Doing the Unthinkable:

Transcription

1 Doing the Unthinkable: Discontinuity in thinking and acting Judith Martens Supervisor: Prof. Dr. J. A. M. Bransen Second supervisor: Dr. M. J. Becker Faculty of Philosophy, Theology and Religious Studies Radboud University Nijmegen May 2014

2

3 DANKWOORD Het schrijven van deze scriptie heeft langer in beslag genomen dan ik van te voren dacht, maar ik heb geen enkele spijt van de extra tijd die ik ervoor heb uitgetrokken. In die extra tijd heb ik ontzettend veel zinnige discussies gehad over het onderwerp, veel extra literatuur gelezen en een veel breder denkkader ontwikkeld waarbinnen ik nu over handelen, intentionaliteit en kwaadaardige handelingen kan nadenken. Dat is terug te vinden in mijn scriptie, maar is vooral iets dat ik met mij mee zal dragen. Dat was niet gebeurd zonder de begeleiding van Jan Bransen. Ik wil hem dan ook hartelijk danken voor de keren dat wij samen kwamen, de inhoudelijke discussies die we hadden, en de door hem gesuggereerde literatuur. Gesprekken waarin hij een enkele keer al ruim vooraf zag dat een argument ergens naartoe leidde (het belang van rollen) om mij vervolgens de ruimte te gunnen om een maand later dezelfde conclusie te trekken. Bedankt voor de balans tussen het vinden van mijn eigen weg en begeleiding. Verder wil ik graag in het bijzonder nog twee mensen bedanken die veel van mijn gedachten hebben aangehoord en mee hebben gedacht over de opbouw en uitwerking van mijn scriptie. Wil Martens en Bas Leijssenaar, bedankt voor jullie tijd en aandacht op bijna alle momenten van het proces.

4

5 Contents 1. Introduction Common Morality Common Morality from a Philosophical Point of View Common Morality from a Psychological Point of View Combining Two Approaches of Common Morality The Unthinkable Choices, Consequences and Personal Character Unwanted Desires and Volitions Volitional Necessities and the Unthinkable Thinking versus Acting Thinkability versus Doability The Stanford Prison Experiment Introduction to the Experiment The Counts: From Bad to Horrible Becoming Prisoner and Guard The Stanford Prison Experiment and the Unthinkable Silencing Virtuous Silencing Evil Silencers Evil Silencers return to the SPE Blind Spots and Biases Role Playing and Ethical Fading Switching Perspectives Biases and the Unthinkable Silencing the Unthinkable References... 54

6

7 1. Introduction Horrible behavior seems to be the one constant factor throughout human history. Wars leave us questioning what damage humans will not do to one another. Up to this day bullying and harassment are everyday practices. These and other examples of horrible behavior leave us puzzled, partly because we cannot imagine that we would display similar behavior if we were to find ourselves in likewise situations. Evil acts seem to stand so far away from our own acts that we find it unimaginable we would act in a similar way. The philosopher Harry Frankfurt has coined the term the unthinkable to refer to those actions wherefore we have all the power, yet which we cannot put ourselves to because of mental obstacles (they are unacceptable and unimaginable). This way of describing evilness appears highly attractive, for it expresses the feeling of incomprehension that we endure when others do evil deeds. Unfortunately, as I will explain below, the concept of the unthinkable fails to grasp what is at stake when people do evil deeds and Frankfurt s approach is incomplete at best. With this thesis, I try to advance our understanding of how people can find themselves acting evilly and how behavior can get out of hand. All people fall under the scope of this investigation, not only some serial killer but you, me, your friends and my parents. It is about all sorts of wrong 1 behavior, from small wrongdoings to evil crimes. Certain actions are generally accepted as good, others as bad. Common morality describes how we have general, shared, ideas about what ought and what ought not to be done. Saying what should be done often has a counterpart; whenever we emphasize something should be done, we imply that there are other things that should not be done. This implies that having a shared sense of morality (commonsense morality) is always accompanied by a commonsense should not of which some actions will be unthinkable. Such actions will generally be seen as wrong, hurtful, and unimaginable. Is knowing what is perceived as unthinkable sufficient to not act upon it? I doubt it is. Both in current times and in history we see people acting in immoral ways that we, but also their contemporaries, find unacceptable and unimaginable. It is scary to see how people can do things that horrify and puzzle others. It makes one wonder if the neighbor, a friend, an enemy or maybe oneself is capable of such actions too. 1 There is much discussion going on about the concept of evil and how it relates to the concept of wrong. People want to distinguish these two from one another for doing something evil intuitively feels worse than doing something that is bad. Around the questions how they are to be distinguished an extensive discussion exists. For the purpose of this thesis this discussion is not relevant, for also wrong deeds are better to be avoided and sometimes seem to happen to us. therefore I will use both terms as were they replaceable. 1

