Akrasia: A Conceptual History

Akrasia: A Conceptual History Although the term akrasia 1 may floor the vast majority of people, when one explains it as lack of self-control or glosses it as something related to the dictum ‘the way to hell is paved with good intentions’ it is immediately recognized to be very familiar and scarcely controversial. It is also well known and well researched and, according to a number of writers, it appears to be, at least, partly analogous to selfdeception.2 For, just as in self-deception, one is purportedly subject to opposing states of mind about what one should think, in akrasia, one is apparently in two states of mind about what one should do. Also, in both cases, there is also an associated sense of remorse which one feels when one ‘comes to oneself’. Indeed, similar solutions have been proposed for both, apparently paradoxical, phenomena. As in the case of self-deception, which has already been examined, the most attractive analysis has been some more-or-less attenuated form of the two-system description in which the situation is seen to be the result of the interaction of two different conceptual systems with one system deliberately concealing or misinterpreting the relevant principles or reasons to deceive the other system into taking the desired, but The Greek word akrasia means ‘lack of strength’ (). The manner in which this has been interpreted differently will be discussed below. 2 Such philosophers include Amélie Oksenberg Rorty, "Self-Deception, Akrasia and Irrationality," in The Multiple Self, ed. Jon Elster (Cambridge: Cambridge University Press, 1986), 115-131; "Akratic Believers," American Philosophical Quarterly 20 (1983): 175-184; David Pears, Motivated Irrationality (Oxford: Clarendon Press, 1984), 15-40; Alfred R. Mele, Irrationality: An Essay on Akrasia, SelfDeception and Self-Control (Oxford: Oxford University Press, 1987), viii-ix; L. Jonathan Cohen, An Essay on Belief and Acceptance (Oxford: Clarendon Press, 1992), 150; and Mary Forrester, "SelfDeception and Valuing Truth," American Philosophical Quarterly 39 (2002): 31-37. 1 Chapter 5: An Overview of Different Attempts to Explain Akrasia 225 unwise, option. Nevertheless, such a resolution cannot be uncritically accepted. For, at the heart of both the problems of self-deception, and of akrasia lies the fact that just one person is involved. The very remorse felt by persons who committed akratic actions indicates that they perceive themselves as the same person, with a single motivational history who, on a particular occasion or through a specific time-period, lost moral selfcontrol in an avoidable fashion. And it is this conceptual possibility of such quitepervasive occurrences that must be coherently and clearly characterized. This chapter would like to explore the case of akrasia, together with a variety of explanations that have been advanced in the literature. The itinerary will commence with a brief historical overview of the ascription of akrasia, followed by a presentation of the philosophical paradox of the classical kind, that appears to be inherent in the orthodox account, and which led some to negate the common ‘garden’ account of akrasia. A number of different attempts at resolving the paradox will then be reviewed with the aim of manifesting their shortcomings. The first account focuses on sidestepping the paradox by offering an explanation that abandons the analysis of the conflict in akrasia along the lines of an interpersonal incongruity of contradictory commands and sees it merely as an opposition between two prima facie guiding principles. However, such an account will be seen to have somewhat trivialized the concept under consideration. The second set of accounts attempts to circumvent the paradox by supposing that, indeed, there are different motivational systems at work, in the mind of the akratic, where the systems battle it out until one overrides the others. The problem here will be shown to be that such explanations attempt to resolve the paradox by an ignoratio elenchi: once again, what one seeks to know is, precisely, how akrasia could occur in a single person, not how the single person could be carved up on the model of two or more people. After having thus surveyed the landscape of explanations, the rather old account advanced by Aristotle will be presented. The Aristotelian elucidation, which will be examined largely through John McDowell’s interpretation, proposes to shift the emphasis towards the character of the moral agent, that is, towards his or her ‘conception’ of a good life, with rather dramatic results. Indeed, Aristotle’s account immediately shows the utility of looking closely at our use of the terms related to akrasia in our normal, common life. Chapter 5: An Overview of Different Attempts to Explain Akrasia 226 Such a way forward would be to consider and see under what conditions one would remorsefully say: ‘I am sorry for what I just did … it’s not me at all’ or ‘I did not mean to but … I just don’t know what came over me’, or even ‘How could I have …’. It indicates, once again, that the conceptual analytic method used by Ludwig Wittgenstein could well prove extremely fruitful. Indeed, it will be interesting to see whether this philosophical question is yet another expression of a web of false analogies or conceptual misunderstandings. 5.1 A Short History of the Use of the Concept of Akrasia Throughout the ages, there has always been the realization that one, quite often, may consciously and deliberately act wrongly. This has expressed itself in a number of ways all of which have a long history. Indeed, akrasia has been under discussion at least since the formulation of the so-called ‘Socratic Paradoxes’. The central one of these paradoxes is usually accepted to be the following: “ς”.3 This position, however, appears to fly in the face of the facts. As E. J. Lemmon argued: “It is so notorious a fact about human agents that they are often subject to a[k]rasia that any ethical position that makes this seem queer or paradoxical or impossible is automatically suspect for just this reason”.4 Aristotle seems to have, at least, partially “[N]one of the wise men considers that anybody ever willingly errs” (Plato, Protagoras 345D-E). All scholars appear to agree that such paradoxes exist. However there does not appear to be any agreement as to how many paradoxes there are, and as to what the paradoxes consist in exactly. David Gallop seems to think that there is only one: “one who knows good and evil will necessarily do the good and avoid the evil if he can” (David Gallop, "The Socratic Paradox in the Protagoras," Phronesis 9 [1964]: 117). Gerasimos Santas talks of two, which he calls the prudential and the ethical paradoxes: the former consists in that no one desires evil things and that all who pursue evil things do so involuntarily, and the latter is that virtue is knowledge and that all who do injustice or wrong do so involuntarily (see Socrates [Boston/MA: Routledge and Kegan Paul, 1982], 183-194); Norman Gulley mentions three: that virtue is knowledge, that no one does wrong willingly, and that all virtues are one (see Norman Gulley, The Philosophy of Socrates [London: Macmillan, 1968], 75-164); Michael J. O’Brien avers that there are six: that no one does wrong willingly, that no one wishes evil, that virtue can be taught, that virtue is an art, that virtue is knowledge, and that vice is ignorance (see The Socratic Paradoxes and the Greek Mind [Chapel Hill: University of North Carolina Press, 1967], 16). J. J. Mulhern (see "Aristotle and the Socratic Paradoxes." Journal of the History of Ideas 35 [1974]: 293-299) notes that Aristotle mentions two Socratic paradoxes in his Nicomachean Ethics. One is a conceptual paradox reported by Aristotle as the identification of the virtues with practical wisdom (see Aristotle, Nicomachean Ethics, VI, 1144b 1430); the other is an empirical paradox reported as the thesis that no one acts against what he judges best except by reason of ignorance (see Aristotle, Nicomachean Ethics, VII, 1145b 26-27). 4 E. J. Lemmon, "Moral Dilemmas," The Philosophical Review 71 (1962): 144. I agree, with Lemmon’s words as quoted here, though not with the facile spirit with which they are used for, as Davidson argued, 3 Chapter 5: An Overview of Different Attempts to Explain Akrasia 227 agreed with the protest that Socrates’ conclusion “plainly conflicts with the given”,5 and attempted, notably in the seventh book of his Nicomachean Ethics, to give an account of the way in which one knows when one is overcome by pleasure. His account is in terms of distinguishing various senses in which somebody is said to know something, thereby sustaining Socrates’ theory that cases of ‘akrasia’ are cases of forms of ignorance or forgetting. At the same time, however, he modifies the Socratic theory in order to bring it in line with what would commonly be said on the subject.6 These ‘Greek’ formulations perceive the problem of akrasia as one about the possibility of a conflict between reason and desire. The question here concerns the possibility or otherwise of acting against one’s own knowledge and in accordance with a certain kind of desire or .7 The philosophical problem concerns the presumption that, nothing ought to be stronger than knowledge, in the process of coming to a practical decision. It is interesting that, while the contemporary discussions on akrasia show lively interest in the ‘Greek’ formulations of the problem considered above,8 within the same debate there is very little concern with the concept as it has been considered in the period between Greek and contemporary philosophy. Indeed, the extremely scanty remarks only deal with what Paul, Augustine and Thomas Aquinas had to say on the matter. Now, this lack of investigation has been blamed on the alleged distinction one can know that one has made a mistake in one’s analysis without knowing what the mistake is and, therefore, the problem remains (HW 23). 5 Aristotle, Nicomachean Ethics, VII, 1145b 28. 6 See Gerasimos Santas, "Aristotle on Practical Inference, the Explanation of Action, and Akrasia," Phronesis 14 (1969): 184. 7 The distinction is made between desires for food, drink, sex and such like which concern and other passions such as anger, the desire for vengeance, the desire for fame, social status, wealth and the like which concern . These differ in the degree to which they are obedient to reason: the latter do, while the former do not, consist partly in views as to how it would be good to behave. In Aristotle’s view, akrasia proper or without qualification (), is seen to be a matter of conflict between reason and the former and not, say, between reason and anger. The latter are only called akrasia because of a similarity. Indeed while, in both cases, physical changes are produced in the agent that can result in the kind of bizarre ‘knowledge’ characteristic of the akratic proper, Aristotle holds that there are significant differences. For, temper, for example mishears reason while following it to some extent since deliberation plays a part in forming the result (see Aristotle, Nicomachean Ethics, VII, 1149a 25-b 3). In such a case, there is no suggestion that such an akratês suffers conflict and acts against his or her ; any dilemma would be within, not against, practical reason’s criteria. Hence, Aristotle would say that this ‘akratês’ succumbs to reason in a way, while the akratês proper acts contrary to his or her , against his or her practical reason’s criteria. (See Justin Gosling, Weakness of the Will [London: Routledge, 1990], 40-45.) 8 Indeed, Charlton notes that perplexity in the face of akrasia as a philosophical problem only revived with the renewed philosophical interest in Greek philosophy in the 1930’s. (See William Charlton, Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 9.) Chapter 5: An Overview of Different Attempts to Explain Akrasia 228 between the Greek and the Christian philosophy of mind.9 William Charlton, for example, sees this divide as indicating two conflicting views of human nature: whereas, in the Greek perspective, there is no purely volitional capacity, in the Judeo-Christian view, the human person is endowed with both a deliberative or reflective capacity, and an executive capacity.10 This divide meant that, while the Greeks saw akrasia as a real problem in need of justification, for the Judeo-Christian world, since it is the will that has the function of ensuring that one does what one should, akrasia turns out to be the standard form of failure. Given that the will is an autonomous principle that can violate the judgment of the intellect, the existence of what was called akrasia by the Greeks11 is obvious and philosophically unproblematic. For instance, the well-known Pauline complaint that “I do not do what I want, but I do the very thing I hate” 12 is seen by Charlton as an avowal of akrasia and as proof that its existence is taken as evident within the Judeo-Christian tradition.13 In further support of this position, he remarks that Augustine (354–430), whom he notes as being the first person to introduce the notion of the will into philosophy and who, therefore, initiated the movement of Christian voluntarism, sometimes viewed the will as a faculty of the mind and, at other times, saw the will as the mind itself in the role of issuing commands. Indeed, within such a word-view, any problem in wrongdoing would be couched in different terms from those of the Greeks. He quotes a passage from Augustine where the Doctor Gratiae asks himself how the mind can give itself orders which it does not obey: The mind gives the word commanding the hand to be moved; and such readiness there is, that the command is scarcely to be discerned from the execution.… The mind 9 See William Charlton, Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 5-6. 10 See Ibid., 177. Risto Saarinen complains against this common prejudice and seeks to show that the contributions of the mediaeval discussion are comparable to the offerings of twentieth century philosophers (see Risto Saarinen, "John Buridan and Donald Davidson on Akrasia," Synthese 96 [1993]: 133-154). 11 This qualification is necessary because the seventh book of the Nicomachean Ethics was only known to the mediaeval West in the 13th century. Nevertheless, while the Greek interpretation of akrasia was not discussed in Augustine’s time, the subject of ‘invitus facere’ or ‘reluctant actions’ was vigorously investigated, particularly due to the influence of heretical movements like Manichaeism and Pelagianism. 12 Rm 7, 15. 13 This passage is discussed by Richard M. Hare as a ‘constantly quoted’ counter-example to a prescriptivist position. (See Richard M. Hare, Freedom and Reason [Oxford: Oxford University Press, 1965], 78.) Note that the point is philosophical. Indeed, the authors who use the Pauline passage as a philosophical example are not normally interested in the contextual historical meaning of the passage. The other well-worn passage in literature, quoted extensively by Hare on the same page, is that of Medea who tries to resist the onset of love for Jason who betrayed her. She famously complains: “video meliora proboque, deteriora sequor” (“I see the better and approve it, but I follow the worse”). Ovid, Metamorphoses, VII, 20. Chapter 5: An Overview of Different Attempts to Explain Akrasia 229 commands mind to will; it is the same, and yet it does not. Whence is this portent, and to what purpose? I say it commands that itself would will a thing; which never would give the command, unless it willed it: yet it does not that, which it commanded.14 The problem here is the occurrence of ‘conflicting wills’ or the situation wherein the person wills something but also has a tendency to will contrary to what he or she wills. It is the Pauline struggle between spirit and flesh transposed into the doctrine of two inner and incomplete wills.15 This partial willing and partial non-willing “imperat animus, ut moveatur manus, et tanta est facilitas, ut vix a servitio discernatur imperium.… imperat animus, ut velit animus, nec alter est nec facit tamen. unde hoc monstrum? et quare istuc? imperat, inquam, ut velit, qui non imperaret, nisi vellet, et non facit quod imperat” (Augustine, Confessions, VIII, 9, 21). This passage is quoted by William Charlton (Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 5). The same passage appears to be used also by Richard M. Hare (Freedom and Reason [Oxford: Oxford University Press, 1965], 81). 15 This doctrine does, of course, suggest an inclination towards Platonic division of the person; however, his talk of the incompleteness of either will also shows evidence of a more Stoic view of the unity of the ruling principle of activity (). For the Stoics, since the  is the rational ruling element of the soul, all other faculties – the five senses, reproductive capacity, and utterance – are imbued with rationality. The sequence is as follows: the soul is affected by the world’s causally creating an impression on the senses (). This results in a presentation (). The soul, can, if it wants, assent () to this presentation, where such assent amounts to apprehension (). See Richard Joyce, "Early Stoicism and Akrasia," Phronesis 40 (1995): 317-318. A presentation may also result in a movement of the soul towards the object (), which in humans involves reason and assent (see Hans Friedrich August von Arnim, ed., Stoicorum Veterum Fragmenta, 2, §826 [Leipzig: K. G. Saur München, 2004], 226). Plutarch quotes Chrysippus as holding that “the  of man is reason prescribing action to him” (De Stoicorum repugnantiis, 1037F [= A. A. Long and D. N. Sedley, eds., The Hellenistic Philosophers, 2 vols. (Cambridge: Cambridge University Press, 1987), 53R]). This means that human passions are, in a fundamental sense, rational movements. This is not to say, however, that struggle and conflict are not possible. For the Stoics, “all forms of conflict are possible until the stage of assent to a hormetic presentation; but of course one cannot assent to two incompatible presentations at once”(Justin Gosling, "The Stoics and Akrasia," Apeiron, 20 [1987]: 199). The strongest example of struggle or conflict would be where a  does not issue in a successful outcome, since not even our bodies are under our control. The situation would concern a case where “our rational  [are] directed ineffectively at an action. This allows for the possibility that as a result of asserting to a presentation of the appropriateness of revenge Medea is subject to the physical effects which constitute the development of anger. Once these are set in train, all that her assent and the judgment that this murder shall be avoided can achieve is an ineffective internal effort to stop. Thus Chrysippus likens people overcome by passion to runners who cannot readily pull up (see Galen, De Placitis Hippocratis et Platonis, IV, 2. 10-18 [=Hans Friedrich August von Arnim, ed., Stoicorum Veterum Fragmenta, 3, §462 (Leipzig: K. G. Saur München, 2004), 113-114]). Thus passion can even be said to carry Medea contrary to her rational , though this would not, of course, be a clash of contemporaneous ; rather, what one  started, a later  would be insufficient to stop” (Justin Gosling, "The Stoics and Akrasia," Apeiron, 20 [1987]: 189190). While there are differences between Augustine and the Stoics in that the latter would see akrasia as a matter of a succession of different assents in rapid conversion (Plutarch, De Virtute Morali, VI, 446F447A; I must here note that this view is not far at all from the position expressed by Terrence M. Penner who holds that at the moment of akrasia a different aspect of the situation flashes itself at the agent, and takes him or her in so that he or she sees the bad choice as good: Penner speaks of “a kind of flip-flopping of aspects” ["Plato and Davidson: Parts of the Soul and Weakness of Will," Canadian Journal of Philosophy 16 Supplementary Volume (1990): 69; see also Terrence M. Penner, "Knowledge vs True Belief in the Socratic Psychology of Action," Apeiron 29 (1996): 200-201, 211-212.]), Augustine is in favour both of the Stoics’ view of the unity of the , and of their view that final assent is characteristically effective (so that the agent can always decide between different presentations and is, 14 Chapter 5: An Overview of Different Attempts to Explain Akrasia 230 is not seen as being so bizarre by Augustine; it is seen, rather, as a sickness of the mind which cannot rise with its whole self to truth because it is heavily overburdened with habit: [A] new will which I now began to have towards the free worshipping and enjoying of thee, O God, the only assured Sweetness; it was not able as yet to overcome my former wilfulness, now hardened in me by so long continuance, Thus did my two wills, one new and tother old, that carnal, and this spiritual, try masteries within me, and by their disagreeing wasted out my soul.16 Charlton imputes the same unquestioning acceptance of akrasia to Thomas Aquinas (1224/5–1274); the former accuses the Angelic Doctor of having been facile when he answered that: “It is clear to experience that many people act against what they know and this is confirmed by divine revelation in a passage of Luke: ‘The servant who knew the will of his master and did not do it will be beaten with many blows’”.17 The English philosopher seems to think that, once again, this ‘facile’ attitude is due to the Christian philosophy of mind that holds that the will is able to act against a rational directive. In fact, however, while Charlton’s explanatory division is intended to provide a justification for the purported lack of interest in akrasia after classical Greek philosophy, his presuppositions are skewed. For, already in Augustine, there is a real concern with the aberrant yet all-pervasive case of ‘reluctant actions’18 and, indeed, an extremely sophisticated discussion where the emphasis on the incompleteness or partiality of willing is intended to safeguard the unity of the ego: finally, responsible for his or her actions). See Justin Gosling, Weakness of the Will (London: Routledge, 1990), 70. 