Socrates and Akrasia: Reading Protagoras 351a – 358d

Aristotle gives two formulations of Socrates’ view of akrasia. One is that he thought it strange that `if when knowledge was in a man something else could master it and drag it about like a slave’. The other is that `there is no such thing as incontinence; no one, …, when he judges acts against what he judges best – people act so only by reason of ignorance’. Aristotle remarks, rather tersely, that this view `plainly contradicts the observed facts’. And this has been, more or less, the assessment of Socrates since then. Except that Aristotle does not say, as the Oxford translator would have it, `observed facts’. He says that Socrates’ view contradicts `ta phainamena’, i.e. what appears or seems to common opinion. Aristotle’s own view of akrasia, like Socrates’, does not remain content with `ta phainamena’, though, characteristically, he tries to make room for them in his moral theory.

Aristotle does not discuss Socrates’ view in detail. His report of it is, I suppose, an extrapolation of remarks made by Socrates in the course of a dialectical refutation of Protagoras’ views about virtue and its teachability in the Protagoras. The purpose of this paper is to suggest that the traditional understanding of Socrates on this issue needs rethinking, and that this is best done by attending to the dialectical context of Socrates’ argument. This reveals that Socrates’ apparent rejection of ta phainamena of weakness is part of a dismantling strategy – a `deconstruction’ if you like. What he deconstructs is a common view of virtue which regards it as fragmented. A dialectical refutation, like all deconstructive practices, is not, strictly, a disproof. The latter requires self-evident or universally accepted premises, while the former aims to stimulate intellectual self-consciousness about a view in those who espouse it or are tempted to do so. It proceeds by revealing that the various strands of belief that `construct’ a view represent conflicting tendencies that the view itself cannot totally control – thus undermining what the view aims to achieve, the unification of these beliefs into a single theoretical vision. A piece of deconstruction can lead to an alternative `construction’, but it need not do so. We must, therefore, be careful not to interpret various theses Socrates employs in the course of his `deconstruction’ of Protagoras, as components of a theory of weakness which gives an alternative to the view of it commonly held. The dialectical lesson may just be that no coherent view of human virtue can allow the picture of weakness that ta phainamena suggest. I try to show that this is precisely what the Protagoras seeks to establish.

Section 1: The Dialectical Setting –

Courage, Confidence and Daring

Just prior to the section of the Protagoras we are concerned with, Socrates had tried unsuccessfully, to persuade Protagoras that wisdom (sophia) is courage (andreia) (348b -351b). This was in response to Protagoras’ claim that not only was andreia distinct from sophia, but that it had nothing to do with the other virtues. A man, Protagoras had said (349), can be most unjust (adikos), impious (anosoios), dissolute (akolastos), and ignorant (amathes) who nevertheless is most courageous. The issue whether virtue forms a unity, and whether it is something teachable, had been the central one since the beginning of the dialogue. This exchange over sophia and andreia reveals a crucial difference between Protagoras and Socrates – one which dictates the course of the subsequent discussion. To fully appreciate its nature we need to bear in mind the following points:

(i)sophos (wise) and sophia (wisdom) have a wider application than the English words which translate them. `Sophos‘ covers (a) a sage or wise old man and (b) skilled artists and craftsmen. The latter are called sophoi by virtue of a skill (techne) that requires intelligence and/or `know-how’. Correspondingly, sophia may involve either wisdom or knowledge (cf., for example, 360d).

(ii)The notion of a sophist (Protagoras admits to being one) is related to sophos in that it suggests the possession of intelligent skill.

(iii)Protagoras had alleged at the beginning of the dialogue that he taught the craft related to being a citizen – being good at conducting one’s own affairs and those of the polis. Protagoras undertakes to make his pupils better citizens. As such he makes some claim to possess sophia, but he does not say whether his instruction makes his pupils wise (sophoi).

(iv)In response to Socrates’ doubts as to whether virtue, unlike other ordinary skills can be taught, Protagoras claims that, of course, good character can be taught. The various virtues which are conjoined in the virtuous person (in those who are good and noble) can be inculcated through upbringing that operates by punishments and rewards.

(v)Protagoras leaves obscure whether traditional upbringing, and his own further style of training, achieve their result (a person who acts well and nobly) by making people capable at something, or in some other way. Thus, the phrase `to become better as a citizen’ is ambiguous between `becoming a better (more desirable or acceptable) citizen’ and `becoming more capable as a citizen’.

