hampton makes a similar point saying, \"Resentment of crimes against others is possible when, but only when, one connects oneself in significant ways to these others.\" Murphy and Hampton, Forgiveness and Mercy, p. 56. 4 6 It is worth noting that Murphy, who does not distinguish revenge from the moral sentiment of resentment, says that there are at least two good consequentialist reasons for institutionalizing the public expression of revenge through punishment. The first is the satisfaction it gives victims. The second reason is the tendency it has to defuse the possibility of private revenge and the resulting social turmoil. He also claims that a purely 71 ways that could influence the public to over-react to crime and, as a result, demand unduly harsh punishment for criminals. For example, some proponents of victims' rights, who claim victims are marginalized by the justice system, use the media to sensationalize violent crimes. By evoking fear and hatred, they aim to persuade the public that harsher punishment of criminals is desirable and necessary. The mutual influence of community and victim, then, may have either a positive or negative effect on the expression of the retributive emotions. On the positive side, the intensity of the victim's resentment may be moderated to become appropriate to circumstances. Victims who have been given a voice, for example, in pilot projects on victim/offender mediation feel empowered by the process and, for the most part, satisfied with the degree of punishment inflicted.47 On the negative side, the community may be inordinately influenced by protesting victims, lose its impersonal perspective and over-react by demanding that laws be changed to increase punishments. As noted previously, our justice system incorporates a strong commitment to fairness. To ensure that the members of the community maintain their impersonal perspective on the crime, that commitment is institutionalized in substantive laws and procedural rules. The laws define the types of conduct that consequentialist case against the institutionalization of revenge \"is going to be fragile.\" Murphy, \"The Role of the Victim,\" pp. 690-1 and footnote 20. 4 7 Coates and Gehm, authors of a major empirical study of victim/offender mediation, report that \"Seventy per cent believed that the offender had been punished adequately. Twenty-four percent indicated the punishment was too little and 5 per cent felt that it was too much.\" Robert B. Coates and John Gehm, Chapter 17, \"An Empirical Assessment,\" Mediation and Criminal Justice, p. 255. 72 are prohibited and limit the sanctions to be applied on violation. The procedural rules set out the framework for a fair process of trial, verdict and sentencing. These laws and rules substantially decrease the possibility of unjust punishment. In the existing system, however, the institutionalization of fairness has been focused almost exclusively on protecting the rights of the accused for two reasons. First, the victim's concerns have been mistakenly interpreted as vengefulness. Secondly, there is a legitimate recognition that the state's prodigious coercive power over the individual - specifically, the accused - must be contained. An imbalance in the social institution has been the result, however. The victim's concerns have, for the most part, been ignored. True fairness requires that they be taken into account. In other words, the social institution should be designed to include all three perspectives - that of the wrongdoer, the community and the victim. In the next section I touch on the theoretical implications of this view. Pluralism about Aims In the preceding sections we saw that each of the stakeholders has a different perspective on the crime and, consequently, differing aims and different audiences for the communication. In short then, the naturalistic account entails pluralism about aims. This approach rejects the idea that punishment's justification can be reduced to a single aim such as, for example, the reform of the wrongdoer and her subsequent reconciliation with the community. On the contrary, this 73 approach is grounded in the notion that, taken together, the aims of all three stakeholders justify the social institution. In particular cases of punishment, of course, the sentence will likely emphasize one of the aims more than the others. Suppose, for example, that the wrongdoer is a repeat offender with a long list of convictions including a number of violent offenses. Clearly, this type of case poses a problem for a theory which insists that attempting to reform the criminal is the essential aim of punishment. It is highly improbable that this hardened criminal will take up self-evaluation and reform. On the naturalistic view, motivating the wrongdoer to take up self-evaluation is one aim, but it is not the only one. This account leaves it open that the sentence could be tailored to accentuate aims such as approval of the victims' resentment, disavowal of the forbidden action and deterrence rather than reform of the criminal. The upshot is that this justification is pluralist. No one aim can be considered the essential one that justifies the practice. Rather, punishment has a plurality of aims unified by the idea that the punishment is a communication of the moral demand for some degree of goodwill. The communication sends different messages from various stakeholders to a variety of audiences. In the next chapter I compare my naturalistic approach with Duffs rationalistic expressivist account. That discussion highlights the explanatory advantages of pluralism about aims. 74 Part II. Clarifications Chapter 3 A Comparison with R. A. Duffs Expressivist View In Part 1 1 explained why we should develop a naturalistic theory of punishment. I also set out the main tenets of a theory explicating punishment as a type of communication from various players to a variety of audiences. It became clear that punishment has a plurality of aims. In Part II, which comprises this and the following two chapters, I clarify and elaborate on the theory. In this chapter I compare my naturalistic expressivist view to R. A. Duffs rationalistic expressivist theory.1 Whereas the naturalistic justification is grounded in the natural moral demand for goodwill, Duffs theory has its seat in the Kantian demand to respect the rationality and autonomy of persons. I argue that Duffs attempt to defend expressivism within a Kantian framework falls prey to the impossibility of satisfying both the Kantian demand with respect to the criminal and the need to address society's concerns. In other words, Duffs Kantian commitments give rise to an essentialist aim of repentance and reform of the criminal, which conflicts with the actual aims of the various stakeholders in the social institution. Recognition of the conflict in fact leads Duff to skepticism about the attainability of his ideal system of punishment within \"any plausibly practicable legal system.\"2 By contrast, the naturalistic version of expressivism 1 Duff, Trials and Punishments. 2 Ibid., p. 11. 75 is defended within the \"general structure or web of human attitudes and feelings. In accommodating pluralism about aims, this framework has the resources to take into account the perspectives and aims of all the players involved in the social institution. Duffs Theory of Punishment: Expression, Penance and Reform In this section I review Duffs expressivist account. I then criticize his view. In the last section I show how the naturalistic theory can deal with the difficulties encountered by Duffs approach. Duff attempts to justify punishment while adhering to the categorical Kantian demand to respect the criminal as a rational and autonomous moral agent.4 This approach has an essential aim, that of attempting to bring the criminal to repent and reform. It is also expressivist. Duff explains his expressivist view in the following way: Punishment has an essentially expressive meaning; it aims to communicate something to the criminal. The nature and purpose of this communicative endeavour is then explained by portraying punishment as a kind of penance which aims to secure and to express the criminal's repentance of his wrong-doing; and thus as a process of reform (or self-reform)....But to talk thus of punishment as a reformative endeavour is not to suggest that it is a contingently efficient means towards a further and independently identifiable end; for the kind of reform at which punishment aims can be achieved only Strawson, \"Freedom and Resentment,\" p. 78. 4 Duffs commitment to the categorical Kantian demand is in no way half-hearted. On his view, any justification of punishment which includes consequentialist aims, such as the aim of deterring the criminal to protect society, is inherently unjust. He says, \"We face here a dilemma which must confront anyone who seeks to apply absolutist and non-consequentialist notions of value and justice to the realm of politics. The categorical demand that we should treat and respect each other as autonomous moral agents forbids us to impose on criminals the kinds of punishment to which they are subjected within our existing legal systems....,\" Duff, Trials and Punishments, p. 296. 76 through the kind of suffering which punishment aims to impose on, or to induce in, the criminal.5 Duffs chief concern is to show that one can maintain respect for the rationality and autonomy of the criminal even while inflicting the suffering and deprivation of punishment. He applies two strategies to this task. First, the criminal is taken to be an active participant in the trial and punishment. And second, punishing is understood as a formal process that is analogous to the informal activity of morally criticizing and blaming a wrongdoer.6 According to Duff, blaming \"is a kind of moral argument with another person.\"7 He illustrates this view with the example of one friend who criticizes and publicly blames another, Jasper, for using his position at work to make sexual advances towards his secretaries. The process of blaming involves engaging Jasper in a moral discussion about his conduct. The friend presents the reasons for his criticism and challenges Jasper to respond to the moral charge. The aim of blaming is to persuade Jasper through rational argument that he acted wrongly in harassing the secretaries and that he should modify his conduct in the future. In general, the \"proper aim\" of moral blame is to bring the other person to modify his future conduct \"only by means of such rational moral persuasion.\"8 But blame also aims to make the wrongdoer suffer. According to Duff, one retains respect for the wrongdoer's autonomy and uses blame for its proper purpose only when one aims \"to induce in him [the wrongdoer] the pain of recognised and 5 Ibid. 6 Ibid., pp. 7-8. 7 Ibid., p. 47. 8 Ibid., p. 50. 77 accepted guilt.\"9 Duff claims that the condemnation of blame may at first cause the wrongdoer to suffer pain which is related only contingently to the wrongdoing but that \"it may be transformed into the kind of pain which is the proper end of moral blame\"10 if it assists the wrongdoer to accept the justice of others' judgments and \"thus to suffer the pain of guilt and remorse.\"11 For Duff, the formal practice of punishment is analogous to informal blaming. Punishment expresses to the wrongdoer the community's condemnation by denouncing and formally disapproving of the criminal act. Duff concedes that punishment may also communicate the wrongness of the act to the public at large and may serve as an authoritative disavowal of the act to the victim but \"its essential expressive aim must be that of communicating to the criminal himself a proper condemnation of his act.\"12 (my emphasis) Like blame, punishment is a kind of rational argument with the wrongdoer. The argument is constituted in the process of trial and punishment in which \"punishment is part of that continuing dialogue with the criminal through which the law aims to guide his conduct by appealing to relevant reasons.\"13 Given this characterization of punishment as part of a rational argument, the question arises as to why Duff insists that the expressive purpose can be achieved only through hard treatment. He admits that it appears to be 9 Ibid., p. 59. 1 0 Ibid. 1 1 Ibid. 1 2 Ibid., p. 236. 1 3 Ibid., p. 238. 78 consistent with his view to imagine a process in which the defendant is convicted and publicly condemned merely verbally or by symbolic punishment. Hard treatment, of course, might add a prudential deterrent but this possibility can play no role in Duffs non-consequentialist justification. Yet Duff maintains that the wrongdoer must suffer hard treatment. Why? As with informal blaming, the answer is to be found in the nature of the relationship between the pain and suffering involved in hard treatment and the kind of pain which expressive punishment is intended to induce in the criminal. In order to explain that relationship, Duff looks to an analogy between punishment and the formal system of penances found in some religious communities. In the context of religion, a penance is an onerous task or painful deprivation undertaken or accepted by the sinner because she understands and admits that she has done wrong. It is a self-imposed suffering which expresses and assists repentance. Its aim is to reconcile the penitent sinner with herself and others. Even if we concede that voluntary penance expresses and assists repentance, however, the question arises as to whether coerced penance can accomplish the same purpose while also maintaining respect for the sinner's autonomy. Duff says coercive penance can achieve its proper purpose but only if the unrepentant or unwilling sinner \"comes, through her subjection to it, to accept it for herself \u00E2\u0080\u0094 to impose it on herself.\"14 Furthermore, it can respect her autonomy only if it addresses her as a rational moral agent. Here, Duff argues that respect for the autonomy of others does not forbid the use of coercive measures per se but it does Ibid., p. 251. 79 forbid the use of manipulative coercive measures - measures which fail to appeal to one's understanding in order to gain one's rational assent.\"15 By analogy, punishment can be seen as a \"compulsory penance\"16 aimed a bringing the criminal to repent her crime and reform. The compulsory penance is justified because it expresses the community's concern for the criminal who, like the sinner in a religious community, is seen to have injured herself as well as others by her crime. This account is an admittedly modest and secularized version of the Platonic claim that the wrongdoer has separated herself from the Good. Duff says, A criminal who flouts the just laws of her community thereby injures herself: she separates herself from the values on which the community and her own well-being depend....If she would only recognise the moral truth about her criminal attitudes and activities, she would see how injurious they are to her true well-being.17 In short, then, Duff maintains that the imposition of punishment on an unrepentant criminal is justified because it is a compulsory penance. The punishment, which \"addresses the criminal as a rational moral agent,\"18 aims at bringing her to understand the nature and implications of her crime. It aims at bringing her to repent the crime. It aims at bringing her to will her own punishment as a penance which expiates the crime and reconciles her with the Ibid., p. 254. Ibid., p. 255. Ibid., pp. 256-7. Ibid., pp. 262-3 80 Good and with the community. Punishment, on this view, is a communicative, retributive and reformative endeavour. Duffs Excessive Rationalism In this section I set out two criticisms of Duffs theory which revolve around his commitment to the Kantian demand. I begin by questioning his explanation of and assurances about respecting the wrongdoer's autonomy and rationality. I then go on to discuss the excessive rationalism of his account. I contend that it gives rise to a narrow scope that focuses exclusively on the criminal and a related essentialism about the aim of punishment. The key feature of Duffs account is the claim that punishment is both a kind of rational argument with the criminal and a coercive penance aimed at bringing about her repentance. Given Duffs assumption that punishment must satisfy the Kantian demand to respect the criminal's autonomy and rationality, this claim is enigmatic. It immediately raises two questions. First, how can a coercive penance be consistent with the Kantian demand to respect autonomy? And second, in what sense can the suffering of the coercive penance be said to be a kind of rational argument? Duff fails to answer these questions satisfactorily. Consider the first question. It is feasible to think of the criminal trial, as Duff does, as an argument with the defendant in which the evidence is examined and discussed in order to decide about guilt or innocence.19 Furthermore, if the 1 9 Brenda Baker criticizes Duffs notion of the trial as a kind of moral argument between two equally autonomous moral agents, the state and the defendant. She suggests that society and the defendant are not on an equal footing. She says the state \"authoritatively determines 81 defendant is found guilty, it is plausible to consider a verbal or symbolic condemnation to be the 'conclusion' of the rational argument. Moreover, this process satisfies Duffs condition that the trial and conviction must maintain respect for the rationality and autonomy of the criminal. Duffs argument becomes problematic, however, when he makes the further claim that we are justified and, in fact, required to impose suffering on the criminal. The suffering of the imposed penance is justified, according to Duff, because the criminal \"injures herself'20 by her crime and because the community aims, in inflicting suffering, to bring her to repentance and eventual reform. The claim that the community must impose suffering on the criminal for her own good is paternalistic. And Duff is certainly aware that it appears to be inconsistent with the Kantian demand. It is illuminating here to look at what it takes, on Duffs view, to satisfy the Kantian demand. He says that respecting another's rationality and autonomy entails responding to him \"as one who is able, and should be allowed, to conduct his own life and determine his own conduct in the light of his own understanding of the values and goals which command his allegiance.\"21 He also says that one may properly bring another to modify his which of the reasons that weigh morally with individual citizens wil l have legal standing.\" Duffs analogy is not perfect but I think it captures the core of what goes on at the trial. It is a presentation of arguments by both sides at which the defendant may, indeed, challenge the authority of statutes related to his case. Baker's criticism overlooks this possibility. It is just such challenges with which the Supreme Court regularly deals. The Court may, in fact, overturn statutes based on legal and moral arguments presented to it by the defendant. Brenda M . Baker, \"Penance as a Model for Punishment,\" Social Theory and Practice, Vol. 18, No. 3 (Fall, 1992), p. 327. 2 0 Duff, Trials and Punishments, p. 256. 2 1 Ibid., p. 6. 82 conduct \"only by bringing him to understand and accept the relevant reasons which justify the attempt.\"22 (my emphasis) In light of this explanation of the Kantian demand, it is surprising that Duff maintains that we are permitted to inflict punishment on the criminal for her own good, even though she may not be \"consciously unhappy\" about her crime and may, in doing the crime, \"attain the ends she actually pursues, and lead the kind of life she enjoys.\"23 Here Duff is attracted to the Hegelian theme that punishment reconciles the criminal with herself whether she realizes it or not. He concedes, however, that this interpretation has the unwanted metaphysical implication of objectifying the criminal's so-called rational will. The solution, says Duff, is to follow McTaggart's reading of Hegel in which punishment, like a coerced penance imposed on an unrepentant sinner, is seen to bring the criminal to understand the nature and implications of the crime.24 This solution, however, evades the problem rather than addressing it. It is unclear how coercion can be said to respect the criminal's rationality or autonomy without the metaphysical interpretation that the punishment is consistent with the criminal's 'real' or 'rational' will. In many, perhaps most, cases the criminal will not, in fact, agree to and accept the suffering imposed. As Duff concedes, the criminal may be leading the life she wants to lead. As I noted previously, the empirical evidence suggests that she Ibid., p. 256. Ibid., p. 259. 83 undergoes a process of desensitization and redefinition before doing the crime. Consequently, she does not feel guilty about her actions. It is possible that she may come to feel guilty when hearing the evidence at the trial. But if she does, then the imposed suffering of punishment is no longer necessary to induce guilt. And if she does not, it is highly unlikely that she agrees to and accepts the suffering that the community imposes on her. Duff seems to be saying that the criminal, through actually experiencing the suffering, will come to accept it as her due. But this interpretation of coercion is problematic. If the criminal does not agree to and accept the suffering at the time it is imposed, it is simply false to say that the imposed coercion is not manipulative. The suffering is imposed against the criminal's will. She has not been convinced by the rational argument at the trial. Therefore, at the time that the suffering is imposed, it is done without her consent. She does not choose to undergo suffering. In short then, the imposition of the so-called 'coercive penance' does not respect her autonomy. Duff could reply that the coercion is not manipulative because the criminal qua rational being can understand why she punished. He, in fact, says that his ideal of punishment aims to induce self-reform. He says punishment \"can, and should, be an attempt to arouse and engage - but not to coerce - the criminal's understanding and his will.\"25 Duff also warns that reformative punishment would become improperly coercive if it were to attempt to \"beat down\" the criminal's will.26 This reply fails, however, to address the second, pressing question which I raised at the outset. 2 5 Ibid. 2 6 Ibid., p. 272. 84 Let us now turn to that second question. If the punishment is supposed to address the criminal qua rational being, why is it necessary to impose suffering on the criminal? Duff says that the suffering is part of a continuing dialogue by which the state aims to guide the criminal's conduct through appeal to relevant reasons. It is clear, however, that the suffering of punishment, whatever it is, is not an appeal to relevant reasons. What, then, can be said about Duffs claim that the imposed suffering is a kind of rational argument that somehow appeals to the criminal's rational understanding? This claim is indicative of Duffs excessive rationalism about which I will have more to say below. Evidently Duff over-intellectualizes the notion of punishment. It is obvious, however, that punishment is not a rational argument. It is psychological, and to some extent, physical suffering. Duff says the suffering expresses condemnation and on that point he and the proponent of the naturalistic theory agree. Nonetheless, the disagreement between the two views is striking. Duff insists that the condemnation does not and should not involve the need for the victim and community to express their retributive sentiments. Nor should the condemnation be imposed with the aim of reducing crime.27 Its only purpose is to bring the criminal to understand and accept the need for reform. It is clear, however, that condemnation can be expressed to the criminal qua rational being in ways that do not involve the coercion or suffering of punishment. Ibid., pp. 23.9-41. 85 The naturalistic theory, in contrast with Duff's account, offers a perceptive and credible explanation of why we impose that suffering. It is, among other things, an expression of retributive sentiments - strong emotions which communicate to the criminal and the public how the victim and community feel about the crime. It is worth noting here briefly how Duff views emotions. In his discussion of blaming, Duff allows that the rational argument with the wrongdoer may appeal to relevant emotional reasons. He insists, however, that those emotional reasons must aim at communicating with the wrongdoer's moral understanding. On Duff's view, any form of communication which does not address the agent only via his moral understanding is manipulative. In other words, Duffs underlying view of the moral agent is one that focuses exclusively on cognitive capacities. Given that this is the case, there is no good reason for Duff to insist on communicating with the criminal qua rational being through physical and psychological suffering rather than through straightforward rational argument. In short then, Duff fails to convince us that his interpretation of punishment as both a coercive penance and a kind of rational argument can adhere to his own canon that punishment must satisfy the Kantian demand. Now let us consider Duffs excessive rationalism. It is the underlying theme of his account and as such has two significant implications. It results in an extremely narrow approach that focuses almost entirely on the criminal with little attention to the victim and community. And as I point out in the next section, it concludes 86 in a skepticism that raises doubts about Duffs uncritical presupposition of the Kantian demand. The narrow scope of Duffs approach is evident in both its retributive element and its essentialist aim. The retributive element consists in impelling the criminal to understand and accept the reasons why her past conduct was wrong. This construal of retributivism leads directly to punishment's essential aim of inducing the criminal to repent and reform. The idea is that punishment \"enables, or forces, the criminal to expiate or atone for her crime.\"28 This narrowness is a product of the account's Kantian framework. The Kantian demand concentrates on the rights of individuals. On Duffs interpretation, it is cashed out as the need to bring the criminal to repent and reform while, at the same time, maintaining respect for the criminal's rationality and autonomy. To be sure, Duff could say that the Kantian demand applies to all members of the community. But then, and this is vital to his theory, he owes us an explanation of how rational and autonomous individuals are to work out their social relationships. The only clue Duff gives us about the nature of social relationships reflects his excessive rationalism. He maintains that blaming and punishing are a kind of rational argument with the wrongdoer. The community must persuade the wrongdoer through rational arguments about moral reasons29 that he is guilty and should undertake penance for his own good. Clearly, this view of human 2 8 Ibid., p. 204. 29 Duff thinks that a community that gives citizens non-moral reasons for compliance with the law, such as reasons aimed to deter, is manipulating its citizens and, therefore, not respecting their rationality and autonomy. Ibid., pp. 181-6. Baker also criticizes Duff on this point. Baker, \"Penance as a Model for Punishment,\" pp. 323-4. 87 relationships is extremely limited. It requires a kind of concern for the criminal -concern that may be expressed only by addressing his cognitive capacities - but is silent about the needs and concerns of those affected by the crime. Moreover, it completely ignores the moral psychologies of all those involved - criminal, victim and community. The naturalistic approach, by contrast, centres on communication of the moral demand for goodwill. The moral demand is grounded in the attitudes and feelings that constitute interpersonal relationships. At this point, in order to illustrate the crucial difference between the moral demand for goodwill and Duffs interpretation of the Kantian demand, it will be helpful to consider an example about our informal practice of blaming. Suppose Jones, a businesswoman who is up for promotion, is invited to give an important presentation at a conference in Toronto. Those who will decide about her promotion will be in attendance observing how she handles herself in stressful situations. It is the first time Jones will be away from her three-year-old daughter and five-year-old son for an extended time. Before leaving for the conference she promises to call her family daily. Once in Toronto, however, she becomes totally involved in her presentation and the other events of the conference. It crosses her mind that she ought to call home but the presentations, the lunches, the business meetings and other social events take priority. Jones does not call her family. When she returns home, her family confronts her and blames her for not calling. The children initially refuse to give her a hug. Her husband angrily expresses his resentment by pointing out how he and the children feel hurt and, in some sense, 88 abandoned. He explains that he and the children looked forward to the calls. They wanted to tell her about their day and hear about hers. They wanted to say that they missed her and hoped she would hurry home. Jones attempts initially to defend her actions by saying that the promotion was important to her. At the time, it made sense to her to focus completely on achieving her goal without being distracted by the feelings and problems that calling home would bring to the fore. Nonetheless, her husband's resentment and her children's hurt feelings quickly elicit in her the guilt she should feel at failing to keep her promise. Jones apologizes to her husband and children, promises never to let it happen again and makes amends by taking a few days off work to spend with the family. In this example, Jones disregards the moral demand for goodwill when she breaks her promise. Her husband and children justifiably feel hurt and angry when she does not call. It matters to them that she does not call. I want to suggest, however, that construing their expression of blame as a rational argument \u00E2\u0080\u0094 as Duff claims we should - is artificial and limiting. The discussion is not a rational argument stating premises and leading to a conclusion. It is an expression of emotions backed up by reasons. That expression is not manipulative if the emotions expressed are genuine and justified. It is worth pointing out, however, that the expression of emotions could become manipulative if, for example, it were feigned merely to produce a reaction from Jones or if it were too strong. Nonetheless, genuine and justified emotions do arise both because of Jones' failure to call and, significantly, because of the existing interpersonal relationships. Those relationships entail the expectation of mutual goodwill and the experience of related reactions when the expectation is either fulfilled or 89 breached. It is when the family expresses resentment and hurt feelings that Jones realizes that it matters to her that they are hurt and resentful. She understands that it is wrong to make promises and not keep them. She understood that fact even as she failed to keep her promise. It is her family's emotional response, however, that motivates her to feel the guilt that she should feel as a result of her broken promise. The family's response matters to her because she empathizes with them. She cares about them and about what they think of her. The example illustrates the fact that the moral demand is not simply about individuals but, significantly, about their interpersonal relationships. Moreover, the moral demand for goodwill incorporates the idea that all parties in an interaction have not only a rational understanding of what is expected but also an emotional engagement that accepts the expectation. It is in virtue of the emotional engagement that one reacts to whether an expectation is met or breached. It is the emotional aspect of our relationships that is not addressed in Duffs overly rationalistic accounts of blaming and punishing. In society, of course, many of our relationships are not close personal ones as in the example but impersonal and distant. Nonetheless, even those impersonal relationships require a minimum degree of goodwill which presupposes at least a minimum degree of emotional involvement. It is stating the obvious to say that punishment is a social institution. It is society's way of ensuring that the minimum requirement for goodwill is met. That minimum requirement is that one must not harm others or their crucial interests. Criminal punishment is the 90 community's appropriate response when that minimum requirement is not met. In other words, criminal punishment is about the wrongdoer's failure to live up to the requirements of her relationship with the other members of society. Any theory which purports to justify punishment should explain the roles of all those involved in the damaged relationship. Duffs theory does not. According to Duff, punishment's essential aim is repentance and reform of the criminal. The reason for this is Duffs narrow interpretation of the Kantian demand. The theory, then, must be judged inadequate because it fails to account for the needs and concerns of the all parties involved, not just the criminal but also the victim and community. Duffs Skepticism In this section I challenge Duffs skepticism regarding the plausibility of justifying punishment in any actual human society. I argue that the specter of skepticism indicates that the Kantian demand is problematic, not - as Duff alleges - that human society is hopelessly flawed. Significantly, it is only after setting out his justificatory theory, which is an ideal one, that Duff discusses its relevance to the conditions that actually exist in society. Predictably perhaps, Duffs excessive rationalism leads to an idealism that results in skepticism about the feasibility of justifying punishment in any actual society. 