8 What is typical about actions that appear unthinkable is that they still appear unimaginable after they have happened, after people acted according to what was thought to be unthinkable. Even after knowing, due to for example World War II, of what horror we are capable, we still find it unimaginable we would act the same. Many people who acted horribly during that time still have trouble explaining why they acted like they did and how they not only acted horrible in one situation, but over and over again. Even if one fears that others could act wrong again, one cannot imagine oneself or those close to oneself to do so. As we will see, the idea of the unthinkable can (on its own) not explain why we both keep on thinking and acting in an unthinkable way (repetitions could be within one person or throughout history). To demonstrate that humans of all kind can lapse into bad behavior in a very short span of time, I will use a description of the Stanford Prison Experiment (SPE). This experiment shows how ordinary college students started to display very disturbing behavior within only five days. Findings in the SPE suggest we are capable of acting in accordance with the unthinkable. Worse even, that we are capable to do so over and over again. I will use the SPE to test Frankfurt s theory and to see what remains of its explanatory power. The SPE will help point at some questions that remain unanswered in Frankfurt s theory, especially those concerning the doablility of the unthinkable. The silencing account, as described by McDowell and Garrard, offers a very natural conception of how certain options of acting become invisible, therewith giving a possible answer to these questions. It serves both to explain heroic acts and fairly evil acts. I will use the idea of silencing, as coined by John McDowell and elaborated by Eve Garrard, to better understand how the unthinkable can become doable and repeatable while remaining unthinkable. Silencing seems to offer a good explanation of the question why we deviate from what we would normally want and expect to do. When silencing, a possible path of acting and the reasons to do so are silenced (no longer seen) because of another overriding reason. This seems obvious for virtuous silencing because, for example, we can immediately understand why many reasons fade away when one s partner is in danger. However, it is far from obvious for evil silencing. Evil silencing requires a separate explanation. I shall argue that social-psychological research on biases provides insights that can explain evil silencing. By applying these insights to evil silencing, I hope to attest that the unthinkable, when seen in a frame of common morality and enriched with an account of silencing, remains a plausible concept that can help us understand why some acts are unimaginable and yet remain actionable. 2

9 I will first argue in favor of the existence of a common morality that is shared by all, mainly using Gert 2 and Haidt 3. Their theories provide conceptions of agents and agency that connect well with Frankfurt s 4 idea of agency. The idea of common morality explicitly argues for shared moral intentions, which Gert and Haidt describe in great detail. Shared intentions can also be found in Frankfurt s theory, but with less substantiation. Frankfurt, on the other hand, develops his concept of agency and the influence of character and volition in greater detail. His conception of human agency will be put to use in chapter three. Based on this conception I will investigate the concept of the unthinkable. This will result in a preliminary but seemingly credible way of explaining how we experience actions as evil deeds. Frankfurt s conceptualization gives, however, rise to some new questions. Frankfurt does especially not explain how we can do the unthinkable, as is demonstrated in the chapter on the Stanford Prison Experiment 5 that shows that we can act in unthinkable ways. The last two chapters are devoted to explaining how intuition and reasoning can be thrown off road, thereby explaining how we can do the unthinkable. First, through a conceptual analysis of reasoning that is out of balance, supported by the concept of silencing. 6 Second, by introducing what social psychologists call biases. A better understanding of how we are capable of doing the unthinkable, and doing it over and over again, can give insight in how we can prevent this from happening. Or so I hope. 2. Common Morality Many moralities and theories on morality explicitly compete with each other on being and representing the moral. This diversity can give one the feeling that opinions on morality are far apart. Common morality provides a general moral framework which is claimed to be shared by all humans. I want to use this framework as a basis for understanding shared unthinkable actions. The focus of common morality on agreement instead of diversity gives a basis for an understanding of why wrong deeds are unimaginable for the majority of people in general. 2 B. Gert, "The Definition of Morality", The Stanford Encyclopedia of Philosophy (Fall 2012 Edition), ed. E. N. Zalta, URL = < 3 J. Haidt, Morality, Perspectives on Psychological Science, 3 (2008) p65-72; Haidt & Kesebir, Morality, in S. Fiske, D. Gilbert, & G. Linzey (Eds.) Handbook of Social Psychology, 5 th edition. Hobeken, NJ: Wiley (2010), pp H. F. Frankfurt, The Importance of What we Care About: Philosophical Eessays. Cambridge University Press, Cambridge 1988; H. F. Frankfurt, Taking Ourselves Seriously & Getting it Right, Stanford University Press P. G. Zimbardo, The Lucifer Effect, Understanding How Good People Turn Evil, Random House Trade Paperbacks, New York J. McDowell & J. G. McFetridge, Are Moral Requirements Hypothetical Imperatives?, Proceedings of the Aristotelian Society, Supplementary Volumes, Vol. 52 (1978), pp ; E. Garrard, The Nature of Evil, Philosophical Explorations, 1 (1998), pp