(vi)In sharply separating andreia from sophia, Protagoras is introducing a split between the two aspects of sophos (distinguished in (i)); one can be courageous without either being wise or possessing some sort of intelligent skill. This throws doubt on whether courage, for Protagoras, implies any intelligent capacity in the person manifesting it.

The question whether all virtue involves some intelligent capacity (dunamis) is linked to another aspect of Greek moral thinking. The good person is someone who does good things (things beneficial to himself and others), as well as noble things (kala). But ordinary morality left it unclear how goodness related to nobility. One might agree that doing good things requires intelligent skill, but does performing noble acts do so? If it does not, how does living well as a human being relate to acting nobly?

These points form the background to the disagreement between Protagoras and Socrates concerning courage. Socrates tries to show that it is in virtue of their specific skill or expertise that people do daring things (`things most men fear’). The confidence shown by the courageous, something fine and admirable (kalon), has as its source the capacity for intelligent assessment of a situation afforded to them by their `know-how’. Foolhardy people, by contrast, dare to do things in ignorance of the risks. Their confidence (Protagoras admits) is insane and, therefore, something base (aischron). The word tharraleos (bold) is usually applied to those who do something dangerous either because they are not deterred by the danger or because they are confident of the outcome. Thus, tharraleos can mean either (a) confident, unafraid or (b) daring, willing to face danger. Now, clearly, the andreios must somehow combine (a) and (b) in a way which excludes insane daring. The problem is to characterise the source of his or her confidence. What is it that gives the courageous – Socrates seems to be asking – their capacity, their `power’, to be daring?

Socrates tries to conclude that it is the wise daring of the knowledgable which is courage, since foolhardy confidence is base. Note here that what Socrates puts to Protagoras is a distinction between two kinds of reason which may motivate a person to face danger; one based on a comparative estimate of the risks involved for the agent in relation to the overall worth of the action, the other based on the agent’s attraction by the worth of the action in disregard of considerations of risks. In calling the latter type of action `mean’ and `base’, Protagoras seems to run together foohardiness and impulsive daring. But foolhardy action is not `reason-less’; it need not be impulsive. It is certainly unintelligent daring, a confidence born of lack of consideration for relevant facts, but it does not have to be acting on `mad’ impulse. Actions of the latter kind can look foolhardy, but they are not due to foolhardiness. The lack of consideration implied by foolhardiness may have all kinds of motivation (for example, phantasy self-images of heroism, a wish for social esteem, and so on) without its being the case that it is action ‘without reason’.

Plato ingeniously captures Protagoras’ fudging in the way he presents the latter’s resistance to Socrates’ conclusion. Protagoras acknowledges that the daring of experts is fine and laudable because their confidence is based on something admirable in itself, skill and `know-how’. But given his view of courage and its relation to wisdom and the other virtues, he cannot allow that courage is due to anything like an intelligent capacity. At the same time he cannot but agree that foolhardy daring is base. But if he accepts that baseness accrues to such daring because it is unintelligent confidence, he will have a hard time arguing that courage does not involve some sort of intelligent capacity – and the connection with sophia will be re-established. By calling foolhardy daring `mad’, he can avoid committing himself on the question whether the daring of the courageous has its source in some intelligent capacity.

Protagoras’ reply to Socrates is significant. He did not agree that the bold are courageous, only that the courageous are bold. And Socrates’ manner of showing that the daring of the courageous is inspired by the confidence they gain from their knowledge will not do. For one could argue, in a parallel fashion, that the strength of the (physically) powerful comes from wisdom; a manifestly false view. The parallel argument is:

(i) the strong are powerful (i.e. capable, for example, to lift weights or wrestle).

(ii) the source of that capacity is knowledge (skill or `know-how’); a person with such skill is more capable than someone without it, and more capable with it than without it.

(iii) it is wisdom which is strength.

But while strength implies capacity, capacity due to skill does not imply strength. You may be physically stronger than me, but my wrestling skill may enable me to defeat you. I am stronger than you only in the context of wrestling; I am not physically stronger than you tout court. Thus, for Protagoras, there are two entirely disparate sources of strength – one that comes from natural bodily constitution and good nurture (eutrophia), and one, in the sense of capacity (dunamis), that comes from skill. Similarly with daring and courage. Daring can come from skill or from rage or madness. But, like strength, courage has its source in `natural’ psychic constitution aided by `good nurture’ of the soul. Building a courageous soul, therefore, is like developing the muscles of the body; virtue, on this view, is the spiritual analogue of body-building – you do it bit by bit. What is unclear, however, is how the `nurturing’ of various psychic `parts’ results in the unity of a good and noble person (`soul’), as distinct from a doer of disparate noble acts.