16 “voluntas autem nova, quae mihi esse coeperat, ut te gratis colere fruique te vellem, deus, sola certa iucunditas, nondum erat idonea ad superandam priorem vetustate roboratam. ita duae voluntates meae, una vetus, alia nova, illa carnalis, illa spiritalis, confligebant inter se atque discordando dissipabant animam meam” (Augustine, Confessions, VIII, 5, 10). 17 Thomas Aquinas, Summa Theologiae, I–II q.77 a.2. (See William Charlton, Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 6.) 18 It is important to note that the subject of ‘reluctant actions’ was discussed by Augustine, both with reference to doing good reluctantly (an example would be an agent’s doing something against her wishes only in order to avoid punishment [see Augustine, De Spiritu et Littera, VIII, 13, and XXXI, 53, 56]), where this bears some analogy to Aristotle’s enkrateia, and with reference to doing evil reluctantly, which concerns us here. Chapter 5: An Overview of Different Attempts to Explain Akrasia 231 Myself when sometime I deliberated upon serving of the Lord my God, (as I long had proposed) it was myself who willed it, and myself who nilled it; it was I myself. I neither willed entirely, nor yet nilled entirely.19 Within this framework, Augustine interpreted the Pauline dictum as meaning that sin is not a separation between one’s own mind, and some alien mind but a division, between the desires of the ego, imposed as a punishment for original sin. 20 Hence, within the ego, there is a battle between the two: there is a conflict between rational desire and sinful concupiscence,21 which is not unlike the Greek conception. There are differences however: unlike the case of the Greeks, there is little concern with philosophical baggage refuting the possibility of akrasia. In Augustine’s view, which must be seen against the background of the debate on original sin, which dominated the discussion at the time, the will is continuously enticed to choose wrong goals and means. His concern is to insist that the agent chooses22 even ‘reluctant actions’ and, therefore, must be regarded as wholly responsible for his or her consent. This does not mean, however, that the doctor from Hippo forsook completely the Greek view of the human person as a rational being. Indeed, he can, at most, be attributed with an indirect “ego cum deliberabam, ut iam servirem domino deo meo, sicut diu disposueram, ego eram, qui volebam, ego, qui nolebam; ego eram. nec plene volebam nec plene nolebam. ideo mecum contendebam et dissipabar a me ipso” (Augustine, Confessions, VIII, 10, 22). See Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan (Leiden: E. J. Brill, 1994), 29. 20 See Augustine, Confessions, VIII, 10, 22. 21 For Augustine, concupiscence is an improper sort of desire – above all, a form of sexual desire – that arises as the result of, and the punishment for, original sin. As such it affects all humankind. It is also the desire that results in the defective condition of the human mind, and of the consequent neglect of the proper goals, a neglect that is unavoidable without the aid of grace. Hence, in Augustine’s theory, weakness of the will has equivocal meanings: it describes both the ontological condition of fallen human nature, and actual akratic acts, where the first is not morally culpable – after baptism –, while the second entails moral blameworthiness (see Ann A. Pang-White, "The Fall of Humanity: Weakness of the Will and Moral Responsibility in the Later Augustine," Medieval Philosophy and Theology 9 [2000]: 51–67). It is interesting to note that Augustinian continentia meant chastity, with incontinentia therefore referring to sexual vice. Thus the Augustinian and the Aristotelian continentia, albeit not coinciding perfectly, became linked through a historical coincidence: the later discussion of the Aristotelian problem of incontinentia was much influenced by the Augustinian view that no human could be continent without divine aid. The difference with Aristotle came to be that many mediaeval interpreters of Aristotle’s Ethics came to see incontinence as a vice distinguished from intemperance only by its subject matter (the latter meant restraint in eating and drinking as contrasted with the former meaning restraint in sexual matters). 22 The terms used are ‘eligere’, ‘consentire’ and ‘arbitrari’ (see Augustine, De Libero Arbitrio III, 7-8; Confessions, VIII, 5, 10-12; VIII, 9, 21-10, 24). Although, the Latin liberum arbitrium to some extent renders the Greek prohairesis (see Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan [Leiden: E. J. Brill, 1994], 24fn.16), Augustine’s notion of choice is different from the Aristotelian use of prohairesis since the agent’s liberum arbitrium leaves space for the possibility where judgment concerning what is best does not issue in pursuit; Aristotle’s prohairesis does not comprise the power of acting or not acting (see Ibid., 36fn.55). The doctrine that the agent chooses even his or her reluctant actions would, however, be in conflict with Aristotle’s since the latter insisted that the akratês does not attain prohairesis or else ignores it; the akratês’ actions are voluntary but not chosen. 19 Chapter 5: An Overview of Different Attempts to Explain Akrasia 232 voluntarism, where the person freely and knowingly consents to a blameworthy action and is responsible for this choice. He insists that this can only happen because of serious defectiveness and ignorance – due to sin – in the agent’s mind.23 In Aquinas, the concern is also to establish the responsibility of the akratês – whom the Latin Scholastics called incontinens. His treatment of incontinentia is true to the spirit, if not the letter, of Aristotle’s account and this, despite the differences between the two deriving from the role awarded, by the Angelic Doctor, to the will. According to Kent,24 Aquinas considered that the incontinens does choose his or her aberrant act, and does so because he or she temporarily sees it as good: what applies to the incontinens is peccare eligens rather than peccare ex electione.25 Hence, the 23 See Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan (Leiden: E. J. Brill, 1994), 42. 24 See Bonnie Kent, "Transitory Vice: Thomas Aquinas on Incontinence," Journal of the History of Philosophy 27 (1989): 199-223. Not everyone agrees with Kent concerning this ‘Augustinian’ interpretation of Aquinas. Some, like, Rega Wood disagree that Thomas proposes that the incontinent act is chosen (see Rega Wood, "Willing Wickedly: Ockham and Burley Compared," Vivarium 37 [1999]: 81), others like Risto Saarinen maintain that the Thomist account is an Aristotelian one on the key point that incontinence results from ignorance. Saarinen proposes a ‘two-step’ explanatory interpretation where the first step is a failure of actual knowledge regarding the prohibiting syllogism, and the second step is the completion of the misleading syllogism. According to the latter, perverted choice is only involved in the second step (see Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan [Leiden: E. J. Brill, 1994], 126). 25 Although there is no substantive difference between Aquinas’ different treatments of the reasoning of the incontinent, the use of this distinction to claim that, although the incontinent does not act from choice (peccare ex electione or secundum electionem), he or she chooses to act (peccare eligens) as he or she does, constitutes a development of Aristotle’s doctrine and is only to be found in Thomas’ theological works (see amongst others Thomas Aquinas, De Malo, q.3 a.12 ad 11: “[D]icendum, quod etiam in peccato infirmitatis potest esse electio; quae tamen non est primum principium peccandi, cum causetur ex passione; et ideo non dicitur talis ex electione peccare, quamvis eligens peccet”; Thomas Aquinas, Summa Theologiae, I–II q.78 a.4 ad 3: “[D]icendum quod aliud est peccare eligentem, et aliud peccare ex electione. Ille enim qui peccat ex passione, peccat quidem eligens, non tamen ex electione, quia electio non est in eo primum peccati principium, sed inducitur ex passione ad eligendum id quod extra passionem existens non eligeret. Sed ille qui peccat ex certa malitia, secundum se eligit malum, eo modo quo dictum est [aa. 2–3]. Et ideo electio in ipso est principium peccati; et propter hoc dicitur ex electione peccare”). The distinction is between an act committed in accordance with the agent’s moral disposition signified by the universal premiss and hence reflecting the agent’s opinion of what is good – which would constitute intemperance – and an act where the error, albeit chosen, is not the actualization of a misguided disposition. It is interesting to note that Aquinas saw Aristotle’s incontinent as judging his or her action to be good (see Thomas Aquinas, Commentary on Aristotle's Nicomachean Ethics, III, lect. 3 [Marietti division §412]; lect. 13 [§§519-520]; VII, lect. 2 [§1314]; lect. 8 [§1430]). Indeed, Kent maintains that the 13th century scholastic saw his peripatetic predecessor as being correct yet incomplete in that he did not draw the conclusion that the incontinent act is chosen. For he does attribute to his predecessor the denial that the incontinent acts ex electione yet he does not find evidence of the claim that the incontinent sins eligens. It is significant here that, in his commentary on 1111b 13-15, which claims that the incontinent does not act eligens, Thomas substitutes non secundum electionem for non eligens (see Thomas Aquinas, Commentary on Aristotle's Nicomachean Ethics, III, lect. 5 [Marietti division §439]) thereby interpreting the Latin translation of Aristotle’s work concordantly with his own doctrine. Here it is also important to note that the scholastics’ use of the Latin peccare broadly signified ‘going wrong’. It also meant an offence against God, but since all acts contrary to nature or right reason are Chapter 5: An Overview of Different Attempts to Explain Akrasia 233 incontinent’s sin is less serious that the intemperant’s – who sins ex electione – for the incontinent only sins in that he or she falls into a transitory ignorance that could have been avoided.26 This means that, while the akratês is responsible for the incontinent action his or her action is partially excusable. Also, despite upholding the possibility of the will’s choosing the akratic action as good by following the akratic ‘syllogism’, the Angelic Doctor cannot be seen as an out-and-out voluntarist. Although he views akratic actions as chosen, he is consistent with the Aristotelian dictum that choice pertains only to the means and with the consideration of the will as a rational appetite in stressing that an essential factor in akrasia is one’s ignorance of the right minor premiss due to passion which partially blinds right reason. Hence, the alternative akratic syllogism is only introduced due to ignorance. Charlton’s imputation, to Aquinas, of a so-called Judeo-Christian view of the will as completely autonomous of reason does not hold. When it comes to the mediaeval Franciscan stream, there appears to be less room to combat Charlton’s position about Judeo-Christian voluntarism. It is curious, however, that the “Franciscan literature on incontinence can only be described as voluminous”.27 The reason may well be that, despite the well-diffused view of the differences between the Franciscans and the Thomists, and despite the fact that the more radical Franciscans did hold that the incontinent person wills to act against the judgment of reason, the majority of Franciscan writers did accept the ‘intellectualist’ view that ignorance is, to some extent, responsible for the evil person’s acts. offences against God, they held that a moral philosopher could consider peccatum as an act contrary to reason (see Thomas Aquinas, Summa Theologiae, I–II q.71 a.6 ad 5). 26 Note that Judith Barad interprets Aquinas’ explanation of incontinence using the assent–consent distinction; that is, that the incontinent’s assent to the universal, and therefore abstract, premiss is right but what is lacking is the agent’s consent to the immediate, particular means in the realm of action, where such a situation occurs when the person consents to what is desired because the desires indirectly move the will by distracting it from what would otherwise be willed in accordance with the general assent. The choice is the agent’s however, since the will must consent to the lower appetite in order for the action to be performed – apart from the consent of the will the desire remains a mere inclination. In the light of this interpretation, Barad sees Aquinas’ interpretation of incontinence as being a case where the will fails to control the desires. (See Judith Barad, "Aquinas' Assent/Consent Distinction and the Problem of Akrasia," New Scholasticism 62 [1988]: 98-111.) Her interpretation seems to me, however, to postulate splits in the agent – between intellect, will and desires – that are much wider than Aquinas would admit to. She also appears to hold an excessively mechanistic view of human action that Aquinas would not maintain. 27 Bonnie Kent, Aristotle and the Franciscans: Gerald Odonis’ Commentary on the Nicomachean Ethics (unpublished dissertation, 1984), 295, as quoted in Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan (Leiden: E. J. Brill, 1994), 7. Chapter 5: An Overview of Different Attempts to Explain Akrasia 234 Interestingly, most – with the exception of Gerald Odonis – agreed with Aquinas that the akratic person acts out of choice.28 It will be clear from this rapid overview that akrasia ceases to be a philosophical problem in two conceptual cases. One concerns the scenario where reason and appetition are thought to be separable. If one ceases to see reason as a kind of desire and appetition as a kind of cognition; if one holds that the will is merely moved by motivational factors like desires, what Locke called ‘uneasinesses’,29 or by factors like Hume’s two principles – the calm desires and tendencies of benevolence and a general appetite to good, or violent passions –, then the philosopher does not need to say much about akrasia. Indeed, the problem posed by akrasia is only understandable if one sees that there is a logical link between wanting to do something and thinking that it would be good to do something. If one were to think that “[t]his order of things, abstractly consider’d, is not necessary … I see no contradiction in supposing a desire of producing misery annex’d to love, and of happiness to hatred”,30 akrasia would never have created any perplexity.31 28 These comments on the Franciscan views are based on the work of Bonnie Kent in her abovementioned dissertation (see Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan [Leiden: E. J. Brill, 1994], 7). 29 See John Locke, "Essay Concerning Human Understanding," Book II, Chapter xxi, Sections 29, 35, in The Works of John Locke, I (London: C. and J. Rivington, 1824), 236, 239-241. For Locke, contemplation of what might be good lacks all practical force since ‘uneasiness’ has no link with evaluation; it is rather a function of anger, fear, envy, shame, hunger, and is directly proportional to pain (see Ezio Vailati, "Leibniz on Locke on Weakness of Will," Journal of the History of Philosophy 28 [1990]: 215). Locke is an externalist; he denies any essential connection between evaluation and motivation, and consequently holds that the judgment that something is good elicits attention only if one happens to care about it (see Michael Bratman, "Practical Reasoning and Weakness of Will," Noûs 13 [1979]: 158-159). 30 David Hume, A Treatise of Human Nature, Book II, Part ii, Section 6, ed. L. A. Selby-Bigge with text revised and variant readings by P. H. Nidditch, 2nd ed. (Oxford: Clarendon Press, 1980), 368. 31 Another clearly externalist thinker is Robert Dunn, who develops an argument proposed by Michael Stocker. In a paper, the latter had maintained that “motivation and evaluation do not stand in a simple and direct relation to one another…. Rather, they are interrelated in various and complex ways, and their interrelations are mediated by large arrays of complex psychic structures, such as mood, energy, and interest” (Michael Stocker, "Desiring the Bad: An Essay in Moral Philosophy," The Journal of Philosophy 76 [1979]: 738-739). The upshot is that “given such moods and circumstances [as accidie, tiredness, illness, despair, and so forth], harming another can be the proper and direct object of attraction. There is no need to posit another object, especially not an egoistic object like pleasure, power over others.… Just as there are desires and appetites directed at harming others, there are desires and appetites directed at harming oneself. In certain moods, such as the self-directed modes of disgust, hatred, guilt, shame, I may seek to humble, abase, or harm myself. Agents, even in the planning and doing of such acts, and certainly afterwards, can believe or know that what is desired is bad” (Ibid.: 748-749). Similarly, Robert Dunn argues that “all-out present-tense summary evaluative thinking about one’s own action is such as to allow any full-fledged instance’s complete dissociation from any accordant volition to act” (Robert Dunn, The Possibility of Weakness of Will [Indianapolis/IN: Hackett Publishing Company, 1987], 13). As he writes repeatedly, “[j]ust as, in certain moods or frames of mind, we can be indifferent or hostile toward another and, hence, not have any concern for or interest in (promoting) the good or interest Chapter 5: An Overview of Different Attempts to Explain Akrasia 235 The other case where akrasia would also be thought to be unperplexing would occur if one were to think that one’s behaviour were caused mechanistically. Indeed if one were to extend a mechanistic explanation like Descartes’ view of animals, as natural machines, to human beings, as Baruch Spinoza (1632–1677) did when he sustained what could be called a token-identity theory of mental and physical events,32 there would be no problem about akrasia.33 Indeed: [I]t would be just as absurd for a circle to complain that God has not given it the properties of a sphere, or a child who is tormented by a stone, that He has not given him a healthy body, as for a weak-minded man to complain that God has denied him strength … and that He has given him a nature so weak that he cannot restrain or moderate his desires.… [T]hat it does not belong to the nature of each man to be strong-minded and that it is no more in our power to have a sound body than a sound Mind, no-one can deny, unless he wishes to contradict both experience and reason.34 Given such a scenario, however, the question is simple: how and why would one ever distinguish between the vicious, the akratic, and the one who behaves compulsively? 5.2 The Paradox of Akrasia There is no doubt that the term akrasia is a specialist one. In addition, as will be seen, it is used in a specialist sense in philosophical contexts. It is important to note, however, that this specialist term could be taken to cover a wide range of cases. Such situations could include procrastination, dilatoriness, lack of courage, failure to follow through one’s plans, choosing what is clearly the worse, or even the utterly inappropriate course, under the influence of passion. What seems to be common to such of the other, so too, we can, in certain moods or frames of mind, be indifferent or hostile towards ourselves and, hence, not have (the promotion of) our own interest or good as an end” (Ibid., 117; see also pages 20, 111, 122). In this light, of course, akrasia is not a problem at all: “there is no logical problem about the possibility of weakness of will … because it is perfectly coherent for us to be unintegrated in a certain way, namely, not to have any volitions to act that correspond to our (felicitous) all-out present-tense summary evaluations about our own actions” (Ibid., 125). Or, to put it differently, “[i]f weakness of will … is problematic, so is strength of will” (Michael Stocker, "Desiring the Bad: An Essay in Moral Philosophy," The Journal of Philosophy 76 [1979]: 753). 32 See Baruch Spinoza, "Ethics," III.2.Sch., in Philosophy of Benedict de Spinoza, trans. R. H. M. Elwes (New York: Tudor Publishing Co., n.d.), 130-134. 33 Given these two reasons, perhaps it is not surprising that philosophy in the modern era has so very little to say about the subject: the way a number of philosophers saw passion, desire and action ensured that no problem was visible. 