10 Wrong acts, acts that we detest when they are performed, have to be acts that people in general find unacceptable. Regarding something to be unacceptable cannot be about single individuals finding an act (or acts) unacceptable. Only when it is a shared idea that a certain act is unacceptable can it be meaningfully retained as being unacceptable after an individual acted in that specific unacceptable way. Say that Joe is a practicing vegetarian. One day he finds himself eating meat. Joe can now think that eating meat is actually not such an unacceptable thing as he had always thought and perhaps he will continue eating meat for the foreseeable future. This is in no way problematic for society (at least not for our society, where the idea that eating meat is unacceptable is not widely shared). Now consider the following case. Say that Jane is against killing, but one day she finds herself bashing someone s head in. This of course has an impact on the individual level (Jane and her poor victim), but also on the level of society. Exactly because the idea that killing is unacceptable is widely shared, it makes sense to retain this idea and to keep thinking that we are not allowed to do such a thing. Just imagine what society would be like if we were constantly thinking that all other people were potential murderers. As Gert states, we will accept moral rules because they do not only restrict people s behavior, but also bring securities that help us avoid suffering. 7 Morality can be described as an essentially contested concept. 8 There exists a widespread agreement on importance and general meaning of the concept, but not on the best realization and implementation of it. Because of the ongoing disagreement about the best realization of morality, users of the concept are often found discussing the use and application of the term, rather than settling on an agreed definition. This contestedness can partly be explained through the two ways morality is used; a normative and a descriptive way. 9 When one defines morality in a normative way one accepts, as part of it, a concept of rational persons that all endorse a specific code as a moral code. 10 When morality is perceived in a descriptive sense it refers to codes of conduct that are put forward and accepted by some group, society or individual. 11 The descriptive perception of morality has recently become more prominent, among others by the 7 B. Gert, Common Morality: Deciding what to do, Oxford University Press, NY 2007, pp W. B. Gallie, Essentially Contested Concepts, Proceedings of the Aristotelian Society, Vol.56, (1956), pp An essentially contested concept is a concept about which many definitions exist that always stride for being the only one. Art, religion, and democracy are concepts that are also said to be contested. 9 B. Gert, Definition of Morality. 10 There are many different codes and they regard other behaviors as immoral, therefore there is much disagreement on the normative definition of morality. Seeing morality as a normative concept implies that all rational persons endorse that concept of morality, independent of their religious beliefs, their upbringing, or the society they are part of (B. Gert, Common Morality, preface). Ethical relativists deny the existence of a universal normative morality; only in the descriptive sense can morality refer to a code of conduct. 11 Being outside of that group, this descriptive definition of morality has no implications on one s behavior. 4

11 work of Haidt, 12 who has been influenced by the views of David Hume and tries to present a naturalistic account of moral judgments. 13 Within this descriptive use of morality, both philosophers and psychologists look for common grounds that are widely, if not universally, accepted by mankind. Those common grounds could, if found, be considered as the normative core of morality. So even when morality is defined in a descriptive way, researchers on common morality are still looking for a common ground that can be perceived as normative or normgiving. To describe a concept as being essentially contested does not only point at disagreement, it also implies there is a level of agreement. This idea of agreement within disagreement is substantiated, among others, by Gert 14 and Haidt & Kesebir 15 as common morality; a shared moral basis that allows for differences. I will use the concept of common morality because it can help explain how the unacceptability of certain acts is shared by many. It can also give insight in how we sometimes fail to do what is right, or even do something considered evil. I will draw on Gert, Haidt, and Haidt & Kesebir to give an account of common morality. Although they take different perspectives on morality, they arrive, I believe, at the same fundamental conclusions. The line of thought that I am developing in this thesis is a combination of the fields of philosophy and psychology, and the authors mentioned above each start reasoning from one of these perspectives. Putting them together is meant to give a broader and more substantiated perspective on common morality, and is certainly not meant to criticize one of the accounts on the basis of the other. The roughly painted picture of common morality will give the right background to show the relevance of my thesis. 2.1 Common Morality from a Philosophical Point of View According to Gert, common morality is the moral system that thoughtful people implicitly use when making moral decisions and judgments. 16 All versions of theoretical and religious morality, as written down by Kant, Mill, or in Christianity or Islam are revisions of this common morality. What makes these revisions problematic is that they try to give one single answer for moral cases, while according to Gert multiple answers are often possible and acceptable. Positing only one of many possible answers to be the correct answer automatically implies that the other answers incorrect. Gert assumes that because we share a basic set of moral ideas there is space 12 J. Haidt, The Emotional Dog and its Rational Tail, A Social Intuitionist Approach to Moral Judgment. Psychological Review, 108 (2001), pp B. Gert, Definition of Morality. 14 B. Gert, Common Morality. 15 J. Haidt & S. Kesebir, Morality. 16 B. Gert, Common Morality, p. 4. 5

12 for the attribution of different degrees of importance for the different aspects of the basic set of moral ideas. 17 If Gert is right, this explains why a diversity of possible and preferable answers exists, while we are still able to stack them all under one theory. Gert s definition of moral people requires that they are also rational people. He claims that a moral agent must have the following three rational characteristics. 18 1) He/she should have a minimal intelligence, including some ability to reason and use past experience to learn from. The minimal intelligence is described as comparable to the intelligence of a ten year old and includes knowledge like knowing that people want to avoid pain and are vulnerable. 2) A moral agent must know the immediate and short-term consequences of most of her actions. 3) A moral agent should possess volitional abilities: evidence should lead to beliefs of how the world is, and functions and acts should be based on these beliefs. Gert assumes that all people are willing to accept such a minimalistic view on the intelligence of mankind. 19 Moral persons are not solely rational. They also have other characteristics, which make it understandable that we actually care about morals and moral acting. We care about which morality others use to guide their actions, the reason being that there might be bad consequences for ourselves and those around us if they do not accept a morality that we find acceptable. 20 Gert assumes that morality is only concerned with behavior that affects other people (causing death, pain, loss of freedom, or loss of pleasure). Furthermore, he assumes that people all share features that benefit from moral and/or pro-social behavior, such as: 1) Humans are vulnerable to suffering harms (like death, pain, disability, or deprivation of freedom and pleasure). 2) Humans want to avoid those things that bring suffering. 3) Accepting rules to avoid suffering is seen as rational. Furthermore, we should acknowledge that 4) rational people are fallible and 5) rational people do not have all the knowledge and hence make mistakes. 21 According to Gert morality is an informal public system applying to all rational persons, governing behavior that affects others, and has the lessening of evil or harm as its goal. 22 Seeing morality as an informal system entails that the moral rules are no universal guide for all moral situations. Differences in the moral frameworks people have are caused by the relative importance they give to different harms. Some may choose a painful operation over an early 17 B. Gert, Common Morality, Preface & pp Ibidem, pp Ibidem, pp Ibidem, pp Ibidem. pp B. Gert, Definition of Morality. 6