The consequence of this way of blocking Socrates’ argument is far-reaching. It replaces Socrates’ distinction between two kinds of reason for acting daringly with a dichotomy between acting courageously (virtuously) and acting by virtue of intelligent capacity. Reason goes on the side of passion or affect: they are both sources of daring but not of courage; the nobility of the latter is not due to them but to something else. And though the daring of the experts is something fine, while that of mania is a disgrace, the nobility or fineness of courage is due to something entirely other. The excellence of courage is, thus, simultaneously cut off from the two main sources of human motivation, reason and passion or affect. What, then, makes courage something attractive in itself?

Section 2. The Framework of Socrates’ Reputation:

Pleasantness, Pleasure and Living Well

Both the content of, and the abrupt transition to, new topics in Socrates’ response to Protagoras have puzzled commentators. Socrates is apparently pressing Protagoras to accept some sort of Hedonism, and when the latter resists, he solicits his support in rejecting the common view that weak action is evidence that knowledge (of good and bad) can be overruled by passion or pleasure. Either Plato has lost control of the dramatic writing of the dialogue or Socrates’ transition to apparently new topics is soundly motivated.

The comments in the previous section ought to help us understand this motivation. For if the nobility of courage is cut off from both intelligence and passion as sources of motivation, two questions become difficult to answer. The first concerns the link between acting admirably (kalos) and the capacity to lead a good life. The second question becomes pressing if our answer to the first suggests that the relation between the capacity to lead a good life and acting nobly is a contingent or an `external’ one. For then the fact that an agent possesses the capacity to estimate the goodness or badness of a course of action or a life plan will not supply a motive or reason for that agent choosing to act courageously or virtuously. So, it becomes possible that choosing to act virtuously (e.g. with courage) is to choose in a way contrary to one’s judgement as to what is best for one, or for one’s life. And though acting courageously is admirable, while being ruled by fear is base and contemptible, the moral psychology of acting courageously, on this view, would be difficult to distinguish from that commonly attributed to those who do bad things being overcome by lust, fear, or some other passion. Can one be `overcome’ by courage into doing fine things in the way one is overcome by fear or lust into doing base things? Sure enough, Protagoras does not think of courage as a passion, but this does not by itself block the possibility that for him, courage acts as a force leading one to act nobly.

In the section (351b 4-353b) prior to his discussion of doing bad things knowingly, being overcome by passion or desire, Socrates carefully plots the considerations that will force on Protagoras the need to retract what he has said about courage and its relation to wisdom and the rest of virtue. It is essential to recognise how Socrates’ remarks on weakness form part of that overall strategy. It makes a world of difference to how we understand the direction Plato means his readers to follow in reflecting about the issue. Central to this strategy, I believe, is the attempt to convince Protagoras of the following: since what has to be made intelligible is how the courageous (or virtuous) person chooses to act, Protagoras’ contrast between courage and the capacity by which people are made confident to face dangers cannot be right. The capacity to estimate the value (the goodness or badness) of things is a manifestation of intelligence in human beings; it is by virtue or excellence in that capacity that their getting things right or wrong itself becomes something valuable. But if the source of courage is, as Protagoras implies, something other than what prompts people to exercise that intelligent capacity, it becomes difficult to see in what way moral excellence is `internal’ to human excellence in general. If courage is an excellence, then choosing to do the courageous thing ought to come out as an instance of choosing well. Consequently being interested in or attracted to choosing excellently, the motive to exercise intelligence in the conduct of one’s life, cannot be separated from the motive to act virtuously. The courageous, being virtuous, must have good reasons for what they do, and they must be interested in having such reasons.

It is as part of this overall strategy that Socrates’ remarks (351b-e) about pleasure and living well need to be understood. If there is such a thing as living a life well or badly then there must be some standard of value or goodness by which one can choose a life, a criterion by which people can judge whether they are well-off in living the way they do. Socrates suggests pleasantness as such a criterion. He has been taken to suggest hedonism, but what he puts to Protagoras, at this stage, is the proposition (1) that those who live well live pleasant lives, since lives marked by unpleasure and misery cannot be good lives. Socrates makes two further claims which have been taken to suggest hedonism. But neither each claim singly, nor their combination, amount to hedonism. Furthermore, both claims are made in response to Protagoras’ reaction to (1).

