by Steve W. Lemke
delivered at the 1998 Southwest Regional meeting
of the Evangelical Theological Society

Socrates and Plato: Framing the Question

        One of the problems raised by Socrates and Plato was that of "akratic" action--that is, why people believe they should do action x, but do action z instead. The term "akratic" is derived from an adjectival transliteration of the Greek word akrasia, often translated "incontinent." Many contemporary philosophers have used terms such as "weakness of will," "lack of self-control," and "moral weakness" to describe this phenomenon. Unfortunately, these translations are interpretive, and may raise other issues not implied in the original Greek nomenclature.(1) This issue of moral weakness was voiced classically in the Christian Scriptures by the Apostle Paul in Rom. 7:12-25, who lamented that he knew what was right but could not bring himself to do it. In this paper, I will survey the major philosophical approaches to this issue, and propose an answer from a Christian voluntarist perspective which takes the Augustinian view of the will very seriously.(2)

        For Socrates and Plato, akratic action presented something of a dilemma. Their rationalist paradigm assumed that given the right information, people will do the rational thing. But in akratic action, persons appear to act against their best judgment. In his dialogue with Protagoras, Socrates denied that true weakness of will exists, because he believes that no one willingly pursues evil. He also found it an absurd explanation that agents fail to do what they believe is best because they are overcome by passion.(3) Socrates recognizes that this doctrine was contrary to public opinion, but his belief was grounded in a firm conviction of the correctness of psychological hedonism. Socrates cannot conceive of someone intentionally doing what they believe will be painful.(4) His explanation, then, for apparently akratic behavior is ignorance--we are misled by appearances to form the wrong beliefs about the situation. We might see pleasure in the short run in drinking alcohol, for example, but we fail to see its long- term painfulness.(5) So, for Socrates, akratic action essentially did not exist.

        Plato (or, as some would have it, the later Plato) offers a somewhat more sophisticated psychology and action theory than had Socrates, in which he essentially endorses the Socratic doctrine concerning akrasia with some important qualifications.(6) Plato describes Leontius as being akratic when his strong desire to observe those who had just been executed overcomes his recognition that this act is shameful and inappropriate. Plato finds this phenomenon not to be an isolated event, for "we often see this elsewhere, when his appetites are forcing a man to act contrary to reason, and he rails at himself with that within himself which is compelling him to do so."(7) Plato accounts for the phenomenon as somewhat of a civil war within the soul between the reason, spirit, and appetite. In at least one reading of Plato,(8) each of these faculties is relatively independent, capable of forming opinions or beliefs and acting on them. Each of the three parts of the soul has its own pleasures, desires, and motivations. So the non-rational elements of the soul have their own goods, but these are often only apparent goods rather than the all-things-considered good demanded by reason. Plato, then, retains the essential elements of the Socratic doctrine, especially the commitment to a rationalistic paradigm.

Aristotle and Akrasia

        Aristotle was obviously aware of the Socratic/Platonic doctrine and affirmed it to some degree.(9) But if it was Socrates and Plato who raised the question of akrasia, it was Aristotle who offered the most robust examination of the subject. Aristotle offered more precise definition of the various expressions of the akratic phenomenon, and offered some careful refinements of the Socratic doctrine.(10) Aristotle defined akrasia in general as a person who, "knowing that what he does is bad, does it as a result of passion."(11) He compared it to a person whose paralyzed limb goes right instead of left as the person intended,(12) or like a city which enacts good laws and then disregards them.(13)

        A fully akratic behavior, in Aristotle's perspective, must meet all four requirements of what might be called the akratic set. Actions which meet some but not all of these requirements are off-color varieties of akrasia which may be called akratic only in a qualified or analogous way. The akratic set consists of the following requirements:

        (a) The Rationality Requirement. To fulfill the rationality requirement, an agent must hold the right beliefs about the situation,(14) and reason appropriately through the practical syllogism,(15) including grasping the correct universal (first) principle relevant to this situation, which becomes the first premise of the practical syllogism.(16) An akratic person grasps and affirms the universal premise (to some degree), but fails to carry it out. Angry or impulsive agents are not fully akratic because they do not take time to comprehend the universal principle.(17) Aristotle views an angry expression of akratic behavior as being less disgraceful than simple moral weakness because anger is less controlled by appetite and more by reason than incontinence. Anger listens to reason but mishears it, like the servant who runs out before hearing his instructions, or the dog who barks before she knows who is at the door. Whereas normal incontinence recognizes a pleasure to be enjoyed, angry moral weakness uses reason misguidedly to avoid pain or harm. Appetitive moral weakness is more unjust than angry incontinence, Aristotle believed, because excessive anger is more natural than excessive appetite.(18) Angry incontinence, however, is a qualified kind of incontinence because its rush to judgment may fail to meet the rationality requirement fully.