13 death, others may not. That morality is a public system does not mean that everyone always agrees on all the decisions that are made, but only that all disagreements occur within a framework of agreement. Basketball can be a public system even if referees can, within limits, disagree in their judgments in calling fouls; but all players know that what the referees call a foul determines what is a foul. 23 Moral agents can, and will, disagree in their moral judgments. These disagreements will, however, fit within a shared framework. This agreement, concerning which rules are moral rules, explains why morality can be a public system even though it is an informal system. Rules about violating moral rules can of course also be fitted within this idea. Cheating, deceiving, injuring, or killing another moral agent in order to gain money for shopping or for a holiday is generally not considered as morally justifiable. 24 Morality does, however, not only forbid certain actions, it also promotes pro-social behavior. Still, the failure to act charitably on every possible occasion does not require justification in the same way that any act of killing, causing pain, deceiving, and breaking promises requires justification. For Gert, morality encourages charitable action, but does not require it; it is always morally good to be charitable, but it is not immoral not to be charitable. 25 Some immoral acts can have a very small impact and hence breaking the moral rule that forbids them can be seen as not such a bad thing. For example, tossing one s soda can on the streets does not necessarily imply the end of the world as we know it, although it might be considered to be immoral. This leaves one with the question why one should still behave moral in those cases. Sometimes, direct reasons for acting moral are simply not convincing. Thus, people might be tempted to use small lies, and sometimes they regard stealing as a proper option to avoid being hungry, even though they would not starve if they would not steal. However, this type of behavior cannot be tolerated, since if everyone would act in this way it would lead to an unsustainable situation. In those cases indirect reasons can bring one to still do the morally right thing. If not just oneself but many people would act in a certain immoral way, the act that was not that bad would no longer be condonable. The consequences of one person breaking a moral rule with small impact are not the problem, but many people breaking the same moral rule could lead to insurmountable problems. It is pure arrogance to think that one has the right to act immoral where everybody else does not have that right. And arrogance is incompatible with the 23 B. Gert, Definition of Morality. 24 B. Gert Common Morality, B. Gert, Definition of Morality. 25 One could argue there are two sorts of pro-social behaviors: 1) those behaviors that are perceived obligatory (saving a drowning child) and 2) those pro-social behaviors that are not obligatory (helping every neighbor with his grocery shopping). Where the line between obligatory and volitional pro-social behaviors should be drawn is not of importance here. 7

14 impartiality required by morality. 26 Arrogance is a character trait that ought to be avoided by all rational people and therefore an indirect reason to act moral. 27 Despite direct and indirect reasons pushing persons towards moral acting, there are still people acting immorally. Every rational person is aware that all people want her, regardless of any ends she might have, not to act immorally, at least with regard to themselves. 28 Every rational person will, most likely, also want all others to do the same. However, some rational people might not want to support common morality. 29 This can be due to religious beliefs or the belonging to a dominant group. In both cases the view of common morality is broken by a view that applies to a certain group. Within larger groups, as religious groups often are, those who follow a certain religious vision are in large numbers. This makes it possible to ignore others without feeling the consequences oneself. According to Gert 30 these dominant and/or religious groups have beliefs that obstruct common morality. Eliminating these beliefs will lead to endorsing common morality again. This can explain why certain groups sometimes act out of line with common morality without noting the problematic consequences for a larger community. To summarize: Gert s conception of morality presupposes that most, if not all, humans possess a minimal rationality that results in the existence of a common morality. This minimal rationality amounts to implicit thinking about which effects one should prefer given one s selfish and social nature. Differences in how we think one should act within specific circumstances can be explained by the relative importance individuals give to the different harms (and goods) that are agreed upon within the framework of common morality. One should keep in mind that a huge overlap remains, even when differences are accepted. It is mainly the implicitly rational prohibition of acts that cause harm and pain that make someone act moral. This framework, Gert thinks, is endorsed by all people, meaning that those acts that are breaking these rules are generally seen as wrong. 26 B. Gert Common Morality, p Gert, as we will see in Haidt as well, makes a differentiation between the questions why be moral and why act moral. The first question gives a broader ground to substantiate reasons for acting morally right. By acting the latter question, one only looks at the moment, potentially missing reasons to act moral that lie in a broader perspective. 28 B. Gert, Common Morality, p Ibidem, pp Ibidem, pp