34 Baruch Spinoza, "Letter LXXVIII," in The Correspondence of Spinoza. Translated and edited with Introduction and Annotations by A. Wolf (London: Frank Cass and Co., 1966), 357-358. Chapter 5: An Overview of Different Attempts to Explain Akrasia 236 ‘garden’ varieties of cases is that there is an element of moral opprobrium involved. Perhaps the passage in Thomas Macaulay’s essay on Sir James Mackintosh, where the author criticizes James II for believing naïvely that the Tories would support his plans because they saw resistance to the royal will as sinful, could serve well to express the familiar view of what is involved in akrasia: Only imagine a man acting for a single day on the supposition that all his neighbours … act up to all that they believe. Imagine a man acting on the supposition that he may safely offer the deadliest injuries and insults to every body who says that revenge is sinful; or that he may safely intrust all his property without security to any person who says that it is wrong to steal. Such a character would be too absurd for the wildest farce…. It might, one should think, have crossed the mind of a man of fifty, who had seen a great deal of the world, that people sometimes do what they think wrong.35 Now, in the view of most people, such an occurrence, of someone knowingly doing what he or she thinks wrong, would be a perfectly possible and, indeed, quite a recurrent feature of their own lives. It would be useful, here, to examine just why quite a number of philosophers have been reluctant to accept common ways of describing and explaining akrasia at their face value. Of course, the first and, perhaps, best-known sceptic was none other than Socrates (470–399 428–347 BC) BC). In the Protagoras,36 the Meno37 and the Gorgias,38 Plato (c. presents Socrates as showing the absurdity of the common opinion that one can desire what is evil.39 In the Protagoras40 he first represents the common 35 Thomas Macaulay, "Sir James Mackintosh," in Macaulay's Critical and Historical Essays, II (London: Longmans Green and Co., 1895), 255-257. 36 See Plato, Protagoras, 351-358. 37 See Plato, Meno, 77B-78A. 38 See Plato, Gorgias, 466-468. 39 The question of which Platonic dialogues present the historical views of Socrates and which merely use Socrates’ character as a speaker in the dialogue is a long and vexed one. It is useful to note that Aristotle believes that he is entitled to distinguish between the two: when discussing the character he normally uses the form ‘the Socrates’ (); when he speaks of the historical person he calls him ‘Socrates’without the article (see Terence H. Irwin, Plato's Ethics [Oxford: Oxford University Press, 1995], 8-9). Moreover, Aristotle notes that Socrates denied the possibility of akrasia (see Aristotle, Nicomachean Ethics, VII, 1145b 22-31; VII, 1147b 13-17; Magna Moralia, II, 1200b 25-32). He also criticizes Socrates for defining all virtues as instances of knowledge (see Aristotle, Nicomachean Ethics, VI, 1144b 17-30; Eudemian Ethics, VII, 1246b 32-36; Magna Moralia, I, 1198a 10-15), and for exaggerating the role of theoretical knowledge in ethics (see Aristotle, Eudemian Ethics, I, 1216b 3-25; Magna Moralia, I, 1183b 8-11). In addition, study of Plato’s dialogues, both on the level of philosophical content, and on that of stylometric investigations (see Leonard Brandwood, The Chronology of Plato's Dialogues [Cambridge: Cambridge University Press, 1990]), is consonant with the difference Aristotle points out between the historical figure and the Platonic character. While giving priority to Aristotle’s views, one can also consider further sources of information about Socrates. For, Xenophon too holds the thesis ‘virtue is knowledge’ as being a fundamental Socratic position (see Xenophon, Memorabilia, III, ix, Chapter 5: An Overview of Different Attempts to Explain Akrasia 237 opinion that “that many, while knowing what is best, refuse to perform it, though they have the power.… and whenever I have asked what can be the reason of this, they say that those who act so are acting under the influence of pleasure or pain, or under the control of one of the things I have just mentioned”.41 Socrates, shows this view to be untenable by first identifying the pleasure with the good and arguing: Let us then lay it down as our statement, that a man does evil in spite of knowing the evil of it. Now if someone asks us: Why? we shall answer: Because he is overcome. By what? the questioner will ask us; and this time we shall be unable to reply: By pleasure – for this has exchanged its name for “the good.” So we must answer only with the words: Because he is overcome. By what? says the questioner. The good – must surely be our reply. Now if our questioner chance to be an arrogant person he will laugh and exclaim: What a ridiculous statement, that a man does evil, knowing it to be evil, and not having to do it, because he is overcome by the good!42 The conclusion is obviously that: [N]o one willingly goes after evil or what he thinks to be evil; it is not in human nature, apparently, to do so – to wish to go after what one thinks to be evil in preference to the good; and when compelled to choose one of two evils, nobody will choose the greater when he may the lesser.43 5). According to Xenophon, Socrates also maintains that it is always the case that people choose, from possible courses of action, what they conceive to be the best for them, and act accordingly (see Xenophon, Memorabilia, III, ix, 4). 40 There are differences between the ‘Protagoras’ formulation and the formulation of this paradox in later dialogues (see Norman Gulley, The Philosophy of Socrates [London: Macmillan, 1968], 119-121). For the ‘Protagoras’ formulation assumes that if a person voluntarily does what is, in fact, wrong, he must invariably be doing what he believes to be right; if he acts contrary to what he either knows or believes to be right, then he must be acting involuntarily. Beginning with the Gorgias, an alternative formulation, which remains the standard interpretation thereafter, is used. Here, the point that no one does wrong wishing to do wrong (see Plato, Gorgias, 509E) is reiterated, but it is coupled with the claim that only what is really good for the agent – and not what seems good to him – is an object of wish (see Plato, Gorgias, 466D-468E). Of course, such an interpretation restricts doing wrong to what is objectively wrong and accordingly restricts the term ‘voluntary’ to actions that are ‘really’ good (and correspondingly ‘involuntary’ to actions that are ‘really’ bad’). It is significant that Aristotle presents Socrates’ thought in terms of the ‘Protagoras’ formulation (see Aristotle, Nicomachean Ethics, VII, 1145b 21-31) and nowhere does he suggest any different Socratic formulation. This view is also confirmed by Xenophon (see Xenophon, Memorabilia, III, ix, 4; IV, vi, 6, 11). It is on these grounds that Norman Gulley maintains that the interpretation of the thesis in the Protagoras represents the original Socratic interpretation (see Norman Gulley, "The Interpretation of 'No One does Wrong Willingly' in Plato’s Dialogues," Phronesis 10 [1965]: 91-96). 41 Plato, Protagoras, 352E. 42 Plato, Protagoras, 355C. 43 Plato, Protagoras, 358C-D. Chapter 5: An Overview of Different Attempts to Explain Akrasia 238 Socrates is here taking ‘pleasure’ as the “all-purpose predicate of favourable assessment”.44 His argument need not necessarily be seen as an endorsement for hedonism;45 his premiss is rather that the good of the chosen and the rejected alternative is of the same kind. And, indeed, if things were so, one could have no reason, however bad, for choosing a course of action that is worse than another would be. Indeed, for Socrates, the weak agent really suffers from a sort of evaluation illusion, very much like an optical illusion: in reality he or she lacks knowledge of good and evil, that is, he or she lacks the art of weighing nearer and further goods against one another. Socrates’ primary motivation is to uphold the value and power of knowledge, where knowledge is a matter of calculating what it would be good to do, whereas passions are beliefs about what it is good to do now. Since both concern what it would be good to do, there is no conflict between the two: it is simply a matter of measuring what it would be better to do in the situation. Indeed, the whole issue is really a question of wisdom; the untutored might short-sightedly be duped into taking the short-term advantage as being the better alternative. On this view, the person who knows the best way to act will always act in that way insofar as it is within his or her power: Socrates was a Leibnizian. He would say that if one were to describe oneself as being overcome by pleasure or pain one would not be giving a correct account of what happened since there is no place for a conflict of opposing maxims within the agent’s mind. In short, akrasia does not exist since, if one knows what is right and good, one will do it. Hence, Socrates’ thesis could be extrapolated46 to maintain that normal accounts of akrasia are not correctly described by invoking conflict within oneself where one’s mind was set on one course and one’s passion on another, conflicting course that 44 David Wiggins, "Weakness of Will, Commensurability and the Objects of Deliberation and Desire," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty (Berkeley/CA: University of California Press, 1980), 255. Indeed, the core of Socrates’ doctrine on the virtues is that each is a special case of wisdom – that is, knowledge of good and evil – applied to specific contexts. Hence, the virtues are, in an important sense, one. Indeed, the discussion in the Protagoras forms part of a sustained argument answering – negatively – to the question as to whether courage is a virtue distinct from the others. 45 See Donald J. Zeyl, "Socrates and Hedonism: Protagoras 351B-358D," Phronesis 25 (1980): 250-269. Note, however, that there have been heated scholarly debates on this issue, with many eminent interpreters of Plato’s works maintaining that a principle of hedonistic prudence which combines ethical and psychological hedonism is needed to give backing to the Socratic doctrine that no one will choose what he knows or believes to be the lesser of the available goods. (See Terence H. Irwin, Plato's Moral Theory [Oxford: Clarendon Press, 1977], 105-106.) 46 This is an extrapolation because Socrates holds that once one possesses a skill one always exercises it without error. Chapter 5: An Overview of Different Attempts to Explain Akrasia 239 eventually won out. At most, such cases could be situations where one has changed one’s mind. For one always acts voluntarily on what one thinks best at the time. Other, more recent, accounts endorse Socratic scepticism about the common account of akrasia even though they propose different reasons for this. They would hold that the common description of akrasia is nonsensical if it means that one holds contradictory value-judgments and argue that the alleged failure in akrasia could only be explained as a situation where the agent makes an insufficient effort to resist contrary desires. They allege that what this really boils down to is, either a change in judgment, or a misjudgment of the amount of effort required. Naturally, this raises problems about the possibility of distinguishing between akrasia and actions committed due to compulsion. And indeed, these accounts conclude that the akratês is unable to resist his or her desires, at the time of action. They may liken akrasia to negligence in that it is non-voluntary, yet blameworthy, and attempt to save the intentionality of the akratês’ action by arguing that even compulsive behaviour can be intentional despite being unfree.47 Alternatively, they would hold that if the agent can offer no reasons for the errant action and “if the available resources for resistance failed, it would be arbitrary to insist that the desire was resistible on the occasion and his action clearly voluntary”;48 “as everything stood the desire does seem to have been as good as irresistible by him then”.49 Similarly, the prescriptivist – who maintains strongly that ‘ought’, in the fullest sense, implies ‘can’50 – would insist that, from the start, since human moral language is the language of weak mortals, the expression ‘I think that I ought’ has the potentiality of being used in an attenuated sense without really equivocating: all along the meaning of the word includes the possibility that one could backslide. 51 This means that one may make a moral judgment and find that it is not in one’s physical or psychological power to act on one’s self-addressed prescription. When it is used in this off-colour way with less than universal prescriptive force, a guilty feeling which is not too strong – so as not to push one into performing the action (!) – steps in to restore the appearance of 47 See Gary Watson, "Skepticism about Weakness of Will," The Philosophical Review 86 (1977): 333335. 48 David Pugmire, "Motivated Irrationality, Part II," The Aristotelian Society Supplementary Volume 56 (1982): 189. 49 Ibid.: 188. 50 Richard M. Hare, The Language of Morals (Oxford: Clarendon Press, 1952), 165-170. 51 See Richard M. Hare, Freedom and Reason (Oxford: Oxford University Press, 1965), 76-77. Chapter 5: An Overview of Different Attempts to Explain Akrasia 240 universal prescriptivity. When the impossibility is psychological, the feeling of remorse is even more apt since, even though it is unable to overcome the particular temptation, it keeps alive the will-power that could overcome lesser ones.52 Once again, the akrasia is seen to be a disguised form of ‘ought but can’t’; it is viewed mainly as a species of compulsive behaviour.53 In these more recent accounts, the point throughout is that, at the time of the action, no distinction could be made between akratic and compulsive action. In both cases, one is unable to conform one’s behaviour to one’s judgments. The general differentiating feature between the two, at most, would be that in akrasia there is the negligent failure on the part of the agent to cultivate reasonable and normal capacities of self-control,54 whereas in compulsion, not even the possession of such normal capacities would enable the agent to resist. One must here note that the relativity involved in this viewpoint means that “[one] could easily be mistaken about whether [one’s] action was weak or compulsive, because [one] could be mistaken about what the relevant standards are. Indeed, in the absence of shared expectations and norms, the distinction might come to lose its force altogether”.55 To sum up, according to these views, the charge against the common understanding of akrasia is that actions contrary to one’s better judgment are not free. One could not have done otherwise at the time of the action. This rapid overview clearly shows that a number of writers have found considerable difficulty in accepting the common description of akrasia. Indeed, they complain that, in reality, it is incoherent. Their feeling is that any account must be impaled on one or another horn of a dilemma. One horn is clearly the difficulty in conceptually distinguishing akratic from compulsive behaviour. Nonetheless, negating the distinction would be a matter of 52 See Richard M. Hare, Freedom and Reason (Oxford: Oxford University Press, 1965), 76-77, 80, 83. See also Richard M. Hare, "Weakness of Will," in Encyclopedia of Ethics, ed. L. C. Becker and C. B. Becker (London: Routledge, 2001), 1790-1791, particularly cases (4) and (4a). The other cases Hare countenances are situations where the prescriptive force of ‘ought’ is somehow weakened, or rendered ineffective. 54 ‘Reasonable and normal capacities’ here refers to the “capacities and skills of resistance which are generally acquired in the normal course of socialization and practice, and which we hold one another responsible for acquiring and maintaining. (See Gary Watson, "Skepticism about Weakness of Will," The Philosophical Review 86 [1977]: 331-332.) 55 Ibid.: 333fn.16. 53 Chapter 5: An Overview of Different Attempts to Explain Akrasia 241 deflating succumbing to temptation into losing control of oneself. John Austin’s (1911– 1960) comments about this are apt and well-known: I am very partial to ice cream, and a bombe is served divided into segments corresponding one to one with persons at High Table: I am tempted to help myself to two segments, thus succumbing to temptation and even conceivably (but why necessarily?) going against my principles. But do I lose control over myself? Do I raven, do I snatch the morsels from the dish and wolf them down, impervious to the consternation of my colleagues? Not a bit of it. We often succumb to temptation with calm and even with finesse.56 The other horn of the dilemma is just as unforgiving for, just as Socrates would have pointed out, if an agent were to choose pleasure voluntarily then the agent has to conceive of it as good. Indeed, it seems ridiculous and even paradoxical that someone could choose pleasure without somehow conceiving of it as good. Hence, the akratic agent must be seen as acting for a reason even if he or she knows that it is the wrong sort of reason. The problem arises in that such an agent not only does not act according to his or her own knowledge; he or she knowingly acts against his or her better knowledge. Once again, however, one cannot accept such sceptical conclusions lying down for they seem to coerce us into denying an extremely common occurrence: we all have found ourselves yielding to temptation in some way. Indeed, “… there is often much in an agent’s own attitude … of detestation of the choice made and of explicit selfreproach that can actually accompany both the choice and the action (‘Here I go again’)” that suggest akrasia, for “[p]eople do appear able to act knowing that they are letting themselves down, that the world is not well lost for this (‘I regret this already’)”.57 The issue, of course is that, if akrasia were merely a question of one’s being unable to fulfil some obligation through some physical or psychological disability, then remorse, criticism, and disapproval would lose their point and any punishment would be positively unjust. To be sure, the sceptical accounts of what would commonly be called akrasia do not correspond with what we want to say: while compulsive behaviour, in general, is not of great practical importance, the opposing sceptical account only provides for a straightforward change of mind. 56 John L. Austin, "A Plea for Excuses," in Philosophical Papers, ed. J. O. Urmson and G. J. Warnock (Oxford: Oxford University Press, 1976), 198fn. 57 David Pugmire, "Motivated Irrationality, Part II," The Aristotelian Society Supplementary Volume 56 (1982): 180. Chapter 5: An Overview of Different Attempts to Explain Akrasia 242 Yet, the paradox is a serious one, since akratic actions typically resist explanation. For, one either explains the occurrence of akrasia by invoking factors that are not reasons, with the upshot that the explanandum loses its character as an action, or one explains it by giving reasons and the akratic action then loses its character as an aberrant action. The challenge is to provide a coherent yet adequate description of the familiar fact of akrasia. 5.2.1 A ‘Moderate’ Account of Akrasia In the face of the conundrum presented by the paradoxical nature of akrasia from an analytic point of view, a number of philosophers have felt that the only way out of the aporia is to construct a conceptual solution that shows that something is wrong with the akratês’ state of mind without the aberration being logically impossible. One moderate solution is to abandon some feature of the analysis of akrasia along the lines of conflicting and irreconcilable maxims. Perhaps the best recent ‘moderate’ account using such a strategy is that proferred by Donald Davidson. In an exceedingly well-known treatment of akrasia “How is Weakness of the Will Possible”, published in 1969,58 Donald Davidson strived to show, both that a noncontradictory description of akratic actions can be given, and that such actions are essentially surd. His strategy is to show that akratic actions are caused and justified by reasons, albeit not by the proper – that is, the best – reasons. His view on the subject appears to be that actions are either caused and justified by the best reasons – and so are rational – or they are not, and so are essentially surd even to the agent himself (HW 42). Davidson presents the problem of akrasia against the background of the conjunction of the following two principles which, he notes, have the air of being self-evident: 58 P1 If an agent wants to do x more than he wants to do y and he believes himself free to do either x or y, then he will intentionally do x if he does either x or y intentionally. P2 If an agent judges that it would be better to do x than to do y, then he wants to do x more than he wants to do y. See Donald Davidson, "How is Weakness of the Will Possible?" in Essays on Actions and Events (Oxford: Oxford University Press, 1982), 21-42. Chapter 5: An Overview of Different Attempts to Explain Akrasia 243 He goes on to note that the problem shows its ugly head when one tries to put these two principles with the persuasive assumption that: P3 There are incontinent actions. (HW 23) In order to show that akrasia is logically possible, Davidson exploits the distinction between conditional or prima facie statements, and sans phrase statements. The former hold, other things being equal; they hold relative to the principles or reasons that ground them.59 The latter hold unconditionally.60 Now, prima facie statements cannot be detached from the principles or reasons that are their ground and, as such, are insulated from action: they are practical only in their subject not in their issue (HW 39). We are told that, in this respect, the situation is like that of reasoning from probabilistic evidence. The probability that it will rain, given that the barometer is falling, is high, and the probability that it will rain, given that the sky is red this evening, is low. The crucial thing is that these two probabilities are not at all irreconcilable, given that the probability that it will rain must be determined in the light of all the relevant circumstances (HW 37-38). Even then, however, the conclusion is a relational conditional judgment. Practical reasoning only arrives at unconditional judgments – which are identical to actions – when one accepts some reason r, where reason r constitutes the ground of one’s judgment that action a is better than action b – and, therefore, the ground of one’s intentional action –, and where one’s reason for doing a rather than b is identical with one’s reason for judging action a better than action b. The important point is that a prima facie statement, that action b is better than action a, all things considered, cannot logically contradict a statement sans phrase that action a is better than action b. Hence it is logically possible for one to deviate from one’s best conditional judgment for one could imagine the possibility of [A]n action, x as incontinent provided simply that the agent has a better reason for doing something else: he does x for a reason r, but he has a reason r' that includes r and more, on the basis of which he judges some alternative y to be better than x. (HW 40) 59 Note that the guiding principles that serve as grounds or reasons for the prima facie judgments may be moral or non-moral; in addition, the problem is not, as traditionally presented, one where passion or desire win at the expense of reason. Davidson’s intent is to show that akrasia is a problem in philosophy of action in general rather than merely in moral philosophy (HW 30). 