        Aristotle found impetuous or excitable akratic people (oi melagcolikoi )(19) less blameworthy than people who were deliberately morally weak, because the impulsive ones never pause to utilize their reason. Excitable people are thus more curable than those who deliberately act against the good. In a sense, the excitable akratic people fail to meet the rationality requirement for akrasia, because they never deliberate about what they should do. Those who deliberate commit moral weakness proper, and are less curable because they do not abide by their deliberations.(20) Theodore Scaltsas describes impetuous weakness (propeteia) as nominal or weak akrasia, because there is no meaningful deliberation before the act; while deliberate moral weakness (asqeneia) he calls strong akrasia.(21)

        (b) The Voluntaristic Requirement. In order to meet the voluntaristic requirement, agents must be cognizant (not ignorant) of the situation at hand,(22) have libertarian freedom (the power to do otherwise), and be capable of changing habits.(23) That which is done in or of ignorance is simply non-voluntary, but if the agent feels pain or regret for the action it is counted as involuntary, but not akratic.(24) Agents under compulsion are not truly akratic.(25) Mixed cases of compulsion, which involve at least some action on the part of the agent, may be akratic. Since it is easier to change one's habits than one's innate nature, Aristotle argued that habitual weakness of will was easier to change than when the moral weakness was grounded one's human nature. Someone could be taught new habits.(26) Brutish and morbid weakness are the extreme or excessive expressions of those who are incompetent to make moral judgments by nature or ingrained habit, living virtually as animals, and hence not truly akratic.

        (c) The Hedonistic Requirement. To fulfill the hedonistic requirement, the agent must have an appetitive object, must feel (in some sense) psychological compulsion to do the best action,(27) and must act excessively compared with most people.(28) Aristotle considered akrasia proper to apply only to appetitive pleasures, such as sex, hunger, thirst, and other such basic needs. Other areas, such as money, gain, honor, or anger, were moral weakness in merely a secondary way.(29)

        (d) The Nonappropriation Requirement. The agent meets the nonappropriation requirement when, despite meeting the criteria of the rationality, voluntaristic, and hedonistic requirements, she acts against the universal principle in a given situation. The morally excellent agent must understand the appropriate act (the conclusion of the practical syllogism) not merely in the abstract, but its relevance to this situation. The akratic, however, fails to grasp or appropriate this conclusion in a way that results in the right action being performed. Aristotle distinguished two "enormously different" ways of knowing--having an abstract kind of knowledge which we do not use, and appropriating knowledge which we put to use. The agent with merely potential knowledge is thus like those who talk in their sleep, or quote famous poetry while drunk, or rave like madmen, or quote other people's lines as actors on stage.(30) The unexercised knowledge is merely potential, but the exercised knowledge is actualized. Potential knowledge can hardly be counted knowledge at all. The akratic person fails to make this transfer from unconscious to conscious knowledge.(31)

Desire/Belief Psychology

        More recently, thinkers such as Donald Davidson,(32) Michael Bratman,(33) and Graeme Marshall(34) have offered proposals similar to the classical paradigm which base action on the best all-things-considered judgment of the agent.(35) For Davidson, alleged weakness of will violates two principles which he holds inviolable:(36)

        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.

Sometimes described as the orthodox or traditional approach, this perspective presupposes rationalistic prescriptivistic internalism.(37) Stan van Hooft summarizes what he calls the "rationalist paradigm" in five key assumptions:(38)

        A1 -- It sees its primary task as that of showing how action might be explained.
        A2 -- It understands acting intentionally in terms of the action being causally explainable
                 by reference to relevant beliefs and desires.
        A3 -- Accordingly, it seeks to subsume all relevant antecedents to action under the
                 categories of beliefs or desires.
        A4 -- It applies the Rationality Principle to intentional action: that is, the assumption
                 that a person will always act rationally under some description.
        A5 -- It assumes the transparency of consciousness.
        A6 -- It understands the elements of intentional action in non-durational terms.