15 2.2 Common Morality from a Psychological Point of View Haidt and Kesebir 31 compare empirical studies of morality, altruism, and pro-social behavior, and bring results of these disciplines together. Based on empirical findings they suggest that morality may be simply too heterogeneous and multifaceted to create one grand unified theory of morality with single answers to every single question. While discussing morality, Haidt and Kesebir 32 think one should return to the question why be moral?, as in their opinion the general tendency to focus on why act moral? is too shortsighted. 33 They are thereby returning to an Aristotelian kind of ethics. Being moral should partially happen through learning by practice, examples, and stories instead of through learning by following rules. Asking the why be moral? question emphasizes practice and habit, rather than knowledge and reasoning. Virtues are then skills of social perception and action that must be acquired and refined over a lifetime. 34 Narrowing the scope of ethics to the question why act moral?, as utilitarianism and deontology have tended to do, would lead us away from stories, character, and habit. 35 Human nature is defined as partially selfish and partially characterized by empathy and reciprocity. 36 Society has to deal with the selfishness of people and morality plays an important part in this. The multiple existing and defensible moralities known over the world are different because there are multiple solutions to deal with selfishness. 37 Moral systems are interlocking sets of values, virtues, norms, practices, identities, institutions, technologies, and evolved psychological mechanisms that work together to suppress or regulate selfishness and make cooperative social life possible. 38 This does not mean it is always easy to repulse selfishness. Haidt and Kesebir distance themselves from the division between emotion (or affect) and cognition. They consider emotions as being a form of cognition. Rather, they distinguish between intuition and reasoning (being both a type of cognition). (a) (Moral) intuition is described as seeing that something is (good/bad, likeable/dislikeable) : the sudden appearance in consciousness, or at the fringe of consciousness, of an evaluative feeling about the 31 Haidt and Kesebir, Morality. 32 Haidt and Kesebir, Morality, pp Ibidem, p P. M. Churchland, Toward a Cognitive Neurobiology of the Moral Virtues, Topoi 17 (1998), pp.84-86; J. McDowell, Virtues and Reason, Monist 62 (1979), pp J. Haidt and S. Kesebir, Morality, p Ibidem, p Haidt and Kesebir use the view on moral pluralism as defined by R. A. Shweder, N. C. Muck, M. Mahapatra & L. Part, The Big Three of Morality (Autonomy, Community, and Divinity), and the Big Three Explanations of Suffering, in eds. A. Brandt & P. Rozin, Morality and health (pp ). New York: Routledge, J. Haidt and S. Kesebir, Morality, p

16 character or actions of a person, without any conscious awareness of having gone through steps of search, weighing evidence, or inferring conclusions. 39 (b) (Moral) reasoning is conscious mental activity that consists of transforming given information about people (and situations) in order to reach a moral judgment. 40 These last processes are intentional, effortful, and controllable, as is their awareness. 41 Intuition is the most used cognitive form, guiding most decisions. Intuition plays a key role in moral decision making as well. Reasoning is a process that often takes place after an intuitive judgment is already made. This first judgment can and will direct the arguments that are used in reasoning about one s judgments. 42 So it can be said that there is a relevant difference between making oneself believe that it would be best to do X and finding oneself to believe that it would be best to do X. Haidt and Kesebir distinguish three functional principles guiding moral judgments: (1) intuitive primacy, (2) moral thinking is for social doing, (3) and morality binds and builds. The first principle, intuitive primacy, states that most of human mental life is made up of automatic intuitive processes, moral intuition being one of them. 43 This signifies that reasoning, at least most of the time, comes after intuition, if there is any reasoning involved at all. In moral thinking reasoning seems to have less power and independence than in other forms of thinking. When reasoning about morals, a variety of motives push people towards finding support for the conclusions already reached by (faster) intuitive processes. Reasoning involves multiple steps, and any one of those steps could be biased by intuitive processes. One famous bias is that people search only for relevant arguments and evidence. 44 Reflection is thus not excluded, but its effects remain rather unclear within Haidt and Kesebir s theory. 45 Haidt and Kesebir, secondly, presume that much of human cognition was shaped by natural selection for life in intensely social groups: moral thinking is for social doing. 46 This second principle that guides moral thinking entails that cognition is socially situated and has social functions. Moral language and reasoning evolved to enable communication, not truth finding. 39 J. Haidt and F. Bjorklund, Social Intuitionist Answer Six Questions about Morality, in ed. W. Sinnott - Armstrong, Moral psychology: Vol. 2. The cognitive science of morality. Cambridge, MA: MIT Press 2008, p J. Haidt, Emotional Dog, p J. A. Bargh, The Four Horsemen of Automaticity. Awareness, Efficiency, Intention, and Control in Social Cognition, In eds. J. R. S. Wyer & T. K. Srull, Handbook of Social Cognition (2nd ed.). Hillsdale, NJ: Erlbaum 1994, pp J. Haidt and S. Kesebir, Morality, pp J. A. Bargh and T. L. Chartrand, The Unbearable Automaticity of Being, American Psychologist, 54,(1999), pp D. Kuhn, The Skills of Argument, Cambridge: Cambridge University Press 1991, pp See J. Kennett and C. Fine, Will the Real Moral Judgment Please Stand up? Ethical Theory and Moral Practice 12 (2009), p J. Haidt and S. Kesebir, Morality, pp