The two claims are:

(2) Things in so far as they are pleasant (i.e., apart from any other features they possess) are good; things insofar as they are unpleasant are bad.

(3) Pleasure is a good thing in itself, unpleasure a bad thing in itself.

Socrates asserts (2) and (3) in response to Protagoras’ remark that he is not sure whether we can identify living pleasantly with living well. But such identification is not what (1) asserts or implies, since to assert such identity between pleasure and goodness is to claim that (4) Things are good, if and only if they are pleasant. Socrates offers (2) and (3) as an explication of the sense in which those who claim (1) are operating with pleasantness as a criterion of goodness. Whatever the implications of that explication are, they do not seem to involve commitment to (4).

Of far more importance dialectically is Protagoras’ reason for resisting the identification of pleasure and goodness he supposes Socrates to be proposing. For living pleasantly to be good, he remarks, the things one enjoyed in life must be admirable or noble (Kala). On this issue Protagoras sides with popular opinion: some pleasant things are bad, some painful things good, and some things neither. For this to be the case, however, it may be that there is a standard of goodness other than pleasure (nobility, for example); but the question whether noble things are in themselves desirable or attractive has not been confronted. The question Socrates is interested in is whether the pleasant can be dissociated from what one regards as good; that is, whether what makes living in certain ways welcome or desirable to an agent are features that make that living attractive rather than repulsive or off-putting. The Greek for `pleasant’ (hedus) and `pleasure’ (hedone) reinforce this point. Both words have connotations which are much wider than gratification of appetite. A Greek would describe a hedu as anything one would be glad to have (as well as sensory and sexual pleasure, also success in politics and games, good esteem, glorious death, and so on). Hedu, then, indicates something desirable because attractive, and, hence, to that extent good to have. This includes gratification of appetite but is not confined to it. At 351d 7-e1 Socrates is careful to say that by hedea (pleasant things) he means things which `either participate in pleasure or produce pleasure’.

Socrates can, thus, assert (2) without implying that the class of good things is co-extensive with the class of pleasant things, or, even, that the former class is co-extensive with the class of things-qua-pleasant. The assertion of (2) and (3) only requires that what is unqualifiedly or absolutely pleasant is good. Thus, the existence of some non-pleasant good things or of some pleasant bad things, which certainly falsifies (4), does not falsify (2). The world could be such that whatever `participated in or produced pleasure’ as a matter of fact also `participated in or produced unpleasure’. In such a world it would be false to assert (4), but (2) and (3) could still be true; in such a world it would be true that good things have the power to attract and bad things the power to repel. It is attractiveness or repulsiveness that Socrates makes a criterion of goodness, not pleasure in the narrow sense. The Greek expressions embrace that wider notion.

The implicit import of this Socratic manoeuvre is to hint that Protagoras’ view of the source of courage is in danger of putting courage outside the space of what makes things attractive or desirable and, hence, outside an agent’s rational capacity to estimate choice-worthiness. Socrates’ next, and seemingly puzzling question to Protagoras is also part of this import. Does Protagoras also agree with the `many’, asks Socrates, in thinking that knowledge can be `overruled’ by pleasure or passion, and be dragged about like a slave? (352b 2). The connection between this question and what has gone on before is this. The `many’, like Protagoras, also reject (2) because, like Protagoras, they mistakenly confuse it with (4), which they do not accept. Indeed, their belief that knowledge in the context of action can be `overmastered’ by pleasure, pain, fear, rage, lust, and the like, not only requires the rejection of (4), but also Socrates’ (2), that if anything is pleasant it is to that extent good. This is because in taking, say, some pleasure to overrule a person’s knowledge of what is good, the `many’ must assume pleasure to be the ruling principle – the arche – of that person’s choice in doing what he or she does. But the weak person is taken to do something bad, hence the principle of his or her choice must also be bad. It follows that anyone who believes that people knowingly act badly because they are `overmastered’ by pleasure, must also believe that sometimes choosing something as pleasant, or as pleasure-producing, is choosing badly. He cannot, therefore, be accepting (2), or, even, (3), as principles of choice. A certain conception of weakness is thus a barrier to acknowledging the connection between goodness and choice-worthiness, attractiveness or desirability, that Socrates wishes to preserve. Let us recall that it is precisely this link that was threatened by Protagoras’ separation of the source of nobility in action (courage) and a person’s capacity to choose intelligently. Socrates’ ensuing argument for the incoherence of the common account of weakness is, I believe, directed at the common threat to this link inherent in how the `many’ and Protagoras think. By inducing Protagoras to side against the `many’ on the question of weakness, Socrates hopes to force Protagoras to recognise that the capacity to choose intelligently cannot be divorced from courage, or from any virtue for that matter. Vice involves bad choosing, (not merely doing something base) and bad choosing is unintelligent choosing.