91 To begin with, it is instructive to look at Duffs argumentative strategy. He does not offer a general explanation or justification of the Kantian demand. Rather, he says his account is intended to \"exhibit its [the Kantian demand's] meaning and show its importance by examining its implications for our understanding of criminal punishment.\"30 The justificatory account produced is an ideal one. Moreover, the so-called ideal deems that, as long as we continue to respond to one another as we actually do, society is in principle unjust. Moreover, punishment as well as other social institutions cannot be justified. If this is the implication of the argument, however, should we not look with suspicion at the unexplained and unjustified major premise - the Kantian demand? What, according to Duff, is the status of his ideal account? On his view, punishment cannot be justified in society as it actually exists because \"the social relationships and shared concerns which constitute a community do not exist.\"31 In particular, he says that \"we do not, and this includes both law-abiding and criminal citizens, have the kind of concern for each other which the idea of a community requires....\"32 It should be clear by now that, for Duff, the \"kind of concern for each other\" that a genuine community requires centres on the need to satisfy the categorical Kantian demand to treat other people as rational and autonomous agents. As I have repeatedly pointed out, Duffs interpretation of this requirement is excessively rationalistic. According to Duff, in cases of blaming and punishing anyway, social relationships should be constituted in and Duff, Trials and Punishments, p. 6. Ibid., p. 292. Ibid., p. 293. 92 limited to rational arguments about moral reasons. Presumably the ideally rational person will be convinced by rational arguments because he must follow the dictates of reason. Ideally, the accused who is guilty will admit it and accept or eventually come to accept punishment as his due - his penance. Yet according to Duff, our actual responses to wrongdoing reveal just how out-of-reach this ideal is. He suggests the ideal would be \"all too easily corrupted\" given \"the kinds of insensitivity, crassness and self-righteousness to which we are all so generally prone....\"33 Accordingly, he says, \"my concern is with the ideal rather than the actual; my aim is not to justify our actual responses to wrong-doing; but to explicate the ideals by which those responses are purportedly informed.\"34 Here, in my view, Duff points to a central flaw in his own account. He discounts the importance of our actual human responses to wrongdoing. On his own assessment, the resultant theory reveals a \"radical gap between the ideal and the actual.\"35 He argues that the gap indicates not that the ideal is misguided but that \"we need to secure deep and far-reaching changes in our social and political relationships, so that our society becomes a genuine community - changes which cannot be simply brought about by political or legislative action.\"36 I want to suggest that the ideal is misguided. Duffs Kantian view of human nature depicts rationality as a limited to cognitive processes which are separate and distinct from human emotions. Let us look more closely at Duffs notion of 3 3 Ibid., p. 295. 3 4 Ibid., p. 72. 3 5 Ibid., p. 293. 3 6 Ibid. 93 ideal rationality or, more precisely, his notion of what constitutes a rational response to the wrongdoer. On Duffs Kantian standard, emotions may play only an insignificant role in one's so-called rational response. Society must confine its response to arguments about moral reasons that appeal to the wrongdoer's capacity to reason cognitively. In his discussion of blaming, Duff does mention emotions in passing. He says, \"[M]oral persuasion appeals to the other person's moral understanding, and thus to the emotions which are involved in such understanding.\"37 Here he focuses on the cognitive content of emotions failing to appreciate the significance of their evaluative and affective components.38 Furthermore, although Duff appears alive to the fact that emotions may play a helpful role in blaming, he has nothing good to say about their role in relation to punishment. Without actually coming right out and saying so, Duff implies that the emotions related to punishment are, more often than not, vindictive. He supports his view that human beings are incapable of having the right kind of concern for one another by focusing on their purported proneness to inappropriate emotions such as \"insensitivity, crassness and self-righteousness.\"39 He suggests these emotions would characterize any attempt to put his ideal of punishment as penance into practice. Here Duff fails to distinguish justified retributive sentiments from mere vengefulness. He fails to 6 1 Ibid., p. 53-4. 3 8 In Valuing Emotions Michael Stocker argues that the notion that \"cool rationality is the best standpoint for inquiry and knowledge is a false truism. He points out that this idea is part and parcel of philosophers' overvaluation of a cool rationality that sets aside emotions as antithetical to reasoning. Stocker calls this view an \"idealization\" of rationality which \"involves serious interrelated misunderstandings about what is idealized - in this case, rationality -and also about what is split off and demonized or is simply not included in the idealization - in this case emotions.\" Michael Stocker, Valuing Emotions (New York: Cambridge University Press, 1996) p. 92. Stocker's overall thesis is that there are important constitutive and epistemological connections between emotions and values. He argues that \"emotions are expressions of, and may even be, evaluative knowledge.\" Ibid., p. 1. 3 9 Duff, Trials and Punishments, p. 295. 94 appreciate that experiencing and expressing resentment and moral indignation may be entirely rational and appropriate ways of responding to harm. On Duffs view, the ideal response to the rational, autonomous wrongdoer must address only his capacity for moral understanding. Duffs ideal of human nature is misconstrued. Human beings are not merely cognitive reasoners, nor should they strive to be. The human moral agent has the cognitive capacity of moral understanding to be sure. But he has much more than that. He has the additional capacities of self-evaluation and empathy. Both of these are also necessary aspects of his moral agency. Duff would likely agree that the capacity for self-evaluation is important. On his view, however, the capacity for empathy has no relevance because the ideal rational and autonomous individual follows the dictates of reason. But the capacity for empathy is relevant. It permits communication that is broader in scope than Duff is willing to allow for. It permits the victim and members of the community to go beyond mere cognitive understanding of the criminal act. It gives rise to their moral sentiments which have cognitive content but, significantly, also have evaluative and affective components. The evaluative component has epistemological value. It provides those affected by the crime with additional information about the wrongdoer's character and action. The affective component is motivating. It gives those affected an additional reason to respond to the situation. Moreover, considering the wrongdoer's perspective for a moment, his capacity for empathy gives him the ability to appreciate and be affected by the moral sentiments expressed. Social relationships require this kind 95 of engagement which moves one to respond to others. In short then, Duffs underlying assumptions about human nature are flawed. Accordingly, it is simply a mistake to subscribe to an ideal such as Duffs - an ideal which, to a great extent, rules out the epistemological and motivational aspects of our natural human responses. In conclusion then, Duffs interpretation of the Kantian demand leads him to produce an ideal theory of punishment. Moreover, he declares that there is such a radical gap between the ideal and the actual that \"Perhaps the ideal is too distant to be a human possibility....\"40 I have argued that the alleged existence of the radical gap should lead us, not to embrace skepticism, but to question Duffs uncritical acceptance of the Kantian demand. In the next section I show how the naturalistic theory resolves the difficulties that Duffs approach encounters. The Explanatory Advantages of the Naturalistic Alternative: Pluralism about Aims Versus Essentialism about Aims It should be obvious that, in spite of their superficial similarity as expressivist theories, Duffs account and the naturalistic theory differ considerably. Duffs theory is both excessively rationalistic and narrow in scope. He focuses on the necessity of communicating with the criminal only through her cognitive capacity of moral understanding. He advocates an essentialist aim, the criminal's 96 repentance and reform. The naturalistic approach, by contrast, embraces a robust concept of rationality and is broad in scope. It calls for interpersonal communication among wrongdoer, victim and community that taps all capacities of moral agency - moral understanding, self-evaluation and empathy. It is pluralist with respect to punishment's aims. It takes into account the perspectives and aims of all the players involved in the social institution. Duffs account is defended on the basis of the Kantian demand. By contrast, the naturalistic theory is grounded in naturalism. It is defended on the basis of the moral demand for some degree of goodwill. In other words, it accepts as a 'given' the attitudes and feelings that are part of our human nature. This framework opens the way to taking a broad approach which supports punishment as a social institution. The moral demand has its seat in the natural human expectation of goodwill from others. When a crime is committed, that expectation is not met. The criminal is implicated of course, but the victim and community are also affected and naturally react with retributive sentiments. The naturalistic theory acknowledges and supports the victim's and community's need to express those sentiments which are based on good reasons. The punishment, which expresses the retributive sentiments - among other things, conveys condemnation and an emphatic message that the wrongdoer cannot get away with devaluing others. It is, then, a strong form of communication \u00E2\u0080\u0094 a way of communicating the moral demand. 97 Punishment is usually also an attempt to persuade the wrongdoer to change her conduct in future. Unlike Duffs account, however, the naturalistic theory does not insist that the aim of bringing the wrongdoer to repent and reform is essential for punishment to be justified. Rather, the theory supports our commonsense assessment that some wrongdoers are beyond reform. In such cases, the wrongdoer is not the primary audience for the communication. In other words, this approach recognizes that punishment is not merely concerned with the desert and hoped for reform of the wrongdoer. It is also an expression of the victim's and community's retributive attitudes. The wrongdoer is one target audience; the community as a whole is another. In short, the theory can acknowledge that, in the case of hardened criminals, punishment will not lead to reform and should focus instead on specific deterrence. In those cases in which the wrongdoer is capable of reform, the community will want to motivate her to take up self-evaluation with the hope that she comes to feel guilt. In contrast with Duffs approach, however, the attempt to get that message across is not limited to addressing the wrongdoer via cognitive capacities. It is directed at all the capacities that constitute her moral agency -cognitive and emotional. Accordingly, the various steps to achieving justice can be viewed as emphasizing different aspects of the communication. The moral and pragmatic reasons presented at the trial and the judgment of her peers on conviction primarily address her capacity for moral understanding. The strong, emphatic condemnation of the sentence gives the victim and community the opportunity to express their retributive sentiments and, in the process, affect the 98 wrongdoer's capacity for empathy. Both of these forms of communication combine to attempt to spur the wrongdoer to exercise her capacity for self-evaluation. It is important, then, that the wrongdoer understand the reasons -moral, pragmatic and emotional - that motivate the victim and community to resort to punishment. It seems fairly obvious that incarceration alone may not be enough to bring about self-evaluation and the motivation to change. For cases in which there is a possibility of reform, society may want to give the wrongdoer the opportunity to gain moral understanding and to develop her capacity for empathy. In other words, penal measures that combine education with incarceration are likely to be more effective than incarceration alone.41 Duffs interpretation of the Kantian demand focuses on the need to respect the rationality and autonomy of the criminal. It results in an ideal account of punishment and a thoroughgoing skepticism about the possibility of achieving that ideal in any actual society. By contrast, the naturalistic theory is grounded in the moral demand to treat others with a certain degree of goodwill. It engenders no conflict with actual human social existence. Rather, it begins with Strawson's insight that human beings are naturally social beings; and given with our natural commitment to social existence is a natural commitment to the whole web or structure of human personal and moral attitudes, feelings, and judgments... . 4 2 In B r i t i s h Columbia there are a number of programs that a im to educate offenders dur ing the period of incarceration. In one program, for example, female offenders are paired w i t h orphaned dogs. E a c h dog is dependent on an offender for care and companionship. The idea is to develop the offender's sense of responsibility. In my view, i t is also a good way to develop her capacity for empathy as a dog natura l ly responds to good care w i t h loyalty and affection. 4 2 Peter Strawson, Skepticism and Naturalism (New York : Columbia Univer s i ty Press, 1985), p. 39. 99 It is worth noting that the moral demand makes no direct reference to rationality or autonomy but rather to goodwill or regard. This fact does not imply that the notions of rationality and autonomy have no role to play in human social existence. It does imply, however, that these notions must be integrated into a picture of social interaction that takes account of emotional as well as cognitive capacities. With regard to rationality, the naturalistic account asserts that the rational response to the wrongdoer is not limited, as is Duffs approach, to rational arguments about moral reasons. As the preceding discussion points out, the rational response can and should attempt to tap the wrongdoer's capacities for moral understanding, empathy and self-evaluation. It can and should make use of moral, pragmatic and emotional reasons. With regard to autonomy, the naturalistic approach focuses on the naturally social nature of the various players involved in the social institution of punishment. Again there is a contrast with Duffs theory which focuses on the autonomy of individuals, particularly the criminal. Although the two theories are in agreement that members of society should have the freedom to determine and pursue their own values and goals, the naturalistic approach makes it clear that one's values and goals are neither determined nor pursued in isolation and independence from others. As social beings we care about the attitudes and feelings that others display towards us. We empathize with one another. This natural emotional engagement may be experienced and expressed in the moral sentiments - among other ways. Moreover, expressing justified retributive 100 sentiments is not manipulative as long as their expression is appropriate - in accordance with moral and social norms. It does not disregard autonomy. It does, however, communicate information to others about the limits of their personal autonomy. In short, the moral demand for goodwill makes no explicit reference to autonomy because the give-and-take of interpersonal relationships is grounded in emotional engagement, not autonomy. In this chapter I have compared R. A. Duffs expressivist account of punishment with my naturalistic account. I have argued that the attempt to defend an expressivist theory within a Kantian framework fails because of the impossibility of satisfying both the Kantian demand with respect to the criminal and the needs and concerns of the victim and community. By contrast, the naturalistic framework provides the resources to explain how social concerns integrate into the justification of punishment. In stark contrast with Duffs account, the naturalistic theory sanctions the victim's and community's expression of justified hostile emotions towards the wrongdoer. This difference between the two theories brings to the fore the question of whether the naturalistic theory can account for the need to respect the wrongdoer's dignity. In the next chapter I explain how the theory accommodates that need. 101 Part II. Clarifications Chapter 4 Punishment and Respect for the Wrongdoer's Dignity In the preceding chapter I showed that Duffs Kantian version of expressivism is inadequate because it fails to account for the needs and concerns of the victim and community. By contrast, the naturalistic theory endorses the victim's and community's expression of retributive sentiments. Can the naturalistic approach deal with the other side of the coin, however? Can it account for the need to respect the wrongdoer's dignity? In this chapter I argue that it can. My argument rests on two claims. First, punishment should be understood in the context of a broadened notion of communication which takes into account the human capacity to communicate nonverbally as well as verbally. I contend that nonverbal communication should not be evaluated on the same paradigm as verbal communication. I propose a provisional paradigm which judges nonverbal communication as appropriate based on moral and social norms. Second, punishment should be judged appropriate on the basis of two normative principles - proportionality and a principle prohibiting degrading punishments. I hold that punishment respects the wrongdoer's dignity when it satisfies this criterion of appropriateness. The Strategic Role of the Emotions I want to begin the discussion by backtracking a bit and making explicit what has been implicit in the discussion of emotions so far. Emotions have been described 102 from two perspectives. From the biological perspective, they are innate propensities to react which have evolved as a result of their adaptive value for the species or the individual. Emotions have a strategic role to play in the human evolutionary story. From the psychological/social perspective, however, they are psychological dispositions having an internal rationale. They have intentional content and are shaped by social norms. When the latter perspective is taken, the question of the appropriateness of emotional expression arises. This justification must address the question of whether punishment is an appropriate expression of retributive sentiments. And that discussion must make reference to psychological and social explanations. Nonetheless, the strategic role of emotions should not be set aside when attempting to understand what an appropriate expression of an emotion might be. Why? Traditional scientific and philosophical explanations of human behavior assume that people must suppress emotions in order to be most effectively rational. Recent sociobiological explanations challenge that view, however. It will be helpful, then, to keep in mind both the strategic and the internal explanations when determining whether punishment is an appropriate expression of retributive sentiments. A Broadened View of Communication: Nonverbal Communication of Emotions In Chapter 2 I argued that punishment is a type of communication from various players to a variety of audiences. There I focused on the players, pointing out that each has a differing perspective on the crime and, consequently, different 103 messages to communicate to various audiences. Here I want to focus on the type of communication that punishment is. In this and the next two sections I defend the first claim in my argument about dignity - the idea that punishment is, in part, a nonverbal communication of retributive emotions. As a starting point, in light of the strategic role of emotions, let us take the expression of retributive sentiments to be a response to harm aimed at communicating information to the wrongdoer and others. I have, in fact, already made this claim in a slightly different way in Chapter 2 where I argued that the unifying theme of this justification is communication of the moral demand for goodwill. In other words, the claim that punishment has a communicative function is, if nothing else, consistent with Plutchik's explanation of emotions as strategic responses that communicate information from one individual to another. As noted however, from the traditional philosophical perspective the expression of emotions - especially hostile ones - is viewed with suspicion. After all, the philosophical paradigm of communication is the logical argument. It is a form of verbal communication and the acknowledged model of rational communication. Yet this paradigm reflects the traditional view that reason and emotion are opposed. It implicitly supports the belief that human beings must suppress emotion in order to be most effectively rational. It supports the belief that they must suppress emotional expression in order to communicate as rationally as possible. 104 Recent interpretations of empirical evidence challenge these beliefs, however. The evidence suggests that the person who is most successful in eliminating emotion from the reasoning process should not be construed as the most rational. Rather, reason and emotion work together to produce rational planning and decision-making.1 In other words, social living requires \"emotional intelligence,\"2 that is, emotion appropriate to circumstances. As I noted in Chapter 1, the experiencing and communicating of emotions calls upon the human capacity for empathy - the capacity to share imaginatively in how another feels.3 Empathy is necessary for successful interpersonal relations 1 In Chapter 11 pointed out that Goleman presents a great deal of psychological and neurophysiological evidence to support his claims about the importance of emotions for social living. For an analysis of the value of emotions from the perspective of theories of rational choice see Robert H . Frank, Passions Within Reason: The Strategic Role of the Emotions, (New York: W. W. Norton & Company, 1988.) For a detailed neurophysiological approach see Antonio R. Damasio, Descartes' Error: Emotion, Reason and the Human Brain, (New York: G. P. Putnam's Sons, 1994.) 2 I have suggested that the traditional philosophical view of rationality belittles the role of emotions. Similarly, Goleman argues that the predominant models of mind among cognitive scientists have failed to acknowledge that \"rationality is guided by - and can be swamped by -feeling.\" Goleman, Emotional Intelligence, p. 41. He.points out, however, that this \"lopsided scientific vision of an emotionally flat mental life\" is changing as psychologists come to recognize the role of feeling in thinking. On Goleman's view, \"emotional intelligence\" includes \"abilities such as being able to motivate oneself and persist in the face of frustrations; to control impulse and delay gratification; to regulate one's moods and keep distress from swamping the ability to think; to empathize and to hope.\" Ibid., p. 34. Of particular interest to my arguments about expressing retributive emotions is Goleman's claim that the goal of emotional life is not suppression but expression appropriate to circumstances. \"When emotions are too muted they create dullness and distance; when out of control, too extreme and persistent, they become pathological....\" Ibid., p. 56. 3 John Deigh points out that the notion of \"empathy capacity\" does not have a settled meaning among those who write about it. In setting out his developmental account of empathy, Deigh attempts to explicate the concept. In particular, he distinguishes mature empathy from immature forms (common in children) such as emotional contagion and emotional identification. Deigh emphasizes an important distinction which I want to reinforce. He notes that mature empathy and emotional identification both involve taking another's perspective and imaginatively participating in the other's life. The difference is that the agent who has a mature capacity for empathy does not lose her own perspective when she empathizes with another. The experience of emotional identification is, by contrast, likely to result in \"a loss of the sense of oneself as separate from the person with whom one identifies.\" This distinction is 105 and, in particular, for morality. Goleman claims that \"The failure to register another's feelings is a major deficit in emotional intelligence, and a tragic failing in what it means to be human. For all rapport, the root of caring, stems from emotional attunement, from the capacity for empathy.\"4 He points out further that Martin Hoffman, who conducts research on empathy, argues that the roots of morality are to be found in empathy, \"since it is empathizing with the potential victims \u00E2\u0080\u0094 someone in pain, danger, or deprivation, say - and so sharing their distress that moves people to act to help them.\"5 Of course, the human capacity for empathy has not been entirely ignored in the philosophical tradition. Adam Smith, among others, recognizes and relies on it in developing his theory of the moral sentiments. He calls it \"sympathy\" saying, \"Pity and compassion are words appropriated to signify our fellow-feeling with the sorrow of others. Sympathy, though its meaning was, perhaps, originally the same, may now ...be made use of to denote our fellow-feeling with any passion whatever.\"6 (my emphasis) In order for one person to come to appreciate how another feels, the other's feelings must be communicated. Empirical evidence indicates not only that appropriate emotions are necessary for rationality but also that they are important in that it supports the idea that the community's moral indignation arises from a perspective that is different from the victim's. It also supports the idea that moral indignation incorporates a normative component that involves approval of the victim's resentment. John Deigh, \"Empathy and Universalizability,\" Ethics, 105 (July, 1995), p. 759. 4 Goleman, Emotional Intelligence, p. 96. 5 Ibid., p. 105. 6 Smith, Theory of the Moral Sentiments, Li.1.5, p. 10. 106 conveyed, to some extent, nonverbally.7 The evidence suggests that nonverbal modes of communication transmit additional, valuable information such as information about others' emotional reactions to oneself and one's actions. To understand and explain fully communication in social living then, we must examine the nonverbal mode. As I pointed out, the paradigm for philosophical communication is a verbal one, the logical argument. We should not judge emotional communication based on it, however, because it implicitly assumes that emotions are antithetical to rationality. In order to evaluate the communication of retributive sentiments through punishment, we must broaden the traditional philosophical notion of rational communication to take account of other modes of expression, in particular, the nonverbal mode. Since philosophers have no paradigm for nonverbal communication, I put forward a provisional one in the next section. My argument to justify punishment as, in ' See Robert M . Gordon, \"Sympathy, Simulation and the Impartial Spectator,\" Symposium on Empathy and Ethics, Ethics (105) July, 1995, pp. 727-42. Gordon argues that the cognitive mechanism of empathy is simulation. Significantly for my purposes, the simulation theory lends support to the claim that nonverbal communication is an essential mode for communicating emotions. Gordon argues for the advantages of the simulation theory over Fodor's \"theory\" theory using Fodor's own example from A Midsummer Night's Dream. In the relevant scene Hermia is speaking with Lysander when she notices Demetrius walk by and look at them in a strange way. She interprets that look along with other evidence as an indication that Demetrius is likely to do harm to Lysander. Fodor claims that Hermia reaches her conclusion by formulating a \"theory\" about Demetrius using logical reasoning. Gordon argues, by contrast, that Hermia's assessment of the situation is acquired through a 'hot methodology\" of \"mirroring\" or \"simulation.