17 And if one would find oneself in discussion, convincing others with arguments might have mattered more than telling the truth. The third principle, morality binds and builds 47, concerns the human capacity for cumulative cultural learning; this capacity can increase the importance of group-level selection and the entitativity of the group an individual belongs to (the perception of a group as pure entity), and thus decrease individual level pressures. It also leaves room for solving problems that arise in social groups, such as the free-rider problem. Five so called hypothetical foundations of morality give substance to these three principles. 48 The five foundations of morality are 1) avoiding harm and caring (we are concerned with suffering of ourselves and others, caring and compassion are part of the care side of harm/care) 2) fairness and reciprocity (general concerns about unfair treatment, cheating and other more abstract notions of justice and right), 3) in-group/loyalty (people have obligations to other group members, such as loyalty, self-sacrifice, and vigilance against betrayal, 4) authority and respect (the various roles we have oblige us to show respect, obedience and so on), 5) purity and sanctity (concerns about both physical and spiritual contagion, including wholesomeness, control of desires and the virtue of chastity). These five foundations that are commonly shared result in general intuitions on morality that all share, in common morality. Summarizing Haidt and Kesebir, humans are social beings that also have egoistic characteristics. Morality is about being social. Moral judgments are intuitive and the intuitions about morality are commonly shared due to five hypothetical foundations. Intuition and reasoning should not be taken as two completely different things; they are both cognitive processes. In general, people will act on their intuitions and reasoning will, most of the time, follow those intuitions with arguments that can underpin the intuitive judgments. Social evolution is at the basis of our moral intuitions. 2.3 Combining Two Approaches of Common Morality Both discussed approaches take a different starting point, which give a different emphasis to their theories. Gert is, first of all, opposed to the idea that there is only one moral theory, or that moral theories exclude each other s solutions. Although there is no moral theory that can tell us exactly what to do in every situation, the diversity of possible moral decisions falls within the lines of common morality. Haidt and Kesebir are keen on pointing out how intuition and reasoning are both cognitive capacities. Intuition comes first in many occasions, directing what 47 J. Haidt and S. Kesebir, Morality, pp Ibidem, pp

18 arguments are used and what our moral judgment is. Starting both from different angles, they use different terminology. This makes some differences between both theories of common morality appear bigger than they actually are. Both approaches disclose a common morality that is shared by (almost) all. The different solutions to moral problems can be attributed to the fact that individuals have their own hierarchy of which sufferings and losses in freedom are more important (Gert) and to the different possible ways in which societies deal with selfishness (Haidt and Kesebir). The framework is largely shared by all because each individual cares about avoiding suffering and accepting rules to avoid suffering (Gert), or by avoiding harm, and advancing fairness, loyalty, respect and sanctity (Haidt and Kesebir). Morality is based on human nature, assuming that all humans have a minimal rationality. This rationality is comparable to that of a ten year old in Gert s writing. Haidt and Kesebir attribute the moral decisions we make to intuition. They separate reasoning from intuitive decision making, but intuition is a matter of cognition. I believe this is an area where one might think the two approaches to be miles apart from one another, purely through the words that they have chosen. If one looks more precisely at the kind of rationality that Gert supposes, this is, however, very close to the intuitive process Haidt and Kesebir talk about: it is about knowing that people avoid suffering and learning from experience. Compared to Gert, Haidt and Kesebir describe some extra characteristics of human nature and agency, making a clear distinction between intuitions and reasoning and stating that intuition plays the lead role. This explicit emphasis on intuition sets reasoning aside as the cognitive capacity that makes us explicitly and consciously decide how to act in moral situations. 49 Both approaches also give insight in why people sometimes fail to do what is moral. According to Gert an individual can be distracted from common morality by following the specific rules and ideas of a dominant group. 50 Haidt and Kesebir focus on the balance between social life and selfishness, saying selfishness is the reason we deviate from the moral path. 51 Gert assumes that reasons of self-interest are normally compatible with acting morally. If they are not, we need 49 Gert seems to be more interested in morality s capacity to provide rational justifications of what we find ourselves wanting to do. This seems to make the apparent differences between Gert and Haidt and Kesebir even smaller. 50 B. Gert, Common Morality, pp J. Haidt and S. Kesebir, Morality, p