        Another prescriptivist approach to this problem substitutes the greatest desire for the best judgment as the strongest motivation for action. R. M. Hare(39), David Wiggins,(40) David Charles,(41) Gary Watson,(42) David Pears,(43) and Stephen Schiffer(44) take this approach. From this somewhat Humean perspective, a rebellious desire "takes over" the will in an act of executive irrationality, whether compulsively (Hare and Watson) or non-compulsively (Pears).(45)

A Voluntarist Critique of the Classical Paradigm

        Before offering a voluntarist alternative account, I would like to mention four critical weaknesses in the classical rationalist and desire/belief accounts. The primary concern is the failure of the rationalist account to account adequately for volition in anthropology. One glaring absence in both the classical rationalist account and the greatest desire account is the agent--more specifically, the agent's will. Even if these accounts were correct, who chooses the greatest good or strongest desire, and who effects the intention to fulfill them? Obviously, neither a belief nor a desire could perform either of these functions. Judgments are not the same thing as choices. What is required is the aspect of personal agency known as the willf.(46) But neither desire nor belief are sufficient alone to cause an event; there must be a corresponding intention (i.e., willd).(47) The greatest desire view also suffers from the problem of accounting adequately for conflicting desires of approximately equal strength, and conflicts between incommensurate best judgments and strongest desires.(48)

        A convincing account of psychological motivation for action is missing in the rationalist account. One of the principal problems in Aristotle's account is his explanation of practical reasoning. The examples of practical reasoning he offers are not consistent, but it is apparent that Aristotle is trying to parallel the practical syllogism with a logical syllogism. Theoretical reasoning leads to a logical conclusion or demonstration by constraint of logic, while practical reasoning leads to an action by psychological constraint. One can err in deliberation if a mistake is made in either the major (universal) premise or in the minor (particular) premise.

        The fundamental problem in applying Aristotle's practical syllogism to ethical issues is that it blurs the distinction between practical and theoretical reasoning. There are at least two consequences of this problem. First, the problem with the examples Aristotle offers as a universal premise is that they are not always true. There are many instances in which dry food, light meat, or sweets would not be good for someone. For instance, it could be that the person was not hungry, or the food could belong to someone else, or should not eat the sweets because he was a diabetic. The only appropriate universal premise would be something on order that "All people should always eat dry food (or light meat, or sweets)." But it is difficult to propose such a premise that would universally obtain; most such premises would be "insane"(49) or "absurd."(50)

        Second, even if one were to grant one of Aristotle's universality premises made absurdly broad, a leap in logic takes place between the logical entailment of the syllogism and the psychological motivation required for an act. Even if an appropriate universal premise could be found that applied in every situation, it does not follow that a person would be psychologically motivated to act in that way. As Norman Dahl comments, "it looks as if a person can draw the conclusion of a practical syllogism and yet not act on it."(51) Even if, for instance, some pill were designed which had universally healthful benefits, a person who had lost her will to live might have rational reasons for refusing to take the pill. Aristotle, unfortunately, assumes that rationality requires action. His confusion of logical and psychological laws respecting human thinking are evident in the "musts" of the syllogism concerning sweets. Anscombe describes this as "a futile mechanistic theory" in which the action of tasting it is mechanically produced if there is nothing to stop it."(52) Milo attributes this move to a lack of clarity: "There is . . . good reason to suspect that Aristotle never clearly distinguished between logical and psychological necessity, between the logical and the psychological "must" . . . ."(53) R. M. Hare has suggested, however, there are some "off-colour uses of the word ought."(54) "Ought" does not necessarily entail "must," "ought" does not entail "want," and "want" does not entail "ought"; it is the will that decides. Richard Reilly reminds us that "answering the question, 'What is it that I now ought to do,' is not to answer the question, 'What shall (will) I do now?'"(55)

        One might suggest many examples of those who knowingly (not ignorantly) violate what they consider the best course of action. For example, consider the radiologist who smokes or the overweight dietician. Each knows, after many years invested in academic preparation, reinforced now by almost daily in clinical experience, the dangers of such behavior. They daily advise their patients against such behavior. But they do the behavior anyway. They know that such behavior is self-destructive, and do not intend to perform it, but they do will to perform it. So their behavior is not out of ignorance, but nor is it irrational in the sense that it lacks justification. Indeed, the dietician and physician daily rationalize their behaviors with reasons to justify the individual instance of the behavior to themselves.(56)