Section 3: Two Models of ‘Being Overruled’

To grasp the full force of Socrates’ argument against the `many’ we need, in the first instance, to notice an ambiguity in how they think of weakness. Suppose the `many’ said to Socrates: `You have misunderstood us. When we say that someone is `overmastered’ by pleasure into doing something he knows is bad we do not mean that choosing pleasure, qua pleasure, is bad. We mean that yielding to it (for the agent in the circumstances) is bad’. This reply introduces two possible models of `yielding’, only one of which, however, is relevant to weakness.

The contrast between the two models is implicit in the image or metaphor Plato uses to characterise the view of the `many’ (352b-c). The model in Socrates’ language is not that of a tug-of-war of forces within a person, one of which emerges the stronger. Socrates does not ask whether the force of love, fear, rage, or pleasure can be stronger in a person than the desire to act according to knowledge or reasoned choice. Clearly they can. There are cases of panic (being `rooted to the spot’ or running uncontrollably), cases of loss of control due to rage or jealousy, obsessive attachments or revulsions, and so on. Clearly these are cases where, as we say, `something stronger in us makes us do what we do’. But equally clearly they are not cases where an agent chooses to act, and, a fortiori, not cases where one chooses to act against his or her better judgement. It is also dubious whether an agent in the grip of such psychic forces judges. Consequently, the formal characterisation of akrasia in Aristotle, that `when one judges one acts against what one judges best’, is inapplicable. The yielding to pleasure the `many’ have in mind cannot be of this sort, since to yield to something as pleasant is not to be bowled over by inner impulse; it is to accept its pleasantness as a reason for doing it, to find it supremely attractive.

As Plato’s language suggests, the model for `being overruled by…’ is not that of a contest of strength but one of command, rule or authority. The defeated (as in the assembly or in war) lose power not in the sense of strength but in the sense of capacity to be in authority, to determine action or policy. Similarly, to be cowardly because one yields to fear is not to be overwhelmed by the emotion, but to allow one’s choice to be governed by how repellent the risks or hardships are in the alternative course of action. The `many’, then, cannot mean that in the cases they have in mind the passion itself becomes the principle of action, that people give in to acting fearfully or lustfully; rather, they must mean that the attractiveness or repulsiveness of the objects of these passions come to determine the agent’s choice. To act lustfully through weakness is not an alternative on the same footing as acting lustfully through desire or lust; it is to yield to choosing badly in the circumstances. But now there is a problem: since such yielding itself represents a choice, namely, to choose in accordance with a standard of choice thought to be bad, then either the standard operating in the agent in yielding to a pleasure is the same as the one operating in the agent’s pursuing that pleasure, or it is not. If it is the same, then the weak agent makes a contradictory application of the one standard of choice – one way to the yielding (that it is bad) and the opposite way to the pursuit of the pleasant object (that it is good). But this renders unintelligible the act as an intentional act. If, however, the standard is not the same for the yielding and for the pursuing, we need an explanation why and how, the agent came to prefer a standard of choice appropriate to the pursuit rather than the one relevant to yielding or resisting. Is it because the standard for the latter is less attractive than the former? And, if so, less attractive in what respect?