\" He says, \"She uses her own perceptual, cognitive, motivational and emotive resources to mirror Demetrius's mind as best she can, and then, continuing in the role of Demetrius, predicts what he wil l do by deciding what to do. This makes it easy to understand how the look she picked up in her own facial muscles, and the underlying emotion she picked up as a result, would work together with other evidence.\" Ibid., p. 733. The relevant point here is that Demetrius's emotions are communicated to Hermia nonverbally. 107 part, a nonverbal communication of retributive sentiments uses this provisional paradigm. Nonverbal Communication of Emotions: Transmitting Additional Information and Supplying Motivation The provisional paradigm, which is suggested by empirical evidence and the analysis of ordinary examples, highlights three salient features of nonverbal communication. First, it transmits additional, valuable information that, in some cases, cannot be effectively communicated verbally - in particular, information about one's emotions. Second, the information communicated has motivational import. Third, to be justified such communication must be appropriate where appropriateness is judged on the basis of moral and social norms. In this section I first look at some further empirical evidence which supports the paradigm generally. I then indicate how moral sentiments fit the paradigm by looking at some common examples. In the sections that follow I focus on punishment showing that, when it fits the paradigm, it respects the wrongdoer's dignity. There is much evidence to support the claim that nonverbal communication transmits information about our emotional reactions. In our everyday experiences we observe friends expressing their affection by smiling, shaking hands and embracing. We see enemies express their dislike with frowns, avoidance of close contact and, in the extreme, the violent contact of pushing, shoving and fighting. According to Goleman, \"People's emotions are rarely put 108 into words; far more often they are expressed through other cues...tone of voice, gesture, facial expression and the like.\"8 Goleman presents convincing evidence in support of the significance of nonverbal communication both in non-human primates and in human beings.9 In humans the evidence comes from patients with neurological problems. Those with certain lesions in the right frontal lobe of the brain were unable to understand the emotional message in another person's tone of voice. A sarcastic \"thanks,\" a grateful \"thanks\" and an angry \"thanks\" all had the same meaning. Patients with injuries in other parts of the right hemisphere were unable to express their own emotions through tone of voice or by gesture. They knew what they felt, but could not convey it in the normal nonverbal way. In both cases, the damaged parts of the brain were areas with strong connections to the limbic system, the area in which the amygdala, seat of the emotions, is located. The evidence leads Goleman to conclude that \"the mode of the emotions is nonverbal.\"10 I suggest it indicates that nonverbal communication supplements the verbal mode adding crucial information about emotions the absence of which leaves the message incomplete. \u00C2\u00B0 Goleman, Emotional Intelligence, p. 96. 9 In one experiment rhesus monkeys were trained to fear a tone by hearing it while receiving an electric shock. They learned to avoid the electric shock by pressing a lever. The monkeys were then placed in separate cages with their only communication being through a T V which allowed each to see the face of the other. When the dreaded tone was sounded for the first monkey but not the second, a look of fear came to the first monkey's face. On seeing that look, the second monkey immediately pushed the lever that prevented the shock. Further research on monkeys shows that certain neurons in the visual cortex of the brain fire only in response to specific facial expressions and gestures. Goleman claims these experiments indicate that \"the brain is designed from the beginning to respond to specific emotional expressions - that is, empathy is a given of biology.\" Ibid., p. 103. While such experiments point to the effectiveness and, indeed, necessity of nonverbal communication of emotions in non-humans, they have not, for obvious reasons, been conducted on human beings. 1 0 Ibid., p. 97. 109 It is worth reiterating here that the communication of moral sentiments, whether verbal or nonverbal, is susceptible to rational justification. Moral sentiments have cognitive content and, as a result, can be rationally justified. They are, as it were, \"intelligent\" emotions in the sense that they convey \"considered\" emotional reactions \u00E2\u0080\u0094 reactions based on reasons and appropriate to the circumstances. Our moral sentiments are expressed through punishment of course, but also through praising, blaming and rewarding. Public pronouncements of praise or blame are usually expressed verbally but they are in no way merely verbal. In praising, for instance, the tone of voice, gestures and facial expressions that accompany the words have as much or more significance than the words themselves. Before looking at punishment then, consider some examples of these other practices. The examples show that our expression of rationally justified moral sentiments often involves nonverbal communication which is essential to the message. Imagine a teacher praising a child for giving an excellent class presentation. The teacher might say enthusiastically \"I'm sure you all agree that Anne did an excellent job on this report\" while at the same time smiling and placing her arm around Anne's shoulder. If the teacher were simply to say the words in a monotone voice with a deadpan expression and no accompanying gesture, the praise would be taken to be, at best, insincere and, at worst, mocking. The nonverbal component is an essential part of the communication. It not only reinforces the teacher's words but also communicates, in a way that the words 110 cannot, her approval of the child and her actions.11 The nonverbal mode also tends to motivate the child in a way that the words cannot. The words address the child's understanding; the nonverbal communication addresses her capacity for empathy. The teacher's nonverbal cues are a powerful way of letting the child know she cares about her. That caring and approval serve to motivate the child to seek further approval through similar actions. Nonetheless, if the praise is too great in proportion to the child's accomplishments, it will be judged insincere just as the merely verbal praise without accompanying nonverbal cues would be. The point is that, as social beings, we have a well-developed, socially reinforced sense of what the appropriate reaction to another's accomplishments should be. Praise out of proportion to accomplishments - too little or too much - is judged inappropriate and unjustified. The example, in the previous chapter, of blaming Jones for failing to call home illustrates in an analogous way the nonverbal component of public blaming. In that example the childrens' refusal to give Jones a hug and her husband's angry tone of voice and gestures all serve to communicate more effectively than mere words the family's negative reaction to Jones' breaking her promise. It is the family's attitudes and emotions expressed through these reactions that motivate Jones to feel guilty, express remorse and make amends by taking a few days off from work to spend with the family. 1 1 Significantly, Goleman points out that \"when a person's words disagree with what is conveyed via his tone of voice, gesture, or other nonverbal channel, the emotional truth is in how he says something rather than in what he says.\" Ibid. I l l It is particularly instructive to look at what might be considered punishment's polar opposite, society's practice of rewarding persons such as humanitarians, public servants and heroes.12 Public expressions of very strong positive moral sentiments are often accompanied by symbolic or material rewards. These nonverbal tokens communicate strong approval of the person and his actions. Furthermore, they are intentional communications. The bestowing of a symbolic or material reward is a communicative act that is intended to mean something to both the recipient and the public at large. Awarding a medal for bravery in battle, for example, declares approval through the community's recognized symbol of honor. Similarly, the international community proclaims approval of persons who have improved the human condition by bestowing the Nobel Prize, a material token of appreciation. These rewards are nonverbal components of the communication of gratitude and admiration. They demonstrate publicly to the person honored and others that the community cares about him and his actions. Furthermore, they are motivating. They serve to encourage the person honored and others to act in ways that benefit humanity. Now consider the trial, conviction and punishment in the light of the proposal that justified communication can be nonverbal as well as verbal. On the broadened view, punishment is, among other things, a nonverbal expression of very strong negative moral sentiments. It, along with the trial and conviction, is part of the stakeholders' communication with their various audiences. Recall that Duff says the trial is a kind of rational argument with the wrongdoer. The naturalistic 1 2 Duff points out that rewards, like punishments, serve an expressive purpose. As the discussion in the previous chapter points out, however, he does not take the practices of punishing or rewarding to be expressions of emotions. Duff, Trials and Punishments, pp. 236-7. r 112 theory agrees that it is a kind of rational argument but rejects Duffs claim that the communication targets only the wrongdoer. The audience is not just the wrongdoer but also the victim and community. At the trial the reasons supporting the wrongdoer's guilt are examined and assessed. The reasons why the criminal action must be condemned are reviewed. If the wrongdoer is guilty, the conviction is the \"conclusion\" of the argument. In other words, the trial is a formal process of mostly verbal communication which publicly sets out the moral, pragmatic and emotional reasons for condemning the wrongdoer.13 Once the wrongdoer is judged guilty, punishment is imposed , in part, as an emotional expression of the victim's and community's rationally justified retributive sentiments. The punishment communicates to the various audiences additional, important information not communicated by the trial and conviction. Undoubtedly, verbal condemnation will have been expressed at the trial and on conviction. The sentence, however, expresses condemnation in a much stronger way - nonverbally. In other words, it is a potent form of communication that is intended to convey to the wrongdoer and others the victim's and community's feelings about the crime. From the victim's perspective, it denounces the wrongdoer and asserts self-respect. From the community's perspective, it denounces the wrongdoer, empathizes with the victim and approves of her reaction. From the wrongdoer's perspective - in cases where he is capable of reform, the sentence may motivate him to take up self-evaluation. If he is 1 6 Because the victim does not have an institutionalized role to play in the existing system, the emotional reasons for condemning the wrongdoer may be, to a great extent, downplayed or even ignored. 113 repentant, certain types of sentences may give him the opportunity to make reparation. Is Punishment Motivating? I have contended not only that punishment provides information about the victim's and community's retributive sentiments but also that it may motivate the target of the communication. Is punishment motivating? The critic might concede that the sentence conveys condemnation. He would insist, however, that it is clearly ineffective in motivating the wrongdoer. The rate of recidivism among criminals lends empirical support to this charge. The objection has force in relation to our actual sentencing practices which are, I submit, aimed primarily at deterrence. Sentencing to deter has usually resulted in incarceration and incarceration alone. And it is naive in the extreme to believe that incarceration by itself will motivate any but a few exceptional wrongdoers to take up self-evaluation. Incarceration sends a clear message of condemnation but is too impersonal and non-specific to elicit guilt and promote self-evaluation. The naturalistic view claims, however, that punishment has a plurality of aims unified under the theme of communication of the moral demand. The recognition that the overall aim of punishment is communicative has implications for sentencing. The goal should be to determine a sentence that has a good chance of communicating successfully with the various audiences. To be sure, incarceration may be necessary to communicate condemnation and to protect the public, particularly when the criminal is prone to violence or is a repeat offender. When there is a possibility for reform, however, the communication 114 should also aim to reach the wrongdoer - by attempting to bring about a change in his moral psychology. Some penal measures are more suitable than others for motivating wrongdoers. Meeting the victim face-to-face in victim/offender mediation, for example, has proven to be motivating for most (carefully selected) offenders who have participated. Victims too have been mostly satisfied with the results of mediation. Proponents emphasize that the face-to-face meeting is a key component of this penal measure. It allows for both verbal and nonverbal communication at a personal level that leaves little room for the offender to deny the victim or the impact of the crime. I suggest that its motivating power arises to a great extent from the nonverbal communication that takes place. In short, the answer to the critic's objection is that some penal measures have a greater probability than others of motivating wrongdoers. In particular, as the examples discussed here and earlier in the chapter show, nonverbal communication can be a strong motivator. It communicates one person's feelings regarding the other person's character and actions to that other person. And other peoples' feelings towards one may be a catalyst for change. For, as Strawson points out, we care deeply about other people's attitudes and reactions towards us. Therefore, although the question is open as to which penal measures are likely to succeed in motivating wrongdoers, the goal of motivating the wrongdoer is one good reason to use nonverbal communication. Another good reason is, of course, that the sentence gives the victim and community an 115 opportunity to express their retributive sentiments in a strong, emphatic and enduring way. So far, I have argued that punishment satisfies two of the features of the provisional paradigm for nonverbal communication. It transmits additional, valuable information about the victim's and community's strong retributive sentiments. And it has the potential to be motivating. Recall, however, that, although human beings inevitably express their retributive sentiments, they have a choice about how to express them. The question that remains to be answered is whether expressing retributive sentiments through punishment is appropriate. Punishment, an Appropriate Form of Nonverbal Communication In this section I defend the second claim in my argument to show that punishment respects the wrongdoer's dignity. I explain how punishment is an appropriate form of, in part, nonverbal communication where appropriateness is judged on the basis of two normative principles.14 The first is the principle of proportionality, that is, the worse the crime is the more severe the punishment should be. The second is the principle that punishment must not be degrading. This principle limits the types of punishments that justifiably may be used. 1 4 The argument in this section owes much to Jeffrie G. Murphy's treatment of the issues. See Jeffrie G. Murphy, \"Cruel and Unusual Punishments,\" Retribution, Justice and Therapy (Boston: D. Reidel Publishing Company, 1979), pp. 223-249. 116 Consider proportionality first. The contemporary notion of proportionality is not the bare retributive principle captured in a crude version of the lex talionis. That version recommends punishment equivalent to the crime, for example, rape for rape, assault for assault. The contemporary version advocated here is intended to embody fairness. It dictates that punishments and crimes are ranked on corresponding scales of severity. The most serious punishments are imposed for the most serious crime, the next most for the next most serious crimes and so on. A punishment will be unfair and unjustified if it is out of proportion to the severity of the crime. For example, imposing a ten-year prison sentence on a pickpocket who stole a hundred dollars would be unfair and unjustified. Proportionality has intuitive appeal because it typifies fairness. It raises the question, however, of how one determines a crime's severity. We may measure severity on the basis of the wrongdoer's intent, the harm done or some combination of the two. It should be clear by now, though, that the naturalistic account widens the scope of justification to encompass the impact of the crime on victim and community. In other words, the severity of a crime is judged not merely on the basis of the wrongdoer's intent but also on the harm done. The question of how to relate these factors to the degree of punishment is a complex one which I take up in detail in the next chapter. For the purposes of the present argument then, it is enough to accept the principle of proportionality based on a norm of fairness, setting aside for now the question of its precise explication. Jeffrie Murphy astutely points out, however, that \"considerations of fairness alone will not answer the question of which punishments will be allowed as the 117 most serious. There will be substantive reasons for not allowing certain punishments (e.g. torture) even if these would satisfy a fairness principle of proportionality.\"15 Murphy's observation highlights the need for the second principle of appropriateness prohibiting degrading punishments. Murphy offers a Kantian argument against degrading or dehumanizing punishments. He says punishments such as torture and mutilation, for example, are \"...addressed exclusively to the sentient or heteronomous - i. e. animal -nature of a person. Sending painful voltage through a man's testicles to which electrodes have been attached, or boiling him in oil, or eviscerating him or gouging out his eyes - these are not human ways of relating to another person. He could not be expected to understand this while it goes on, have a view about it, enter into discourse about it, or conduct any other characteristically human activities during the process \u00E2\u0080\u0094 a process whose very point is to reduce him to a terrified, defecating, urinating, screaming animal.\"16 Murphy's argument against degrading or dehumanizing punishments is based, ultimately, on Kant's metaphysical distinction between the natural and rational -the phenomenal and noumenal - selves. Clearly, that distinction is not available to this justification which is firmly rooted in a natural description of human beings. Murphy's argument can be adapted, however, to work with the naturalistic explication of dignity set out in Chapter 1. Moreover, the claim that punishment is a type of communication with a variety of audiences provides additional resources, untapped by Murphy, with which to rule out degrading punishments. In Chapter 11 put forward a norm of human dignity based on a naturalistic description of the moral agent. Here I shall show that punishment may respect 1 5 Ibid., p. 232. 1 6 Ibid., p. 233. 118 the wrongdoer's dignity explicated in this way. I begin by arguing that cruel punishments such as torture, mutilation and, I shall claim,'corporal puriishments do not respect the wrongdoer's dignity and, therefore, are not justified. I then argue that penal measures routinely imposed such as loss of freedom and hard labor normally respect the wrongdoer's dignity. I caution, however, that even these penal measures can be imposed in ways that are degrading and that, therefore, even they may violate dignity. On the naturalistic approach, an agent's dignity is grounded in his natural capacities of moral understanding, self-evaluation and empathy. In contrast with the Kantian notion, this notion does not focus exclusively on rational capacities that are explicated as, somehow, distinct from or superior to other human capacities. In particular, this notion of dignity does not require people to overcome their natural commitment to the attitudes and feelings that constitute their social relationships. It allows that respecting another's dignity involves communicating with her in ways that tap all the capacities that underlie her moral agency including her capacity for empathy. It recognizes that the human psychological make-up must be taken into account in setting out and explicating moral concepts such as dignity. What, then, are we to make of the expression of hostile emotions through punishment? Recall that the cognitive content of a retributive sentiment consists in a belief, based on reasons, that the wrongdoer failed to meet the moral demand and a corresponding evaluation of her as uncaring of the victim's value. The negative reaction to the wrongdoer's disregard of the moral demand vis va vis 1 1 9 the victim is, in fact, an affirmation of the victim's dignity and a call for the wrongdoer to respect it in future. Both resentment and moral indignation, however, also reflect an emotional 'connection'17 with the wrongdoer. They are natural negative reactions towards one who is judged capable of fulfilling the moral demand but disregards it. As Strawson notes, the \"partial withdrawal\" of goodwill from the wrongdoer is part and parcel of continuing to view her as a member of the moral community, \"only as one who has offended against its demands.\"18 In other words, the wrongdoer in virtue of her humanity is accepted as an appropriate target for the attitudes and feelings of the community. The expression of retributive sentiments towards her is a reflection of the fact that she is judged capable both of understanding the reasons for the negative reaction and of caring about others. Underlying the reaction is the implicit recognition that she may well be affected by others' attitudes and feelings. In short, the expression of retributive sentiments is an attempt to communicate with another intentional being who can both be reasoned with and be affected by the emotional reactions of her fellows. Does explicating dignity in this way leave it open for victim and community to express strong retributive sentiments through cruel punishments such as torture and mutilation? No, it does not. Adapting Murphy's argument against such 1 7 Goleman assembles and interprets a huge amount of psychological and neurophysiological evidence that supports the notion of an emotional 'connection' among human beings which involves subtle, often nonverbal communication. Goleman says that \"Emotions are contagious....Most emotional contagion is ...subtle, part of a tacit exchange that happens in every encounter. We catch and transmit moods from each other in what amounts to a subterranean economy of the psyche in which some encounters are toxic, some nourishing. This emotional exchange is typically at a subtle, almost imperceptible level...\" Goleman, Emotional Intelligence, pp. 114-17. 1 8 Strawson, \"Freedom and Resentment,\" p. 77. 120 punishments to the naturalistic approach, I contend that a penal measure treats the wrongdoer with dignity only if it addresses her on the basis of the three capacities that constitute her moral agency. Any punishment which does not violates her dignity and is not appropriate. Penal measures such as torture and mutilation, for example, do not address the wrongdoer via her capacities. As Murphy so vividly points out, these punishments treat humans as mere animals.19 It is true that the trial may address the wrongdoer's moral understanding regardless of what punishment follows. When the punishment that follows is torture or mutilation, however, the reasons given at the trial which back up the community's negative reaction lie unused and useless. The inflicting of monstrous physical pain or ineffable psychological suffering overwhelms the wrongdoer's capacities to understand, to feel empathy, to evaluate. The capacities become inoperative. A wrongdoer who is paralyzed with fear of the impending punishment or actually suffering the excruciating pain of torture or mutilation is incapable of empathizing with the victim or reflecting on the reasons for rejecting the criminal conduct in future. Moreover, the effects of torture and mutilation may well be both physically and psychologically damaging in permanent ways - ways that cannot, in some cases, even be predicted by the punishing authority. In other words, there is an appreciable risk that some wrongdoers punished in this way may never regain their capacities to understand, to feel empathy, to evaluate. In short, punishments such as these are degrading. They do not communicate with the In fact, for other moral reasons, we would rule out treating animals in the way described. 121 wrongdoer qua moral agent and, therefore, do not respect her dignity. Consequently, they are ruled out as inappropriate. The critic would object, however, that the naturalistic approach sanctions nonverbal communication and that torture and mutilation are, in fact, extreme forms of nonverbal communication. It is true that torture and mutilation are, for the most part nonverbal. Yet they do not communicate with the wrongdoer when they are inflicted. As noted, they incapacitate him to such a degree that he becomes incapable of receiving the communication.20 Nonetheless, the critic could insist that once the torture or mutilation stops, the wrongdoer is no longer incapacitated and , therefore, is in a position to recall and reflect on the nonverbal message. At least two important responses can be made. First, as I pointed out above, there is considerable risk that punishments such as torture or mutilation will permanently incapacitate the wrongdoer. He may actually become incapable of moral understanding. For example, psychological torture or actual mutilation of his brain such as frontal lobotomy may impair or destroy his capacity to reason. Another probable result is that his capacity for empathy will be severely impaired or even destroyed. It seems reasonable to expect that the wrongdoer's reaction to being deliberately tortured or mutilated by the state will be to lose his receptivity to others' expressions of approval or disapproval. That is, the 20 In The Body in Pain Elaine Scarry analyzes the elements of great pain such as that inflicted through torture. She points out that one of the aspects of great pain is its \"obliteration of the contents of consciousness. Pain annihilates not only the objects of complex thought and emotion but also the objects of the most elemental acts of perception.\" Elaine Scarry, \"The 122 degrading treatment from fellow moral agents will be such an emotional shock that it will permanently dull his capacity for empathy. The critic could insist, however, that the punishing authority could devise measures to ensure that wrongdoers would not be permanently incapacitated. This possibility prompts my second response. Suppose methods of torture or mutilation or other cruel punishments were carefully devised so that the wrongdoer was not permanently incapacitated. He would then have the capability to reflect on the nonverbal communication after the fact. The problem is that cruel punishments communicate the wrong message. On the naturalistic approach, one aim of punishment is to communicate condemnation to the wrongdoer because of his crime. Cruel punishments convey to the wrongdoer a message that goes far beyond condemnation, however. They declare the wrongdoer to be worthless - an object rather than a moral agent. This distinction between condemnation and degradation is an important one which is at the heart of what is and is not to be conveyed through punishment. Jean Hampton captures the import of the distinction when she points out that resentment is an emotional protest against one's victimization which may be a way to assert that one really is better than one's wrongdoer, \"not in the sense that one is morally more valuable than he is, but in the sense that he merits moral disapproval in a way his victim does not.\"21 (my emphasis) In other words, we want the punishment to communicate condemnation, not worthlessness. Structure of Torture,\" The Body in Pain: The Making and the Unmaking of the World (New York: Oxford University Press, 1985), p. 54. 2 1 Murphy and Hampton, Forgiveness and Mercy, p. 60. 123 To put the point another way, cruel punishments are indicative not merely of the \"partial withdrawal\" of goodwill to which Strawson refers but of its complete withdrawal. With a partial withdrawal of goodwill we continue to view the wrongdoer as a moral agent, but as one who deserves disapproval. With a complete withdrawal of goodwill, we no longer view the wrongdoer as a moral agent at all. Cruel punishments send the message that the victim and community view the wrongdoer as having little or no moral value. Cruel punishments communicate the wrong message not just to the wrongdoer but to the other audiences as well. The infliction of such punishments proclaims to all that it is fitting and right for the state to respond to wrongdoers with brutality. This message of brutality affects the public's capacity for empathy in the wrong way. It dulls the capacity for empathy rather than educating it to be sensitive to the suffering of others. The spectacle of the state first threatening a human being with excruciating pain and suffering and then carrying out the threat tends to corrupt one's capacity to care about one's fellows. In so doing, the state not only condones but actually promotes the acceptance of brutality towards others. The critic might concede that the naturalistic theory has the resources to rule out cruel punishments such as torture and mutilation but respond that lesser corporal punishments are justified by this approach. It is true that some people consider all corporal punishment as varieties of torture or mutilation. Let us concede, however, that a distinction can be made between 'mere' corporal 1 2 4 punishment and extreme punishments such as torture or mutilation, even though the distinction may be difficult to characterize precisely. Corporal punishments, then, would be physical punishments that cause the wrongdoer pain and suffering but are not extreme enough to be classified as torture or mutilation. The proponent of such punishments could say that they are a nonverbal communication of the victim's and community's retributive sentiments. Moreover, they are desirable because they convey the message of condemnation quickly and 'succinctly.' Like torture and mutilation, however, corporal punishments communicate the wrong message. Such punishments communicate more than the strong disapproval of condemnation. They communicate the message that the wrongdoer is morally less valuable than the other members of society. Degradation come in degrees, after all. Corporal punishments may not be as degrading as torture and mutilation, but they convey a degree of degradation nonetheless. When we discuss corporal punishment in the abstract, it may not be obvious that the message is one of degradation rather than condemnation. When we think about actual examples, however, the message of degradation becomes evident. Caning, whipping or beating a wrongdoer is likely to result in various physical injuries such as bruising, torn flesh, bleeding, broken bones, scarring or other permanent disfigurements. These punishments communicate the message that brutality towards this person, expressed through the violence of the corporal punishment, is permissible. I submit that the message communicated is one of degradation. 