        The rationalist account fails to account adequately for volition in an intuitively compelling action theory. At the beginning of book 7 of the Nicomachean Ethics, it appears that both akratic and enkratic behavior will be given thorough discussion. But in fact, only akrasia receives detailed treatment. To Norman Dahl, this "suggests that akrasia is in need of explanation in a way in which enkrateia is not."(57) It is greater emphasis on the faculty of will, I think, which would rectify Aristotle's account. In the absence of a stronger role for volition, akrasia became something of a tar baby for Aristotle. That we are morally weak, that we yield to temptation, that we fall short of our ideals, is indisputable. But this human moral fallibility is not merely because of ignorance, compulsion, desire, belief, or insufficient resolve. We are morally weak because we will to do so. The problem is not a weak will, but an evil one. It is primarily a voluntaristic issue, not merely an intellectual one. From the voluntaristic perspective, what is amazing is not that people sometimes act akratically, but that they ever act enkratically.

        Second, the rationalist account weakens moral accountability. Aristotle's definition of compulsion allows habitual behavior to merge into compulsion, thus violating his own definition. Habit would seem to fall under the "mixed" category, which is nearer voluntary than involuntary but takes account of external pressures. At its worst, blurring the distinction between compulsion and habit opens the door to the possibility of excusing oneself for a wicked deed simply by asserting that this act was the result of a habit for which one could no longer really be held accountable.

        Cases of compulsion that I believe should be ruled out from being considered as weakness of will can be grouped under two categories: psychological compulsion (including clinically diagnosed irresistible psychoses and phobias of mental illness such as obsessive compulsive disorder, multiple personality disorder, narcolepsy, and kleptomania), and physiological compulsion (such as surgical brain stimulation, forced physical movement of limbs against one's will, chemical imbalance which causes mental illness, and drug-induced states caused by drugs given against one's will). Each of these involve judgments which should be made by professionals to distinguish a milder resistible case from an irresistible case. Most people face milder examples of each of these three categories. For example, most people experience lesser forms of psychological compulsion such as mild depression from time to time in their lives, but they are still held accountable for their actions. But even the person being held at gunpoint has a choice to do or not do what he or she is being pressured to do, even though the alternatives are very unpleasant. We feel pity for the person who is victimized by being forced to submit to compulsion against his or her will, but we blame the person who could have resisted. Any resistible case cannot be counted as compulsion.

        There is no middle ground, I believe, between compulsion and choice; in the final analysis one either wills or one doesn't. To illustrate this point, consider the three options offered by both Frank Jackson(58) and Gary Watson(59) for a potentially akratic situation such as agent A drinking an alcoholic beverage before she is about to drive. In all three cases the agent A knows that she should not take the drink, but in case (a) she recklessly or self-indulgently drinks anyway; in case (b) she acts "weak willed" against her better judgment; and in case (c) she is a compulsive alcoholic who cannot resist her desire to drink. Jackson and Watson consider agent A to be morally accountable in case (a); not morally accountable in case (c), because agent A could not resist the compulsion to do so; and to exhibit true weakness of will only in case (b). Aristotle would classify (a) as wickedness, I think, except for her knowing what was the right thing to do. He would agree that (b) was akratic, and that (c) was exempted because of compulsion. I, however, would disagree with Jackson, Watson, and Aristotle in their analysis of cases (b) and (c). Regarding case (c), while I acknowledge some genuine cases of compulsion for which one should not be held morally accountable, alcoholism is ordinarily resistible (as evidenced by the fact that thousands of people overcome this addiction each year). Thus (c) is not a case of true compulsion, and agent A is accountable for her actions. I also deny that case (b) counts as a genuine case of weakness of will because regarding volition there is no middle ground between cases (a) and (c) when (c) is a genuine case of compulsion. The will is functioning just fine; it is simply willing evil things.

        Third, I am concerned that Aristotle's "normal pleasure standard" is that it is too
relativistic to be very useful in determining an appropriate action. A number of problems arise in comparing my action with those of others. Agent A might be a person who is characteristically self-controlled, but has a weakness in a particular area, or characteristically weak but very self-controlled in one area. One must therefore distinguish an akratic act from an akratic person. Moses, for example, is described as a man with the character trait of unusual meekness or self-control, "more than any man who was on the face of the earth" (Num. 12:3). Yet, sorely grieved by the disbelief and belligerence of his peoples, he acted uncharacteristically, losing his temper and speaking rashly (Exo. 17:1-7), disqualifying himself from admission into the Promised Land (Exo. 27:12-14). Moses is thus an enkratic person (in character) who committed an akratic act.