60 Davidson identifies such a kind of judgment as the agent’s formulation of the intention on which he or she acts, or attempts to act. It is a judgment that “corresponds to, or perhaps is identical with, the action” (Donald Davidson, "Intending," in Essays on Actions and Events [Oxford: Oxford University Press, 1982], 98). This sort of judgment is understood as the position that the desirable characteristic of the given proposed action is enough to act upon. Chapter 5: An Overview of Different Attempts to Explain Akrasia 244 Indeed, Davidson insists that even if a judgment is made in the light of all relevant reasons – that is, when an agent makes an all-things-considered judgment in the fullest sense of the word –, it still does not constitute an unconditional judgment. The latter is reached when the agent performs an action intentionally. In this light (P 1) and (P 2), which imply that if the agent judges a to be better than b, then the agent will do a rather than b, hold only if ‘better than’ is used in an unconditional sense. Of course, this is logically compatible with the agent’s judging b to be better than a all things considered, and thereby being incontinent: “[t]he logical difficulty has vanished because a judgement that [b] is better than [a], all things considered, is a relational, or p[rima] f[acie] , judgement, and so cannot conflict logically with any unconditional judgement” (HW 39).61 The question, however, arises as to whether this logical possibility is also a real possibility, which question is only treated exceedingly briefly in the article under examination.62 For, how could anyone actually judge conditionally that b is better than a, on the basis of all the reasons available, and yet judge unconditionally that a is better than b, sans phrase? The reply involves an appeal to the distinction between reasons that logically justify an action and reasons that physically cause it. It is an appeal to Davidson’s theory of the dual character of causality.63 A normal, wholesome, rational action is caused by the mental event that is a reason for it. However, it might happen that the reasons involved in a prima facie all-things-considered judgment lack causal power. In such cases, the action they justify – that is, the action they are reasons for – fails to 61 G. F. Schueler makes the point that this logical gap, exploited in akratic action, comes at a high price, for it “has the effect of totally disconnecting practical reasons from practice…. on Davidson’s account of practical reasoning it is not even logically possible for one’s reasons to support a judgment on which one can act, nor is it logically possible to act on a judgment supported by one’s reasons” (G. F. Schueler, "Akrasia Revisited," Mind 92 [1983]: 583). Sarah Buss too notes that “[b]y stipulation, the [unconditional] judgment is completely ungrounded; it is not supported by any additional considerations, nor does the agent think it is” (Sarah Buss, "Weakness of Will," Pacific Philosophical Quarterly 78 [1997]: 31). C. C. W. Taylor makes the somewhat related point that whatever the nature of the connection between what an agent judges it better to do and what he wants more to do, it appears that it must hold as well for conditional, prima facie, relational or all-things-considered judgments, as for sans phrase or unconditional judgments (see C. C. W. Taylor, "Plato, Hare and Davidson on Akrasia," Mind 89 (1980): 500; "Reply to Schueler on Akrasia," Mind 93 [1984]: 585). 62 Brief references are made in HW 41-42. 63 See Jeanne Peijnenburg, "The Dual Character of Causality: How Davidson's Theory of Action Explanation Deals with "Akrasia"," Dialektik 2 (1998): 71-81. Chapter 5: An Overview of Different Attempts to Explain Akrasia 245 materialize. Instead, the akratic action is caused and justified by other reasons that do not constitute the best reasons. The result is that the action is not caused and justified by the right reasons. Of course, the agent has a reason for doing a, not b; what is lacking is a reason for not letting the better reasons for doing b, not a, prevail (HW 42fn.25). This occurs when such aberrant reasons that cause and justify the akratic action violate what Davidson has called the ‘principle of continence’ which tells us to “perform the action judged best on the basis of all available relevant reasons” (HW 41).64 It is the ignoring of the principle of continence that constitutes the irrationality; such reasons cause but they do not justify the rejection and violation of the principle. Hence, what happens is that “[i]n the cases of irrationality, the causal relation remains while the logical relation is missing or distorted. In the cases of irrationality … there is a mental cause that is not a reason for what it causes” (PaIr 298). This mental cause, which justifies the akratic action, causes both the action and the violation of the principle – which it does not justify – and hence overrules the causal power of the all-things-considered judgment. Now, it is clear that, according to this account, akrasia concerns the specific failure of the agent to act in accordance with his or her best reasons where the irrationality constituting akrasia is engendered due to the agent’s violation of the principle of continence without justification. Now, there are two interrelated problems with such an account. The first is that, while this elucidation of akrasia maintains that the reasoning that causes the action without rationalizing it is still intentional, it trivializes the issue. The reason is that it ignores completely the element of wrongness that is a constitutive part of akrasia. Indeed, ironically, according to Davidson’s account, if one were to ask the akratic whether, although he or she does hold evaluative reasons in favour of abstaining, he or she judges it best sans phrase to do what he or she, in fact, is doing, the akratic will reply in the affirmative. Now this seems counterintuitive; in reality, the akratic would sincerely disavow any such judgment.65 This principle is the analogue of the Hempelian ‘requirement of total evidence for inductive reasoning’, that is, the principle that one should give one’s credence to the hypothesis supported by all the available, relevant evidence (see Carl G. Hempel, Aspects of Scientific Explanation and Other Essays in the Philosophy of Science [New York: The Free Press, 1965], 397-403). 65 This point was made in Michael Bratman, "Practical Reasoning and Weakness of Will," Noûs 13 (1979): 160. Robinson notes that “[i]f he really does judge all out that B is better than A, and knows what he is doing, as we are supposing, then the agent does not merely recognize the possibility that he is wrong about A, but dismisses the appearance that A is best as illusory and misleading. He really believes (not merely wishes or hopes) that B is better than A. Thus, contrary to our initial supposition, his judgment in favour of A, though it will perhaps be in some objective sense his best judgment, will not be regarded by him as such” (Kirk Robinson, "Reason, Desire, and Weakness of Will," American Philosophical 64 Chapter 5: An Overview of Different Attempts to Explain Akrasia 246 For surely, whereas the akratic does move from a negative to a positive attitude towards the aberrant action, at no point would he or she say categorically ‘this is the best thing to do’. The Davidsonian description would leave no space for real remorse. A second problem appears to be that, on the Davidsonian model, the nature of the conflict the akratês finds himself in is understated. Indeed, in reality, Davidson’s discussion cannot cater for hard cases, where the agent is unable to reconcile different action-alternatives with one another and, indeed, chooses to do what is clearly worse. The reason is that, if one were to think it best, all things considered, to do something, one must think that it is best sans phrase to perform the action. There can be no knowing cohabitation between a better all-things-considered judgment and an aberrant unconditional judgment66: [I]t is logically impossible that one should without extreme logical incoherence make an unconditional judgment favouring b when one is fully conscious that the totality of the evidence so far favours a, unless one thinks (reasonably or unreasonably) that the addition of further evidence, so far unidentified, would tilt the balance in favour of b … [which] possibility must be regarded as irrelevant to our discussion of incontinence.67 Quarterly 28 [1991]: 290). Pears too thinks that Davidson’s position, that if an agent does an action intentionally and avoidably, then he must judge it best to do that action, implies the questionable doctrine maintaining that the akratic agent acts not with the thought that it is not wrong to act in such a way but with the thought that acting in such a way is best (see David Pears, "How Easy is Akrasia?" Philosophia 11 [1982] 38). 66 This point was made by Paul Grice and Judith Baker, "Davidson on "Weakness of Will"," in Essays on Davidson: Actions and Events, ed. Bruce Vermazen and Merrill B. Hintikka (Oxford: Clarendon Press, 1985), 27-49. It was anticipated in Michael Bratman, "Practical Reasoning and Weakness of Will," Noûs 13 (1979): 160. 67 Paul Grice and Judith Baker, "Davidson on "Weakness of Will"," in Essays on Davidson: Actions and Events, ed. Bruce Vermazen and Merrill B. Hintikka (Oxford: Clarendon Press, 1985), 44. For “[e]ven if, as Davidson believes, an adequate account of practical reasoning requires agents to form all-out unconditional evaluative judgments as well as all-things-considered evaluative judgments, the former, no less than the latter, are made in light of everything the agent considers” (Sarah Buss, "Weakness of Will," Pacific Philosophical Quarterly 78 [1997]: 31); if not, “it seems that, on Davidson’s own account, the weak-willed agent cannot help but regard the action as something that just happens to her” (Ibid.: 30). Grice and Baker note that, according to the ‘naïve view’, the natural place to locate the failure of reasoning involved in akrasia is, in fact, between the unconditional value judgment and the intention (see Paul Grice and Judith Baker, "Davidson on "Weakness of Will"," in Essays on Davidson: Actions and Events, ed. Bruce Vermazen and Merrill B. Hintikka (Oxford: Clarendon Press, 1985), 49). Of course, Davidson’s account leaves no room for the naïve view to get a foothold. In order to accommodate such a view Stephen Williams has split the principle of continence into two further principles constitutive of rationality: (C1) Judge best that course of action which all one’s reasons support; (C2) Intend that course of action which one judges best. He holds that the naïve view is thereby accommodated since, the irrationality of akrasia consists in the violation of (C2) (see Stephen Williams, "Belief, Desire and the Praxis of Reasoning," Proceedings of the Aristotelian Society 90 [1989]: 141). In this fashion he avoids this line of criticism although, of course, the problem pointed out by Grice and Baker resurfaces at the stage where the violation occurs. Alfred R. Mele takes a similar line, holding that “unconditional judgments are not (pace Davidson) identical with intentions” (Alfred R. Mele, "Akrasia, Reasons, and Causes," Philosophical Studies 44 [1983]: 358). What happens, in akratic actions that do not involve change of mind, is that one retains an unconditional judgment without retaining an intention. The reason Chapter 5: An Overview of Different Attempts to Explain Akrasia 247 And, even though, in his reply to the objection, Davidson attempted to maintain that an agent can fail to pursue a course all the relevant reasons support68 – just as a theoretical thinker can flout Hempel and Carnap’s requirement of total evidence – it certainly seems paradoxical that someone could think it best to do one thing when one knew that all the relevant reasons demand doing another.69 The only possibility appears to be that “to suppose someone to make an unconditional judgment in a direction opposed to an ‘all things before me’ judgment, we must also suppose the latter judgment not to be ‘fully present’ to the judger, on some suitable interpretation of that phrase”.70 There does not seem to be any space for an agent’s acting, knowingly and willingly, wrongly and being sorry for it. Thus, while such an account may cover for cases where the term ‘akrasia’ is used loosely, it does not appear to cover the strong sense of the term which one feels is a tight and accurate, and yet paradoxical, description of certain human situations. 5.2.2 Other Proposed Resolutions: the Postulation of Multiple Cognitive Systems at Work in the Mind of the Akratês It seems that the only way one can describe such limit cases of akrasia fittingly is through the use of some kind of a multiple-systems analysis and, indeed, in this section, three extremely influential such accounts will be presented. The obvious advantage of such accounts is that they apparently safeguard the intuition that, in akrasia, the agent is subject to two different states of mind about what he or she should do. for such a divergence is that “[u]nconditional judgments and intentions are sometimes formed on the basis of evaluations of one’s reasons which are not in line with the motivational force of these reasons” (Ibid.: 363). The problem, however, is that all such accounts apply equally well to an agent’s acting on compulsion. Indeed, whenever evaluation and motivation are seen as coming apart, we seem to be facing a case of compulsion (see Sergio Tenenbaum, "The Judgment of a Weak Will," Philosophy and Phenomenological Research 59 [1999]: 885). 68 Donald Davidson, "Replies to Essays I-IX," in Essays on Davidson: Actions and Events, ed. Bruce Vermazen and Merrill B. Hintikka (Oxford: Clarendon Press, 1985), 205. 69 Charlton argues that the theoretical parallel would be to refuse to accept a hypothesis when one not only thinks that the evidence supports it but that the evidence makes it irrational not to accept it. (See William Charlton, Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 125.) 70 Paul Grice and Judith Baker, "Davidson on "Weakness of Will"," in Essays on Davidson: Actions and Events, ed. Bruce Vermazen and Merrill B. Hintikka (Oxford: Clarendon Press, 1985), 44-45. Chapter 5: An Overview of Different Attempts to Explain Akrasia 248 The first such account is that by Davidson himself who, in a later paper (in 1982),71 admits that a version of the Freudian notion of the divided mind in terms of mental functional sub-systems is necessary in order to explain the gap between an allthings-considered judgment, that is the proximate outcome of reasoning, and an unconditional judgment that indicates what the agent actually does. As explained above, what happens is that the aberrant reason, after being overruled, and tasting defeat, reappears to overrule the second-order principle of continence. Davidson, here, acknowledges that: “[o]nly by partitioning the mind does it seem possible to explain how a thought or impulse can cause another to which it bears no rational relation” (PaIr 303). Just as in the case of his explanation of self-deception, Davidson avers that the sub-sections forming such ‘partitions’ of the mind are semiautonomous and temporary, each being characterized by “a supporting structure of reasons, of interlocking beliefs, expectations, assumptions, attitudes and desires” (PaIr 300). The intentional motive causing the akratic break causes the right intentions together with the principle of continence to be placed, temporarily, in one structure and the unwarranted intentions to be placed in a second structure. Hence, Davidson’s description of akrasia depends on a particular theory about the nature of the working of the mind and of practical reason.72 71 See Donald Davidson, "Paradoxes of Irrationality," in Philosophical Essays on Freud, ed. Richard Wollheim and James Hopkins (Cambridge: Cambridge University Press, 1982), 289-305. 72 Indeed, Davidson’s account not only depends on the idea of a partitioning of the mind, but also rests on the postulate of anomalous monism. The term ‘monism’ shows that on the ontological level (the level of tokens), mental events are identical to physical events. However, they resist translation into physical terms (they are anomalous on the semantical level of types). Now, natural laws, by definition, can only be written in physical terms, hence mental events resist any predication or explanation on the basis of laws. Davidson accepts that reasons are rational causes. This means that to say that an action is intentional is to say that it was caused by the beliefs-desires that rationalize it. However, in accordance with anomalous monism, one cannot form psycho-physical or psychological laws. (See Donald Davidson, "Mental Events," in Essays on Actions and Events [Oxford: Oxford University Press, 1982], 214-215.) There are, however, various queries about this position, which seems to have become the new orthodoxy. One is admitted by Davidson himself: “Beliefs and desires that would rationalize an action if they caused it in the right way – through a course of practical reasoning, as we might try saying – may cause it in other ways. If so, the action was not performed with the intention that we could have read off from the attitudes that caused it. What I despair of spelling out is the way in which attitudes must cause actions if they are to rationalize the action” (Donald Davidson, "Freedom to Act," in Essays on Actions and Events [Oxford: Oxford University Press, 1982], 79). He admits that he does not know whether it is possible to eliminate wrong causal chains. The bigger problem, however, concerns the possibility of identifying any causal chain as right or wrong! (See Anthony Kenny, Will, Freedom and Power [Oxford: Basil Blackwell, 1975], 121fn.) In addition, of course, the physiologist and the psychologist will slice the world into quite different events (see Jennifer Hornsby, "Which Physical Events are Mental," Proceedings of the Aristotelian Society 81 [1980-1981]: 76-85; "Physicalism, Events and Part-Whole Relations," in Actions and Events: Perspectives on the Philosophy of Donald Davidson, ed. E. LePore and B. P. McLaughlin [Oxford: Basil Blackwell, 1988], 444-458). Chapter 5: An Overview of Different Attempts to Explain Akrasia 249 While Davidson’s way of describing akratic actions changed (from a more conceptual consideration of the possibility of such irrational actions, to his consideration – in order to explain why the person did this, rather than that – of the necessity of a functional and dynamic partitioning of the mind into person-like structures that interact with one another), he consistently viewed such actions as being a matter of entanglement of reasons.73 His account is idiosyncratic in two respects. First, he maintains throughout that he proposes: 73 Another account which, although distinct, can be considered as Davidsonian, holds that one can distinguish different kinds of rightness or wrongness, and that these different kinds may be indexed. Thus, to give an example: that a person is innocent renders it right to release him; that a mob is awaiting him outside to lynch him renders it wrong to release him. The point would be that agglomeration does not hold “across different kinds of reasons” (Susan Hurley, "Conflict, Akrasia and Cognitivism," Proceedings of the Aristotelian Society 86 [1986]: 26). In addition, such indexed conclusions are pro tanto right or wrong; one can understand the releasing of the innocent prisoner even when an all-thingsconsidered judgment would dictate otherwise. Here, precisely because such reasons are pro tanto, not prima facie, the idea is that “[t]he relationships between different pro tanto reasons are, in interesting ways, like relationships between individuals whose wills clash.