125 Moreover, corporal punishments communicate the wrong message - a message of violence - not just to the wrongdoer but also to the other audiences for the communication. In Punishment and Modern Society David Garland suggests that the state has a civilizing mission. He notes further that the state's social institution of punishment may have a significant influence on the public. He says, \"...social institutions can provide a sentimental education for a population, bringing about a refinement of feeling and a growing sensitivity to the rights or the sufferings of others, other more reactionary policies can begin to undo the civilizing process, and to unleash aggressions, hostilities, and selfishness in the sphere of public life.\"22 (my emphasis) When the state uses corporal punishments, it condones the use of violence towards others. As with cruel punishments, the violence of corporal punishments dulls the capacity for empathy in both the wrongdoer and the public at large. When a corporal punishment is the sentence \u00E2\u0080\u0094 a whipping for example, someone must impose it. The person who inflicts it both brings the suffering and pain directly to the wrongdoer and observes close at hand its immediate effects - torn flesh, blood, screams of pain. In short, the punisher is bound to be desensitized and dehumanized by the experience of inflicting the whipping, as will the public who observe or are made aware of the punishment. The point is that state-sanctioned corporal punishment has a dehumanizing effect not only on the wrongdoer but 22 David Garland, \"Punishment and Sensibilities A Genealogy of'Civilized' Sanctions,\" Punishment and Modern Society (Chicago: the University of Chicago Press, 1990), p. 246. 1 2 6 also on the person who inflicts it and on the public in general. Therefore, in using such punishments, the state not only fails to respect the dignity of the wrongdoer but also fails to fulfill its responsibility to educate the public and to exhibit moral leadership.23 The question remains as to whether the penal measures that are routinely imposed on the criminal - loss of freedom and hard labor, for example - respect his dignity. There are two worries underlying this question. First, do these penal measures communicate with the wrongdoer in a human way, that is, through the capacities that constitute his moral agency? Or are they degrading like cruel punishments? And second, can the naturalistic aim of \"satisfying\" the victim's and community's need to express their retributive sentiments in this way be shown to respect the wrongdoer's dignity? In other words, can this aim be 2 3 This discussion of corporal punishment raises, the question of whether the death penalty is justified on the naturalistic approach. Adversaries in the debate over the death penalty agree that it cannot be not ruled out as a punishment for murder on the basis of proportionality alone. The question to be addressed, then, is whether it is a degrading punishment. The answer to that question is a matter of controversy. Ernest van den Haag, for example, contends that it is not. His opponent, Jeffrey H . Reiman, argues that it is. The debate between these two is lengthy and comprehensive but inconclusive. In my view, the naturalistic theory provides a new way of approaching the issue which would throw some light on it and lead to a resolution. The issue is complex, however, and would require a wide-ranging review of the literature and lengthy argumentation that is beyond the scope of this work. For a comprehensive discussion of both sides of the issue, see the following works. Ernest van den Haag, \"In Defense of the Death Penalty,\" Philosophy of Law, Fifth Edition, ed. Joel Feinberg and Hyman Gross (Belmont, Ca: Wadsworth Publishing Co., 1995), pp. 742-747. Stephen Nathanson, \"Should we Execute Those Who Deserve to Die?\" Philosophy of Law, Fifth Edition, ed. Joel Feinberg and Hyman Gross (Belmont, Ca: Wadsworth Publishing Co., 1995), pp. 748-756. Jeffrey H . Reiman, \"Justice, Civilization and the Death Penalty: Answering van den Haag,\" Philosophy and Public Affairs Reader, ed. A. John Simmons, Marshall Cohen, Joshua Cohen, Charles R. Beitz (Princeton, N . J : Princeton University Press, 1995), pp. 274-307. Stephen Nathanson, \"Does It Matter If the Death Penalty is Arbitrarily Administered?\" Philosophy and Public Affairs Reader, ed. A. John Simmons, Marshall Cohen, Joshua Cohen, Charles R. Beitz (Princeton, N . J : Princeton University Press, 1995), pp. 308-323. Ernest van den Haag, \"Refuting Reiman and Nathanson,\" Philosophy and Public Affairs Reader, ed. A . John Simmons, Marshall Cohen, Joshua Cohen, Charles R. Beitz (Princeton, N . J : Princeton University Press, 1995), pp. 324-335. 127 defended against the Kantian charge that the wrongdoer is being used merely as a means and not as an end in himself? Consider first the worry about whether loss of freedom and hard labor, the most severe punishments that the Canadian justice system imposes, communicate with the wrongdoer in a human way. Per se neither of these penal measures overwhelms the wrongdoer's capacities in the way that torture or mutilation does. Nonetheless, each has the potential to do so. Here I want to take a closer look at loss of freedom. The discussion of how hard labor could be degrading would follow along the same lines. In our justice system loss of freedom ranges from electronic monitoring while living and working in the community to incarceration in a maximum security prison for up to twenty five years. Certainly electronic monitoring, a system in which the offender's whereabouts in the community are carefully controlled by the authorities, does not disable his capacities. The rationale for this measure is that the offender can remain a productive member of the community and continue to participate in family life. In fact, this penal measure may give the offender the opportunity to exercise capacities which have been lying dormant -for example, in a case where he had turned to crime because of substance abuse. Incarceration in prison greatly restricts personal freedom but does not normally impair an offender's capacities. Offenders are customarily permitted access to media, books and various educational and counseling programs. They may even be encouraged to develop and exercise their capacities through the various 128 programs offered. Incarceration does, however, have the potential to be degrading depending on how it is administered. For example, in recent years in the United States extreme maximum security institutions called 'control unit' prisons have been built ostensibly to house the most violent prisoners. These prisons share so-called 'administrative' features aimed at controlling inmates. Prisoners are kept in solitary confinement in tiny (usually 6' by 8') cells for 22 to 23 hours a day. There are few or no opportunities for work. There is no congregate dining, exercise or religious services. These features are justified as administrative necessities, not as a form of punishment. It is likely, however, that these 'administrative necessities' are degrading. They are not short term disciplinary measures but the normal routine. An inmate sentenced for five, ten or twenty five years and sent to a control unit prison will endure these conditions on a daily basis perhaps for the entire sentence. The continual isolation and extraordinary confinement may well lead to ineffable psychological suffering that interferes with normal capacities. Dowker and Good, representatives of a committee to end control unit prisons, contend that the following results are probable, \"It is known that control unit conditions produce feelings of resentment, rage, and mental deterioration.... Prisoners will have been so deprived of human contact that it will be hard for them to cope with social situations again. The inhumanity of control units cannot reduce violence; it can only increase it.\"24 ^ 4 Fay Dowker and Glenn Good, \"The Proliferation of Control Unit Prisons in the United States,\" Journal of Prisoners on Prisons, Vol. 4, No. 2, 1993, p. 102. 129 Furthermore, even if these 'administrative' measures are deemed by some to be within the bounds of the humane, the attitudes towards inmates that they foster set the stage for obviously dehumanizing treatment. For example, Paul K. Delo, state executioner and Warden at a control unit state prison in Missouri, says of his duties as executioner, \"...its just like your own house. Nobody likes to take out the garbage but somebody has to....\"25 And at that prison the following 'disciplinary' procedure is used: \"...prison officials apply the 'double litter restraint' to recalcitrant prisoners. The prisoner's hands are cuffed behind his back; his ankles are cuffed; and he is forced to lie face-down on an army cot, his head turned to the side. A second cot is then tightly strapped upside-down over the prisoner and the ends are strapped shut, totally enclosing and immobilizing him.\"26 In short, although penal measures involving loss of freedom are not inherently degrading, the existence of control unit prisons shows that they can be. Safeguards must be incorporated into the administration of justice to ensure that sentences are not implemented in a degrading manner. Now consider the second worry. Do the punishments routinely imposed 'use' the wrongdoer merely as a means to satisfy the victim's and community's need to vent their emotions? This objection is closely related to the worry that any expression of retributive emotions is mere vengefulness. And, significantly, it points to the heart of the difference between an account like Duff s which focuses narrowly on the wrongdoer and the naturalistic account which claims that punishment has a plurality of aims, no one of which is the sole, exclusive consideration that justifies the social institution. 2 5 Ibid., p. 98. 2 6 Ibid. 130 The naturalistic view rejects the idea that punishment's justification should focus primarily on the wrongdoer's perspective. It takes into account the perspectives of all the stakeholders in the social institution. In other words, it attempts to balance the needs of the victim and community with the needs of the wrongdoer, keeping in mind that the wrongdoer has disregarded the moral demand. The critic's interpretation of the need to express retributive sentiments as merely \"using\" the wrongdoer misconceives the nature of the communication. It overlooks the fact that the victim's resentment and the community's moral indignation are reasonable reactions to deliberate harm done - moral sentiments based on reasons and justifiable. Burglary, assault, rape and murder are grievous harms, after all. It should come as no surprise that they arouse strong retributive sentiments which need to be expressed. The intensity of those sentiments can be communicated accurately only through strong responses. As noted, the nonverbal mode is well-suited to the task. It is a strong form of communication and is also, as Goleman indicates, a natural mode for communicating emotions. The critic might well concede that there is a need to communicate various strong messages to the wrongdoer and others and hold, nonetheless, that the punishments routinely imposed by our justice system are inappropriate ways to communicate. He would argue that the victim's and community's various 131 messages could be conveyed more humanely using mostly verbal forms of communication such as public blaming. When a crime is committed, however, the severity of the response is guided not only by normative moral principles such as the need to respect the wrongdoer's dignity but also by social norms. Would public blaming communicate the victim's and community's various messages? It, unlike punishment, is largely verbal. As the analysis earlier in the chapter indicates, mere verbal condemnation is unlikely to transmit the strong retributive sentiments that victim and community want to convey when faced with the grievous harm of a crime. Still, the critic could point out that public blaming would incorporate nonverbal communication through tone of voice, gestures and facial expressions. Nonetheless, most would agree that, in our social milieu, even this form of nonverbal communication is inadequate for expressing strong retributive sentiments. Blaming is a form of communication used to show disapproval for minor failings such as lying. It seems far too mild a response to crimes. The critic could argue, however, that a severe form of official public blaming could incorporate stronger nonverbal expressions of negative emotions than informal blaming does. In official public blaming, for example, the official charged with the task could use not only tone of voice, gestures and facial expressions but also ritualistic expressions \u00E2\u0080\u0094 perhaps burning or hanging in effigy \u00E2\u0080\u0094 to communicate condemnation. 132 Ritualistic public blaming might be an appropriate form of communication in a small, self-contained society where everyone knows everyone and there is no opportunity to leave. It would serve both as a strong expression of retributive sentiments - for some crimes anyway - and as motivation for the criminal to take up self-evaluation. In such a society everyone in the community would know what the criminal had done and would be likely to keep it in mind for a long time. In other words, the ritualistic blaming would not be the end of the criminal's 'sentence.' She would be expected to demonstrate on a daily basis over a long period of time that she was remorseful and had reformed.27 We must take account of our own context, however. In large, impersonal, modern societies such as ours condemnation through official public blaming would fail to convey the depth and intensity of the denunciation intended for crimes. It would be too short-lived and too narrow in the scope of its social impact to convey the victim's and community's persisting condemnation. It would barely create a ripple in the criminal's life. After the conviction was carried out, she could simply move to another neighborhood or city where she was unknown - her reputation, lifestyle and moral psychology unaffected. Furthermore, the victim and other audiences for the communication would be well aware that the community's official response to crimes had little substantive effect on the lives of criminals. That awareness would leave victims feeling devalued. It would leave the public ^ ' In Chapter 6 I discuss the use of an alternative mode of sentencing, the healing circle, which has been used in a number of small, isolated, culturally homogeneous, Canadian aboriginal communities. This alternative mode incorporates some of the elements of public blaming mentioned here. In the healing circle, the offender is confronted face-to-face by his victim and interested members of the community. Moreover, he is not incarcerated. He serves his sentence within the community usually doing some form of community service. 133 feeling vulnerable to crime. In short, it would undermine public confidence in the justice system. It is, in fact, difficult to imagine any society in which the appropriate sanction for crimes against the person - assault, rape and murder - would be mere ritualistic blaming. Such crimes have a long-term or permanent effect on the victims, their families and the community as a whole. In their aftermath, the stakeholders in the justice system need to and want to respond. They must stand up for the victim, disavow the criminal action, affirm the community's values and, where possible, attempt to bring about a change in the wrongdoer's moral psychology. Suppose, for instance, the wrongdoer is a serial murderer. It is simply naive to think she would take ritualistic public blaming seriously. It is the suffering and deprivation of long-term penal measures that communicate how the victims and the community feel about the crime. In short, given our human psychological make-up, mere public blaming is not a strong enough sanction to accomplish these various aims. Given the serious harm of the actions in a criminal wrongdoing, punishment is an appropriate communicative response in our social context. It does not merely 'use' the wrongdoer. Communication with the wrongdoer, where possible, is one of punishment's aims but not the only one. The naturalistic account can concede that some wrongdoers are beyond reach. As long as the wrongdoer is treated as a moral agent who must be addressed via his capacities for moral understanding, empathy and self-evaluation, his dignity is not violated. 134 In summary then, punishment satisfies the third and justifying feature in the paradigm for nonverbal communication. It is an appropriate expression of retributive sentiments because it satisfies both moral norms that ensure fairness and circumscribe the types of punishment allowed and social norms regarding appropriate expression. Punishment, appropriately administered, does not threaten the wrongdoer's dignity; it confirms it by showing that she is a member of the moral community who is open to moral assessment. Conclusion In this chapter I presented a two-part argument to show that punishment respects the wrongdoer's dignity. In the first part I argued that punishment should be understood in the context of a broadened notion of communication which takes into account the human capacity to communicate nonverbally. I put forward a provisional paradigm which judges nonverbal communication as appropriate or not based on moral and social norms. In the second part I defended a criterion of appropriateness for punishment based on two normative principles - the principle of proportionality and the principle against degrading punishments. I established that punishment respects the wrongdoer's dignity when it satisfies this criterion. In the next chapter I consider the question, raised here, of how the naturalistic account relates the degree of punishment to the severity of the crime. 135 Part II. Clarifications Chapters Wrongful Intent, Harm Done and the Sentence The Kantian notion that wrongful intent is crucial to determining the wrongdoer's desert is deeply imbedded in legal tradition.1 A harm that occurs accidentally, for example, is not punished at all even though it is similar to a harm that is intentionally done.2 Our justice system also takes harm done into account, however. The sentence for an attempted crime, for example, is less severe than that for a completed crime even though the wrongful intent is similar in both. Should the sentence be sensitive to the harm done as well as to the wrongdoer's intent? In this chapter I argue that it should. 1 Mewett and Manning discuss the concept of mens rea in detail from its origin in English common law to its inclusion as a \"general principle\" in the Canadian constitution. They mark the significance of the concept in this way, \"The term mens rea is used as a portmanteau description of the mental element that is a necessary part of most criminal offenses and it can be said that it is a basic principle of the criminal law of Canada that criminal liability requires the existence of a culpable state of mind.\" Alan W. Mewett and Morris Manning, Mewett and Manning on the Criminal Law (Toronto and Vancouver: Butterworths Canada Ltd., 1994), p. 169. 2 The exception is strict liability offenses. In Canada some statutes which are part of the criminal law do not require that the offender have wrongful intent. For the most part, such statutes \u00E2\u0080\u0094 often called \"public welfare statutes\" \u00E2\u0080\u0094 are in the area of regulatory criminal law which is distinguished from the so-called proper criminal law. The practice of classifying any criminal offense as a strict liability offense is a subject of controversy in legal theorizing. Recent case law in Canada supports the notion that wrongful intent is a necessary condition for criminal punishment. In Sault Ste. Marie, for example, the city of Sault Ste. Marie was charged with discharging materials that polluted the water supply. The incident was a strict liability offense but the Supreme Court held that the city was allowed a defence of \"due diligence.\" In other words, the city could avoid criminal liability by showing that it had taken all reasonable care under the circumstances. J . Dickson characterized all reasonable care as \"consideration of what a reasonable man would have done under the circumstances. This defence wi l l be available i f the accused reasonably believed in a mistaken set of facts which, i f true, would render the act or omission innocent, or if he took all reasonable steps to avoid the particular event.\" The test for \"due diligence\" marks off a degree of wrongful intent sufficient to shift the offense from being one of strict liability to being one that requires wrongful intent. Ibid., pp. 228-40. 136 I begin by considering traditional consequentialist and Kantian approaches to punishment claiming that neither provides good theoretical support for our commonsense sentencing practices. Consequentialist theories focus on preventing similar harms in future. Kantian accounts concentrate on giving the wrongdoer his just deserts. Each brings in reasons for punishment that are important, but neither explains fully why both wrongful intent and harm done should affect the sentence. The naturalistic approach furnishes the missing theoretical support. It explains how both the degree of wrongful intent and the gravity of the harm done affect the intensity of the victim's and community's retributive sentiments. In order to communicate them accurately, the sentence should reflect their intensity. I conclude that this inquiry into factors affecting our retributive sentiments supports the naturalistic claim that the wrongdoer is not the only audience for the communication of punishment. Problems for Traditional Theoretical Approaches in Explaining Sentencing In this section I analyze why neither traditional approach to punishment fully explains our sentencing practices. I then consider Adam Smith's lesser-known approach which I classify as naturalistic and retributive. Smith's account serves as an introduction to the next section. There I explicate my naturalistic approach to sentencing which, in contrast to Smith's, is pluralist. 137 Consequentialist approaches to punishment focus on its forward-looking benefits for society. The deterrence theorist, for example, argues that punishment is beneficial because it will specifically deter the wrongdoer and generally deter others who might be tempted to do harm. Even when the consequentialist adopts a goal of benefiting the wrongdoer through rehabilitation or reform, his justification incorporates and is grounded in the notion that such a change in the wrongdoer will benefit society. On what basis do consequentialists determine the sentence? In theory one can calculate the costs and benefits of various punishments by a common measure such as satisfaction or frustration of desires weighted according to the strength of the preferences they express. One can then choose punishments which provide the greatest benefit for society at the least cost. Deterrence theorists, for example, would determine a punishment to fit the crime by selecting one severe enough to deter both potential and actual offenders and yet lenient enough to maintain the best proportion of benefits over costs.3 Roughly speaking then, deterrence is accomplished by making the punishment just severe enough to prevent similar harms in future. 3 Another approach which takes a wholehearted consequentialist tack is that which would treat the wrongdoer for psychological illness - abandoning the assumptions of guilt and desert altogether. One criticism put forward against such an approach is that different wrongdoers would require differing amounts of treatment even for similar crimes. The proponent of treatment need not see this implication as problematic, however. Another key criticism is that the duration of criminals' sentences would be indefinite because no definite time could be set for the amount of treatment needed. See Herbert Morris, \"Persons and Punishment,\" Punishment and Rehabilitation, ed. Jeffrie G. Murphy (Belmont, California: Wadsworth Publishing Company, 1985), pp. 24-41. 138 The usual criticism of consequentialist approaches is that, for reasons of social utility, the wrongdoer's dignity might not be respected or innocents might be punished.4 In other words, Kantian worries about respecting the dignity and autonomy of the accused drive the debate. On the theoretical level, consequentialists address those worries by acknowledging that wrongdoers' and innocents' rights must be protected.5 On the practical level, that acknowledgment translates into prohibitions on cruel punishments and the adoption of procedural safeguards aimed at protecting anyone accused or convicted of a crime. There is, however, another criticism which can be leveled at consequentialist accounts. It is seldom mentioned because the traditional debate focuses narrowly on the tension between backward and forward-looking reasons for punishment. The victim's dignity may be sacrificed for reasons of social utility. To see this, it will be helpful to make a distinction between the harm done to society and that done to the victim. On the consequentialist view, crimes such as violence against prostitutes or homosexuals, for example, may be judged as having little harmful impact on society as a whole.6 The victims, who are often classified by society as offenders themselves, traditionally have not been valued. 4 McCloskey, \" A Non-Utilitarian Approach to Punishment,\" pp. 119-34. 5 Matti Hayry argues that these traditional objections, which take aim at utilitarian theories of punishment, lose most of their force when applied to a liberal form of utilitarianism as opposed to the classical form. See Matti Hayry, \"A Defence of the Utilitarian Theory of Punishment,\" Retributivism and Its Critics: papers of the special Nordic conference held at the University of Toronto, 25-27 June 1990/ Canadian Section of the International Society for Philosophy of Law and Social Philosophy, ed. Wesley Cragg (Stuttgart: Franz Steiner Verlag Stuttgart, 1992), pp. 129-147. 6 Fletcher discusses the fact that the justice system incorporates and perpetuates society's prejudices against certain groups such as gays and women. Fletcher, With Justice for Some, pp. 9-36, pp. 109-112. 139 As a result, the sentences for crimes against them may be less severe even though the impact of a crime is just as devastating for these victims as for other, more valued ones. In such cases, sentences reflect society's values and concerns and deprecate those of the victims. In other words, society may not stand up for the victim. It may be deemed beneficial to downplay or even ignore the harm done to the victim. In short then, consequentialist approaches to sentencing may be criticized for focusing on the community's needs and neglecting those of the wrongdoer and victim. Even modified consequentialist accounts, which acknowledge the rights of wrongdoers and innocents, fail to mention victims' rights. This is because the traditional debate has been narrow in scope concentrating on Kantian worries. Kantian approaches to sentencing place the wrongdoer at centre stage. She is said to deserve punishment because she is the author of her wrongful intent. Such approaches take their lead from Kant's conviction that the only thing that matters in our moral assessment of another person is the quality of her will.7 What she accomplishes through her will, the good or bad consequences, is morally irrelevant. Underlying this view is the condition of control \u00E2\u0080\u0094 the notion, articulated by Nagel, that Kant says, \"The good wil l is not good because of what it effects or accomplishes or because of its adequacy to achieve some proposed end; it is good only because of its willing, i . e., it is good of itself....Usefulness or foiitfulness can neither dimmish not augment this worth.\" Immanuel Kant, Foundations of the Metaphysics of Morals, trans. Lewis White Beck (Indianapolis: Bobbs-Merri l l Educational Publishing, 1959), s. 1, par. 3. 140 \"Without being able to explain exactly why, we feel that the appropriateness of moral assessment is easily undermined by the discovery that the act or attribute, no matter how good or bad, is not under the person's control.\"8 This condition has intuitive plausibility because clear cases in which we exempt an agent from moral assessment illustrate a lack of control - cases of involuntary movement, physical force and non-culpable ignorance, for example. Nagel notes, however, that what one does depends in many ways on factors not in one's control - factors such as the temperament and abilities with which one is born, the circumstances and opportunities life presents and the specific outcomes of one's action. And these external factors are not usually thought to exempt one from moral assessment. Nagel points out that Kant's approach to moral assessment is an attempt to address the fact that much of what one does depends on factors not in one's control. On the face of it, it appears that the condition of control is fulfilled if we morally assess a person only on the basis of her willing. And Kant, well aware that consequences are not completely in a person's control because they are subject to fortune, reduces the scope of moral assessment to what purportedly is in her control \u00E2\u0080\u0094 her willings. On an exclusively Kantian approach to punishment then, the wrongdoer would be punished based only on her wrongful intent. The idea is that she is not completely in control of the harm done and, therefore, her just deserts should not be based on it. The severity of the sentence would be geared to wrongful intent without consideration of the harm done or the potential for future harm. In 8 Thomas Nagel, \"Moral Luck,\" Free Will, ed. Gary Watson (New York: Oxford University Press, 141 straightforward cases, in which the criminal plans the crime, carries out the plan and succeeds, the harm done will be a reflection of her intent and the punishment will appear to take into account not only intent but also harm done. The problematic aspect of the Kantian approach becomes apparent, however, when a criminal with similar bad intent attempts a similar crime and fails. On the view that it is wrongful intent and only wrongful intent that matters, the sentence for an attempted crime should be the same as that for the completed crime. Yet in our justice system it is not. While there can be no question that the Kantian focus on intent \u00E2\u0080\u0094 the need for a culpable state of mind \u00E2\u0080\u0094 is a basic principle of our system, it also takes harm done into account. The Kantian could, of course, bite the bullet and argue that attempts should be punished the same as completed crimes. Most people, though, would find such a conclusion intuitively unsettling.9 This problem with intuitive plausibility gives us good reason to question the entire Kantian rationale. Nagel does so quite effectively. He points out the Kantian must explain how one can morally assess another on the basis of the stripped down acts of the will itself. He argues that the Kantian cannot account for control of acts of will because they too are products of antecedent circumstances outside of the will's control. He says, \"Everything seems to result from the combined influence of factors, antecedent and posterior to action, that are not 1982), p. 175. 