        But by what universal principle or norm shall we judge agent A? Like Aristotle, Gary Watson applies the rather vague and relativistic moral standard that a weak person is one defeated by a temptation "that most people could have resisted," or when "one has failed to meet standards of reasonable or normal self-control" of more temperate persons. But in the same context he acknowledges that "weakness is relative to expectations and norms, and it is conceivable that a whole community could fall short of these."(60) These appear to be two standards which could easily conflict with each other. Who is to determine whether agent A should be judged by what "most people" would have done, or if the "whole community" has fallen short of the standard? What if agent A grows up in a drug-infested ghetto--should she determine her behavior by assessing the typical behavior of the people around her? Who determines what are "reasonable" standards? Indeed, agent A felt her choice was reasonable, because she had compelling reasons for her choice.

        By these standards, who can be precise, for example, about what is incontinent eating or drinking? Should we determine the national average of weight or daily caloric intake? But what if agent A were a marathon runner, or a football lineman, or a diabetic? What if agent A had a low metabolism rate, such that she could eat less than agent B (who has a high metabolism rate), but still gain weight? Likewise, what if agent B can "hold his liquor," while agent A gets drunk after a couple of beers? Would much greater self-control in eating and drinking be required of agent A than agent B? Or, ironically, if agent A were overweight because of a lack of self-control in her eating, her body weight would allow her to drink more alcohol without getting drunk, and thus her lack of self-control in one area made less self-control necessary in another area. Is it not the case, then, that standards of self-control are person-specific? Such relativistic standards will never provide an adequate objective basis for distinguishing akratic from non-akratic acts, much less appropriate moral behavior.

        Fourth, the rationality account has an anthropology that is too optimistic. I share Aristotle's conviction that we are not accountable if we are ignorant of the circumstances of the case. But the rationality requirement of the traditional Socratic-Aristotelian paradigm appear too high for normal decision making. For virtually never do we have all the facts before us before we make a decision. Decisions are usually inductive subjective judgments made under time pressure well before all the evidence is in.(61) Our examination of the evidence is thus usually rushed, and we often have to make forced decisions under a time limit which does not permit careful deliberation. As we hastily examine the evidence, some reason stands out as "good enough" to justify stopping deliberation to decide. In parliamentary terms, the motion is made for nominations of potential reasons to cease. We determine that this reason or motive is of sufficient weight for us (or give it sufficient weight) such that we estimate that nothing is likely to outweigh it. This "good enough" reason becomes the winner, and we decide to pursue this course of action. Once we set out on a course, we are not easily dissuaded. Our minds are made up, and it may take overwhelming evidence or obstacles to change our minds. Such inductive judgments are susceptible to false beliefs, overlooked options, misassessed probabilities, mistaken judgments, and cognitive dissonance. Sometimes we even commit what Robert Audi calls "doxastic incontinence," that is, deciding against what an objective observer would describe as a belief that flies in the face of evidence and lacks sufficient evidential warrant.(62)

        Virtually every choice we make is made out of at least partial ignorance, thus one could argue on these grounds alone that no pure case of weakness of will exists. We find ourselves in more the situation of Heidegger's "being-thrown-into-the-world," than sitting atop an Archimedean perspective from which we may know the first principles with absolute objectivity. Retrospectively, we may look back over such choices out of incomplete or incorrect knowledge with regret, making expressions such as "I didn't think about it at the time," or "If I had only known . . .," but at the time we chose these options because they appeared to offer the best consequences. But deliberation is not an endless activity; it always ends in a decision (even if that decision is not to act right now). Aristotle's rationalistic approach does not appear realistic. As Ronald Milo asserts, it is a "fantastic exaggeration of human rationality" to presuppose that people "are so rational as never to do what is unreasonable."(63)

A Voluntaristic Alternative

The following assertions offer an outline of the voluntaristic perspective being advocated here:

        (a) Divine Voluntaristic Agency. The universe is ultimately moved only by agents. Nothing happens apart from activity of an agent--everything is created, upheld, or directly moved by an agent. From this theistic perspective (shared by all the major Western religious traditions), all motion, events, causes, and contingent things were ultimately brought into being, maintained, and caused by a personal agent, namely, God, the Unmoved Mover and First Cause. As Aquinas affirms, "every movement both of the will and of nature proceeds from God as the Prime Mover."(64) Divine agency is the ultimate cause of all things, but the proximate cause of some but not all things. No thing or event could exist or continue to exist without Him, however, He is not the proximate cause of all things or events.