… they are manifestations of the conflicting values that may continue to exert their discrete influences on us even when we have arrived at all-things-considered judgments” (Ibid.: 34, 37). Hence, such pro tanto reasons guide us in picturing the formation of the boundaries between the functional sub-systems described in Davidson’s later work. Now, the idea that the notion of goodness is not unitary is attractive, for it appears natural to say that different kinds of reasons render good in different ways: one can imagine conflict between reasons having to do with justice and reasons pertaining to commiseration. However, it seems that akrasia is not about such a putative conflict: it is not about a hypothetical dilemma one might find oneself in because one thinks that it might be the case that one has an absolute duty to do both x and y and these two actions are not compossible. In addition, it is difficult to see how one can pass, from rankings of alternatives under specific and incommensurable reasons, to rational judgments as to what it would be better to do. One way forward, that would avoid such difficulties, holds on to the idea that it is not the case “that there is just one evaluative dimension , and one quantitative measure m, such that -ness is all that matters, and all courses of action can be compared with one another by the measure m in respect of -ness” (David Wiggins, "Weakness of Will, Commensurability and the Objects of Deliberation and Desire," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 256). Yet, such an account would insist that there is no over-reaching consideration – such as the maximization of the subject’s total utility – that is decidable against “an indifference curve on which each possible choice can be located for purposes of comparison in respect of utility with the other possible choices” (Ibid., 259). It would aver that the judgement that one course of action is better than another is arrived at within the background of the agent’s conception of the life that he sees it would be good to attempt to realize: “A man usually asks himself ‘What shall I do?’ not with a view to maximizing anything but only in response to a particular context. This will make particular and contingent demands on his moral or practical perception, but the relevant features of the situation may not all jump to the eye” (David Wiggins, "Deliberation and Practical Reasoning," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 232). Only within such a preexisting background conception could one put a value upon and have a reason to choose whether one course of action represents a greater good than some other course (see David Wiggins, "Weakness of Will, Commensurability and the Objects of Deliberation and Desire," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 259-261). The same applies – perhaps surprisingly – even to pleasure; pleasures differ in kind so that “[o]ne’s notion of what is pleasant is not external to one’s conception of the good life” (Julia Annas, "Aristotle on Pleasure and Goodness," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 289); one cannot get the pleasures of, say, the just man without being just, nor can one appreciate the pleasantness of some virtuous action in a way that makes no reference to the Chapter 5: An Overview of Different Attempts to Explain Akrasia 250 to divorce [the] problem entirely from the moralist’s concern that our sense of the conventionally right may be lulled, dulled or duped by a lively pleasure…. [i]n approaching the problem of incontinence it is a good idea to dwell on the cases where morality simply doesn't enter the picture as one of the contestants for our favour—or if it does, it is on the wrong side. (HW 30) Secondly, his analysis makes the question of motivation completely irrelevant, indeed he avers that: “[i]t is no part of the analysis of weakness of the warrant or weakness of the will that the falling off from the agent’s standards is motivated (though no doubt it often is), but this is integral to the analysis of self-deception” (DD 85). Now, the issue as to whether the agent’s diversion from the moral good be motivated by some desire, or by the desire for the avoidance of pain has occupied an important position in the traditional treatment of the question of akrasia and, indeed, the failure to make such aspects relevant to one’s treatment of akrasia requires further explanation. Davidson’s reason for such a position appears to be his conviction that the explanation of such an aberrance must retain the core of the Plato Principle – that is, that each of the maxims involved in the conflict is, in some sense, rational (PaIr 294) – so that the irrationality is intentional, while going beyond the said principle so that the intention is not perfectly rational. To a good extent, I want to go along with the kind of reasoning that upholds the core of the Plato Principle for I too want to say that the agent acts intentionally, or at least voluntarily, when she acts akratically; I too hold that akratic action is within the space of the agent’s reasons or at least within the space where she would not find the quest for reasons senseless. I too do not want to say that, when a person acts akratically, it is a matter of “an alien force overwhelm[ing] his or her will.… [as] happens when Medea begs her own hand not to murder her children. Her hand, or the passion of revenge behind it, overcomes her will” (PaIr 294).74 For that would mean that the agent’s doing was not an action; from her point of view, it would be as if another person had moved her. Nevertheless, I feel that Davidson’s failure to grant any relevance at all to questions of motivation in his account of akrasia constitutes a deficiency. For, while on the one hand, even the desires we share with animals are coloured by the use of agent’s conception of the good life (Ibid., 287, 290). Now this type of account is extremely promising in helping to elucidate akrasia and is, indeed, consonant with the way my account will be developed. 74 Davidson uses this case as a paradigm for what he calls the ‘Medea Principle’. Chapter 5: An Overview of Different Attempts to Explain Akrasia 251 concepts that could only be possessed by creatures employing a certain linguistic prowess, on the other hand, sensations like hunger, thirst, weariness, and sexual desire are noteworthy persuasive elements. At this stage, one real concern I want to mention is that, in his eagerness to treat the conflict in akrasia as a conflict of reasons, Davidson refuses to treat akrasia as an ethical issue. He avers that, since “[t]here are numerous occasions when immediate pleasure yields to principle, politeness, or sense of duty and yet we judge (or know) that all things considered we should opt for pleasure”, therefore, one must avoid “succumb[ing] to the temptation to reduce incontinence to such special cases as being overcome by the beast in us, or of failing to heed the call of duty, or of succumbing to temptation” (HW 30). This is the reason he gives for dwelling on cases where ethical reasons, either simply do not enter, or enter the picture on the side contrary to one’s better judgment. Now, traditionally, the question of akrasia has always been treated as an ethical problem. Indeed, after our consideration of Aristotle’s account of akrasia, Davidson’s specific failure to consider akrasia as an ethical issue will be seen to be a central reason for the inadequacies in his account. Related to this aspect, a significant shortcoming I see in his account is his neglect of the discontinuity between predictive judgments and intentional actions; the latter, unlike the former, do not rest on any inductive or other evidence. One simply sees what one is to do, not in the light of any self-observation, but in the light of what one wants. And indeed, intimately related to this disregard is his Janus-like view of causality as applied to the relation between the mental and the physical through anomalous monism. This view of causality applied to the mental, and to akrasia in particular means that the latter is seen as a matter of the agent’s having a reason that does not cause that which it renders reasonable because, simultaneously, there is a mental cause that successfully produces an action for which it is not a reason. He avers: “a person is irrational if he … has a reason which does not cause what it is a sufficient reason for” (PaIr 299):75 75 Despite the subsequent quotation, there appears to be a difference here between the account in this essay and the one in his 1969 essay. In the latter work, he held that the akratic action is caused and justified by reasons that do not constitute the best reasons (HW 42fn.25); in Paradoxes of Irrationality, he Chapter 5: An Overview of Different Attempts to Explain Akrasia 252 If r is someone’s reason for holding that p, then his holding that r must be, I think, a cause of his holding that p. But, and this is what is crucial here, his holding that r may cause his holding that p without r being his reason; indeed, the agent may even think that r is a reason to reject p. (HW 41) But this means that such a ‘reason’ operates, as it were, behind the back of the agent; it is independent of whether the agent actually wants it or not. Of course, Davidson wants to insist that it is always a reason that leads the agent to act, even if the latter does not justify the action – he wants to insist that the akratic action is intentional –, however in deed, such a ‘reason’ plays no role in the agent’s choice of the particular action. For, as a cause that produces an action for which it is not a reason, it is a blind force and belongs to the category of the non-rational. Clearly, Davidson’s characterization of akrasia applies equally well to an agent’s acting on compulsion76 and, equally clearly, this is not the way in which we use the concept.77 For it is essential to the conceptual possibility of akrasia that the agent acts for reasons that he or she excludes or sets aside, or would do so had he or she thought about it. And this is radically distinct from any account of some intention or reason forcing the establishment of partially autonomous, but still intentional, temporary structures in the mind in order to separate the warranted intentions and principle (of continence) from the unwarranted intentions and compelling the agent to perform his surd, intentional (sic!) action. To be sure, in view of the common picture of akrasia, other multiple-systems analyses, that have portrayed a conscientious system battling it out with pleasure- appears to hold that the irrational motive does cause the akratic action, but it is not a reason for it (PaIr 298, 300). 76 See Sergio Tenenbaum, "The Judgment of a Weak Will," Philosophy and Phenomenological Research 59 (1999): 880. 77 Richard Holton, for one, while making a distinction between akrasia and weakness of will (he holds that, while akrasia is a kind of weakness of reason, weakness of will arises when agents are too ready to reconsider their intentions [see "Intention and Weakness of Will," The Journal of Philosophy 96 (1999): 241]), thinks that it is not obvious that compulsive acts cannot be classified among weak willed acts (Ibid.: 261). He holds that to claim that, say, a kleptomaniac is weak willed would be inappropriate rather than false. He maintains that it would, perhaps, not be wrong to say that “compulsives are pathologically weak willed” (Ibid.: 261). My position, of course, is that if it were not senseless to say that a human action was compelled, one would say that such an unfree agent committed an action that was not willed, rather than weak willed. As Aristotle averred: “those in whom nature is the cause of such a state no one would call incontinent” (Aristotle, Nicomachean Ethics, VII, 1148b 30); the incontinent person “acts willingly (for he acts in a sense with knowledge both of what he does and of the end to which he does it), but is not wicked, since his purpose is good; so that he is half-wicked” (Aristotle, Nicomachean Ethics, VII, 1152a 15-20). Chapter 5: An Overview of Different Attempts to Explain Akrasia 253 seeking systems, and that have sought to evade the problem of portraying akratic actions as irresistible, have been proposed. Indeed such accounts can be seen as philosophical descendants of Plato’s division of the soul into three parts;78 according to this account, weakness of will occurs when, in a conflict of desires,79 an irrational desire issuing from one part of the soul impedes a rational desire coming from another part of the soul: the result is that one acts contrary to the opinion, embodied in the rational desire, of what it is best to do in the circumstances.80 Although they make various restrictions on this basic outlook, it will be seen that these accounts are fundamentally Platonic. The difference between Davidson’s and these accounts is that, while the former subdivides the agent’s reason, the latter partition the agent into rational and irrational parts.81 Their advantage is that they are able to portray the sparring, that is so familiar in akrasia, vividly and uncompromisingly. One exponent of such a kind of multiple-systems analysis is David F. Pears (1921–). He agrees with Davidson that a person may, in fact, reject what he believes is the better course of action. However, they present diverse versions of how this could be possible. In Pears’ own terminology, for Davidson, the akratic fault must occur within the agent’s reasoning82 and cannot occur between the last line of reasoning and the 78 See Plato, Republic, 436C-437E; 439D; 441A-C. See also James Jerome Walsh, Aristotle's Conception of Moral Weakness (New York: Columbia University Press, 1963), 28, 35. 79 See Plato, Republic, 436B-C. 80 See Terrence M. Penner, "Plato and Davidson: Parts of the Soul and Weakness of Will," Canadian Journal of Philosophy 16 Supplementary Volume (1990): 42. See also James Jerome Walsh, Aristotle's Conception of Moral Weakness (New York: Columbia University Press, 1963), 28-59, for an account of the development of Plato’s thought. Plato distinguishes three kinds of desires with three kinds of pleasures and, similarly, three appetites and controls (see Plato, Republic, 580D): desires of reason (boulêsis), desires of spirit (thumoeides), and desires of appetite (epithumêtikon). These last two kinds of desire are irrational, blind to all considerations of good: “Let no one then, disconcert us when off our guard with the objection that everybody desires not drink but good drink and not food but good food, because [the argument will run] all men desire good” (Plato, Republic, 438A1-4). 81 The difference with Davidson, in this regard, will by now be clear. Unlike Davidson’s functional partitions, which relate with one another on logical terms – that is, one can see the presence of logical relations, like negation and substitution, in their interrelations, and in their relations with reality –, these thinkers make a clear distinction between irrational and rational desires. In this respect, while Davidson’s own account appeals to a modified version of ‘the Plato Principle’ (in Davidson’s account, each of the desires involved in the conflict is, in some sense, rational: both desires are for what the agent thinks best), these accounts (and, confusingly, Plato’s account too) appeal to ‘the Medea Principle’: they attempt to explain akrasia by appeal to alien irrational desires. (Donald Davidson speaks of the ‘Plato’ and the ‘Medea’ Principles in "Paradoxes of Irrationality," in Philosophical Essays on Freud, ed. Richard Wollheim and James Hopkins [Cambridge: Cambridge University Press, 1982], 294-295.) 82 Pears holds that Davidson exploits what might be called ‘latitude’, or a gap the agent thinks may be present between the considerations he has and all the relevant considerations there are. If there were no such latitude, the agent “could not make his irrational move without conscious self-contradiction” (David Pears, "How Easy is Akrasia?" Philosophia 11 [1982]: 38). Chapter 5: An Overview of Different Attempts to Explain Akrasia 254 action83 – barring the case of change of mind – on pain of contradiction between two unconditional judgments. Pears wants to hold that the passage from evaluation to action is not logically closed; that is, such an occurrence need not be a case of selfcontradiction.84 The latter presents a multiple-systems analysis of the human agent in which a reason-dominated system battles it out with desire-dominated systems.85 He avers that the will is not fully integrated for reasoned judgment is not the only constituent of the will. He states that “there are other constituents of the will that evolved earlier [than reasoned judgment] and they can still take their seat at the control of intentional action without using force”.86 Such constituents include physical appetites and emotions like fear and anger and, when they take control of the will, the action remains unforced. The basic idea is the following: [W]hen a desire that has been defeated in deliberation takes over control of intentional action, it does not put external pressure on the will. Rather, it constitutes the will in spite of the fact that it does not conform to the agent’s better judgment and does not even pretend to do so. To put the point in another way, the executive element in the agent is, for the time being, the rebellious desire, even if his reasoned judgment is that he should not identify with it. The will, which is the source of intentional action is not the natural servant of a single master, reason, and so intentional action that does not fit the agent’s reasoned view of the best thing to do is not, therefore, forced action.87 Hence, the cause of conscious last-ditch akratic actions is the presence of conflicting forces ‘behind the will’ that “can break out in intentional action without the 83 See David Pears, "How Easy is Akrasia?" Philosophia 11 [1982]: 36. Contradiction is avoided because the heterogeneous misfit between a value judgment and an action ensures that the connection between the two cannot be explained on the analogy of the relation between two beliefs: “the relevant part of the definition of quasi-truth [attributed to actions] merely makes it necessary that a quasi-true action should fit the value-judgment but it does nothing to show that it is impossible to combine the value-judgment with an action that does not fit it” (David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 [1982]: 173). 85 Pears wants to establish the possibility of what he calls ‘underivative brazen akrasia’ (see David Pears, "How Easy is Akrasia?" Philosophia 11 [1982]: 33-50) or ‘conscious last-ditch akrasia’ (see David Pears, Motivated Irrationality [Oxford: Clarendon Press, 1984], 113-117, 154-252). He is examining the possibility of a case that is underivative because the fault is intrinsic to the move that is actually being made, and brazen because the person himself knows that the move is irrational (see David Pears, "How Easy is Akrasia?" Philosophia 11 [1982]: 36). It is what he calls a ‘limiting case’ of irrationality where there is no deficiency in the agent’s deliberation and where he or she acts consciously against his or her own better judgment (see David Pears, Motivated Irrationality [Oxford: Clarendon Press, 1984], 22). 86 David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 168. 87 Ibid.: 168. 84 Chapter 5: An Overview of Different Attempts to Explain Akrasia 255 endorsement of a revised valuation”.88 For, in action, it is the will, not the world, that sets the standard of fit and, in the case of the will, unlike in the case of the world – which must be seen as fully integrated –, “there is conflict behind the throne”,89 that may result in “an unforced change in the immediate cause of intentional action”.90 Pears, in a number of papers, and in Motivated Irrationality, proceeds to examine the question as to the causation of such actions. He avers that, even if a particular desire has usurped the role of intentional action, it is a generally appropriate cause for human action and so could issue in such an effect even if the agent is aware of its operation.91 He suggests that the success of the rebellious desire is often attributable to secondary qualities like last-minute intensity.92 Hence, for such reasons, Pears believes that motivated irrationality in action is easier than motivated irrationality in belief-formation. Pears’ position, therefore, is that akrasia occurs when a desire issues in action without passing through the checkpoint of reason at all. The very possibility that desires could lead to actions, even when there is no moral or prudential judgment, shows that there is no necessary backward connection running, from doing an action intentionally and avoidably, to judging it prudentially and morally best to do it:93 “[i]f there are any intentional non-compulsive actions that issue from mere desires, the backward connection [to prudential or moral valuation] is not necessary”.94 In such cases, the backward connection would only hold for the weak valuation of the immediate desire or 88 David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 176. 