9 Nagel too believes that Kant's attempt to restrict moral assessment to what is purportedly in the agent's control is intuitively unacceptable. He says, \"The mens rea which could have existed in the absence of any consequences does not exhaust the grounds of moral judgment. Actual results influence culpability or esteem in a large class of unquestionably ethical cases ranging from negligence through political choice....If one negligently leaves the bath running with the baby in it, one wi l l realize, as one bounds up the stairs toward the bathroom, that i f the baby has drowned one has done something awful, whereas if it has not one has merely been careless.\" Ibid., p. 179. 142 within the agent's control.\"10 Kant himself, of course, had an answer to Nagel's challenge. In his transcendental philosophy the moral agent is the genuine originator of his willings. Kant's defence of the will as an uncontrolled controller depends, however, on obscure metaphysical assumptions. The upshot is that the Kantian attempt to satisfy a strong condition of control by limiting moral assessment to an agent's intent fails. Duffs theory, discussed in Chapter 3, illustrates the difficulties that a Kantian approach to punishment encounters. Recall that Duffs expressivist theory takes the wrongdoer to be the only audience for the communication of punishment. The narrow scope of Duffs approach results in an essentialism about the aim of punishment which fails to account for the aims of the other stakeholders - the victim and the community. In short then, Kantian approaches to sentencing can be criticized for focusing exclusively on the wrongdoer. The reason they do stems from the underlying view of moral agency. It attempts to restrict moral assessment to an agent's intent by assuming a strong but unattainable condition of control. The two well-known approaches just discussed are the ones commonly cited in attempting to explain and justify our sentencing practices. The one focuses on protecting the community from future harm; the other concentrates on giving the wrongdoer his just deserts while protecting his autonomy. Neither explains fully why we take both wrongful intent and harm done into account in sentencing. At 1 0 Ibid., p. 183. 143 this point it is worth considering Adam Smith's lesser-known approach because it attempts to deal with both. Smith's theory is a naturalistic, retributive one that focuses on the victim's and community's retributive sentiments. According to Smith, those sentiments are affected not only by the quality of the wrongdoer's will but also by the harm done. Their intensity increases with the gravity of the harm done. In other words, Smith acknowledges that the consequences of actions make a difference to our moral assessment of others. He concedes that this fact leaves agents subject to fortune but, unlike the Kantian, makes no attempt to eliminate factors not in an agent's control. He recognizes the futility of trying artificially to narrow the scope of moral assessment solely to what is purportedly in an agent's control. He claims instead that the \"irregularity of sentiments\"11 that leads us to take harm done as well as wrongful intent into account when we morally assess ourselves and others is an aspect of human nature which has utility. He says it is a good thing that we do not experience retributive sentiments merely in reaction to another's intentions because if we did, \"Sentiments, thoughts, intentions, would become the objects of punishment; and if the indignation of mankind run as high against them as against actions; if the baseness of thought which had given birth to no action, seemed in the eyes of the world as much to call aloud for vengeance as the baseness of the action, every court of judicature would become a real inquisition. There would be no safety for the most innocent and circumspect conduct. Bad wishes, bad views, bad designs might still be suspected....\"12 Smith, Theory of the Moral Sentiments, II.iii.intro.6, p. 93. Ibid., II.iii.3.2, p. 105. 144 In contrast with Kantian approaches then, Smith's naturalistic account explicates practices such as punishment, not in terms of an agent's control of her willing, but rather in terms of moral sentiments that others experience in reaction to her intentional actions and the consequences resulting from them. Yet Smith's justification of punishment also stands in direct contrast to consequentialist accounts. Although he acknowledges that punishment may have utility, his justification of it is retributive. On his view, society's retributive practices require no justification beyond the fact that the victim and community experience retributive sentiments. The human desire for retaliation for harm done is fundamental, inevitable, and in need of no forward-looking justification.13 Human beings have no choice about experiencing retributive sentiments and no choice about expressing them, though they have some limited choice about how to express them through different types of punishments.14 Smith's acceptance of the role of retributive sentiments results in a theory that has the resources to be sensitive not only to wrongful intent but also to harm done. Nonetheless, the theory is flawed because it attempts to justify punishment based solely on the retributive reason that it is an expression of the Smith says, \"The horrors which are supposed to haunt the bed of the murderer, the ghosts which superstition imagines, rise from their graves to demand vengeance upon those who brought them to an untimely end, all take their origin from this natural sympathy with the imaginary resentment of the slain. And with regard, at least, to this most dreadful of all crimes, Nature, antecedent to all reflections upon the utility of punishment, has in this manner stamped upon the human heart, in the strongest and most indelible characters, an immediate and instinctive approbation of the sacred and necessary law of retaliation.\" Ibid., II.i.3.1, p. 71. 1 4 Paul Russell discusses another possible and stronger interpretation of Smith's position. He points out that Smith could be viewed as claiming that, as with the retributive sentiments, we also have no choice about our retributive practices. In other words, they are involuntary and 145 victim's and community's justified moral sentiments.15 Justifying these sentiments themselves, however, does not justify their expression through punishment. They could be expressed in ways which did not involve the deliberate infliction of suffering and deprivation on the wrongdoer. Smith says then-expression through punishment is justified based on the \"necessary law of retaliation.\"16 But this appeal to necessity does not justify the practice. It simply states that human beings have no choice but to punish. On the face of it, this claim is false. And if Smith actually believes we have no choice but to express our retributive sentiments through punishment, he owes us an argument for that claim. Moreover, if retaliation is the aim of punishment, Smith also owes us an explanation as to how his theory would deal with society's needs to respect the wrongdoer's dignity and to prepare for her future in the community. In sum then, Smith's exclusively retributive justification must be rejected. Nonetheless, his understanding that our moral assessment of others is based on the consequences of their actions as well as their intent is significant. It opens the way for us to take a naturalistic view of moral agency which acknowledges that human beings are subject to fortune. In other words, a moral agent's willing and the consequences that result from her actions are on a par in the sense that neither can satisfy the strong Kantian condition of control. Both are subject to factors outside the agent's control. In the sections that follow I make use of spontaneous, not a matter of choice at all. Russell concludes, however, that the weaker interpretation is the correct one. Russell, Freedom and Moral Sentiment, p. 147. 1 5 As noted, Smith suggests that the retributive sentiments have indirect utility. That is, punishment satisfies the victim's and community's desires for retribution. He does not justify punishment on that basis, however. 6 Smith, Theory of the Moral Sentiments, II.i.3.1, p. 71. 146 Smith's insight to develop an account of sentencing which, unlike Smith's, is not retributivist but pluralist. Wrongful Intent: A Necessary Condition for Punishment On the naturalistic approach, an agent is open to moral assessment in virtue of her capacities of moral understanding, self-evaluation and empathy, not in virtue of her control of her willing. Accordingly, this account has the resources to take a broader view of the agent acting in the world. Specifically, it can accommodate both wrongful intent and harm done in determining the sentence. Let us look first at wrongful intent. In keeping with our strong commonsense judgment, this naturalistic approach takes it to be a necessary condition for punishment. Why? The significance of wrongful intent lies in an analysis of what it means to disregard the moral demand. The moral demand requires some degree of goodwill in social relationships. In its criminal statutes society marks out conduct that does not meet even the minimum requirement for goodwill. Any moral agent who does not meet that requirement has criminal intent.17 Such blatant disregard of the moral demand is a necessary condition for criminal punishment because it is taken to reveal the quality of the agent's will. His disregard implies a failure to feel empathy, to apply moral reasoning and to evaluate his attitudes, feelings and goals with even minimum regard for others. Since he has the relevant capacities for moral agency, he is open to moral assessment for that attitude of disregard. The upshot of this view is that if - for 17 Beings who do harm but do not have the requisite capacities for moral agency - animals, children and the insane, for example - are not moral agents and not candidates for criminal punishment. 147 reasons such as accident, coercion or manipulation, for example \u00E2\u0080\u0094 an action or attempt which seems indicative of wrongful intent does not, in fact, flow from an attitude of disregard, the agent is not open to negative moral assessment for it. Clearly, wrongful intent need not be criminal. It may vary from indifference to outright malevolence. I have suggested - following Feinberg's characterization -that a crime is conduct against persons or property which produces serious harm or unreasonable risk of harm to others or conduct which causes harm to the public in general. Criminal intent, then, encompasses grievous wrongful intent which aims at doing robbery, assault, rape, murder or other actions such that harmful consequences will, in the ordinary course of events, probably or possibly occur. The \"other actions\" mentioned here are those classified as criminally negligent.18 It does not include wrongful intent which is expressed, for example, merely by an insulting remark or sneering glance. Society has no interest in imposing criminal punishment on persons whose bad will is expressed in minor ways which do not cause harm or unreasonable risk of harm to others. Why? This question leads us to consider again the principle of proportionality. Proportionality Again In the previous chapter I tentatively accepted the principle of proportionality on the basis of its intuitive appeal. In preparation for considering the role that harm 1 o One is criminally negligent and a candidate for punishment i f one acts such that harmful consequences wil l , in the ordinary course of events, probably or possibly occur. The wrongdoer's intent is constituted in an awareness of probabilities or possibilities, not in an actual purpose. In other words, the community expects moral agents to evaluate the probable impact of their proposed actions before they act. In Canada, the definition of criminal negligence is: \"Every one is criminally negligent who (a) in doing anything, or (b) in omitting to do anything it is his duty 148 done plays in determining sentences, I now want to look more closely at the relationship of wrongful intent and harm done to proportionality. Proportionality dictates that the most severe punishments be imposed for the most serious crimes. But it does not tell us how severe the punishments should be. I argued, therefore, for an upper limit on proportional punishments. That is, no matter how evil the crime is, some severe punishments are ruled out as responses. Degrading punishments such as torture and mutilation are ruled out because they fail to respect the wrongdoer's dignity. It will be helpful to think of severe punishments as being at the upper end of a continuum of negative responses to disregard that has blaming at the lower end. Beyond a certain point on the continuum the less severe responses are not criminal punishments. Those responses, which include minor punishments and blaming, are our usual reactions to expressions of minor disregard. The principle of proportionality on its own does not, however, rule out the possibility that responses at the upper end of the continuum could be imposed in reaction to minor disregard such as wrongful thoughts or non-harmful moral failings. Interpreted in the broadest sense, it simply says that the worse the conduct is the more severe the negative response should be. Still, just as some punishments are ruled out because they are degrading, some conduct should be ruled out of the domain of criminal punishment. Such conduct is of two types. One, as mentioned previously, is conduct which does not flow from an attitude of disregard but results in harm done anyway. It to do, shows wanton or reckless disregard for the lives or safety of others.\" (my emphasis) 149 encompasses harms done by accident with no negligence involved or as a result of coercion or manipulation. We do not punish or even blame people for harms that are unintentionally done. That is, wrongful intent is a necessary condition for punishing or blaming. The other type of conduct not in the domain of criminal punishment is that which flows from an attitude of minor disregard. In such cases, the wrongful intent may be expressed in a variety of disparaging ways that fall short of doing or attempting harm. But why not punish such conduct? To put the point succinctly, neither the target of the conduct nor the community have good reasons to inflict criminal punishment on people for less serious wrongful intent and minor moral failings. They are not motivated to punish because the retributive sentiments aroused are not intense. Roughly speaking, the intensity of the retributive sentiments varies in proportion to the disregard expressed by the wrongdoer. When the disregard is extreme, as in a violent crime, they are intense; when the disregard is minor, as in failing to keep a promise, they are correspondingly less intense. The community may be motivated to blame a person or even to impose minor punishments for minor moral failings. These less severe responses are appropriate given the degree of disregard expressed. They reflect and communicate the lower intensity of the retributive sentiments experienced in reaction.19 Mewett and Manning, Mewett and Manning on the Criminal Law, p. 197. 1 9 To put this point more generally, there is a gap between law and morality - between what is legally prohibited and what is considered morally wrong. The question of where the line should be drawn between law and morality is controversial. Feinberg poses the question in this way, \"What sorts of conduct may the state rightly make criminal?\" Feinberg, Harm to Others, p. 3. He presents a deep and thorough analysis of this and various related questions in the four-150 Why the Harm Done Should be Taken into Account in Sentencing Let us now consider why, in the case of criminal activity, the gravity of the harm done matters. And why, as I shall argue, it should affect society's response. Smith calls attention to the problem that concerns us regarding wrongful intent and the resultant consequences. He suggests that it is the \"intention or affection of the heart\" for which an agent must answer, not consequences, but immediately notes there is a problem with this analysis saying, \"But how well soever we may be persuaded of the truth of this equitable maxim, when we consider it after this manner, in abstract, yet when we come to particular cases, the actual consequences which happen to proceed from any action, have a very great effect upon our sentiments concerning its merit or demerit, and almost always either enhance or diminish our sense of both.\"20 There are two points to note. The first is that our justice system and Smith agree that intent matters in our assessment of the wrongdoer. As I just explained, wrongful intent is indicative of the quality of the wrongdoer's will. And it is through his will that disregard for others is expressed. The second point is that our justice system and Smith also agree that consequences matter. Smith explains why. They \"have a very great effect upon our sentiments concerning its [the action's]...demerit.\" Smith's observation raises several questions. How and why does the harm done increase the intensity of our retributive sentiments? When is that increase justified? And what is the role of reflection in damping down or intensifying those sentiments? To answer these questions, it will be helpful to look at some examples. volume work, The Moral Limits of the Criminal Law. Joel Feinberg, The Moral Limits of the Criminal Law, 4 vols. (New York: Oxford University Press, 1984-90). 2 0 Smith, Theory of the Moral Sentiments, II.iii.intro.3, p. 93. 151 Consider first an example in which there is harm done but no wrongful intent.21 I have claimed that such cases are not in the domain of criminal punishment. Yet, given Smith's analysis, we might think that the harm done would intensify our retributive sentiments to the point where we would want to punish the agent who did the harm. A truck driver en route to a delivery is driving carefully and within the speed limit. A child runs directly in front of the truck from between two parked cars. The driver has no time to stop or to avoid hitting the child who is killed in the accident. Initially the child's parents feel deep resentment towards the driver and want him punished. And some members of the community feel moral indignation. Once the facts of the incident are examined though and it becomes clear that the truck driver was not negligent, both the victim's family and the community reflect on the incident and realize that he was not at fault. There was nothing he could have done to avoid hitting the child. The driver may feel regret. The family will feel intense sadness at the loss and perhaps even guilt for not teaching the child to be more careful when playing. But the family's and community's unjustified retributive sentiments should dissipate. If they do not, they will be considered unreasonable and unjustified. In short, when the harm done is grievous but there is no wrongful intent, reflection on the facts should be successful in damping down unjustified retributive sentiments. Now consider an example which compares our reactions when there is wrongful intent with harm done and with no harm done. Two truck drivers whose duties 21 Judith Jarvis Thomson discusses similar examples in \"The Decline of Cause,\" Philosophy of Law, ed. Joel Feinberg and Hyman Gross (Belmont, Ca.: Wadsworth Publishing Co., 1995), pp. 512-19. 152 include maintenance and routine inspections of their trucks are reminded to check their brakes before setting out on deliveries. Neither driver bothers to check. Both trucks do, in fact, have faulty brakes. A child steps into a crosswalk in front of the first driver's truck. He cannot stop in time, hits the child and kills her. He would have been able to stop in time if the brakes had been in good working order. He is charged with criminal negligence, tried and punished. He acted such that harmful consequences would, in the ordinary course of events, possibly occur. His wrongful intent is constituted in an awareness of that possibility. The second driver reaches his destination without incident. When he returns to the truck depot, the supervisor, who has been told that the driver did not check his brakes, suspends him from work for a week. The two drivers had similar wrongful intent. At first glance it may seem that we should blame or punish them equally. Certainly that would be Kant's view. In fact, we do not. And it is the consequences of their actions that make the difference in our responses to them. In the first driver's case it is not just the wrongful intent but also the actual harm done, the child's death, that accounts for the family's and community's intense retributive sentiments. The family's intense resentment is justified. If the driver had checked his brakes, their child would not have been killed. They accurately evaluate him as uncaring and negligent about the safety of others. The community's moral indignation is also justified. The members of the community evaluate the driver as uncaring, sympathize with the victim's family and approve of their resentment. 153 In the case of the second driver, who has similar wrongful intent, there is no victim and, as a result, no response of resentment. Members of the community who are aware of the incident may experience moral indignation. But it would be much less intense than that experienced towards the first driver. What about the role of reflection though? After all, reflection was effective in damping down unjustified retributive sentiments in the case where there was harm done but no wrongful intent. On the face of it, it seems that reflection on the second driver's wrongful intent and on the consequences that could have resulted would intensify the community's moral indignation to the same level as that directed at the first driver. Reflection is unlikely to have this effect. Why? The difference between the first and second case is the impact on the victim and her family. In the case where the child is killed, the members of the community, in virtue of their capacity for empathy, experience more intense moral indignation. In both cases the cognitive component of the moral indignation is similar. It consists in the belief that the driver, in failing to check the brakes, had negligent disregard for the safety of others. But the affective component differs. In the first case it consists in both sympathy for the victim and a related need to get back at the driver on behalf of the victim. In the second it is considerably less intense. There may be a need to get back at the second driver for the sake of deterrence but sympathy for the victim is not a factor since there is no victim -simply a risk that there could have been one. Moreover, the evaluative component may be affected by this difference in intensity. In this example it is. The community's evaluation of the first truck driver is more severe because 154 actual harm was done. There are two points to consider here. Is that evaluation justified? Is it fair? The evaluation is justified because the first truck driver acted with disregard for the moral demand when he took the unsafe truck on the road. As a moral agent he had the capacity to understand that the unsafe truck had the potential to kill innocent people. He had the capacity to evaluate the possible or probable impact of his actions. In other words, he had the capacity to understand the risk he was taking with the lives of others and did not care enough to take the time to avoid it. The result was that he did kill a child. Although this evaluation is justified, it may seem unfair in comparison with the less severe evaluation of the second driver. Let us set aside the question of fairness for a moment and consider the community's evaluation of the second driver. To understand it, we must look to the role of reflection. Suppose members of the community were to reflect on the possibility that the second driver could have killed a child. Initially reflection about a hypothetical victim may be somewhat effective in increasing the intensity of the affective component. The effect of that reflection is likely to be short-lived, however. The reason is that reflection about a hypothetical victim and a hypothetical impact is unlikely to elicit sustained sympathy. It is clear that there is no permanent impact. Even a person with a vivid imagination is unlikely to conjure up lasting sympathy for hypothetical victims facing hypothetical grief, fear and financial burdens. Not only are the effects of imagination likely to be short-lived; they could also be easily undermined by contrary imaginary scenarios or reflection on the reality that no harm was done. 155 The upshot is that the affective component remains less intense and that fact affects the evaluative component. The community's evaluation of the second driver is less severe. Moreover, the milder evaluation is justified. Given that the two drivers had similar wrongful intent, are these evaluations fair? The suspicion that the evaluation of the first driver is unfair only arises on comparison with that of the second driver who did no harm. But we saw that the evaluation of the second driver is justified. Moreover, it strikes us as unfair to imagine consequences that might have occurred and evaluate him on that basis. This point raises a related question though. In the case of attempted crimes, we do punish the wrongdoer even though, as is in the case of the second truck driver, little or no harm is done. Why are these types of cases treated differently? Recall that wrongful intent comes in degrees. In the case of the negligent second truck driver, it is constituted in the fact that harmful consequences could possibly occur. In the case of attempted crimes, the wrongful intent is greater. It is constituted in the intent to commit a crime - murder, for example. In other words, the wrongful intent, in and of itself, is grievous enough that it arouses intense retributive sentiments. The strong communication of punishment is the appropriate way to communicate that intensity. It is worth pointing out that attempted crimes require an actus reus or guilty act as well as wrongful intent. The emphasis, however, is placed on the grievous wrongful intent.22 In short, we criminally punish attempted crimes even if little or no harm is done because of the gravity of the wrongful intent. Mewett and Manning, Mewett and Manning on the Criminal Law, p. 311. 156 To sum up the case of the two truck drivers, then, there is a significant difference between them. The stakeholders and audiences differ in the two cases. In the second case there is no victim as stakeholder or as audience. Consequently, the community has no need to communicate agreement with and sympathy for a victim. Let me summarize my naturalistic position on wrongful intent, harm done and the retributive sentiments. Punishment is, in part, a communication of the victim's and community's retributive sentiments. The intensity of those sentiments is affected by the degree of wrongful intent and the gravity of the harm done. Roughly speaking, it increases in proportion to the magnitude of the disregard and the seriousness of the harm. Wrongful intent is a necessary condition for the retributive sentiments to be justified because it reveals the quality of the wrongdoer's will. In the absence of wrongful intent, reflection should be effective in damping down unjustified retributive sentiments. The intensity of the retributive sentiments is also affected by the harm done because it has an impact on the victim and community. That impact has a direct affect on the victim's resentment. It also affects the affective component of the community's moral indignation and may affect the evaluative component. If there is no victim or the victim is unharmed, the affective component is naturally less intense and the evaluative component may be less severe. If the wrongful intent is grievous, however, the evaluation may be severe even when there is no victim. Reflection is likely to be ineffective in increasing the intensity 157 of the retributive sentiments when there is no harm done. In short, the effect of reflection is asymmetrical. It should be effective in damping down retributive sentiments which are unjustified; it is likely to be ineffective in increasing their intensity when there is no harm done. In this section I have explicated the relationship of wrongful intent and harm done to the intensity of the retributive sentiments. Their intensity has significance for the social institution of punishment because the sentence is, in part, a communication of those sentiments to various audiences. In order to communicate them accurately, the sentence should reflect their intensity in some way. Objections and Replies In this section I consider two challenges to the foregoing analysis of the affect of harm done on the intensity of the retributive sentiments. The defence against these objections brings to light some implications of this naturalistic view. The first challenge is an example suggested by Judith Jarvis Thomson which is, on the face of it, puzzling for any analysis of cases like that of the two truck drivers.23 Thomson's example is a variation on a civil suit in California, Summers v. Tice. Summers, the plaintiff, went hunting with the two defendants Tice and Simonson.24 A quail was flushed out and both Tice and Simonson allegedly fired in 2 3 Thomson, \"The Decline of Cause,\" pp. 518-19. 2 4 The fact that this is a civil suit and not a criminal case is not relevant to my analysis which is concerned with the question of how we would morally assess the parties. 