        (b) Human Voluntaristic Agency. Personal selves (perhaps including some or all conscious beings) share to some degree in the creativity and agency of the Creator Agent, but never ex nihilo. All human (and subhuman) agency must deal with (redesign, reconfigure, compound) things already in existence. One primary expression of agency, selfhood, and personhood is a variety of libertarian freedom, which is expressed in the willf. It is the will (not mere rationality or desire) which performs an executive function within the psychosomatic unity that is a personal agent.(65) Such a view of freedom meets Robert Kane's condition of ultimate dominion:

        An agent's power (or control) over a choice at a time t satisfies the condition of sole or ultimate dominion if and only if (i) the agent's making the choice rather than doing otherwise (or vice versa, i.e., doing otherwise rather than making the choice) can be explained by saying that the agent rationally willed at t to do so, and (ii) no further explanation can be given for the agent's choosing rather than doing otherwise (or vice versa), or of the agent's rationally willing at t to do so, that is an explanation in terms of conditions whose existence cannot be explained by the agent's choosing or rationally willing something at t.(66)

        Harry Frankfurt, along with Bigelow, Dodds, and Pargetter, virtually equates second-order desires with the will, which in turn is that which sets the personhood of human agents apart from other animals (who also are capable of desires, motives, and choices).(67) Aristotle made a similar claim when he spoke of humans as rational animals. This human rationality was evidenced by the fact that while humans and animals both have basic needs, humans desire them in a higher way (not merely biological needs, but a more qualitative satisfaction).(68) This freedom is not absolute; it is limited to the options available.(69)

        Some thinkers such as William Rowe and Peter van Inwagen assert that libertarian freedom occurs seldom, if at all. Rowe suggests that there some pressures override volition,(70) and van Inwagen argues that genuine acts of libertarian freedom are rather rare.(71) But I argue that in virtually every circumstance, the will is exerted to some degree.(72) Even those actions which appear nearly reflexive in a stimulus-response reaction, such as w1, are in fact split-second, spur-of-the-moment decisions against which there are no countervailing reasons we can think of at the moment. Decisions are irreducibly volitional, not reducible to other states.

        (c) Agent Initiated Action. This account of freedom is agent determined or self-determined not in the sense of being oblivious to the givens and influences of life,(73) but in the sense that the willd is the ultimate determinant of action. Agency action could be described as indeterministic,(74) adeterministic,(75) or undeterministic(76) in the sense that human actions are not ruled by nomic causal determinism. While they may be vaguely predictable, human decisions are ultimately antinomian, influenced by but not governed by any law, even a law of nature, and may not be reduced to genetics, sociology, or psychology.(77) Determinists persistently attempt to subsume agent-initiated action under categories such as nomic necessity, the laws of nature, or nomic causation. All of these fail to do justice to the unique nature of the will of an agent.(78) One need only reflect on Aristotle's distinction between the stone, the staff, and the person who wields the staff.(79) The staff moves the stone, but is itself moved. It is the person who initiates the action. Likewise, determinists would like to apply covering-law explanation to history, reducing history to biology. But one need merely reflect on the difference between an account of the explosion of a volcano and an account of the reasons leading up to a war to note that human decisions are not subject to laws in any meaningful way. Because they are the product of human decisions, historic events are unrepeatable and ideographic, not repeatable and nomothetic.(80)

        The most definitive threat to volitional agency raised by nomic causal determinism. Causal determinism is the doctrine that all events can be explained by causal antecedents.(81) As Roderick Chisholm reminds us, the concept of causation is nomological, presupposing "physical necessity, a concept that is usually expressed by reference to the laws of nature."(82) While the laws of nature obviously affect humans, they do not govern human decisions.(83) Reducing all explanation to physical causation is a leap of categories and epistemic imperialism. It may be the product of an outdated Newtonian worldview, which raises the universal law of causation (ULC) to the level of cosmic principle, not unlike the role of the hypothetical grand unifying theory (GUT) of post-Einsteinian physics. When the determinist asks the question framed in such a way that demands a causal answer, the question presupposes the wrong answer and does not permit the correct answer. Like asking "Have you stopped beating your wife?," asking what nomically caused a decision admits of no appropriate answer. One might as well ask, "What color is a decision?"(84)

        The account of freedom being advocated here could also be called anti-indeterministic, however, because while not determined by nomic causation, neither does this view propose the kind of randomness exhibited in the indeterminacy of modern quantum physics (which poses just as much of a threat to meaningful choice as does determinism).(85) Human choices are not random accidents, but willful determinations.