The point is simply that though objects of desire are, as such, attractive, and objects of aversion are, as such, repulsive, as objects of pursuit or avoidance, they may be either attractive or repulsive. The `pleasant’, one might say, is both the `formal’ intentional description of an object of pursuit and the `substantive’ intentional description of an object of desire. Similarly with `unpleasant’, `object of avoidance’ and `object of aversion’. We note that Protagoras’ resistance to saying that the pleasant life is the life lived well, on the grounds that it is so only if the things pursued as pleasant are also noble (worthy of respect), leaves it open that one may pursue noble things not qua pleasant, not as things one enjoyed in life. But this makes it difficult to understand what, for him, is the `formal’ intentional characterisation of the noble as object of pursuit. As Socrates remarks in a revealing passage at the start of his argument against the `many’: `do you, men, say that what happens is other than (allo ti gar) that very often eating, drinking and sex, being pleasant (hedeon onton), overpower people into doing these things knowing they are wicked’ (353c 4-8, my emphasis). It is the pleasantness of these activities that overpower people, not the blind impulse to engage in the activities themselves. If it were the activities themselves we could not say that people pursued them by way of yielding to them. The question now is whether the `many’ can give a coherent account of such pursuit. Socrates thinks not. It is instructive to see why.

Section 4: Socrates’ Argument (353d1–356c4)

The stage is now set for Socrates to show that there is an incoherence in how the `many’ explain weakness; it is their account of the occurrence Socrates attacks, not the occurrence itself. The two previous sections have brought out the two essential props in Socrates’ dialectic. First, that pleasantness is a criterion of the good, in the sense that it is a criterion of choice-worthiness or pursuit. And, secondly, that cases of knowledge being overruled by pleasure or passion fall within the ambit of an agent choosing. However, since, ex hypothesi, the agent thinks of his or her act as bad, he or she must hold, before, during and after the act, that the choice involved in the act is bad. Now, the explanation the `many’ give of this case is that the agent’s knowledge of what is bad is `overruled’ by pleasure or passion. But, as we saw previously, it is not the pursuit of pleasure or the avoidance of unpleasure that the agent thinks is bad. By itself the pursuit of these activities satisfies Socrates’ principle of choice (i.e., (2)). Therefore, the knowledge that is supposed to be defeated must be the agent’s knowledge that the choice exemplified in the act is bad. Consequently, the explanation the `many’ give must make clear how the choice to choose badly is made. If the answer `overruled by pleasure’ is to do the job, it must explain how this choice, the choice to choose badly, offends against the principle of choice.

Socrates suggests that the only way it could do so is on consequentialist grounds (353d and repeated a number of times at 354d-e). For the choice to choose by pleasure to be a bad choice, the choice by pleasure must be believed to have unpleasant consequences. It will not do for the `many’, or anyone else, (e.g. Protagoras), to complain that the badness they have in mind is that of the act (on the grounds that it is, say, base, ignoble, or even, harmful). For, even granting that these characterisations of the act are true, it would not settle the question of what made the agent’s choice bad as a choice. The weak `offend’ in a special way: they offend by the way they choose, not by what they do; e.g. pursue pleasure or avoid unpleasure.

It is important to realise that what looks like an arbitrary foisting of consequentialism on the `many’ by Socrates is nothing of the sort; nor is the hedonism, which, as we saw, the `many’, in common with Protagoras, reject. Socrates’ point is that since the `many’ need to give an account of the badness of how the weak choose, the description `overruled by pleasure’ must be interpreted as a description of making a choice badly. It is, then, as a consequence of a particular kind of explanation of this that hedonistic consequentialism comes in. For, if the `many’, in their explanation, appeal to the fact that pleasure comes to dominate the weak person’s choice, they must think that it is this dominance which shows what makes that choice bad as a bit of choosing. And the only way it can do so is by supposing that this dominance involves a misapplication of the principle of choice – for what else does `bad choosing’ mean but `making a mistake in deliberation or in how the choice is arrived at’?

I hope this makes clear why Socrates claims that the `many’ could not mean by `bad’ anything other than `things that have greater unpleasant consequences than the immediate pleasures they generate’. He claims this on behalf of the `many’ not because they (or Socrates) believe in hedonistic consequentialism, but because this is required by their explanation of the weak person’s badness of choice. Socrates’ proposal that the `many’ substitute `pleasant’ for `good’ and `good’ for `pleasant’ in their explanation is similarly grounded. This is because in saying that the weak choose badly because they are `overruled by pleasure’, the `many’ imply that it is with respect to pleasure and unpleasure that the choice is bad – that `badness’ here can only mean `choosing (overall) unpleasure for the sake of (immediate) pleasure’. Otherwise they could not give any sense to the phrase `knowing it is bad’, where the object of knowledge is not the activity (eating, drinking, sex, etc.) itself but the pursuit of that activity (in the circumstances). According to the `many’ what the weak know is that their choice is bad, and that the weak make it because considerations of pleasure dominate in how they choose. It follows that for the `many’ this dominance of considerations of pleasure must count as a criterion of badness of choice.