158 Summers' direction. One of the two wounded Summers, but Summers was unable to prove which one because both had fired similar pellets from similar guns. Summers won the case. The court ruled joint and several liability against the defendants. Thomson notes that most people think this ruling fair. Thomson asks us to suppose, however, that during the trial conclusive evidence is presented to show that the shot that caused Summers' injury came from Tice's gun. At that point, the suit against Simonson would be dismissed. Nonetheless, Thomson claims that our moral assessment of Simonson would not shift. She finds that reaction puzzling in light of how we assess the second truck driver who was negligent but did not cause harm. She tentatively suggests two reasons for our differing responses. First, Simonson nearly caused the very same harm that Tice caused whereas the second truck driver did not nearly cause the very same harm as the first. But she says that if he had - if both truck drivers were driving side by side when the child was hit - we would view the case in the same way as Tice and Simonson. We would think no worse of what the one driver did than of what the other did. This claim leads to Thomson's second point. She says, \"what seems to come out\" is not merely that actual consequences affect our sense of demerit about an act but that the higher the risk of bad consequences the greater the demerit of the act. Thomson finds this result perplexing in light of the fact that Simonson's negligence is the same in both scenarios as is the second driver's. Thomson's analysis is, in effect, an objection to the naturalistic view that our retributive sentiments would be more intense towards the one who actually did the harm. 159 In reply I want to dispute Thomson's claim that our moral assessment of Simonson remains unchanged. Our assessment is not nearly as definitive as Thomson suggests. When we find out that Simonson did not injure Summers, what we really experience is ambivalence. And the naturalistic approach has the resources to explain why. Consider the scenario in which evidence is produced to prove that Simonson did not do the harm. Thomson correctly notes that we view this case differently from that of the second truck driver who is negligent but does no harm. Why is it that our moral indignation towards Simonson initially remains intense even when we learn that he did not do the harm? Both the second driver and Simonson are negligent in a way that could have caused harm to others but did not. The difference is that in Simonson's case actual harm was done though it was not done by Simonson. Nonetheless, there is a victim with whom the community can sympathize. And significantly, it is easy for them to imagine how Simonson might have been the one to harm that victim. As a result, initially anyway, the affective component of their moral indignation towards Simonson is as intense as that towards Tice. Is Thomson correct to assume, however, that the community's moral indignation towards Simonson would not decrease even after reflection? As I mentioned, what people really feel in this case is ambivalence. Reflection may well be effective in damping down their moral indignation. It is true that Simonson had negligent disregard similar to Tice's. The cognitive component of their moral indignation acknowledges that fact. The affective component consists in both sympathy for the actual victim, who - it is easy to imagine - could have been 160 Simonson's, and a desire to stand up for him. But when one reflects on the fact that it was Tice's bullet that caused the harm to the victim, it becomes difficult to maintain the intensity of one's moral indignation towards Simonson. To see this it is helpful to consider what happens to the victim's resentment when he learns that it was Tice and not Simonson that shot him. It is unlikely that Summers would continue to resent Simonson with the same intensity as Tice. His resentment would become focused on Tice, the person who caused his injury and subsequent suffering. Moreover, when the community reflects on the victim's response and the fact that it was Tice who caused the actual harm to that victim, their moral indignation would also become focused on Tice. I conclude that the intensity of their moral indignation towards Simonson would decrease. Consider another example which presents a challenge to this naturalistic analysis.25 Smith lures Jones to an isolated location. Once there, three other members of Smith's gang turn up. The gang members proceed to play a version of Russian roulette in which they take turns pointing the gun at Jones' head and pulling the trigger. Each of the gang members has had at least one turn when Smith eventually fires the shot that kills Jones. The four gang members are arrested, but the police cannot determine who fired the fatal shot. All four are charged with conspiracy to murder.26 The community's moral assessment of all This example was suggested by Paul Russell. Mewett and Manning point out that, as far as the provisions of the Criminal Code are concerned, a conspiracy to commit an offense is considered to be as serious as the offense itself. They note that the Code states, \"Every one who conspires with any one to commit an indictable offense...is guilty of an indictable offense and liable to the same punishment as that to which an accused who is guilty of that offense would, on conviction, be liable.\" Mewett and Manning, Mewett and Manning on the Criminal Law, p. 330. 161 four is the same. At the trial, though, it comes to light that it was Smith who fired the shot that killed Jones. The question is whether the fact that Smith actually did the harm would make a difference to our moral assessment of the other three. The fact that the other three gang members conspired with Smith to do the murder is significant. Those involved in a conspiracy to commit a crime have a common goal to commit the crime. In this case the goal was reached. Jones was murdered. Smith fired the fatal bullet, but any of the others could have. In short, all four acted as a group to reach a common goal. As a result, then, it is likely that our moral assessment of the other three would be the same as our assessment of Smith. It is important to understand why this example differs significantly from the previous one. In the case of Tice and Simonson, the wrongful intent was far less grievous. More importantly, however, there was no conspiracy. When the facts of that case came to light, it was clear that Simonson was not involved in the injury to Summers. In the case of the gang members, however, all four not only had grievous wrongful intent but also were intimately involved in carrying out the criminal act. A question for the naturalistic analysis is whether the fact that Smith actually fired the bullet that killed Jones would make a difference from the perspective of the victim's close family or friends? Some of the family of some victims might be more resentful towards Smith because he actually fired the fatal shot; others would consider that fact inconsequential. In other words, I want to suggest that 162 the answer is indeterminate in the sense that it would depend on the individuals involved. Would the fact that Smith actually did the harm make a difference from the community's perspective? As the foregoing analysis suggests, the members of the community would view all four gang members as taking part in committing the murder. These facts, in themselves, entail that the community's moral indignation would be intense towards all of them. In addition, the affective component of the community's moral indignation would reflect sympathy for the members of the victim's family and a desire to stand up for them. Given these facts, it likely that the community's moral indignation towards all four would be similar in intensity. Does this example show, then, that this naturalistic analysis is flawed? I have claimed that both the degree of wrongful intent and the harm done affect the intensity of the retributive sentiments. In this example, all four gang members had similar wrongful intent, but Smith fired the fatal shot. On the face of it, then, it may seem that the naturalistic analysis should claim that our retributive sentiments towards Smith would be more intense. I submit, however, that to come to this conclusion is to take a narrow view of who did the harm. In this example, all four gang members did the harm. All four conspired as a group to reach a common goal. All four had similar wrongful intent and all four participated in doing the harm. Therefore, the naturalistic analysis accords with our commonsense view that the intensity of our retributive sentiments towards all four would be similar. 163 The analyses of challenges to the naturalistic view show that the harm done does makes a difference to the intensity of our retributive sentiments. The victim does not react merely to the wrongdoer's attitude. He also reacts to what happens to him as a result of that attitude. In other words, the scope of the victim's moral assessment is not confined to some kernel of willing - of criminal intent - which is purported to be under the wrongdoer's complete control. The victim reacts to both wrongful intent and harm done, not because either is immune to fortune, but because both affect him in a hurtful and long-term way. As with the victim, the intensity of the community's reaction also increases with the gravity of the harm done. The members of the community are affected because they care about the impact of the crime not only on the victim and but also on society as a whole. And, like the victim, the scope of their assessment is not confined to a judgment about the quality of the wrongdoer's will. It takes into account the harm done for two reasons. First, it is likely that an actual harm will have a long-term, deleterious impact not just on the victim but also on the whole community. For instance, if health care is socialized, it is the community that must support any psychological counseling which the victim may need as a result of the crime. Second and more importantly, as the examples show, actual harm to an actual victim vividly brings home the horror of a crime in a way that imagined probable harm cannot. It is the harm done that triggers sympathy for the victim justifiably increasing the intensity of the community's moral indignation. 164 To sum up, the naturalistic approach takes a broadened view of moral assessment which acknowledges Smith's insight that \"the actual consequences which happen to proceed from any action, have a very great effect upon our sentiments concerning its merit or demerit.\" Wrongful intent is a necessary condition for punishment but harm done also matters. The reason is that the intensity of our retributive sentiments is affected by the harm done. Reflection has an important role to play in adjusting our sentiments to appropriate intensity and its effect is asymmetrical. It should be effective in damping down retributive sentiments when the agent has no or little wrongful intent. It is likely to be ineffective in intensifying them when the agent has done no or little harm. In other words, ceteris paribus, the intensity of our retributive sentiments naturally \"tracks\" the harm done to the victim and community. Since the sentence, in part, communicates the retributive sentiments, it should convey their intensity and convey it to the appropriate audience. In the next section I briefly consider the implications of this view for some of our existing practices. Implications for Sentencing The naturalistic theory supports our practice of less severe punishment for crimes in which the harm done is less grievous - others things being equal. It justifies punishing attempted crimes less severely than completed crimes. In cases of criminal negligence, it justifies taking degree of harm done as well as degree of wrongful intent into account in sentencing. 165 In cases where wrongful intent is similar, the difference in the intensity of our reactions is accounted for and justified by our sympathy for the victim who usually suffers the greatest impact from the harm done. On the theory's broadened view of communication, the sentence is a nonverbal expression of those intense retributive sentiments. It, along with the trial and conviction, is part of an overall communication with the various audiences involved in or concerned with the crime. If this nonverbal component is to be accurate and meaningful, it must reflect the intensity of the retributive sentiments. In other words, if they are very intense because the wrongdoer's crime resulted in grievous harm done as in violent, brutal rape, for example, the punishment should be more severe than for an attempted rape where far less harm has been done. I have suggested that this naturalistic approach provides the missing theoretical support for our commonsense sentencing practices, in particular, for the fact that we take harm done into account in sentencing. The analyses of the various cases in the previous section revealed the complexity of our reactions, however. The reality of that complexity raises a further question for the naturalistic approach. How can the social institution of punishment effectively communicate such complex reactions to its various audiences? The answer is that, while the theory does indeed support the existing practice of taking harm done into account, it points to the need to broaden the scope of the sentencing process. Given punishment's plurality of aims, the sentencing process can be and should be directed towards the various audiences. In other words, it should not be focused exclusively on the wrongdoer. Redressing the harm done, 166 for example, need not always translate into a longer sentence for the wrongdoer. The justice system could speak directly to the victim's needs. From the victim's perspective, punishment as communication will be most satisfying if it elicits a sympathetic response from the wrongdoer, the community or both. Here the theory opens the way for sentencing practices which incorporate restitution or compensation for the victim. Obviously, some crimes are such that the damage cannot be restored, but in many cases it can. Moreover, even when direct restoration is not possible, most victims would appreciate some tangible acknowledgment of their loss and, perhaps, the opportunity for psychological counseling. In contrast to the consequentialist focus on society and the Kantian focus on the wrongdoer, the naturalistic theory emphasizes the relationships among all those affected by the crime \u00E2\u0080\u0094 community, wrongdoer and victim. In doing so, it provides theoretical support for our commonsense sentencing practices. I have also suggested that the theory points to the need to broaden the scope of the sentencing process. In the next chapter I take up that suggestion arguing that it provides a good framework for explaining and evaluating various reforms in sentencing as well as in pretrial and trial proceedings. 167 Part III. An Application of the Theory Chapter 6 The Role of the Victim The naturalistic theory makes it clear that the victim is both a stakeholder in the social institution of punishment and an audience for the communication. As such, it provides a standpoint from which one can assess the victims' rights movement. Proponents of that movement contend that victims are not well served by the existing system. I have agreed with them, arguing that victims should be given a distinct role in the social institution. It is not just victims and their supporters who criticize the justice system, however. Advocates of prison reform point out that sentencing practices, which rely heavily on incarceration, are a dismal failure. The majority of offenders return to the community hardened by the experience and likely to offend again. In recent times, the justice system has responded to criticisms from both groups by introducing several reforms, many of them controversial. In this chapter I take a critical look at representative reforms \u00E2\u0080\u0094 some proposed, others already implemented. The analysis provides an opportunity to test the naturalistic theory's explanatory value by applying it to actual and proposed practices. In contrast with traditional theories, this theory affords a useful framework for explaining and evaluating these reforms. This application of the theory illustrates its superior explanatory value. 168 The Attack on Traditional Sentencing Practices Traditionally, victims of crimes have been all but disregarded by liberal democracies with roots in the British system. Accordingly, in 1980 William Clifford, Director of the Australian Institute of Criminology, stated: \"The victims of crime are, indeed, the forgotten tribe of our criminal justice system. It is quite remarkable that, for so long, the interests of victims (except as witnesses for the prosecution) have been given so little public attention.\"1 More recently, however, the victims' rights movement has been somewhat successful in calling attention to victims' concerns. The justice system now permits victims to submit impact statements in some cases. It has also experimented with various types of alternative sentencing. Those alternatives usually incorporate restitution for the victim. At the same time the system has had to face the rather obvious fact that incarceration on its own often fails to secure any lasting benefit for offenders or society. Proponents of alternative sentencing claim, however, that the new modes of sentencing are helpful not only to victims but also to offenders. They point out that experimental programs using these alternatives have been mostly successful in reintegrating offenders into the community. Spurred, apparently, by the success of the various practical initiatives, some contemporary philosophers have begun to argue for reforms either in pretrial and trial proceedings or in sentencing. What is curious about their arguments, 1 William Clifford, \"Foreword,\" Proceedings of a Seminar on Victims of Crimes, Institute of Criminology Sidney University Law School, 1980, p. 9. 169 however, is that they are advanced in the absence of adequate theoretical support. Jeffrie Murphy, for example, takes up the case for victims' rights arguing for the need to consider \"the legitimacy of hatred and desires for revenge as operative values in a system of criminal law.\"2 Yet, on the face of it at least, this position conflicts with his avowed Kantian views about punishment. For there can be no doubt that Kant would have opposed the inclusion of any emotions in the practice of punishment. George P. Fletcher, an advocate for reform of the American criminal justice system, also argues for changes designed to give the victim a role in the social institution of puriishment. He says, Solidarity with victims represents a new way of thinking about criminal punishment. Thus an alternative way emerges in the centuries-old debate between those who advocate deterrence of future offenders and those who yearn for retribution by making the punishment fit past crimes. Each of these traditional views has something to offer but none adequately accounts for punishment in a time when deterrence seems not to work and the promise of abstract retribution rings hollow. The imperative of punishing the guilty springs not from our personal duties to high ideals but from our relationship with the humbled victim in our midst.3 Here Fletcher points out that the traditional theoretical framework lacks the resources to explain the victim's role. His own argument for giving the victim a role is, however, based on practical considerations. Specifically, Fletcher focuses on those victims of crimes who are also victims of historical discrimination by society - \"Blacks, Jews, Gays and women.\"4 He 2 Murphy, \"The Role of the Victim,\" p. 685. 3 Fletcher, With Justice for Some, pp. 6-7. 4 Ibid., p. 6. 170 contends that the fact that these minority groups have been discriminated against as victims of crimes shows that there are grave defects in the way the justice system treats all victims. He analyzes a number of recent high profile cases in which the victim is a member of one of these groups. He says, \"On the whole, these are cases of failure, cases where the system has failed to hear the victim.\"5 He contends that these victims have been unjustly dealt with because the legal system is \"rent by structural defects.\"6 He argues for a number of practical reforms designed to give all victims, not just those who belong to minority groups, a voice. Fletcher's arguments are convincing. Moreover, the reforms he proposes are promising. I consider them in the sections that follow. What Fletcher does not do, however, is analyze the existing theoretical framework in which punishment is justified. As the long quotation above indicates, he suggests the framework is inadequate but has nothing more to say about it. In other words, he analyzes the symptoms of the problem, not the underlying cause - a theoretical framework that is fixated on backward and forward-looking reasons for punishment and the seemingly intractable tension between them. Taking a different tack, Wesley Cragg argues for reform on behalf of offenders.7 On Cragg's view, the aim of the social institution of punishment should be conflict resolution. To that end, he does attempt to develop a theoretical justification for 5 Ibid., p. 7. 6 Ibid., p. 5. 7 Wesley Cragg, The Practice of Punishment: Towards a theory of restorative justice (New York, N Y : Routledge, 1992). 171 his \"restorative\" approach. As Conrad Brunk points out, however, Cragg's account \"is very close to the 'hybrid account' offered by H. L. A. Hart whose view he criticizes strongly.\" According to Brunk, Cragg's view, like Hart's, is ...a hybrid of utilitarian and retributivist views. He too, holds that the primary objective of punishment is the forward-looking aim of maintaining the system of law as a formal structure of conflict resolution. The account he gives of the three primary functions of punishment as demonstrative, persuasive and enabling explains them all as ways of maintaining respect for, and compliance with, the system of law....Cragg also insists on certain retributive principles most importantly the principle that only responsible moral agents who have voluntarily disobeyed the law should be punished.8 Brunk notes that Cragg fails not only to distinguish his position from Hart's but also to develop adequately his notion of restorative justice. Cragg shies away from exploring the retributive aspects of restorative justice and the related sentencing possibilities, in particular, the prospects for addressing the victim's concerns through innovative sentencing. Brunk suggests that Cragg avoids the retributive element because he believes it requires unrealistic assumptions about the moral consensus underlying the law in modern pluralistic societies.9 These theorists and others are on the right track in calling for reforms to give victims a voice or to improve offenders' chances for successful reintegration. They are hampered, however, by a traditional framework which focuses on and is consumed by the intractable debate between retributivists and consequentialists. What they need is a fresh approach - a framework which can accommodate both the retributive sentiments of victims and the need to get through to those \u00C2\u00B0 Conrad G. Brunk, \"Restorative Justice and Punishment,\" Dialogue 35 (1996), p. 596. 9 Ibid., p. 597. 1 7 2 offenders capable of reform. The naturalistic theory has the explanatory resources to deal with these contemporary concerns. As we have seen, it brings to light two elements that are central to an understanding of punishment. The first is that punishment is an expression and attempted communication of retributive sentiments, among other things. And the second is that punishment's plurality of aims is best explained as a type of communication from the stakeholders in the institution to its various audiences. In this framework with these elements in hand, the theorist has the means to evaluate or even to invent alternative approaches to trial proceedings and sentencing. In the next section I first review briefly my arguments for the need to give the victim a role. I then discuss some proposed reforms in pre-trial and trial proceedings and some actual reforms in sentencing. These reforms either are proposed or have been implemented for good pragmatic reasons. They have not yet been given an adequate philosophical justification. My analysis in the next section and the following one shows that the naturalistic theory gives us the resources to explain and evaluate these reforms. Including The Victim's Perspective in the Social Institution of Punishment In Chapter 2 I argued that the victim should be given a distinct role in the social institution of punishment. Recall that, in contrast with traditional accounts \u00E2\u0080\u0094 even those that centre on punishment's expressive function, the naturalistic theory explains why the victim's perspective should be included. Specifically, the theory acknowledges the significance of the retributive emotions to the 173 justification of punishment. It distinguishes the victim's resentment, a moral sentiment susceptible to justification, from mere vengeance. It makes it clear that, when the wrongdoer commits the crime, it is not just his relationship with society as a whole that is affected. The victim, too, has a substantial and justifiable stake in what happens to him; Yet traditionally the victim has no distinct institutionalized role to play. For the most part, he is included only if the prosecution brings him in as a witness.10 In cases where there is no trial because the offender pleads guilty, the victim may be unaware of the sentence imposed or, for that matter, of whether the offender was even apprehended. The onus is on the victim to find out. For example, Mark Wright, a proponent of victim/offender mediation, notes that \"Victims often would like to see what their offender looks like, to express their feelings to him or her, and to ask questions: 'Why did you pick me? Can any of what you stole be recovered?\"11 In other words, victims have personal needs that are not met either at the trial or in traditional sentencing. Meeting them would be a way to recognize and perhaps alleviate their suffering. To reiterate, two further reasons for including the victim's perspective were also discussed in Chapter 2. The victim's resentment and the reasons for it may provide important information about the crime. And giving the victim a role i U As already noted, the victim has been given a somewhat expanded role in recent years. The use of Victim Impact Statements is one example - albeit a controversial one. H Mark Wright, \"Introduction,\" Mediation and Criminal Justice: Victims, Offenders and Community, ed. Mark Wright and Burt Galaway (Newbury Park, Ca: SAGE Publications, 1989), p. 1. 174 within the social institution provides society with an opportunity to influence him to adjust his resentment to an appropriate level. Following on from that argument, I now want to consider what type of role the victim might be given. I analyze two proposals designed to give the victim a voice \u00E2\u0080\u0094 one in the pre-trial and trial proceedings, the other in sentencing. In the course of the discussion it becomes clear, surprisingly perhaps, that giving the victim a voice may also be beneficial for those offenders capable of reform. I shall argue, in particular, that giving the victim a distinct role in sentencing through victim/offender mediation enhances communication with the offender. Consider first Fletcher's proposal that the victim play a part in pre-trial and trial proceedings. Fletcher, in fact, opposes giving the victim a role in sentencing because he believes that would open the way to exercising vengeance. I discuss his objection in the next section where I deal with objections to each type of proposal. What specifically does Fletcher suggest? With respect to pre-trial proceedings, he argues that the victim should be given the right to accept or reject any plea-bargain. Plea-bargaining is a common practice in which the prosecution agrees to a lesser charge in return for the defendant's cooperation and waiver of his right to trial. The practice effectively puts the case in the control of the police and prosecutor. It shuts the victim out entirely, disregarding his concerns and insulting his dignity. Fletcher describes the current situation in this way, The presumption should be that the prosecution will enforce the law as written and as interpreted in the courts. Any deviation from this standard should meet with the approval of all parties involved - the court, the 175 prosecution, the defendant, and the victim. Of all these parties, only the victim is treated as below the threshold of procedural empowerment - as someone whose objections carry no legal weight.12 At first glance it seems the practice of plea-bargaining should be dropped altogether. Fletcher points out, however, that the system would grind to a halt if all those who now plead guilty as a result of plea-bargains went to trial instead. He argues, therefore, for revising the process by giving the victim the right of rejection. Fletcher believes that his model for plea-bargaining may also serve the aims of restorative justice by leading to reconciliation. He notes that prosecutors would have an incentive to try to facilitate understanding between the victim and offender. He surmises that victims would be unlikely to exercise vengefulness when approaching the offender from the position of strength that his model gives them.13 From the standpoint of the naturalistic theory, Fletcher's model for pre-trial proceedings is practical and fair in the sense that it involves all stakeholders, that is, the community - represented by the prosecutor, the offender and the victim. Moreover, it communicates with all three audiences. In particular, it asserts the victim's value by giving him a powerful role. With respect to trial proceedings, Fletcher looks to the model used in some European countries, Germany for example. According to the German Code of Criminal Procedure, the victim of a crime of the person or property may join the proceedings as a private prosecutor. As such, he has the opportunity both to try 1 2 Fletcher, With Justice for Some, p. 193. 1 3 Ibid. , p. 248. 176 to gain reparation from the accused and to further the criminal prosecution. Overall the German system differs from ours, however, in that the judge has greater control over the proceedings. She reads the dossier before the trial, calls the witnesses, filters the questions asked and keeps the official record of the trial. Fletcher sums up these differences saying, \"[The judge] is definitely at the center of a wheel of justice with many spokes. There is no conceptual problem in allocating one of those spokes to the victim.\"14 Our justice system, however, is \"bipolar.\" Fletcher characterizes it \"not so much a wheel as a seesaw operated by two lawyers.\"15 The prosecutor represents the public. The defence attorney represents the accused. The victim may be included as a witness for the prosecution if it suits the prosecutor, but that decision is not up to the victim.16 Given this bipolar system, it would be problematic to introduce the victim as another prosecutor - \"like having two pitchers throwing possible strikes at the batter.\"17 In particular, one of the strong and important guarantees of this system is the right of the accused to confront and question the witnesses against him. Permitting the victim to act as 1 4 Ibid., p. 195. 