89 Ibid.: 176. 90 Ibid.: 176. It is interesting to note here that, on his own admission, Frank Jackson’s account is similar to the way David Pears sets up the problem; both would agree that “weak-willed action is action arising from wants and desires that have not evolved according to the dictates of the agent’s reasons” (Frank Jackson, "Weakness of Will," Mind 93 [1984]: 14). Indeed, Jackson is trying to show how it could be possible “that a reason someone has for acting can have a causal efficacy in the production of his action that is greater than the efficacy it has qua reason” (Christopher Cordner, "Jackson on Weakness of Will," Mind 94 [1985]: 275). Of course, this question arises originally from a well-known feature of Davidson’s view (HW 41). However, Jackson undertakes the challenge in a manner that is more reminiscent of Pears and Mele than of Davidson. 91 See David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 165; see also David Pears, "How Easy is Akrasia?" Philosophia 11 (1982): 48. 92 See David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 167. See also David Pears, Motivated Irrationality (Oxford: Clarendon Press, 1984), 173-179. 93 The ‘forward connection’ would run in the opposite direction; from judging an action to be the best to doing it. Pears holds that Davidson maintains that both the forward and the backward connections are necessary in intentional action. He ‘reinterprets’ Davidson’s account in terms of the necessity of weak valuation, in order to allow for the possibility of conscious last-ditch akrasia. (See David Pears, Motivated Irrationality [Oxford: Clarendon Press, 1984], 183-256.) 94 Ibid., 209. Chapter 5: An Overview of Different Attempts to Explain Akrasia 256 the rebellious yen.95 And, this can happen, even when the agent’s moral or prudential judgment shows the act to be bad: Reason does not check, and is not generally believed to check, every desire that issues in intentional action without compulsion. Because reason is known not to operate a check-point through which all desires are filtered, however imperfectly, human agents do not even pretend that their rebellious desires always speak with the voice of reason. True, the introduction of the concept of weak-valuation allows us to say that they express themselves in value-judgments. But this is only because this is a different sense of the phrase ‘value-judgment’. It is obvious that these weak value-judgments are not rationally derived from the original considerations of reason. It is less obvious, but equally important, that they are often not even irrational side-blows of reason, and so the agent does not claim that that is their origin.96 Alfred F. Mele (1951–) has developed another multiple-systems account of akrasia, which is extremely articulate, rather formidable, and conceptually very similar to that offered by Pears. Indeed, both authors prise evaluation and motivation apart in such a manner that what characterizes the limit cases of akrasia is the complex interaction of many psychological factors. Both refuse to identify a person with his or her reason;97 for better judgments could be based on, and supported by, considerations including appetites, emotions and so forth:98 hence, the self of self-control is “not properly identified with reason”.99 They would both say that the will of the human agent is not fully integrated; one’s desires have access to one’s intentional actions outside, and independently of, one’s reasoned value-judgments.100 Both insist that, when one acts akratically, the main problem is that, at the time of action, the balance of one’s motivations lies on the side of the akratic action101 so that one’s best judgments are rendered inefficacious.102 They also both aver that, although the desires that make themselves felt in akratic actions are rebellious and independent of one’s valuejudgment, they are not properly described as alien forces; they do not make the agent a 95 See David Pears, Motivated Irrationality [Oxford: Clarendon Press, 1984], 196. Ibid., 197. 97 See Ibid., 196-197; Alfred R. Mele, "Autonomy, Self-Control and Weakness of Will," in The Oxford Handbook of Free Will, ed. Robert Kane (New York: Oxford University Press, 2002) 532. 98 See Alfred R. Mele, "Errant Self-Control and the Self-Controlled Person," Pacific Philosophical Quarterly 71 (1990): 51. 99 Ibid.: 56. See also David Pears, "How Easy is Akrasia?" Philosophia 11 (1982): 47-48. 100 See David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 175; Alfred R. Mele, Autonomous Agents (Oxford: Oxford University Press, 2001), 27-29. 101 See Alfred R. Mele, "Is Akratic Action Unfree?" Philosophy and Phenomenological Research 46 (1986): 679; David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 168-169. 102 See David Pears, "How Easy is Akrasia?" Philosophia 11 (1982): 47; Alfred R. Mele, "Akratic Action and the Practical Role of Better Judgment," Pacific Philosophical Quarterly 72 (1991): 45. 96 Chapter 5: An Overview of Different Attempts to Explain Akrasia 257 prisoner or make the action unavoidable. They are, rather, other constituents – besides reasoned judgment – of the agent’s will.103 Despite the many similarities, the differences between the two philosophers are twofold. One is that Mele, unlike Pears, holds that akrasia must be caused by a motive at its point of origin; the latter, but not the former, holds that one cannot equate akrasia with motivated irrational action.104 The other difference is that Mele supplements a ‘Pears’ kind of account by proposing psychological explanations of akratic, and other, actions.105 Mele’s account characterizes akrasia as occurring in cases where: The motivational force of evaluated items is out of line with our evaluations, and with the better judgments and intentions formed on the basis of our evaluations. In cases of this sort, intentions – even here-and-now intentions – may be overridden by competing motivations.… For evaluation and preponderant motivation sometimes part company.106 According to this account, therefore, akratic actions are matters of best judgments, which have a motivational dimension and which rationally commit an agent to action, being thwarted from issuing in a corresponding intention by competing motivations.107 The motivational strength of a desire depends on factors like the strength of various supporting sub-desires; the strength of various contrary motives; and cognitive factors such as the vividness of one’s imagination of the situation, the immediacy or otherwise of the benefits, the force of habit, and the salience of the course of action.108 Mele insists, however, that the disparity between “evaluation and the preponderant causal force of pertinent items … need not render the agent a helpless 103 See David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 (1982): 168-169; Alfred R. Mele, Autonomous Agents (Oxford: Oxford University Press, 2001), 6. 104 Pears maintains that Aristotle held that factors like misperception, faults of character, and slips in reasoning that are uncharacteristic of the agent, could all cause akratic actions (see David Pears, "Motivated Irrationality Part I," The Aristotelian Society Supplementary Volume 56 [1982]: 159). 105 See Alfred R. Mele, "Akratic Action and the Practical Role of Better Judgment," Pacific Philosophical Quarterly 72 (1991): 42-44. 106 Alfred R. Mele, Irrationality: An Essay on Akrasia, Self-Deception and Self-Control (Oxford: Oxford University Press, 1987), 104. See also Alfred R. Mele, "Incontinent Believing," Philosophical Quarterly 36 (1986): 218. 107 See Alfred R. Mele, "Akratic Action and the Practical Role of Better Judgment," Pacific Philosophical Quarterly 72 (1991): 43-45. 108 See Alfred R. Mele, Irrationality: An Essay on Akrasia, Self-Deception and Self-Control (Oxford: Oxford University Press, 1987), 84-93. Chapter 5: An Overview of Different Attempts to Explain Akrasia 258 victim. [The agent] may have at her disposal a variety of techniques for bringing her feeling-state into line with her better judgment”.109 The point is that, at least in some instances of akrasia, the fact that the motivational state of the agent is out of line with his or her evaluative assessment does not compel the agent. For it is “open to the agent to prevent his being in this motivational condition by exercising his powers of selfcontrol”,110 where this is a matter of “purposive intervention into one’s own motivational condition”.111 In all the strategies the agent could employ, however, the appeal to motivation appears to be compulsory.112 Hence, while Mele insists throughout that the postulation of any intentional motive or any governing principle – such as the Davidsonian principle of continence – to resist or prevent forming certain intentions, does not tell the whole story,113 and that the recourse to second-order desires or preferences is not necessary,114 he replaces such a narrative by the agent’s powers and skills of self-control where, if she is motivated to exercise self-control, she can work at trying to manipulate her own motivational condition. It is important to note that, as in Pears’ account, even though Mele insists that “[p]assions, emotions, and the like that run counter are rarely plausibly seen as alien forces”,115 it is clear that the self is seen as a rather loose confederation of a very wide range of masters. What Mele has in mind, in his explanation, is a battle of motivational factors inside the person, where the default intentional outcome is that which figures in 109 Alfred R. Mele, "Akratic Feelings," Philosophy and Phenomenological Research 50 (1989): 286-287. Alfred R. Mele, "Is Akratic Action Unfree?" Philosophy and Phenomenological Research 46 (1986): 679. 111 Alfred R. Mele, Autonomous Agents (Oxford: Oxford University Press, 2001), 64. 112 Mele makes a distinction between two different kinds of resistance to aberrant desires. It is a distinction between skilled resistance, where the agent manipulates his or her motivational condition, in various possible ways: by uttering self-commands, adopting focusing strategies, altering the situational environment, and so forth; and brute resistance, where the agent simply forms or retains a contrary intention. The latter form of resistance, however, only works if the agent is strong enough – so that weaker persons have to rely more on skilled resistance – where, what distinguishes stronger from weaker persons is presented as a function of the power displayed in the sheer effort of will in resisting temptation and the greater strength of motivation that sees to it that the better judgments and the motivations they involve are effective. However, according to Mele, ‘powers’ or ‘skills’ of self-control are not enough; what is crucial is that the agent be motivated to exercise self-control (see Alfred R. Mele, "Is Akratic Action Unfree?" Philosophy and Phenomenological Research 46 [1986]: 678-679; Alfred R. Mele, Autonomous Agents [Oxford: Oxford University Press, 2001], 80-83). 113 See Alfred R. Mele, Irrationality: An Essay on Akrasia, Self-Deception and Self-Control (Oxford: Oxford University Press, 1987), 44-48. 114 See Alfred R. Mele, Autonomous Agents (Oxford: Oxford University Press, 2001), 65-80. 115 Alfred R. Mele, "Errant Self-Control and the Self-Controlled Person," Pacific Philosophical Quarterly 71 (1990): 56. 110 Chapter 5: An Overview of Different Attempts to Explain Akrasia 259 the motivated best judgment, but where the intentional outcome may be different owing to unchecked recalcitrant desires or other motivational factors. Indeed, when faced with multiple-systems accounts, the problem, of who the self engaged in such a struggle is, is evident. To be sure, one is surprised that such attempts can be seen as resolving the paradox. In the later Davidson and in Mele, what one finds is more a causal psychological and mechanistic type of explanation than a conceptual analysis. Indeed, despite their protestations to the contrary, it is not always clear, if the actions they are portraying are – at the time – compulsive rather than akratic. All three accounts, moreover, seem merely to use our common and everyday mentalist vocabulary – together with Freudian terminology in Davidson’s later account – in order to explain the picture of the akratês rather on the model of a group of people disagreeing rather violently with one another about what it would be best to do. In all three kinds of multiple-systems accounts what seems to be the concern is the endeavour to find a way to divide the person’s mental activity so that no one system alone is guilty of akrasia; the aberrant condition is attributable to the combination of the systems.116 As with the multiple-systems accounts of self-deception, here too, the problem throughout appears to be one of ignoratio elenchi. What one wants is clarity about a problem in which one person – not two or more individuals – is involved. Any Such a failing is also manifested by Amélie Oksenberg Rorty’s account. Her portrayal of akrasia uses the metaphor of a plan of a city. One level of the city is the older mediaeval city of relatively autonomous neighbourhoods of quite different kinds and with diverse organization, imposed over which is the grand plan of arterial roads emanating from the centre to the outskirts (see Amélie Oksenberg Rorty, "SelfDeception, Akrasia and Irrationality," in The Multiple Self, ed. Jon Elster [Cambridge: Cambridge University Press, 1986], 115-116). The first ‘city’ represents one’s intellectual habits, patterns salient in one’s perception, ways of interpreting one’s particular situation (see Amélie Oksenberg Rorty, "Akratic Believers," American Philosophical Quarterly 20 [1983]: 175-184), one’s capacity to be moved by empathy or excitement of societal trends (see Amélie Oksenberg Rorty, "Akrasia and Conflict," Inquiry 22 [1980]: 193-212), one’s moods, emotions, needs, fantasies, and so forth. They are, as it were, protointentional motives (see Amélie Oksenberg Rorty, "Self-Deception, Akrasia and Irrationality," in The Multiple Self, ed. Jon Elster [Cambridge: Cambridge University Press, 1986], 121). The second ‘city’ represents the judicial, legislative, and executive branches of the self that control rational agency. The latter is not the ruler or surveyor of the former; but rather, it depends on the former for its very exercises of rationality. In this respect, the political and social practices within which the individual lives are so central to the formation of systematic disorder in the agent’s character that “[w]hether we succeed in resisting the lure or pressures of akrasia is largely a matter of political and economic luck” (Amélie Oksenberg Rorty, "The Social and Political Sources of Akrasia," Ethics: An International Journal of Social, Political, and Legal Philosophy 107 [1997]: 656). Hence, in one sense, the ‘self’ is a loose configuration, made up of a cacophony of voices (see Ibid.: 655), and in another the ‘self’ is an agent who “takes the unification of his traits, his thoughts and his actions, as a central project” (Amélie Oksenberg Rorty, "Self-Deception, Akrasia and Irrationality," in The Multiple Self, ed. Jon Elster [Cambridge: Cambridge University Press, 1986], 131). Once again, however, it is hard to see how these two pictures of the self can be reconciled in such a way that the same agent accepts one judgment, yet acts on another. 116 Chapter 5: An Overview of Different Attempts to Explain Akrasia 260 depiction of a number of independent motivational, or other, systems fighting for hegemony is inherently unsuitable for a fitting clarification of the situation. After all, the person, if he or she is to feel remorse, must be the same person, with a single history, who hears the dictates of conscience, who feels the illicit pleasure, who decides accordingly, and who feels sorry for his or her action. Any other conceptualization would be inherently false to the situation. 5.3 An Old Approach: The Way Forward? Given that mediaeval and contemporary discussions of akrasia are deeply influenced by what Aristotle (384–322 BC) had to say about the subject and that “we have probably more to learn on the subject from him than from any other pre-twentiethcentury philosopher”,117 it is important that his position be treated in some detail. In Aristotle, the aim of ethics is to acquire practical knowledge: “we are inquiring not in order to know what virtue is, but in order to become good”.118 Indeed, reasoning itself does not motivate to act; only reasoning with a view to some desired end does.119 The latter kind of reasoning is called ‘practical reasoning’ where the aim in such reasoning is to discover what one should do in order to realize one’s end. Hence, in contrast with theoretical reasoning, where the good state is truth, the good state of practical reasoning is “truth in agreement with right desire”.120 It is clear that, for Aristotle, the concept of practical reasoning is connected intimately with the concept of desire, where his generic term for desire is . Indeed, Aristotle goes on to distinguish three different varieties of desires, which division appears to derive from Plato’s tripartite view of the soul. Hence, is desire for what reason presents to be good,121  corresponds to the ‘spirited’ part of the soul – it is the seat of anger and of shame – and covers desire for fame, social status, vengeance and honour, and  is desire for pleasure.122 Now, the latter two kinds of  develop before reason and, being unreasoned, one is in need of training – through guided 117 William Charlton, Weakness of Will. A Philosophical Introduction (Oxford: Basil Blackwell, 1988), 34. 118 Aristotle, Nicomachean Ethics, II, 1103b 27-29. 119 See Aristotle, Nicomachean Ethics, VI, 1139a 35. 120 Aristotle, Nicomachean Ethics, VI, 1139a 31-32. 121 See Aristotle, Eudemian Ethics, II, 1223b 7. 122 See Aristotle, De Anima, II, 414b 5-6. Chapter 5: An Overview of Different Attempts to Explain Akrasia 261 practice and habituation – in order to direct them in the right manner and to the right kinds of objects. This, however, does not mean that they are non-cognitive; rather is grounded in an evaluation of the object as noble, just or disgraceful, whereas  is oriented towards its object as something pleasant.123 Both are, therefore, modes of awareness of their object. The process of moral maturation is to learn to do what is virtuous, that is, to make it a habit or second nature to come to take the appropriate pleasure in doing virtuous actions: the virtuous person has the right dispositions and therefore enjoys doing virtuous actions for their own sake. Hence, in the person of virtue and practical wisdom, all responses and evaluations “are integrated … [and] infused and corrected by the reasoned scheme of values”,124 in such a way that “the virtuous person’s conception of what is truly pleasant is now shaped by his independent, reasoned conception of what is good …[so that] nothing will tempt or lure him as much as the temperate … action itself.125 Nothing else will seem as pleasurable”.126 Obviously very few of us are like that! The conceptual problem remains, however, even in our imperfect lives, because Aristotle draws a very strong conceptual connection between the intellectual virtues of  and  and the moral virtues.127 The virtuous one is able to live according to right reason. The truly virtuous person (the , being in this case the ), however, in contrast with the merely continent person (the ), does not have an appetite for self-indulgence For Aristotle, as for Plato in Timaeus (70D-71D),  does not have any cognitive powers. This position, however, has to do with the source of such an evaluative response; the contrast here is with the rational part of the soul. It is not that  has no intentionality: it is oriented to a ‘conception’ of an object as something pleasant. 124 M. F. Burnyeat, "Aristotle on Learning to be Good," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty (Berkeley/CA: University of California Press, 1980), 80. 125 Aristotle holds that  – or paradigmatic akrasia – is epithumetic akrasia, that is concerning actions involving the appetitive responses in the human person. In the temperate man, the appetitive element harmonizes with reason so that the temperate man desires the things he ought, as he ought, and when he ought, and this is what reason directs (see Aristotle, Nicomachean Ethics, III, 1119b 13-18). Akrasia that involves honour or indignation, that is, akrasia concerning actions involving the spirited responses of the human person are only called so by courtesy of an extension of the term (they are  126 M. F. Burnyeat, "Aristotle on Learning to be Good," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty (Berkeley/CA: University of California Press, 1980), 88. 127 See Aristotle, Nicomachean Ethics, VI, 1144b 15-25. In Nicomachean Ethics, II, 1106b 36-1107a2, Aristotle states that “[Moral] virtue, then, is a state of character concerned with choice, lying in a mean, i.e. the mean relative to us, this being determined by a rational principle, and by that principle by which the man of practical wisdom would determine it”. 