It is by combining the criterion of (good) choice – Socrates’ (2) – with the `many’s’ criterion of (bad) choice, required by their explanation of weakness, that Socrates derives the incoherence or absurdity of that explanation. By the criterion of (good) choice to pursue something as good is to pursue that thing insofar as it is pleasant. Hence, to pursue something bad is to pursue that thing insofar as it is unpleasant. This characterisation of good or bad choice is independent of how the things chosen turn out to be. Nevertheless, the criterion of whether someone chooses well or badly depends on whether they considered the thing chosen as being (overall) more pleasant than unpleasant. But on the `many’s’ view of weakness, the weak in choosing pleasure choose badly. It follows that the pleasure they pursued is (overall) more unpleasant than pleasant. However, the weak, ex hypothesi, know that they are choosing badly since they pursue something against their judgement of what is best. Consequently, what makes their choice a bad one is not the feature (overall unpleasantness over pleasantness) that makes any chosen item bad. Rather, the weak choose badly because, on the `many’s’ view, they chose to choose `in accordance with pleasure’ when they knew or believed that they should not. In other words, the sense of `choosing badly’ as applied to the weak by the `many’ ought to be different from the sense of that phrase when applied to the goodness or badness of the items chosen – to how they turned out to be. The badness of choice in the case of the weak is, thus, not independent of how the item chosen is thought of by the agent. Believing that in pursuing pleasure they choose badly, the weak, according to the `many’, must be applying to their choosing the criterion of (good) choice. But this cannot be done without incoherence. For, according to this criterion the pursuit of some pleasure is bad because it is overall more unpleasant than pleasant. But the weak are supposed to choose that pleasure knowing this. If follows that, on this view, the weak pursue a pleasure qua unpleasant, otherwise they could not think of that choice as a bad one. But this renders incoherent the choice of the weak as a choice. As Socrates puts it (355d 3-7): the `many’ must think that for the weak the `good’ (= the `pleasant’) does not count as worthy to win (axion nikan) over the `bad’ (= the `unpleasant’).

Socrates’ point is well taken, but we need to distinguish its substance from its dialectical force. The former concerns the absurdity of a certain conception of what makes bad the way the weak chose. The latter hints at some consequences of this result for Protagoras’ view of virtue (cf. the next section below). Concerning the substance of the point, Socrates is suggesting that what, according to the `many’, accounts for the badness of akratic choice yields an incoherent account of the choice. For if the expression `being overcome by X’ designates the value that came to rule the choice of the akrates (the pleasantness of some pleasure, say), then, on the view of the `many’, the akrates chooses an action which he judges, in terms of the very value operative in his choice, to be less choice-worthy than some other alternative. Thus, it makes sense to say that people in their pursuit of pleasures often miscalculate the overall pleasantness or unpleasantness of some pursuit, or that they misjudge the occasion, the extent, and the appropriateness of the pursuit. What is being said is that due to a mistake (perhaps itself due to lack of skill or capacity to evaluate things intelligently) such people receive a lesser value in place of a greater one. But on the `many’s’ view of akrasia, what makes the choice of the weak bad is not that they receive a lesser value for a greater one; rather, it is that they act as if choosing to act contrary to the principle of (good) choice is more worthy than acting according to the principle, where the criterion of worthiness invoked (pleasantness) is the very same as the one employed in the principle of (good) choice itself. Thus, if `being overruled by pleasure’ is meant to explain wherein lies the specific badness of the choosing in akratic action, the explanation is absurd. It attributes to the weak an incoherent, self-contradictory, employment of the criterion of choice-worthiness. The crucial point in Socrates’ strategy is the insistence that what we need to explain with respect to the weak is their badness as choosers; we need to explain it not with reference to their obtaining a lesser value instead of a greater one (this is just ordinary miscalculation), but with reference to their choosing to act contrary to what they know is the principle of (good) choice. It is this which the `many’s’ `being overcome by pleasure’ fails to explain coherently.

We note here that Socrates’ suggestion as to what would count as an account of bad choice does not commit him to claiming that this is the only feature of the `phenomenon’ the `many’ seek, incoherently, to explain. It only sets out what is required if what needs explanation about the phenomenon is confined to its being an instance of choosing badly. If to explain weakness is to explain a species of choosing badly, then the phrase `overruled by pleasure’ gives no coherent form of such an explanation.