19 indirect reasons to be drawn to the moral choice. 52 However, this implies that self-interest can, and sometimes will be a reason to act immoral in his theory too. The discussed approaches leave an important point disregarded: sometimes we experience that our intuitive choices deviate from what we think we ought to (and also would like to) do. In extreme situations, such as World War II, we see people acting against their own expectations. Doing something immoral can sometimes be a choice, like collaborating with the enemy in order to protect oneself, that obviously goes against common morality, yet can be explained through self-interest. But what happens in those situations where we do what we never expected to do, for example turning in Jews without there being an obvious reason for it? Do we want the moral behavior, but are we unable to act accordingly? Or do we often not realize our choices are nonmoral or immoral? The next chapter will investigate what it means to do something we want to do, but also, more importantly, what it means to do things that we do not want to do or, even more, find unthinkable, but that we still do anyway. 3. The Unthinkable Through using Frankfurt s theory on agency, intentionality, and the unthinkable, I will combine and extend two important points that I have derived, in the previous chapter, from the different approaches concerning common morality. These points are: (1) (almost) all human beings share the capacity to intuitively and implicitly think and act morally, 2) due to human nature these intuitions are shared to a large extent among all humans. Frankfurt agrees that some values are shared by (almost) all humans, 53 but he argues that we differ in the relative importance we give these values. In addition to their interest in staying alive, people generally have various other similarly primitive and protean concerns as well, which also provide them with reasons for acting. For instance, we cannot help caring about avoiding crippling injury and illness, about maintaining at least some minimal contact with other human beings, and about being free from chronic suffering and endlessly stupefying boredom. We love being intact and healthy, being satisfied, and being in touch. We cannot bring ourselves to be wholly indifferent to these things, much less categorically opposed to them B. Gert, Common Morality, pp H. F. Frankfurt, Taking Ourselves Seriously, p Ibidem, p

20 Since these desires come naturally to all human beings, they are regarded as necessary human aspects. 55 This does not mean they are equally important to everyone. Individuals differ in the importance these desires have in their practical reasoning. 56 All this fits perfectly with the idea of a common morality as described in chapter two. While parts of Frankfurt s account fit nicely with the theories of common morality, he also provides deeper understanding of what happens on the individual level when people choose certain actions, especially where he attempts to explain why certain acts are considered to be unimaginable ( unthinkable in Frankfurt s terminology) by human beings. People, sometimes without even knowing why, do things that they themselves and other people consider unthinkable to do. However, if these things are really considered to be unthinkable at both the group level and individual level, how come that individual people can do them? We saw that, from the perspective of theories of common morality, we cannot explain how individual people intuitively or deliberatively can choose to do what they and others find unthinkable. In short: we cannot explain how people can choose to do what is perceived as lying outside the domain of possible choices for acting. Intuitively, this shortcoming of common morality leads me to think that a distinction ought to be made between what is perceived as unthinkable but can become doable, and what is unthinkable and undoable. It is my guess that Frankfurt s theoretical framework provides enough conceptual building blocks for me to come up with such an account. Frankfurt s attention to the different weight people s desires have in their practical reasoning, his specific attention to people s volitions and reflections, and his development of the concept of the unthinkable make that his theory provides a great starting point for an attempt to understand why and how we do evil deeds. We saw that Haidt and Kesebir pointed to the different values people attach to their five hypothetical foundations of morality. 57 Frankfurt also offers an explanation why people sometimes differ in their choice of action and sometimes even fail to act in line with moral rules. Different people have different desires and goals, causing conflicting desires. Sometimes they strive for the same hard to obtain desires and goals to be fulfilled, causing competition. This implies that moral rules sometimes can and will go against what we actually want to do, or want to achieve. 58 What an agent wants can go together with, or stand in opposition to, the rules of the society we live in and common morality. What we care about is very individual and hence often different from moral rules that are generally applicable. 59 If we agree with Frankfurt s 55 H. F. Frankfurt, Taking Ourselves Seriously, p It might be superfluous to add, but the balance of reasoning that is talked about is not of a conscious reflective kind (at least it need not be). These processes are highly automated. 57 Haidt and Kesebir, Morality, pp H. F. Frankfurt, Taking Ourselves Seriously, pp Ibidem, pp

21 argument, the conclusion follows that we will not always choose moral action over our (personally) preferred action. A major difference between the discussed approaches on common morality (Gert, Haidt & Kesebir) and Frankfurt s theory on agency is the attention given by the latter to volitions and reflections. Although Frankfurt s explanation of action is based on building character and making intuitive decisions, his theory also includes a moment of distance in which one reflects on one s character. A character is not only developed through what we experience (what happens to us), but also through what we want to do with what we experience: we are reflective beings. Frankfurt argues that the unthinkable can only exist against a background of things we care about. The things we love imply that certain actions that would hurt or damage what we love are unacceptable and therefore unthinkable. When interpreting the concept unthinkable as such we can explain why people cannot bring themselves to doing evil deeds, and cannot imagine they will ever be able to act in evil ways. A highly intuitive idea, one that can help us to further understand why we attempt to avoid harmful behavior and why this is so much part of our human nature. Based on Frankfurt s conception of character and the conception of the unthinkable that he has built on it, I will try to understand how we can do what is commonly deemed unacceptable and unthinkable. First, I will introduce Frankfurt s opinions on volitions, choices, and character, elements that are new in the sense that they were not discussed by me in the section on common morality. Second, I will turn to the concept of the unthinkable as Frankfurt develops it. Ultimately, a conclusion will follow on what it entails to do the unthinkable. 3.1 Choices, Consequences and Personal Character Caring and loving others, things, and activities is at the heart of Frankfurt s theory. Frankfurt sees, of course, that being as free as possible is valued highly in contemporary society. He emphasizes, however, that free choices always also involve necessities. Every choice we make (because of things we care about) has determining consequences for ourselves. Such consequences are an expression of our freedom, according to Frankfurt. 60 Saying that choices lead to necessary consequences does not show that we are not free. Rather, it illustrates that building personal character involves boundaries and setting boundaries. Our necessities result in acts we associate with as well those acts we dissociate from. 60 H. F. Frankfurt, On the Necessity of Ideals, in eds. G. G. Noam, T. E. Wran; The Moral Self: Building a Better Paradigm. Cambridge, Massachusetts, The MIT Press 1993, pp