1 5 Ibid. 1 6 Fletcher discusses the case of Rodney King, a victim of police brutality against Blacks. He notes that King did not appear at the Simi Valley state trial against the four Los Angeles police officers accused of the beating. Because King had a criminal record, the prosecutor decided that the risk of cross-examination outweighed the advantage of the jury hearing from him. According to Fletcher, this was a clear way of saying that it was not King's case at all. \"He was merely a conduit of proof that the prosecution, the People, could use or not use as they saw fit.\" Ibid., pp. 193-4. At that first trial the accused were acquitted. In the subsequent federal trial King testified on his own behalf. Fletcher sums up that decision to have him speak in this way, \"This man [King] could be beaten but not silenced. King's victory at trial was not only seeing two of his tormentors convicted but gaining the opportunity to be heard. He emerged from the trial as a man who could be trusted, as a citizen capable of telling the truth.\" Ibid., p. 67. 1 7 Ibid., p. 195. 177 co-prosecutor rather than as a witness for the prosecution leaves it unclear how cross-examination of the victim would take place. Fletcher suggests that the problem could be avoided if the victim is allowed, acting alone or through an attorney, merely to ask questions of the prosecution and defense witnesses. He concludes that in an adversarial system such as ours, in which procedural safeguards are needed, this is the most that could be done. With Fletcher's proposal for trial proceedings, the victim would have a distinct role and some communication with the accused. He would not merely be a witness for the prosecution but also have an opportunity to find out what he wanted to know about the crime. Moreover, if it turns out that the accused is guilty, she will have been confronted by the victim at the trial. The naturalistic approach would support the proposal to give the victim a role in the trial because it is part of the overall communication with the various audiences. The question is whether permitting the victim merely to ask questions does enough to address his concerns. I discuss this issue in the next section. One weakness of Fletcher's proposals is their limited scope. They do not address cases in which the offender pleads guilty and that plea is not the result of a plea-bargain. In such cases there is no trial and, with traditional sentencing models, no way for the victim to confront his offender. It is probably for this reason, in part anyway, that the justice system's innovations have been in sentencing rather than at the pre-trial and trial stage. The advantage to giving the victim a voice in sentencing is that this process applies to all offenders. 178 Consider now this second type of proposal. Innovations in sentencing include victim/offender mediation, offender/community restitution, sentencing circles and offender/community adoption programs. These alternatives have several common features. They include all the stakeholders. They involve personal communication. They require the wrongdoer to be accountable for the harm done and to redress it in some way. They provide the wrongdoer with personalized support from the community during the period of reintegration.18 I use the example of victim/offender mediation in this discussion because it best illustrates these common features.19 Proponents of victim/offender mediation have a quite different view from Fletcher as to the role the victim should play in sentencing. Cragg, for example, points out that the justice system has traditionally blocked dialogue between victim and offender by interposing itself between them. The state imposes the sentence and pronounces the matter resolved effectively disregarding \"the attitudes of those who have been adversely affected by the offending behaviour.\"20 With victim/offender mediation, however, 'resolution' of the 'conflict' is achieved through direct communication between the victim and offender with facilitation by a trained mediator who represents the community. Most commonly, mediation has 1 8 Baker points out these five common features. Here I paraphrase from her account. Brenda M . Baker, \"Improving our Practice of Sentencing,\" Utilitas, Vol. 9, No. 1, March, 1997, p. 105. Baker follows Cragg in emphasizing that various alternative sentences are forms of restorative justice which benefit the offender and community. Nonetheless, both Baker and Cragg also mention potential benefits for the victim. 1 9 This innovation has been introduced through various pilot projects. As such, it is not an established procedure in the social institution of punishment. 20 Cragg, The Practice of Punishment, p. 173. 179 been used with offenders and victims of non-violent crimes but not always.21 The mediation is a mode of sentencing and, in some cases - usually first offenses, may completely determine the sentence. In other cases, the offender, who is already serving a sentence, is selected by justice authorities as a good candidate and asked if he is willing to participate. In other words, the process and its outcomes may become part of the sentence. How does it work? Mark Chupp sums up the three basic aspects of mediation as \"facts, feelings and restitution.\"22 The trained mediator first meets separately with each party to offer support and encouragement on behalf of the community. According to Chupp, the victim and the offender each in his own way feels alienated from the community.23 In the separate meetings the community's representative attempts to break the cycle of mistrust by being non-judgmental, validating the trauma each party has experienced and extending support. The tenor of these initial meetings also helps the mediator to determine the best approach for fostering communication at the joint meeting. In the face-to-face meeting first the victim and then the offender tells his story. Each participant is encouraged \"to state factual information and express their past and present emotions.\"24 (my emphasis) At the end of the storytelling, the ^ Mark S. Umbreit chronicles several cases of violent crimes in which victim/offender mediation was used and found helpful for both victims and offenders. Mark S. Umbreit, Chapter 7, \"Violent Offenders and their Victims,\" Mediation and Criminal Justice: Victims, Offenders and Community, ed. Mark Wright and Burt Galaway (Newbury Park, Ca: SAGE Publications, 1989), pp. 99-112. 2 2 Chupp, \"Reconciliation Procedures and Rationale,\" p. 61. 2 3 Ibid. 2 4 Ibid. 180 victim should understand the offender's motivation for the crime, the impact of the arrest and court proceedings and the offender's response to the victim's story. The offender should understand the various impacts the crime has had on the victim such as physical loss, resentment, fear, anxiety, mistrust and secondary victimization by the system. He should also understand the victim's response to his story. Reconciliation is not forced on the participants but Chupp reports that often the offender does apologize to the victim even if he entered the meeting vowing not to. When this occurs, many victims in turn accept the apology or offer forgiveness. After the factual and emotional part of the meeting is finished and any reconciliation has taken place, the mediator shifts attention to coming up with an agreement for redressing the harm done. The offender is encouraged to offer tangible restitution based on the victim's earlier description of the impact of the crime. This part of the meeting involves negotiation, however. The victim may say what he would like and the offender may respond according to his willingness and ability. The mediator does not intervene unless there is a stalemate. Once the parties agree, the mediator writes the agreement as a contract with an expected completion date. The contract may include any combination of financial payments, work for the victim or community, or even an agreement about future behavior. 181 Consider an actual example.25 Gerald Zuk was attacked in a washroom and kicked in the head repeatedly by his attacker Richard Bosch. Bosch stole $300 and broke Zuk's nose, jaw and most of the bones on one side of his face. Bosch was found guilty and jailed but that was not enough for Zuk. He felt Bosch should not be permitted to keep the stolen money. Zuk, a self-confessed right-winger, managed to arrange a meeting with Bosch in prison without the benefit of a mediator. At the meeting, he \"poured out his vituperation and scorn.\" Bosch agreed to pay him back in installments but defaulted after the first payment. Zuk proceeded to track down Bosch, who was in prison again by the time Zuk found him. At that point both agreed to participate in victim/offender mediation. They met twice with a mediator present. Bosch had always pleaded guilty to his crimes so had never met any of his victims other than the persistent Zuk. Commenting on the mediation, Bosch said, \"I got to see how I affected him. I had never really thought of it. That's why this program has kind of woken me up to how I affect other people.\" He said he not only gained insight about how his actions affect others but also felt better after explaining to Zuk about his own past. He said meeting his victim had made him want to \"jump into programs.\" Zuk is not ready to shake Bosch's hand but said he was saddened when he watched a videotape of how he humiliated Bosch at that first, unmediated meeting. \"Revenge isn't what its cracked up to be,\" he said.26 2 ^ This story is summarized from an article in The Globe and Mad. Sean Fine, \"Victims, offenders, and healing,\" Toronto Globe and Mail, December 23, 1993. 26 Zuk's reaction to mediation is somewhat circumspect compared to the other victim that Fine describes in his article. After hearing her offender's story at mediation, Diane, a rape victim, has forgiven him and says, \"I would go to great lengths to help him. Not because of any wild 182 Empirical evidence from various pilot projects shows that, in selected cases and with proper preparation, the actual face-to-face communication of victim/offender mediation is beneficial to the victim, the offender and the community.27 Chupp points out a key benefit for the victim saying \"the expression of emotions encourages people to direct their feelings to the cause rather than lashing out at society or others working in the criminal justice system.\"28 Another obvious benefit is the resultant agreement for reparation of the victim by the offender. In addition, the face-to-face meeting tends to break down stereotypes. The victim sees the offender as a person rather than a monster to be feared and hated. This realization sets the stage for integrating the offender back into the community. For his part, the offender gains an appreciation of the impact of his crime on the victim. Clearly, all these benefits are good for the community as well. What is the aim of victim/offender mediation? The primary aim is reconciliation through personal communication between the victim and offender. The process is designed to empower the victim, offender and community to resolve the conflict fantasies. I just think that the more I help him, the better I feel about myself and what happened.\" 2 7 Launay and Murray discuss the benefits of victim/offender groups in which victims meet, not their own offenders, but offenders convicted of similar crimes. Like mediation, the meetings were beneficial with one exception. Launay and Murray say, \"It soon became clear that meetings involving hardened professional criminals...were unlikely to benefit victims or offenders. The victims understandably were not reassured in any way by meeting professional burglars, who were untypical since most burglars are young people. The professional burglars in turn seemed little affected by accounts of the victims' suffering.\" Gilles Launay and Peter Murray, Chapter 8, \"Victim/Offender Groups,\" Mediation and Criminal Justice: Victims, Offenders and Community, ed. Mark Wright and Burt Galaway (Newbury park, Ca: SAGE Publications, 1989), p. 122. 2 8 Chupp, \"Reconciliation Procedures and Rationale,\" p. 63. 183 created by the crime and its aftermath through their active participation. From the perspective of the offender, this active role is strikingly different from the passive one she is assigned in traditional sentencing. In this process she is encouraged to apologize and redress the harm. From the perspective of the victim, an apology acknowledges the harm done and affirms his dignity. That acknowledgment is authenticated by the mutually agreed upon restitution. Chupp notes, for example, that when restitution involves the offender working for the victim, the relationship between them is usually strengthened as the offender completes the task. From the perspective of the community, this type of innovation is beneficial because it both satisfies the victim's needs and assists in reintegrating the offender. In short then, reconciliation is the primary aim of victim/offender mediation but restitution for the victim and rehabilitation of the offender go hand-in-hand with it. It is striking how well this innovation fits within the naturalistic framework. Victim/offender mediation comprises a face-to-face meeting of stakeholders which focuses on communication of facts and feelings. The naturalistic theory says that punishment is a type of communication, in part nonverbal, from the stakeholders to various audiences. Moreover, it emphasizes that punishment communicates the victim's and community's retributive sentiments, among other things. Victim/offender mediation stresses the need for the offender to apologize and redress the harm done to the victim or community. The naturalistic theory says that harm done should be taken into account in sentencing because of its impact on the victim and community. Victim/offender mediation incorporates a plurality of aims. Reconciliation, which involves 184 communication of the victim's retributive sentiments and the offender's acknowledgment of them, is one. Restitution for the victim and reintegration of the offender are others. The naturalistic theory says that punishment has a plurality of aims. In summary then, the two innovations discussed have been put forward in response to criticisms of the justice system by victims' rights groups on the one hand and advocates for prison reform on the other. The first gives the victim a voice prior to sentencing; the second involves the victim and offender in sentencing. Both proposals differ from the traditional process in that they personally involve the victim, offender or both. These innovations fit well with the naturalistic approach which has the resources to explain and justify them. Objections and Replies In this section I raise some important objections to the two proposals described in the previous section and reply to them in the naturalistic context. Consider first Fletcher's proposal for a change in trial proceedings. He holds that the victim or her attorney should be allowed to ask questions of the witnesses. This innovation gives the victim a distinct role. She would at least have an opportunity to find out what she wants to know about the crime. This proposal is a step in the right direction but it does not go far enough. With Fletcher's modification of the German model, the victim would not necessarily have a chance to tell her story. Only if the prosecution called her as a witness 185 would she have that opportunity. It is important, however, that the victim be given the opportunity to express her views. As noted, her perspective on the crime may furnish important information about its impact that would not otherwise come to light. On the one hand, then, Fletcher rejects the German model because it is incompatible with our adversarial system. On the other hand, he also rejects the current practice which usually permits the victim to submit an impact statement at the time of sentencing. That practice could be made a right, after all. If it were, the victim would always have the opportunity to speak on her own behalf. Yet Fletcher is adamantly opposed to giving the victim a voice at the sentencing hearing. He contends that permitting the victim or her family to submit unsworn testimony at sentencing is likely to promote expressions of vengeance and distort judgment at precisely the wrong time - when the punishment is about to be determined.29 Let us consider each of Fletcher's worries in turn. His worry about importing procedures from the German to the adversarial system without modification is well-founded. It raises the question, however, as to why Fletcher takes the adversarial approach as a 'given.' As I noted earlier on, Fletcher criticizes the traditional theoretical framework in which punishment is justified but offers no alternative. He then goes on to argue for reforms based on practical considerations alone. It may be that proposing such reforms in the context of the 29 Commenting on sentencing hearings in the United States, Fletcher notes, \"... the unsworn testimony of literally everybody is admissible at a sentencing hearing. The ordinary rules of 186 existing theoretical framework constrains his reflection on what might be done. In the context of the naturalistic justification, one need not take the adversarial system as a given. The theory leaves it open to call for reform of the adversarial system itself based on the need to address the aims of the various stakeholders in the social institution. Fletcher's second worry is also well-founded. When the victim's input is submitted at the time of sentencing in the form of a victim impact statement, it is not open to public scrutiny in the same way as it would be at the trial. There is no procedural control on the victim's statement in the sense that it is unsworn testimony. There is no cross-examination and usually no jury to judge the appropriateness of the victim's reaction given the particular circumstances of the crime. To see this point, consider an example. Fletcher describes a sentencing hearing in the state of California for murderer, Richard Allen Davis. At the hearing Marc Klaas, whose daughter was murdered by Davis, accused Davis of molesting his daughter and raged that Davis should be condemned to Hilter's circle in hell. Klaas then lunged towards Davis and had to be taken outside. According to Fletcher, the prosecution deliberately exploited Klaas's anger in order to persuade the jury to recommend the death penalty for Mr. Davis.30 evidence are suspended...In reaching a decision on the proper sentence, the judge allows everyone in the community to express their opinion....\" Fletcher, With Justice for Some, p. 198. 3 0 George P. Fletcher, \"Victims Rights and Wrongs,\" The New York Times, OP-ED, Sunday, September 29, 1996. 187 Nonetheless, the question arises as to why Fletcher does not even consider changes to procedures at the sentencing hearing designed both to give the victim a role and to safeguard the rights of the offender. Procedures could be changed to require that the victim's impact statement be a sworn statement which would be open to cross-examination by the offender's lawyer. Such a modification would give the victim a voice and provide some procedural safeguards for the offender. Placing procedural controls on the victim's testimony in the traditional sentencing hearing is just one possible way of giving the victim a role at the time of sentencing. Let us turn now to alternative modes of sentencing. Fletcher's objection to giving the victim a role at the sentencing hearing gives us a hint as to the various objections that might be raised regarding alternative sentencing. Before discussing some objections, it is worth pointing out that innovations such as victim/offender mediation, sentencing circles and offender/community restitution programs do not fall neatly into the categories established by the traditional model, that is - trial, sentencing hearing, sentence. These innovations are an amalgam - both a mode of sentencing and an alternative type of sentence. Victim/offender mediation, for example, may be a way to determine the sentence either totally or in part. In addition, however, the process of mediation itself may be viewed as part of the sentence. The offender is confronted by the victim and forced to face up to the impact of his crime. He must take an active role in the process. He is expected to cooperate with the victim on an agreement to redress the harm done. 188 There are at least three objections that critics of alternative sentencing would raise. First, it allows the personal, variable and often vengeful nature of victims' reactions to affect the sentence. Second, the individualized sentences that result from it violate the norm of fairness which says that similar crimes should receive similar punishments. Third, alternative types of sentences are not punishment at all but alternatives to punishment. Consider the first objection. Fletcher criticizes giving the victim a voice in sentencing because, in his words, \"...informal testimonials by the angry and aggrieved could generate excessive sentences serving primarily the need for revenge.\"31 This objection, though, is aimed at traditional sentencing hearings. It targets the fact that they allow unsworn testimony about the impact of the crime by the victim or her next of kin. As discussed above, I agree with Fletcher that the permitting the victim to give unsworn testimony at the sentencing hearing is problematic. Alternative sentencing differs markedly from traditional sentencing, however. It does not permit the victim free rein to exercise vengefulness. To see this point, consider again the example of victim/offender mediation. In this proceeding the roles of all three players are well-defined. In particular, the mediator is charged with guiding the participants towards resolution of the conflict by promoting communication. Moreover, empirical evidence indicates that victims choose to participate, not in order to exercise revenge, but rather to recover their loss, to 3 1 Ibid. 189 help the offender and to convey to him that his crime hurt others and that he should be held accountable.32 We have seen that the traditional sentencing hearing dispenses with some of the procedural protections of the trial leaving it open for the victim to indulge in vengefulness. In victim/offender mediation, by contrast, the mediator focuses on eliciting communication that is collaborative rather than adversarial. She keeps the complementary goals of reconciliation, reparation for the victim and reintegration of the offender at the forefront. The victim is encouraged to tell his story and express his feelings, but the non-adversarial atmosphere discourages inappropriate reactions. Moreover, the opportunity for the victim to communicate his resentment in a personal and appropriate manner may be therapeutic opening him to the possibility of reconciliation. Furthermore, the communication is two-way. The offender has the opportunity to tell his story and express his feelings. As the story of Zuk and Bosch illustrates, face-to-face, personal communication is more likely to tap the offender's capacity for empathy than the impersonal communication of traditional sentencing. It has a better chance of getting through to the offender thereby motivating him to take up self-evaluation. As such, the process can be therapeutic for both parties. In fact, it is worth noting that this and other types of alternative sentencing tend to blur the line between punishment and \"therapy\" which traditional theories have emphasized. This fact may be viewed as a strength because it is likely to benefit 3 2 Robert B. Coates and John Gehm, \"An Empirical Assessment,\" p. 252. 1 9 0 the various audiences for the communication.33 The idea that alternative modes of sentencing may serve as a form of therapy for both victim and offender is best illustrated by another alternative mode, the sentencing circle. \"Sentencing circle\"34 or ''healing circle\" is the name given to a community-based approach to justice which has been implemented in a growing number of aboriginal communities in Canada. The approach attempts to deal with endemic problems that permeate such communities as a result of historical injustices. In order to understand the aims of this alternative mode, one must first appreciate the circumstances of the aboriginal communities in which it has been implemented. Yet it is difficult for mainstream society to even begin to comprehend the social conditions in such communities and the historical reasons for them. Rupert Ross sums up the historical injustices in this way, As she [an Ojibway woman who was taken to a residential school from ages six to sixteen] spoke, it became clear that residential schools were not the solitary cause of social breakdown amongst Aboriginal people. Rather, they were the closing punctuation mark in a loud, long declaration saying that nothing Aboriginal could possibly be of value to anyone. That message had been delivered in almost every way imaginable, and had touched every aspect of traditional social organization. Nothing was exempt, whether it was spiritual beliefs and practices, child-raising techniques, pharmacology, psychology, dispute resolution, decision making, clan organization or community governance. In time, even economic independence was stripped away as governments built community schools, which made it impossible for families to tend traplines often a hundred kilometres back in the bush. Even the law added its voice to the degradation, making it illegal to possess medicine bundles, vote in Canadian elections, hold a potlatch to honour assistance of others or (difficult as this is to believe) hire a lawyer to even ask a court to d d See Morris's objections to the \"therapy\" model of punishment. Morris, \"Persons and Punishment.\" 3 4 The process is referred to as a \"sentencing circle\" because the actual court proceeding takes place in two circles, one within the other. Those who wish to observe sit in the outer circle. Those who wish to speak sit in the inner one. 1 9 1 force governments to honour their treaty obligations.35 Aboriginal communities which suffered such thoroughgoing degradation are the heirs to debilitating social problems such as widespread sexual abuse. For example, in Hollow Water, a village of 600 people on the eastern shore of Lake Winnipeg, those implementing the healing circles estimate that 80% of the population, both female and male, have been victims of sexual abuse and that 50% of the population, again both male and female, have sexually abused someone else. In the face of the admitted failure of the justice system to deal with the problems, these communities turned to traditional aboriginal teachings to meet the challenge. The healing circles in Hollow Water provide a good example of how the process works. The process involves a team of both lay people from the community and professionals who work with the victim, offender and others affected by the crime. The team's involvement begins with the initial disclosure of the abuse and moves through protecting the victim, confronting the victimizer, assisting the non-offending spouse and family, assisting the victimizer to admit and accept responsibility, creating a healing contract (sentence), providing community-based support to the offender in carrying out the contract and, finally, to holding a cleansing ceremony to mark the completion of the contract. If possible, the healing contract is agreed upon before the actual sentencing circle at court takes place. The process bears similarities to victim/offender mediation but 3^ Rupert Ross, Returning to the Teachings: Exploring Aboriginal Justice (Toronto: Penguin Books Canada Ltd, 1996), pp. 46-7. Ross has worked as an Assistant Crown Attorney in Aboriginal communities. Before writing Returning to the Teachings he undertook a three-year secondment with Justice Canada examining and experiencing aboriginal modes of justice firsthand. 192 also includes traditional native ways of dealing with conflict such as the use of the Tiealing' circle where the victim, offender and others affected sit in a circle with the team and, in turn, share their stories and feelings about the abuse. The aim of this process is to end the cycle of abuse by giving the victim and community an opportunity to express their retributive sentiments while, at the same time, offering support to the offender. The idea is that the offender cannot understand what he did until he appreciates the impact of his crime on the victim and community. The victim's and community's communication of their retributive sentiments in the healing circle is a key component in bringing him to understand that impact. Ross characterizes the process as one in which the offender learns firsthand in a feeling way how people were affected by the crime. Ross notes that ...every effort is dedicated to putting offenders through processes where they cannot stay distant from the harm they have caused. They are not permitted to hide as they can by simply going to jail. Instead, everything is aimed at making them actually feel some portion of the pain, grief, outrage, sorrow or other emotions that they have caused in others.36 There can be no doubt that this process is a form of therapy for the victim, offender and community. 'Healing' is explicitly identified as its aim. The sentence is actually called a healing contract. It is important to note, however, that proponents of this approach are well aware that healing cannot take place until the victim's and community's retributive sentiments are expressed, acknowledged and dealt with in some way. Ross remarks, \"The word 'healing5 3 6 Ibid., p. 168. 193 seems such a soft word, but...Hollow Water's healing process is anything but soft. In fact, jail is a much easier alternative, because it does not require the victimizer to face the real truth about abuse.\"37 In other words, the emphasis on healing or therapy is in no way intended to absolve the offender from responsibility or dispense with the notions of guilt and desert. Rather, the process aims to make him accountable. Accordingly, Ross says, \"In the vision of Hollow Water, it is entirely natural that victims speak in angry voices and push for angry responses to their abusers. This is not a surprise, for abuse causes anger.\"38 In the healing circle, though, the victim is assisted by others in expressing his resentment in an appropriate way. He is encouraged to speak about the impact of the crime on him and his family and discouraged from making judgmental comments about the offender. Ross concludes that \"...careful heart speaking, with its nonjudgmental disclosure of feelings, no matter how intense, is ultimately irresistible to the vast majority of offenders.\"39 While this type of approach to crime is suitable for specific applications such as the one described, there is no suggestion that it ought to be applied outside aboriginal communities or for all crimes within those communities. Communities such as Hollow Water typically are small, isolated, culturally homogeneous and in crisis. The justice system had admittedly failed to solve their problems or even to make progress. As a result, such communities have reached agreements with the justice system to implement sentencing circles for less serious crimes and for 3 7 Ibid., p. 37. 3 8 Ibid., p. 177. 3 9 Ibid., p. 170. 