123 Chapter 5: An Overview of Different Attempts to Explain Akrasia 262 and feels no pleasure in acting contrary to reason. Analogically with the continent, and in contrast with the vicious one (the here being the ), the incontinent (the ) thinks and knows that he or she ought to live according to reason despite contrary appetites. Hence, in contrast with both the virtuous and the vicious, both the continent and the incontinent are conflicted. Aristotle expresses the problem in the following manner: Now we may ask how a man who judges rightly can behave incontinently. That he should behave so when he has knowledge, some say is impossible; for it would be strange – so Socrates thought – if when knowledge was in a man something else could master it and drag it about like a slave.128 The counter-evidence, of course, makes it difficult not to attempt to resolve the dilemma. In his solution to the dilemma, Aristotle makes use of the basic distinction between actual and potential knowledge to bring out different senses of ‘knowing’. In four consecutive passages, in fact, he distinguishes various senses in which one can be said to ‘know’ something. In 1146b 31-36 he states that: [S]ince we use the word ‘know’ in two senses (for both the man who has knowledge but is not using it and he who is using it are said to know), it will make a difference whether, when a man does what he should not, he has the knowledge, but is not exercising it, or is exercising it; for the latter seems strange, but not the former.129 In 1146b 36 - 1147a 10 he points out that one can know the universal premiss ‘dry food is good for every man’, but either not possess, or not actualize the particular premiss that the food before one is food of that kind. In 1147a 11-24 he notes that one may have knowledge merely in the sense that one possesses knowledge in the same way as when one is asleep or mad or drunk, and further, that using the language of knowledge when one is in such a condition is no proof that one actively possesses knowledge. In the fourth passage (1147a 24 – b 18), Aristotle gives what has been termed a ‘psychological analysis’130 where he provides an example of conflicting motivations 128 Aristotle, Nicomachean Ethics, VII, 1145b 21. Aristotle, Nicomachean Ethics, VII, 1146b 31-36. 130 See Alfred R. Mele, "Aristotle on Akrasia and Knowledge," The Modern Schoolman 58 (1981): 139. 129 Chapter 5: An Overview of Different Attempts to Explain Akrasia 263 where two pieces of practical reasoning, involving two are present; the –motivation conflicts with the – motivation. Now, a great scholarly debate has raged regarding the exact uses Aristotle makes of the potential-actual distinction and whether he is giving four independent solutions or whether the last one is his real solution to the problem.131 Whatever the interpretation, it appears that Aristotle seems, at this point, to stand on the side of Socrates’ position in positing some kind of partial ignorance in the akratic. The key to the Aristotelian solution to the dilemma is his notion of the so-called ‘practical syllogism’. Indeed, in this context, which is the only place in the Nicomachean Ethics where Aristotle explains what he means by a practical syllogism, its introduction is an explanation of what to know something, without exercising that knowledge, means. Now, this is a hugely complex issue of which I shall only be giving, here, a general overview. Aristotle appears to set much store by the parallel between the theoretical and the practical syllogism. He notes that, like the former, the practical syllogism consists in two kinds of premisses. The first is a major premiss that pertains to something wanted, or needed, or good, and is universal in that it shows that such and such a kind of man should do such and such a kind of act. 132 The second is a minor premiss that relates to the particular or factual situational aspects of what, the person can see, falls under the major premiss. In different places, he notes that premisses (in 131 See James Jerome Walsh, Aristotle's Conception of Moral Weakness (New York: Columbia University Press, 1963), 118-158, for a review of the disputes. Burnet holds that Aristotle is presenting several solutions to one problem, Stewart that he is giving several solutions to several problems, Gauthier and Jolif that he is giving one solution in several stages. Also, many commentators, including most notably John Cook Wilson (Aristotelian Studies, I [Oxford: Clarendon Press, 1879], 19-23, 48-56, 61-63), have suggested that the test must be divided and amended. Gerasimos Santas maintains that Aristotle is, throughout, answering a logical, not an explanatory, query. Hence, Aristotle, in all four passages, is explaining what not having ‘full’ or ‘active’ or ‘explicit’ knowledge’ could mean. In the first three passages he explores what could go wrong with the knowledge provided by practical reasoning and in the fourth passage he shows that something goes wrong with the practical reason-wish motivation, and that the supplanting opinion-desire motivation explains the action. (Santas avers that Aristotle does not flesh out the possibility, of why, in some people, the desire causes the non-functioning of the practical-reasonwish motivation, because he only wants to answer the logical question ‘How is it possible for one to act contrary to his knowledge?’, and not the explanatory question ‘How can we explain an actual act of akrasia?’). Therefore, he holds that it is the fourth passage that gives Aristotle’s most complete elucidation. (See Gerasimos Santas, "Aristotle on Practical Inference, the Explanation of Action, and Akrasia," Phronesis 14 [1969]: 178-182.) 132 See Aristotle, De Anima, III, 434a 16. The point is that generalizations of such a degree of abstraction characterize reasoning and constitute the difference between animal desires (which can be verbalized in singular sentences such as ‘I want to drink’; ‘This is drink’ [see Aristotle, De motu animalium, VII, 701a 31-33]) and characteristically human practical reasoning. Chapter 5: An Overview of Different Attempts to Explain Akrasia 264 practical matters) are of two kinds, universal and particular,133 and that “premisses of action are of two kinds, of the good and of the possible”.134 The parallel between the theoretical and practical syllogisms is drawn out further when he insists that: The one opinion is universal, the other is concerned with the particular facts, and here we come to something within the sphere of perception; when a single opinion results from the two, the soul must in one type of case [theoretical reasoning] affirm the conclusion, while in the case of opinions concerned with production [practical reasoning] it must immediately act.135 Hence, the two premisses in the practical syllogism represent a logical structure that conceptualizes the dynamic structure of the soul. The conclusions or actions “are the expression of deliberative desire which is the result of the coalescing of desire … with the reasoning set out in the premisses”.136 The major premiss expresses an inclination or want while the minor premiss expresses some (constituent or productive) means related to the good wanted. Now this approach could well appear to be one where “practical principles form a closed, consistent deductive system, beginning with a priori first principles concerning the essence or nature of man”, where the system is notable for its lack of freedom since, from the a priori first principles, there follows “a closed, consistent hierarchy of rules of practice, covering both the moral and non-moral sides of human life, down to the smallest details: the decision to eat some candy, the need to make a cloak”. In deciding what to do, one need only “subsume the situation under the relevant rule, plug it into the right place in the hierarchy”. It is also important to such an interpretation that, “[s]ince it is to be a consistent system, there should be no incommensurable claims. In every situation there will be one appropriate response, and for every right action only one most appropriate justification”.137 133 See Aristotle, Nicomachean Ethics, VII, 1147a 1-4 and 1147 a 25-26. Aristotle, De motu animalium, VII, 701a 23-25. 135 Aristotle, Nicomachean Ethics, VII, 1147a 27-31. See also Aristotle, De motu animalium, VII, 701a 1017. 136 Anthony Kenny, Aristotle's Theory of the Will (London: Duckworth, 1979), 158. 137 Martha Nussbaum, Aristotle's "De motu animalium" (Princeton/NJ: Princeton University Press, 1978), 167. One must note that Martha Nussbaum radically opposes this ‘deductivist’ account. 134 Chapter 5: An Overview of Different Attempts to Explain Akrasia 265 This account is linked to the understanding of the practical syllogism advanced by D. J. Allan.138 The latter’s basic position is that Aristotle’s most mature and best thoughts on the nature of moral reasoning are not contained in Book III of the Nicomachean Ethics; rather, they are to be found in the sixth book of the same work. The difference, according to this influential article, is that whereas in the third book Aristotle writes as if all human action might be analysed according to a means–end model, by the sixth book, he has separated deliberation from choice and exemplified two irreducibly distinct patterns of practical syllogism. According to Allan, the point is that, in the sixth book, Aristotle limited deliberation () to instrumental (means–end) thinking, and averred that  or “choice with no ulterior end in view” has to do with “the man’s general aim or policy, reflecting his disposition of character”.139 In addition, he claims that Aristotle separated the kind of practical syllogism pertaining to instrumental thinking from that relating to overall disposition of character and insists that such a distinction is essential if Aristotle is to accommodate his insight that there is intrinsic value in ethical actions. For this latter type treats an action as an instance of a rule; in Allan’s words, “actions [are] … subsumed by intuition under general rules”.140 Such an approach might prove appealing but it does not stand. For one thing, even on a purely exegetical level, Aristotle, quite often, associates deliberation with practical reason.141 For another, Allan’s approach assimilates practical reasoning to the deductivist understanding that is characteristic of theoretical reasoning. In other words, it makes practical reasoning a question of acceptance of rules and of their ‘top-down’ application to particular cases and situations. What is interesting, and extremely pertinent here, is the use Aristotle makes of a parallel between the geometrical method of analysis and deliberation () in 138 See D. J. Allan, "The Practical Syllogism," in Autour d'Aristote: Recueil d'Études de Philosophie Ancienne et Médiévale offert à Monseigneur A. Mansion (Louvain: Publications Universitaires de Louvain, 1955), 325-340. The acceptability of such a picture of the application of rule/case practical reasoning as perception-mediated deduction is also taken on board by such authors as David Charles (in Aristotle’s Philosophy of Action [London: Duckworth, 1984]); and Terence H. Irwin (in "Some Rational Aspects of Incontinence," The Southern Journal of Philosophy 27 Supplement [1988]: 49-88). 139 D. J. Allan, "The Practical Syllogism," in Autour d'Aristote: Recueil d'Études de Philosophie Ancienne et Médiévale offert à Monseigneur A. Mansion (Louvain: Publications Universitaires de Louvain, 1955), 338-339. 140 Ibid., 336. 141 See Aristotle, Nicomachean Ethics, VI, 1141b 12-23, 1142a 11-23, 1142b 21-26. Chapter 5: An Overview of Different Attempts to Explain Akrasia 266 the third book of the Nicomachean Ethics.142 It is significant that Aristotle speaks here of a hypothetical starting-point of deliberation in terms of its aim () and describes how one tries to reason ‘backwards’, in the light of the aim, to ways and means to bring this about. It is also significant that he appears to draw a parallel between the auxiliary constructions which one posits in order to solve, analytically, a geometrical problem, and where there is no possible a priori guarantee of success, and the means) by which the desired result is to be brought about.143 Also significant is the weight Aristotle gives to the claim that the proper state of the practical intellect () is akin to a perceptual capacity (): [P]ractical reason () is concerned with the ultimate particular (), which is the object not of scientific knowledge () but of perception () – not the perception of qualities peculiar to one sense but a perception akin to that by which we perceive that the particular figure before us is a triangle; for in that direction as well as in that of the major premiss there will be a limit. But this is rather perception than practical wisdom, though it is another kind of perception than that of the qualities peculiar to each sense.144 [W]hen we speak of judgement and understanding and practical wisdom and intuitive reason we credit the same people with possessing judgement and having reached years of reason and with having practical wisdom and understanding. For all these faculties deal with ultimates, i.e. with particulars; and being a man of understanding and of good or sympathetic judgement consists in being able to judge about the things with which practical wisdom is concerned … Now all things which have to be done are included among particulars or ultimates; for not only must the man of practical wisdom know particular facts, but understanding and judgement are also concerned with things to be done, and these are ultimates … the intuitive reason involved in practical reasonings grasps the last and variable fact, i.e. the minor premiss. For these variable facts are the starting-points for the apprehension of the end, since the universals are reached from the particulars; of these therefore we must have perception, and this perception is intuitive reason ().145 It is important to see here that Aristotle’s parallel with the geometer’s challenge – in which there is no routine procedure for analysing a given figure into component figures that can then be used in constructing the said figure using rule and compass – does not lend itself well to a rule/case model. For, according to the latter model, practical reasoning is seen simply as a matter of having the right maxims or precepts 142 See Aristotle, Nicomachean Ethics, III, 1112b 15ff. See Jaakko Hintikka and Unto Remes, The Method of Analysis: Its Geometrical Origin and Its General Significance (Dordrecht: D. Reidel Publishing Company, 1974), 87. 144 Aristotle, Nicomachean Ethics, VI, 1142a 26-30. 145 Aristotle, Nicomachean Ethics, VI, 1143a 25-1143b 5. 143 Chapter 5: An Overview of Different Attempts to Explain Akrasia 267 and putting them together with the results of perception where perception merely contributes facts, presumably available to anyone. Such a model, however, would see ‘practical perception’ as disjointed from one’s practical intellect. It would posit a conceptual distance between the choice of a particular action with a view to  or ,146 and  or  itself, a conceptual distance of which Aristotle clearly disapproves. For, if one understands Aristotle carefully, the particular practical judgment implicit in a  is at no conceptual distance from the concept of ; the latter is precisely what, in a , one sees what one is doing as being.147 In addition, such a model would imply that the general conception of a good life () is codifiable in some format that would enable deduction of specific practical conclusions. On the contrary, however, Aristotle appears to hold that the fact that some factors rather than others strike a person as practically relevant is what determines which concern he or she acts upon in a given situation. Hence, it seems that one cannot give a deductive argument in favour of seeing some feature rather than another as what matters: the concept of perception marks the point at which one has run out of discursive justifications or reasons.148 In John McDowell’s words: In terms of the “practical syllogism” what the special capacity that Aristotle appeals to yields is not awareness of the truth of the minor premiss (that is presumably afforded by ordinary cognitive capacities), but its selection from among other features of the situation as minor premiss – as what is practically significant. To use the terminology of De Motu Animalium 701a 9ff., the “premiss of the good” is the content of the correct conception of doing well (cf. 1144a 31-33); in the light of that, one feature rather than another of the situation has its description serve as “premiss of the possible” i.e., as  (that is ) just is (see Aristotle, Nicomachean Ethics, I, 1095a 1820, 1098b 20-22). See also James Jerome Walsh, Aristotle's Conception of Moral Weakness (New York: Columbia University Press, 1963), 146-148. 147 Aristotle insists that choosing some specific action with a view to  is choosing it as an instance of  (or ). For  or is constituted by the doings, undertaken for their own sake, which make up the good life; it is not some further aim that someone who lives the good life has in mind (see Aristotle, Nicomachean Ethics, VI, 1139a 32-1139b 5). 148 This may well be the reason why, in terms of the well-trodden question as to whether virtue could be taught, Aristotle shuns Socratic intellectualism and emphasizes the importance of beginnings and the gradual development of good habits () of feeling and acting. Even in the absence of argument, however, one can show the good life by reminding people who share it, listing virtues, and providing character sketches of those who possess them. Aristotle holds that only when one possesses the beginning – the that – without explanation as second nature, could one move on to the because of one’s conviction about what is noble and good. (See M. F. Burnyeat, "Aristotle on Learning to be Good," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 6992.) 146 Chapter 5: An Overview of Different Attempts to Explain Akrasia 268 pointing to something that can be done to gratify the orectic state whose content is the premiss of the good.149 McDowell continues to warn, justly, that the doxastic capacity that selects the minor premiss cannot be separated from the orectic state whose content is expressed in the major premiss. The reason is that the orectic state cannot be identified with maxims or principles that one could work out independently of their application to particular cases or situations. He concludes that: [T]here is nothing for [the correct conception of doing well, or the orectic state] to be except the capacity to get things right occasion by occasion: that is, the perceptual capacity (in a pregnant sense of “perception”) that singles out the right minor premiss. The two premisses are reflections of a single state, which can be viewed indifferently as an orectic state and as a cognitive, specifically perceptual, capacity.150 On this view, therefore, the ‘conception’ of a good life one has resides and shows itself in one’s evaluations of various situations, in one’s various projects and doings, in the level of care and concern one shows in one’s actions and words, and most fundamentally in one’s ‘deliberative wantings’ or choices.151 In the light of this treatment of the ‘practical syllogism’ one can now attempt to see what exactly goes wrong in the case of akrasia. Indeed, the fact that the has the capacity to select from the vast number of features present in a situation, and see correctly what is practically relevant, and only register through ordinary perception – but attribute no practical significance to – what is not, shows why akrasia proves so difficult to accommodate. For, one would like to say both that the akratês matches the  in his or her view of what is practically salient in the situation, and that he or she succumbs to a temptation that the  would see to be practically irrelevant. One has to clarify how the akratês sees things, both as the  would, and differently. Aristotle’s famous statement on the question is the following: 149 John McDowell, "Comment on Irwin's "Some Rational Aspects of Incontinence"," Southern Journal of Philosophy 27 Supplement (1988): 93-94. I am grateful to John McDowell for this interpretation of Aristotelian deliberation in the light of the weight Aristotle attaches to the claim that the practical intellect is a perceptual capacity. 150 Ibid.: 94. 