22 Frankfurt states that in general we do not choose what the objects of our love and care are. Loving and caring happen to us. These desires and wants that befall upon us are called firstorder desires (or volitions). It is impossible to explain why my neighbor likes to take care of his chickens and I like to read books; these preferences in first-order volitions somehow pop up and develop. 61 Yet these cares are at the basis of our character. What we care about and love can arise for many reasons. Cares can come from imperatives of tradition, from style, intellect, or ambition. This does not mean that our first order volitions are fully beyond our will. We do have influence on how the objects of love and care appear in our actions, through what Frankfurt calls second-order volitions (and higher-order volitions). Higher order volitions are volitions about volitions and volitions to act on volitions. 62 We can confirm or deny first order desires by directing actions through selective identification with certain attitudes and preferences. 63 In doing so, we take responsibility for these attitudes and preferences. Humans cannot literally produce their character, but they do take responsibility for it through their higher order volitions. 64 Every person develops reactions to his/her first order volitions. Both identification with and dissociation from first order volitions derive from the capability of forming higher order volitions as reactions to our first order volitions. While our first order volitions are not chosen, it does not follow that our relationship to what we care about is wholly non-cognitive and without reason. There is considerable room for reason and argument in the clarification of ideals and in the evaluation of their worthiness. 65 The actions of an agent reinforce the caring, creating automaticity in reactions to situations. These automatic, intuitive actions are then in line with their (higher-order) volitions. Frankfurt formulates a theoretical framework in which agents attempt to have a fully integrated self. His answer to the question of how we can integrate all the different (and often conflicting) volitions a person can have is the idea of a hierarchy of volitions. 66 A fully integrated self has harmony in all (levels of) volitions. Harmony implies a) knowing the desires, b) not deceiving oneself about them, and c) acting on them. 67 Conflicts between volitions that do arise have to be resolved in order to be a fully integrated self; this can be done by searching for (even) higher order volitions and outlawing deviating lower ones. 68 An agent that has constituted 61 H. F. Frankfurt, Necessity of Ideals, pp H. F. Frankfurt, Freedom of the Will and the Concept of a Person, Journal of Philosophy, LXVIII 5-20 (1971), pp H. F. Frankfurt, Necessity of Ideals, pp H. F. Frankfurt, Taking ourselves seriously, pp Ibidem, p H. F. Frankfurt, Importance, pp Ibidem, pp Ibidem, pp

23 himself/herself without ambivalence is also called wholehearted. 69 Self-knowledge is crucial in acting wholeheartedly; only by knowing one s volitions one can act and react on them through higher order volitions. 70 Frankfurt s argument, for as far as we have discussed it, gives us an integrated picture of the influence of both automatic and reasoned strivings on producing our actions. 3.2 Unwanted Desires and Volitions What are the implications of Frankfurt s ideas for our understanding of evil actions? As we generally cannot decide which first order volitions we have, we also cannot decide having or not having unwanted (possibly evil) first order volitions. 71 Frankfurt mentions that we have a dark side consisting of rebellious desires that we simply do not want to identify with. 72 Like all first order volitions, these unwanted volitions, of which some could be called dark or evil, befall on us. Frankfurt, however, also explains how higher order volitions can help us to overcome intensely unwanted, despised, desires. Through higher order volitions we can try to keep the influence of unwanted desires as small as possible. Dissociation (by not endorsing) of volitions diminishes their power to function as motives to act on. The fact that, in general, humans do not want to endorse despised desires, can be explained within the framework of common morality: there are certain acts that we all want to avoid and hence, we would also rather not experience the desire to do these acts. The sort of unwanted desires talked about here, is of course not about, say, eating fifteen cookies in one minute. Although this might be unwanted, it is not the type of unwanted looked for. Killing fifteen people in one minute, just for fun, is unwanted in the way meant here. How do our varieties of volitions lead us to actions? Frankfurt calls reasons that are motivating actions practical reasons. 73 Everybody has goals in life, and practical reason helps tell if our attempts to reach these goals will benefit from certain actions. Whether or not a reason, or a combination of reasons, is sufficiently motivating for an action depends on a combination of factors. Not every kind of reason has equal force. Wishes, things we care about, and things we love, are examples of different kinds of reasons with different forces. A goal is far more effective in motivating action than a wish, for we identify ourselves more with our goals. They are part of our personality. We will be more willing to put a lot of energy into reaching a goal we care much 69 H. F. Frankfurt, Importance, pp What we care about can and will change throughout live, meaning that if someone is a fully integrated self at some point, he will have to reevaluate and reorder his higher order volitions over time. H. F. Frankfurt, Importance, pp H. F. Frankfurt, Taking Ourselves Seriously, pp ; pp Frankfurt, Taking Ourselves Seriously, pp Ibidem. pp Frankfurt, Taking Ourselves Seriously, pp