194 some serious crimes such as sexual abuse which are at the heart of the social breakdown. Since the introduction of sentencing circles, progress has been striking. In Hollow Water over the past nine years only five offenders out of forty eight have been incarcerated. The five failed to participate adequately in the healing program. Out of the remaining forty three only two have re-offended. One re-offended at an early stage before sentencing had taken place. The other re-offended when the program was in its infancy. Since then he has completed the healing program and is now a productive member of the community.40 This alternative mode is an example of how the communication of punishment may be therapeutic for all the audiences. The example shows that even when the victim's and community's retributive sentiments are extremely intense, as they tend to be in cases of sexual abuse, they can be expressed in a way that is not vengeful. In short then, alternative modes of sentencing are far less likely to elicit inappropriate reactions from victims than traditional sentencing hearings because they are non-adversarial and personal in nature. They promote two-way communication rather than confrontation or mere venting of emotions. These aspects of alternative sentencing fit well with the naturalistic approach which emphasizes appropriate communication of the moral demand. 4 0 Ibid., p. 36. 195 The second objection to alternative sentencing is that the individualized sentences which result may violate the norm of fairness - that is, same crime, same sentence \u00E2\u0080\u0094 that traditional sentencing has, to some extent, observed. Alternative sentencing is flexible, accommodating sentences that involve restitution, reparation, custodial treatment, supervised parole etc. It does result in individualized sentences. This result should not be considered problematic, however. It is a strength of alternative sentences that they do not focus solely on the offender. They include the other audiences. Furthermore, this result is not a violation of the norm of fairness. It is simply an indication that the assessment of fairness must be made over a wider domain \u00E2\u0080\u0094 one which includes all the audiences for the communication. An individualized sentence is, in fact, likely to communicate more effectively with the various audiences for the communication because it is sensitive to the concerns and needs of those most directly affected by the crime. Such sentences usually take harm done into account in a tangible way, providing reparation to the victim or community. They speak to the offender by confronting him with the impact of his crime on his victim making him directly and personally accountable. In addition, they benefit the community because they assist in reintegrating the offender and may compensate the victim. The question that remains to be addressed, then, is whether alternative sentences adhere to the naturalistic principle that more serious crimes should 196 receive more serious punishments. As long as alternative sentencing is confined to sentencing for less serious crimes - burglary and other non-violent offenses, particularly first offenses - there is little worry that the proportionality principle will be violated. Even in traditional sentencing, the judge has considerable flexibility in determining sentences for these less serious crimes. In alternative sentencing that flexibility is turned over to the stakeholders. Since they collaborate to determine a sentence that is satisfactory to all of them, there is little likelihood that it will be grossly out of proportion to the crime. There can be no question, however, that the use of alternative sentencing for more serious crimes, particularly crimes involving violence, raises difficulties. Determining an alternative punishment that is severe enough for a serious crime is problematic, as is the need to maintain public safety. For example, if the offender is convicted of a violent assault, anything less severe than a term in prison would be unacceptable to many members of the community. What alternative would be appropriate? Would supervised parole and mandatory attendance at group therapy be acceptable? Neither the victim nor community would agree that it was. Given our norms of appropriateness, none of the usual alternatives to prison would be severe enough. In the case of more serious crimes then, a judge has the expertise needed to come up with appropriate terms in prison. Nonetheless, traditional sentences for more serious crimes could be combined with alternatives such as victim/offender mediation. Incarceration would communicate the greater condemnation intended 197 and would protect the public.41 Mediation could still be used to answer the victim's needs but only if the victim wanted to participate. In such cases mediation may be considered successful even if no agreement for reparation is reached because the victim has the opportunity to express personally his resentment and to hear the offender's story. With regard to reintegrating offenders, other alternatives could be introduced to educate those offenders capable of reform in the course of their incarceration. In sum then, individualized sentences are fair. They address all the audiences for punishment, not just the offender. Moreover, the worry that such sentences cannot be made severe enough to fulfill the naturalistic proportionality principle can be met in two ways. The first way is to use alternative sentencing, on its own, only for less serious crimes. The second is to use traditional sentencing for more serious crimes but to gain some of the advantages of alternative sentencing by combining certain of its features with incarceration. The third objection to alternative sentencing asserts that it is not punishment at all but an alternative to punishment. In fact, some proponents of alternative 4 1 At first glance, this conclusion appears to rule out using sentencing circles in cases of grievous sexual abuse. As noted however, aboriginal communities differ markedly from mainstream society. They are isolated and homogeneous. Everyone in the community knows the offender and knows what his healing contract requires. In these circumstances, the non-custodial sentence does convey the required and appropriate condemnation. As Ross notes, \"the very public nature of the sentencing circles has created another way to satisfy the law's demand that there be a strong public denunciation of such offenses [sexual abuse]: standing before two hundred of your fellows, each of whom has a chance to speak to you and all of whom are going to hear any promises you make and demand that you keep them is a more effective denunciation than sitting in a jail cell, perhaps hundreds of miles away.\" Ibid., p. 215. 198 sentencing argue that it is not punishment.42 They claim it is a type of restorative justice which aims at reconciliation and reintegration. Are alternative sentences punishment? One way to approach the question is to determine whether such sentences satisfy the defining features of punishment. What are those features? On the naturalistic approach, criminal punishment is physical or psychological suffering or deprivation intentionally imposed on an offender for an offense that is a crime. The punishment is intended to express condemnation, among other things. Do alternative sentences have these features? An alternative sentence is imposed on an offender for a crime. Alternatives that result in restitution for the victim or service to the community are coercive, usually require the offender to work without being paid and restrict her liberty. Furthermore, these alternatives are nonverbal expressions of denunciation. They are intended to confront the offender with the impact of her crime and impel her to redress the harm done. Proponents of alternative sentencing concede that such sentences involve coercion but downplay that aspect. Significantly, they also tend to disregard or deny the condemnation involved, focusing instead on the aim of reconciliation. The point is, however, that reconciliation is unlikely to take place unless the victim's retributive needs are satisfied. Alternative sentencing attempts to do just that in two ways. First, the victim is afforded the opportunity personally to Baker, \"Improving our Practice of Sentencing,\" pp. 113-4. 199 express his justified resentment to the offender. Second, the harm done is 'repaired' by the offender in some tangible way. Both of these features involve denunciation of the criminal and the crime. I argued above that alternative sentencing works well for less serious crimes but not for more serious ones. The reason is that the denunciation of an alternative sentence is mild compared with that conveyed by a term in prison. For more serious crimes no alternative to prison seems severe enough to express the strong denunciation that our social context demands.43 I conclude that alternative sentences are a form of punishment because they do satisfy its defining features. They are a type of suffering or deprivation, which expresses condemnation - among other things, intentionally imposed on the offender for a crime. Conclusion In summary then, the innovations discussed in this chapter have been proposed in response to criticisms of the justice system by proponents of victims' rights on the one hand and advocates for offenders on the other. I have discussed two types of proposals - changes before sentencing at the pre-trial and trial stage and alternative sentencing. Both types provide a way for the victim to communicate his retributive sentiments. Alternative sentencing also affords an opportunity for the victim to obtain reparation for the harm done. In addition, it incorporates features that assist the community and offender with reintegration. In other 4 3 As I pointed out, the social context of aboriginal communities differs from that of mainstream society. 200 words, the proposals discussed have the potential to benefit all the audiences for the communication of punishment. The analysis of these proposals highlights the explanatory advantages of the naturalistic theory. The foundation in naturalism provides the resources to take into account all the capacities of those affected by the crime, emotional as well as cognitive capacities. The theory taps those additional resources in explaining and justifying the innovative practices proposed. In clarifying that punishment is, among other things, the communication of strong but justified retributive sentiments, it helps us to understand why and how the victim should be included in the process of achieving justice. Specifically, the theory gives us an understanding of why practices such as victim/offender mediation, with its focus on communication of facts and feelings, addresses the victim's needs. Finally, the theory's theme of pluralism about aims comes to the fore in the assessment of our actual practices. This theme accords with our commonsense view that the social institution is a complex one involving various stakeholders who have differing motives and needs. In assessing the two types of proposals, the theory has the advantage of not being bound by a narrowness that focuses on one of the audiences to the detriment of the others. Rather, it emphasizes the need for communication with all those affected by the crime - wrongdoer, victim and community. 201 Conclusion In this final chapter I review the overall methodology of this study in order to make clear the relationship of the various aspects of the theory to its two basic themes - naturalism and pluralism about aims. This new approach to punishment was motivated by a recognition that contemporary theorists working in the traditional framework for justifying punishment are caught in an intractable debate. Those theorists, in clinging unyieldingly to their conflicting intuitions, have neither explained fully various important reasons for punishment nor integrated them into one theory. More specifically, they have failed to resolve the tension between retributivist and consequentialist reasons for punishment. Coincident with the stalemate in the theoretical debate, justice systems in liberal democracies have been faced with criticisms about punishment from both victims' rights groups and advocates for offenders. They have responded with various practical reforms in sentencing centred on promoting communication among the those affected by a crime. This emphasis on communication in the practical domain supports claims made by expressivist theorists that punishment has an expressive function. Spurred, then, by the theoretical impasse and by the success of practical initiatives focused on communication, this study has taken heed of the insight that punishment is an expressive and communicative endeavour. This approach contrasts sharply, however, with contemporary expressivist approaches in that 202 it rejects the traditional framework and adopts a naturalistic one. The naturalism of this study, which is one of its two fundamental themes, consists in the fact that it takes an empirically informed descriptive approach to the problem of justifying punishment. The development of an expressivist line in the context of naturalism provides the theory with the resources to advance a deeper and far more subtle expressivist view. To begin with, this naturalistic analysis has revealed that moral sentiments play a significant role in the practice of punishment \u00E2\u0080\u0094 a role that has been all but ignored in contemporary justifications of punishment. One reason for the oversight may be that, overall, the role of emotions in moral theorizing has been downplayed in the first half of the twentieth century. Yet the study of human emotional capacities offers an abundance of relatively untapped explanatory resources. This approach then, because it is naturalistic, is well-placed to take advantage of various historical and contemporary insights concerning the role of emotions in the practice of punishment. The theory builds on Smith's and Strawson's claims that punishment is, in part, an expression of negative emotional attitudes. It pays heed to Damasio's discussion of the importance of emotions to rationality and Goleman's findings regarding the need for emotional intelligence in social living. These theorists reject the idea, dominant in philosophical thought, that reason and emotion are opposed and that rationality requires the suppression of emotions. They argue, on the contrary, that rationality requires appropriate 203 emotions and that rational action may involve the expression of appropriate emotions. This theory's recognition and acknowledgment of the role of retributive sentiments has significant implications. It brings to light that a justification of the social institution of punishment must take account of the moral psychologies of all those involved in the practice - victim, community and wrongdoer. That finding makes it clear that there may be various perspectives on a crime and its aftermath which need to be addressed. The resultant justification of the social institution, then, must avoid exclusive focus on just one of those perspectives. It must account both for the need to express retributive sentiments and for the other concerns of all those involved. In other words, regard for the diverse perspectives of the various stakeholders in the social institution generates a plurality of aims. All of them are important; no one of them is the sole, exclusive consideration that justifies the social institution. Pluralism, however, does not rule out the fact that, in a particular instance of punishment, communication of some or one of the aims may be sufficient to justify the particular instance. In short, pluralism about aims is the second fundamental theme of this justification. In order to develop an expressivist line in a way that accounts for pluralism about aims, this justification shifted the emphasis from the notion of expression to that of communication. Interpreting punishment as a form of communication has two important advantages. First, it brings to the fore the idea that punishment communicates various messages from the stakeholders in the social institution to a variety of audiences - not merely wrongdoer but also victim and community 204 at large. Second, it provides the opportunity to emphasize that the communication with punishment's various audiences is intentional. The shift in emphasis from expression to communication is particularly important for this naturalistic approach because of the role that retributive sentiments play. An expression of emotion may be done for its own sake - as a way of venting the emotion. In the case of the social institution, however, the punishment is intended to do more than merely express the victim's and community's retributive emotions \u00E2\u0080\u0094 among other things. The communication is intended to have meaning and that meaning is intended to be apparent to the victim, the community and the wrongdoer. This naturalistic theory, because it takes into account the different perspectives of those actually involved in the practice, has practical implications. This result is especially significant at a time when the existing social institution faces criticism from apparently opposing sides. Victims' rights groups and a broad segment of the public claim that the justice system is too lenient towards offenders. Offenders' advocacy groups and their supporters point out that the system is failing miserably in its stated aim of reforming and reintegrating offenders. The naturalistic approach not only confirms the need for certain types of reforms but also points to the direction they should take. The theory says punishment is a type of communication from those affected by a crime to distinct and varied audiences. The arguments put forward for this view have emphasized in a general way the need to design penal measures that 205 promote communication among wrongdoer, victim and community. More specifically, however, they have underlined the fact that the victim has been disregarded both in traditional theories of punishment and, to a great extent, in the social institution itself. By bringing to light and explaining the victim's need to express his justified resentment, the theory calls for reforms designed to address the victim's concerns. Implementing pretrial and trial proceedings that give the victim a role and establishing alternative sentencing methods, which both involve the victim and offer restitution, are measures which should be adopted. The call to give the victim a role does not rule out attention to offenders' needs, however. The theory's pluralism about aims permits enough flexibility to address concerns raised by offenders' advocates. Reforms which incorporate a variety of communicative measures may be helpful to those offenders who are capable of reform. In particular, alternative sentences which require an offender's active participation may both assist in reintegration and make him accountable for his crime. Nonetheless, pluralism about aims leaves it open for the justice system to be sensitive in sentencing to the fact that different offenders have differing capacities as well as differing needs. If an offender is incapable of reform, a serial murderer for example, the system need not emphasize the admittedly hopeless aim of reform. It could and should focus instead on satisfying the victim's and community's retributive sentiments and their concerns about public safety. In addition, the naturalistic theory points to a number of areas worthy of further philosophical investigation. This study has focused on explication of the victim's moral psychology because it is the victim who has been neglected by traditional 206 approaches. Further reflection on the moral psychologies of the other audiences for punishment is sure to improve our understanding of the types of reforms likely to be successful. In fact, the whole question of what reforms should be made in light of the naturalistic theory is one rich with issues that begin with the theory and spill over into areas of concern in applied ethics. For example, the fact that the naturalistic theory concedes that some offenders simply are not capable of reform raises a number of difficult questions. How can we determine fairly which offenders are capable of reform? How should we treat those offenders who are incapable of reform? Of course, these problems are not unique to the naturalistic theory. This account's pluralism about aims places them front and centre, however, by admitting outright that some offenders are not capable of reform. The question of whether the death penalty is ever justified is another highly controversial issue in both theoretical and applied ethics. The naturalistic theory, by opening the discussion to consideration of the fact that the sentence communicates with a variety of audiences, offers a fresh and promising approach to this problem. To summarize then, the naturalism of this account brings an empirically informed description of human nature to bear on the problem of punishment. That description reveals that the experiencing of retributive emotions and the need to express them underlie and, to a great extent, explain the practice of punishment. The recognition that an explanation of the practice must make reference to human moral psychology points to the type of justification required. That justification must take account of both the different perspectives of all 207 those involved and the need for the victim and community to express their retributive sentiments. In other words, the justification will have a plurality of aims and will emphasize communication. In justifying punishment as a type of communication from various stakeholders to a variety of audiences, this naturalistic justification does what is required. In short, it explains why it is justifiable for citizens of a civilized and humane society to punish a fellow human being. 208 Bibliography Bacon, Francis. \"Of Revenge.\" The Essayes or Counsels, Civill and Morall. Ed. Michael Kiernan. Toronto, Oxford University Press, 1985. Baker, Brenda M. \"Penance as a Model for Punishment.\" Social Theory and Practice. 18, 3 (Fall, 1992), 311-31. . \"Improving our Practice of Sentencing.\" Utilitas. 9,1 (March, 1997), 99-114. Brunk, Conrad G. \"Restorative Justice and Punishment.\" Dialogue. 35 (1996), 593-8. Chupp, Mark. Chapter 4. \"Reconciliation Procedures and Rationale.\" Mediation and Criminal Justice: Victims, Offenders and Community. Ed. Martin Wright and Burt Galaway. Newbury Park, Ca: SAGE Publications Ltd., 1989, pp. 56-68. Clifford, William. \"Foreword.\" Proceedings of a Seminar on Victims of Crimes. Institute of Criminology Sidney University Law School, 1980. Coates, Robert B. and John Gehm. Chapter 17. \"An Empirical Assessment.\" Mediation and Criminal Justice: Victims, Offenders and Community. Ed. Martin Wright and Burt Galaway. Newbury Park, Ca: SAGE Publications Ltd., 1989, pp. 251-63. Cragg, Wesley. The Practice of Punishment: Towards a theory of restorative justice. New York, NY: Routledge, 1992. Damasio, Antonio R. Descartes' Error: Emotion, Reason and the Human Brain. New York: G. P. Putnam's Sons, 1994. Dowker, Fay and Glenn Good. \"The Proliferation of Control Unit Prisons in the United States.\" Journal of Prisoners on Prisons. 4 (1993), 95-110. Deigh, John. \"Empathy and Universalizability.\" Ethics, 105 (July, 1995), 743-63. Duff, R. A. Trials and Punishments. New York: Cambridge University Press, 1986. Fattah, Ezzat A. Understanding Criminal Victimization: An Introduction to Theoretical Victimology. Scarborough, Ontario: Prentice-Hall Canada Inc., 1991. Feinberg, Joel. Chapter 5. \"The Expressive Function of Punishment.\" Doing and Deserving. New Jersey: Princeton University Press, 1970, pp. 95-118. 209 . The Moral Limits of the Criminal Law , 4 volumes. New York: Oxford University Press, 1984-90. Fletcher, George P. With Justice for Some: Victims'Rights in Criminal Trials. Don Mills, Ontario: Addison-Wesley Publishing Company, 1995. Frank, Robert H. Passions Within Reason: The Strategic Role of the Emotions. New York: W. W. Norton & Company, 1988. Frankfurt, Harry G. \"Freedom of the Will and the Concept of a Person.\" Free Will. Ed. Gary Watson. New York: Oxford University Press, 1982, pp. 81-95. Garland, David. Punishment and Modern Society: A Study in Social Theory. Chicago: the University of Chicago Press, 1990. Goleman, Daniel. Emotional Intelligence: Why it can matter more than IQ. New York: Bantam Books, 1995. Gordon, Robert M. \"Sympathy, Simulation and the Impartial Spectator.\" Ethics, 105 (July, 1995), 727-42. Greenspan, Patricia. Emotions and Reasons: An Inquiry into Emotional Justification. New York: Routledge, Chapman and Hall Inc., 1988. Hampton, Jean. \"A New Theory of Retribution.\" Liability and Responsibility. Ed. R. G. Frey and Christopher W. Morris. New York: Cambridge University Press, 1991, pp. 377- 414. Hart, H. L. A. The Concept of Law. Oxford: Oxford University Press, 1961. . \"Prolegomenon to the Principles of Punishment.\" Punishment and Responsibility: Essays in the Philosophy of Law. New York: Oxford University Press, 1968, pp. 1-27. Hayry, Matti. \"A Defence of the Utilitarian Theory of Punishment.\" Retributivism and Its Critics: papers of the special Nordic conference held at the University of Toronto, 25-27 June 1990/ Canadian Section of the International Society for Philosophy of Law and Social Philosophy. Ed. Wesley Cragg. Stuttgart: Franz Steiner Verlag Stuttgart, 1992, pp. 129-47. Honderich, Ted. Chapter 5, \"Compromises,\" Punishment: The Supposed Justifications. London: Hutchinson and Company Ltd., 1969, pp. 138-43. Kant, Immanuel. Foundations of the Metaphysics of Morals. Trans. Lewis White Beck. Indianapolis: Bobbs-Merrill Educational Publishing, 1959. Launay, Gilles and Peter Murray. Chapter 8. \"Victim/Offender Groups.\" 210 Mediation and Criminal Justice: Victims, Offenders and Community. Ed. Martin Wright and Burt Galaway. Newbury Park, Ca: SAGE Publications Ltd., 1989. pp. 113-31. Mackie, J. L. Chapter XV. \"Morality and the Retributive Emotions.\" Persons and Values. Oxford: Oxford University Press, 1985, pp. 206-19. McCloskey, H. J. \"A Non-Utilitarian Approach to Punishment.\" Philosophical Perspectives on Punishment. Ed. Gertrude Ezorsky. Albany, New York: State University of New York Press, 1972, pp. 119-134. Mewett, Alan W. and Morris Manning. Mewett and Manning on the Criminal Law. Toronto and Vancouver: Butterworths Canada Ltd., 1994. Morris, Herbert. \"Persons and Punishment.\" Punishment and Rehabilitation. Ed. Jeffrie G. Murphy. Belmont, California: Wadsworth Publishing Company, 1985, pp. 24-41. Murphy, Jeffrie G. Retribution, Justice, and Therapy. Boston: D. Reidel Publishing Company, 1979. . \"Getting Even: The Role of the Victim.\" Philosophy of Law. Ed. Joel Feinberg and Hyman Gross. Belmont, Ca: Wadsworth PubHsliing Company, 1995, pp. 685-96. Murphy, Jeffrie G. and Jean Hampton. Forgiveness and Mercy. New York: Cambridge University Press, 1988. Nagel, Thomas. \"Moral Luck.\" Free Will. Ed. Gary Watson. New York: Oxford University Press, 1982, pp. 174-86. Nathanson, Stephen. \"Does It Matter If the Death Penalty is Arbitrarily Administered?\" Philosophy and Public Affairs Reader. Ed. A. John Simmons, Marshall Cohen, Joshua Cohen, Charles R. Beitz. Princeton, N. J: Princeton University Press, 1995, pp. 308-23. . \"Should we Execute Those Who Deserve to Die?\" Philosophy of Law. Ed. Joel Feinberg and Hyman Gross. Belmont, Ca: Wadsworth Publishing Co., 1995, pp. 748-56. Nozick, Robert. \"Retributive Punishment.\" Philosophical Explanations . Cambridge, Ma.: The Belknap Press of Harvard University Press, 1981, pp. 363-97. Plutchik, Robert. The Psychology and Biology of Emotion. New York: Harper Collins College Publishers, 1994. 211 Primoratz, Igor. \"Punishment as Language.\" Philosophy, 64 (1989), 187-205. Reprinted in Philosophy of Law. Ed. Joel Feinberg and Hyman Gross. Belmont, Ca.: Wadsworth Publishing Co., 1995, pp. 602-12. Rawls, John. \"Two Concepts of Rules.\" The Philosophical Review, 44 (1955), 3-13. Reiman, Jeffrey H. \"Justice, Civilization and the Death Penalty: Answering van den Haag.\" Philosophy and Public Affairs Reader. Ed. A. John Simmons, Marshall Cohen, Joshua Cohen, Charles R. Beitz. Princeton, N. J: Princeton University Press, 1995, pp. 274-307. Ross, Rupert. Returning to the Teachings: Exploring Aboriginal Justice. Toronto: Penguin Books Canada Ltd., 1996. Russell, Paul. Freedom and Moral Sentiment: Hume's Way of Naturalizing Responsibility. New York: Oxford University Press, 1995. Scarry, Elaine. Chapter 1. \"The Structure of Torture.\" The Body in Pain: The Making and the Unmaking of the World. New York: Oxford University Press, 1985, pp. 27- 59. Smith, Adam. The Theory of the Moral Sentiments. Ed. D. D. Raphael and A. L. Macfie. Indianapolis, Indiana: Liberty Fund Inc., 1982. Solomon, Robert C. The Passions. Garden City, New York: Anchor Press/Doubleday, 1977. . Chapter 5. \"Justice and the Moral Sentiments.\" A Passion for Justice: Emotions and the Origins of the Social Contract. Don Mills, Ontario: Addison-Wesley Publishing Company, Inc., 1990, pp. 198-241. Stent, Gunther S. \"Introduction.\" Morality as a Biological Phenomenon: The Presuppositions of Sociobiological Research. Ed. Gunther S. Stent. Berkeley and Los Angeles, California: University of California Press Ltd., 1980, pp. 1-18. Strawson, Peter. \"Freedom and Resentment.\" Free Will. Ed. Gary Watson. New York: Oxford University Press, 1982, pp. 59-80. . Skepticism and Naturalism. New York: Columbia University Press, 1985. Stocker, Michael. Valuing Emotions. New York: Cambridge University Press, 1996. Thomson, Judith Jarvis. \"The Decline of Cause.\" Philosophy of Law. Ed. Joel Feinberg and Hyman Gross. Belmont, Ca.: Wadsworth Publishing Co., 1995, pp. 512-19. 212 Umbreit, Mark S. Chapter 7. \"Violent Offenders and their Victims.\" Mediation and Criminal Justice: Victims, Offenders and Community, Ed. Martin Wright and Burt Galaway. Newbury Park, Ca: SAGE Publications Ltd., 1989, pp. 99-112. van den Haag, Ernest. \"In Defense of the Death Penalty.\" Philosophy of Law. Ed. Joel Feinberg and Hyman Gross. Belmont, Ca: Wadsworth Publishing Co., 1995, pp. 742-47. . \"Refuting Reiman and Nathanson.\" Philosophy and Public Affairs Reader. Ed. A. John Simmons, Marshall Cohen, Joshua Cohen, Charles R. Beitz. Princeton, N. J: Princeton University Press, 1995, pp. 324-35. Wallace, R. Jay. Responsibility and the Moral Sentiments. Cambridge, Ma: Harvard University Press, 1994. Watson, Gary. \"Free Agency.\" Free Will. Ed. Gary Watson. New York: Oxford University Press, 1982, pp. 96-110. Williams, B. A. O. \"Conclusion.\" Morality as a Biological Phenomenon: The Presuppositions of Sociobiological Research. Ed. Gunther S. Stent. Berkeley and Los Angeles, California: University of California Press Ltd., 1980, pp. 275-85. Wolf, Susan. Freedom Within Reason. New York: Oxford University Press, 1990. Wright, Mark. \"Introduction.\" Mediation and Criminal Justice: Victims, Offenders and Community. Ed. Mark Wright and Burt Galaway. Newbury Park, Ca: SAGE Publications, 1989, pp. 1-13. 213 "@en . "Thesis/Dissertation"@en . "1998-11"@en . "10.14288/1.0099380"@en . "eng"@en . "Philosophy"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "A naturalistic justification for criminal punishment"@en . "Text"@en . "http://hdl.handle.net/2429/9610"@en .