151 See Aristotle, Nicomachean Ethics, III, 1111b 6-7. Chapter 5: An Overview of Different Attempts to Explain Akrasia 269 Now, the last premiss () both being an opinion about a perceptible object, and being what determines our actions, this a man either has not when he is in the state of passion, or has it in the sense in which having knowledge did not mean knowing but only talking, as a drunken man may utter the verses of Empedocles. And because the last term is not universal nor equally an object of scientific knowledge with the universal term, the position that Socrates sought to establish actually seems to result; for it is not in the presence of what is thought to be knowledge proper that the affection of incontinence arises (nor is it this that is ‘dragged about’ as a result of the state of passion), but in that of perceptual knowledge ().152 Much ink has been spilt on the exact interpretation of ‘’ which somehow is ignored or forgotten during the state of passion, where the views can largely be divided into two camps: the more traditional camp interpreting it as the minor premiss and a relatively more recent case for the position that it refers to the conclusion.153 The main general difference between these two groups is that, whereas on the first interpretation, Aristotle is not far at all from the Socratic position of explaining all wrongdoing as being due to ignorance, 154 on the second view, clear-eyed 152 Aristotle, Nicomachean Ethics, VII, 1147b 9-18. It is important to note that the differences between the two camps turn on questions of interpretation of the conceptual interrelations between key concepts like ‘’, ‘o’ and ‘’ rather than on different psychological mechanisms. The point is that those in the first camp hold that, since the akratês is ignorant of the minor premises, no good results. Those in the second camp maintain that there is a distinction between the conclusion of the practical syllogism and the action it indicates. Hence the latter must explain how it is that the choice that is the deliberate desire of something within one’s power that emerges as result of deliberation, and which normally connects the result of deliberation with the action that is the outcome of the deliberation, is voluntarily () acted against. (See Risto Saarinen, Weakness of the Will in Mediaeval Thought: from Augustine to Buridan [Leiden: E. J. Brill, 1994], 12-17.) 154 The interpretation of akratic ignorance within this first camp can be split into two: the more traditional understanding has been the view that passion simply inhibits the akratês from knowing the minor premiss of the ‘good’ syllogism with sufficient clarity (supporters of this view include David Ross, Aristotle [London: Methuen and Co. Ltd, 1977], 223-224; James Jerome Walsh, Aristotle's Conception of Moral Weakness [New York: Columbia University Press, 1963], 149-158; John Lloyd Ackrill, Aristotle's Ethics [London: Faber and Faber, 1973], 31-33 [on one interpretation he provides]; Richard Robinson, "Aristotle on Akrasia," in Articles on Aristotle: Ethics and Politics, II, ed. J. Barnes, M. Schofield, and R. Sorabji [London: Duckworth, 1977], 82-83, 86-87; William Francis Ross Hardie, Aristotle's Ethical Theory [Oxford: Oxford University Press, 1980], 276-293; David Wiggins, "Weakness of Will, Commensurability and the Objects of Deliberation and Desire," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 247-251; and William Charlton, Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 46-47); another view, however, has maintained that what happens, under the influence of passion, is that the akratês does actually know both the major and the minor premiss of the ‘good’ syllogism’, yet fails to connect them properly, with the result that the proper conclusion is not reached. A number of these scholars have seen support for this, second view in Aristotle’s treatment of error in Analytica Priora II 21, 67a 31-67b 6, where he admits the possibility of possessing both the major and the minor premiss but failing to connect them properly (Harold H. Joachim, Aristotle: The Nicomachean Ethics, ed. D. A. Rees [Oxford: Clarendon Press, 1951], 223-229; Raphael Demos, "A Note on Akrasia," Ethics 71 [1961]: 196; 153 Chapter 5: An Overview of Different Attempts to Explain Akrasia 270 akrasia is allowed for by the peripatetic philosopher, since the akratês does reach the conclusion of the practical syllogism, yet he or she voluntarily violates it. 155 It has also been suggested that Aristotle himself did not intend to exclude any of the possibilities, but wanted to explain both instances in his treatment of the question.156 John McDowell, holds that Aristotle focuses on a breakdown at the minor premiss of practical reasoning. He holds that the Aristotelian would clearly see that the fact that bodily pleasure is illicitly available is practically irrelevant and counts for nothing. Indeed, he maintains that the clear perception of the requirements of virtue of the means that what would otherwise be possible reasons or desires for acting differently are not merely overridden, but are silenced. The latter simply do not count as reasons for acting in that manner.157 While he continues to hold this interpretation of the , McDowell shows a certain development in his interpretation of Aristotle’s akratês. In a 1978 article he states: and René Antoine Gauthier and Jean Yves Jolif, Aristote: l’Ethique à Nicomaque: Introduction, Traduction et Commentaire, II [Louvain: Publications Universitaires de Louvain, 1959], 602-617 [although the view of the latter is somewhat more complex: they maintain that, under the influence of appetite, the akratês fails to connect the particular premiss – which can be subsumed under either of the two major premises – with the moral rule, but connects it with the major premiss of appetite], have been seen as upholding this view). 155 Kenny and Pears say that Aristotle allows clear-eyed akrasia (see William Charlton, Weakness of Will. A Philosophical Introduction [Oxford: Basil Blackwell, 1988], 41; Anthony Kenny, "The Practical Syllogism and Incontinence," Phronesis 11 [1966]: 180-184; Aristotle’s Theory of the Will [London: Duckworth, 1979], 160-166; and David Pears, Motivated Irrationality [Oxford: Clarendon Press, 1984], 233-240). It is interesting to see that Burnet too holds that Aristotle allows for clear-eyed akrasia. His view is that, in akrasia, the moral syllogism can be completed but, in the absence of , to which it can present itself, nothing happens. For,  alone moves nothing in the absence of or  . In the akratês,  occupies the soul (see John Burnet, The Ethics of Aristotle [London: Methuen and Co. Ltd, 1900], 303). Charles maintains that Aristotle would argue that, in intentional akrasia, it is the conclusion that is not properly grasped; such an akratês reaches the conclusion but holds it in an off-colour fashion and does not act on it (see David Charles, Aristotle's Philosophy of Action [London: Duckworth, 1984], 120-121). It is worthy of note that, while Alfred Mele does not hold that  means ‘conclusion’, he still sustains that the weak akratês acts against the deliberative conclusion (see Alfred R. Mele, "Choice and Virtue in the Nicomachean Ethics," Journal of the History of Philosophy 19 [1981]: 416fn.19; "Aristotle on Akrasia and Knowledge," The Modern Schoolman 58 [1981]: 147-157). Dahl too holds that  is not the conclusion (for he maintains that Aristotle’s position is that the conclusion of a practical syllogism is an action). He sustains that it refers to something that expresses the agent’s recognition of what he or she ought to do and, in this light, he maintains that Aristotle allows for genuine conflict of motives in weak akrasia (see Norman O. Dahl, Practical Reason, Aristotle, and Weakness of the Will [Minneapolis: University of Minnesota Press, 1984], 190, 206-207). 156 See David Charles, Aristotle's Philosophy of Action (London: Duckworth, 1984), 124-128. 157 See John McDowell, "Are Moral Requirements Hypothetical Imperatives?" The Aristotelian Society Supplementary Volume 52 (1978): 26-29. Chapter 5: An Overview of Different Attempts to Explain Akrasia 271 The way out is to attenuate the degree to which the continent or incontinent person’s conception of a situation matches that of a virtuous person. Their inclinations are aroused, as the virtuous person’s are not, by their awareness of competing attractions: a lively desire clouds or blurs the focus of their attention on ‘the noble’.158 In a later article, however, he clarifies and amplifies his interpretation, stating that one could contemplate two different kinds of flaws. One flaw would regard the content of the akratês’ knowledge of the dictates of practical reasoning and their application in particular circumstance, the other is a flaw where the akratês’ grasp itself is defective; it is not yet fully integrated into his or her habituated tendencies with respect to his or her emotions. Hence, one could contemplate a situation where the content of the akratês’ practical thinking is correct – in an attenuated sense – with regards to both his general conception and the choices he ‘sort of’ makes, in knowing that his appetite has no practical significance in the present circumstances, and where the akratês nevertheless acts on the latter. McDowell holds that Aristotle is committed throughout to conceiving “the development of the state of the practical intellect as inseparable from the development of the proper state of ethical character”. 159 The problem of the akratês is that, unlike in the case of the , the practical thinking with the sort of content he or she is akratic about is not yet fully ingrained into his or her character. The unfortunate result is that, when one is akratic, one lets oneself be swayed by considerations which one’s own practical thinking shows one to be practically irrelevant in the circumstances.160 Perhaps one of the greatest contributions of this interpretation of Aristotle’s account of akrasia is that it clearly shows that one cannot understand moral choices in terms of individual moments in an agent’s life. It embeds Aristotle’s examination of akrasia in the larger context of the character () and the disposition () that a 158 John McDowell, "Are Moral Requirements Hypothetical Imperatives?" The Aristotelian Society Supplementary Volume 52 (1978): 28. 159 John McDowell, "Comment on Irwin's "Some Rational Aspects of Incontinence"," Southern Journal of Philosophy 27 Supplement (1988): 97. McDowell refers, here, to M. F. Burnyeat, "Aristotle on Learning to be Good," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty (Berkeley/CA: University of California Press, 1980), 69-92. 160 Dahl’s position is extremely similar. He notes that: “an akratês can know what he ought to do in the situation that faces him. His knowledge of general principles is sufficiently integrated into his character to insure that he will often act on them, but it is not sufficiently integrated to prevent either conflicting desires from arising, or to prevent action on strong conflicting desires should they arise, In this way his knowledge differs from both that of the enkratês and the phronimos” (Norman O. Dahl, Practical Reason, Aristotle, and Weakness of the Will [Minneapolis: University of Minnesota Press, 1984], 216). Chapter 5: An Overview of Different Attempts to Explain Akrasia 272 moral agent acquires through doing and through becoming habituated to doing. It recognizes that Aristotle is not being bland in linking virtue and practice, but rather, that he holds that practice has cognitive features in enabling the agent to learn what is noble and just. In so doing, McDowell recognizes two central features of Aristotle’s account. The first is that a constitutive aspect of the proper state of the practical intellect is precisely that of habituated inclinations with respect to feelings and emotions.161 This means that his is a recognition that Aristotelian virtue is a disposition for acting and for feeling, a recognition that allows him to escape the intellectualistic point of view that sees practical reasoning as divorced from the training of emotions and desires and that views non-cognitive training as a mere aid in ensuring that one is able to listen carefully and without any distractions to what practical reasoning – which has its own, independent, content – is saying. The second, central, feature McDowell points out is precisely the recognition that one cannot divorce an agent’s choices from his or her character. He keeps the important Aristotelian dictum that “choice () cannot exist without reason and intellect or without a moral state”162 in mind, in maintaining that both and demand reference to the larger context of and , that is, to the larger context of the goals and habits of the moral agent. Hence, for a person who is perfectly formed by the exercise of temperance to live such a good life, on a specific occasion where this person sees that he must sacrifice an otherwise attractive prospect of pleasure, it is not a question of seeing that the gain in acting temperately outweighs the loss of pleasure incurred. It is rather that, in the circumstances, missing the pleasure is no loss at all. And this is why Aristotle can assert that the fully mature person of virtue and practical wisdom cannot be akratic; he or she simply has no reason to be. The akratic, hopefully, is on the way to such integration of character, in learning to take appropriate pleasure in good activities. 5.4 A Comparison with the Account Proferred by Donald Davidson As a final note, it is important to compare briefly Aristotle’s conceptual account with the one proferred by Davidson in his 1969 article. After all, Davidson’s analysis is, This aspect of Aristotle’s account has been beautifully developed by L. A. Kosman ("Being Properly Affected: Virtues and Feelings in Aristotle's Ethics," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty [Berkeley/CA: University of California Press, 1980], 103-116). 162 Aristotle, Nicomachean Ethics, VI, 1139a 33-34. Italics are mine. 161 Chapter 5: An Overview of Different Attempts to Explain Akrasia 273 perhaps, the most influential contemporary account in the literature. In addition, his early account is a conceptual one, based on a characterization of moral conflict and, prima facie, his work is a rather radical development of the Aristotelian analysis. As explained above, the analogy explored by Davidson in making his conceptualization of akrasia is that between a judgment of probability relative to the contextual evidence, and a decision taken in relation to judgments of the desirability of an action relative to such and such contextual facts. In both cases, the analogy between the requirement of total evidence for inductive reasoning and the principle of continence ensures that the more exhaustive the set of considerations that issue in the conclusive judgment is, the more rational the resultant unconditional judgment detached from the premisses will be. It is clear, however, from our consideration of Aristotle’s work, that this appealing analogy is marred by a crucial and fatal simplification. For, Davidson assumes that there are formal criteria that allow for the possibility of the conjunction of premisses of the competing syllogisms in order to form all-things-considered judgments, which take the form ‘that p makes it better to  than not to ’ where p is the conjunction of all available propositions which the agent judges to be both true and relevant (HW 38-41). Hence, in his view, the most rational practical syllogism is such because it embraces more circumstances and considerations than its rivals. The problem appears as soon as one sees that there are no such formal criteria to compare competing syllogisms. Indeed, evaluation of possible actions is, of its essence, dialectical and is directed by the practical perceptions, reasonings, and purposes of the agent that gave rise to the syllogisms in the first place. There is no question at all of any predictive power of practical syllogisms; for an agent asks herself “what shall I do?”, not with a view to maximising anything (whether it be rationality or pleasure, and so forth), but in response to a particular context that makes demands on her practical perception, where a high degree of practical, situational appreciation (what Aristotle would call ) is called for. For, the conception of excellence of the agent, itself, places her in a particular world of morally relevant facts and faced with a prima facie conflict the agent’s moral task is to make out how things really are, and what action to posit, in the light of her conception of the good life. Hence, it is not a question of maximising some bundle of goods that are recognizable as such from outside any view Chapter 5: An Overview of Different Attempts to Explain Akrasia 274 of moral excellence. There is no possible external decision procedure, applicable from a morally neutral standpoint. And even within the agent’s conception of  and , her concerns do not form a closed, complete, and consistent system and the value of the different concerns is not necessarily fixed a priori. What an agent often faces is a question of choosing what to do in the face of a plurality of possible ends and goods. Even if the conception of and  is a unity insofar as the results of the agent’s efforts at discernment tend to hang together, this is not always the case and the concerns and actions in specific situations might, indeed, contribute towards the agent’s modifying her conception of a good life. What is in play is a question of acting in the light of one’s conception of the good life that is also one’s self-making. This does not mean, of course that all decisions are agonizing ones; some are pretty clear and easy to take for most agents: in a technical arena like medicine, for instance, a doctor might well agonize about the specific treatment to administer to a particular patient in a difficult case but, in most cases, she will not need to do so. Indeed, certain procedures in crafts and technical practices and certain virtues in a person’s life demand that certain facts count for nothing in the situation. Such could happen, for example, in a situation where the courageous agent sees that the likelihood of suffering grievous injury is silenced in the face of the defence of the innocent against an aggressor.163 However, there obviously are situations where the practically relevant facts are only dimly recognized by the agent and, as it were, she needs to peer more closely in order to see what to do. The point throughout here is that there is some important way in which the cognitive appraisal of a situation differs relative to the character of the agent: there is often no independent account of the action, i.e., no account without reference to the agent’s character or lack thereof, that shows it to be rational or not. Davidson’s account, on the lines of Hempel’s logic of explanation, is a top-down view of practical reasoning. His aim seems to be to fit akrasia into a complete conceptual account of deliberation and action that could constitute the materials for a theory that would enable a prediction of the action. What must, rather, be sought is a framework which articulates 163 See David Wiggins, "Deliberation and Practical Reasoning," in Essays on Aristotle's Ethics, ed. Amélie Oksenberg Rorty (Berkeley/CA: University of California Press, 1980), 234. Chapter 5: An Overview of Different Attempts to Explain Akrasia 275 the mutual relations of an agent’s concerns and her perception of how things are in the world, and which enables one to understand the complex yet unitary ideal the agent tries, in the process of living her life, to make true within the concrete situations in which she makes specific choices. The connected flaw is that, in Davidson’s account, the question of the agent’s conception of the good is irrelevant. Indeed, akrasia, in his view, would be the somewhat trivial matter of an agent’s deciding in light of some reasons when she knows that there are more, or better, reasons that she ought to take into consideration which advocate otherwise. The whole emphasis of his account is in terms of the idiosyncratic moment in the agent’s life. And there, again, it focuses on the individual act as divorced from the state of character of the agent that enables and situates her choices and the complex human practices that give the act intelligibility. It points towards the irrationality of akrasia, on the lines of an improper inference from a conditional allthings-considered judgment that a certain course of action is best to an unconditional judgment that another action is best, rather than the agent’s awareness of the wrongness or foolishness of her action. It is clear that one must continue to press forward for an analytic account that brings out the conceptual connections between the agent’s conception of the good, his or her perception of how things are in the world and the real possibility of akratic action. 5.5 Shewing the Fly the Way out of the Fly-bottle Again After having seen various descriptions of akrasia, it has been evident, throughout the discussion, that the formulation of an adequate conception of akrasia shows itself to be heavily dependent on the grammar of one’s use of terms. This has been especially evident in the presentation of Aristotle’s account, where one could immediately appreciate the additional clarity gained from the different point of view advanced. It is obvious that more clarity is needed for it is clear that many philosophical accounts of akrasia are disguised in some fashion by a certain amount of nonsense. Thus, there is an urgent need to resolve the confusion and, seeing that the use of terms Chapter 5: An Overview of Different Attempts to Explain Akrasia 276 related to akrasia is grammatically related to the description of one’s character, a good way to start could be to investigate this concept more thoroughly. One should investigate the circumstances and presuppositions of use of such words and learn therefrom in order to achieve more conceptual clarity. The difficulty, of course, is that the philosophical prejudice is certainly not a stupid one. Perhaps, to a greater degree than in the case of self-deception, this quest is one for the patient and the brave!