Section 5: The Import of Socrates’ Dialectic

I conclude with some comments on the general thrust of Socrates’ argument about akrasia and its broader implications for a theory of virtue.

As I suggested at the beginning, the target of Socrates’ dialectic is Protagoras’ theory of virtue which involves a fragmented conception of it. Virtues, for Protagoras, can be possessed independently of each other. As we saw, Protagoras’ separation of courage from wisdom was based on a dichotomy between the source of courage in people and the sort of ‘power’ or capacity to be confident that intelligent skill generates. Socrates’ procedure is to show that this entails an effective divorce of the cognitive content of reasons for acting courageously from the desire to so act on the grounds that courage is recognised as a component of a life plan that is attractive overall. The discussion of the common explanation of weakness is relevant to Socrates’ strategy because that explanation assumes the same sort of split between the cognitive content of choice and the affective and/or connational sources of action. People’s knowledge of what is good (or bad) can be ‘defeated’ by pleasure or passion. In showing the common explanation of weakness to be incoherent as an account of a kind of choice, Socrates undermines the assumption on which it is based – thereby forcing Protagoras to recognise that his ground for separating courage from wisdom runs the risk of making the choice to act courageously insulated from what makes courage attractive or desirable.

One interesting implication of Plato’s discussion is the suggestion that what makes akrasia incoherent is the split between the cognitive and the affective aspect of choice, a split essential to how the `many’ see weakness. In the realm of `pure’ cognition, of rational calculation or decision, there is no conceptual room for akrasia, only for error or mistake. Akrasia is only possible where choice is possible; and choice cannot be divorced from rational desire, the object of such desire being to achieve a good by thinking well about what it is, and about the conditions of its occurrence. It is the centrality of rational desire to all virtue that Protagoras’ view misses. And because of this he also underestimates what is involved in the claim that virtue is teachable. If that claim is true it must show how that desire can be instilled, or be made an operative force in a life. If, like Protagoras and the `many’, we separate the cognitive and the affective, while believing that virtue is teachable, we end up with isolating the training of character from the development of intelligence. We cannot, then, avoid being pessimistic about how realisable the ideal of virtue is in human beings. The attribution of weakness to intractable and recalcitrant factors in `human nature’ is a `front’ for a conception of a moral agent that sees his or her rationality as morally unmotivating and his or her moral motivation as non-rational. To think thus, however, is to deny that a rational desire is essential to virtue because it is required by all choice. It is this denial which underpins the fragmented view of virtue.

Another interesting complication of Socrates’ argument is that what makes akrasia coherent is not the assumption of a plurality of values that are incomparable, such as nobility and well-being or pleasure concerning which an agent may be conflicted. Such conflicts may indeed exist, but an irrational resolution of such conflicts, going for the one when one thinks one ought to go for the other does not amount to akrasia. Weakness requires that, even within the constraints of a single value, the choice of the agent went against what on the agent’s own principle of good, rational, choice was the more attractive alternative. For this to be possible something like a displacement of rational desire from its proper object or concern, viz. pursuit of what is good overall, to a partial, temporary or limited good, is required. How such displacement comes about is another question. But whatever is the answer it must conceptually incorporate the idea that the `desiring Reason’, necessary to all virtue and intelligent choice also operates in the akratic. This, I believe, is the force of Socrates’ remark at the conclusion of the argument: `… no one pursues bad things willingly (hekon), or those things that he thinks are bad. It is not in human nature, it seems to want to pursue what one thinks is bad rather than what is good. And when someone is compelled to choose between two bad things no one will choose the greater when he may have the lesser’ (358c 6-d 4).

My argument has been that for Plato akrasia does not require the rejection of this principle of choice; it is operative in the akratic, just as much as in the virtuous. What separates the two cannot be that the one has, while the other lacks, the capacity to rationally desire the good. The difference between them must lie elsewhere, otherwise the akratic cannot be said to choose what he or she does. The weak, then, do not prefer the lesser to the greater good, even though their acts look as if they do. Akrasia is a surface phenomenon, an appearance, in overt behaviour, of a faulty psychic structure. The Protagoras does not tell us what that is, but it effectively stops us from looking for the explanation of akrasia in places that would render choosing to act virtuously a mysterious and non-rational thing.