        (d) Teleological Intelligibility. While the emphasis of this perspective is voluntaristic, it is not irrational. Agent A could offer a teleological personal explanation for why she chose a particular course of action, drawn from a complex nexus of often mutually competing desires, motives, needs, beliefs, convictions, and a plethora of other competing influences. Such teleological explanations provide reasons which satisfy the Principle of Sufficient Reason, but do not meet the stricter requirements of the Universal Law of Causation. Teleological intelligibility provides epistemic justification, and even moral justification of an act, but not ontological causation. Hence one could call this perspective "agent causation" only at the risk of a misunderstanding which might impose nomic causal determinism on it.

        (e) Deliberation and Decision. My approach to action theory is in essential agreement with the model advocated by Hugh McCann, in which deliberation yields decision and intention formation, which in turn leads to execution of intention.(86) Antecedent givens, both external (genetic, socio-economic, and environmental) and internal (psychological) factors may strongly influence or predispose one to particular choices, but are not absolutely determinative. To use Leibniz' phrase, such a motive "inclines the will without necessitating it."(87) The will thus functions with a creaturely autonomy in which antecedent circumstances, including the condition of the agent, leave some of the things we will do undetermined: they are determined only by our choices, which are motivationally explicable but not themselves causally determined. Although many of the external and internal conditions of choice are inevitably fixed by the world and not under my control, some range of open possibilities is generally presented to me on an occasion of action--and when by acting I make one of those possibilities actual, the final explanation of this . . . is given by the intentional explanation of my action, which is comprehensible only through my point of view . . . .(88)

        Deliberation is the process of assigning weights to various potential reasons for acting, drawn from a complex nexus of possible motives and reasons, including biological and psychological needs, desires, beliefs, emotions, pragmatic considerations, habits of interpretation, correct and incorrect information, and socio-economic pressures. Perceptions of consequences may make some choices appear more attractive or pragmatic than others. These factors do not come with pre-assigned weights, but we assign weights, grant priorities, and give precedence to some factors over others in a process Nozick describes as "non-random weighting":

        The reasons do not come with previously given precisely specified weights; the decision process is not one of discovering such precise weights but of assigning them. The process not only weighs reasons, it (also) weights them . . . . [T]here is no prior causal determination of the precise weight each reason will have in competition with others.(89)

        The willd may thus be compared to a senator deciding how he shall vote on an important issue.(90) He might vote for or against a bill, even if it were against his own principles or morally repugnant to himself, because of an almost infinite variety of possible influences--moral convictions (his own principles and those of others); desires, needs, and pragmatic reasons (voter popularity, a pragmatic compromise that is less than ideal, donor pressure, lobbyist pressure, a bribe, an assurance of a federal project in his district, pressure from the whip of his party); emotional or motivational reasons (anger, love, duty, hurt, revenge, loyalty); and cognitive reasons (facts, intuitions). He may not be consciously aware of all his reasons, but his decision will be a conscious one. We are more aware of our own privileged reasons for deciding than we can be of others. We may hypothesize about the reasons of others, but we cannot know with certainty what counted as reasons for them, or even what alternatives they considered as live options.(91)

        Character and habit have a significant role to play in weighting of potential reasons as well. There is something of a hermeneutical spiral in the decision making process whereby character and act are interrelated. Character becomes a hermeneutical filter for new actions, but new actions can play a role in reshaping character. Sometimes one's choices undergo small changes (based on new input) and even a paradigm shift which readjusts one's default presuppositions. Choices tend to be consistent with one's worldview. Character and habits provide a default tendency which nevertheless may be overridden by free choices. Nozick's analogy of legal precedents expresses this dynamic well:

        The weights of reasons are inchoate until the decision. The decision need not bestow exact quantities, though, only make some reasons come to outweigh others . . . . The bestowed weights
. . . set up a framework within which we make future decisions, not eternal but one we tentatively are committed to. The process of decision fixes the weights reasons are to have. The situation resembles that of precedents within a legal system; an earlier decision is not simply ignored though it may be overturned for reason.(92)

        Deliberation is seldom exhaustive because we ordinarily make decisions under time pressure. Our examination of the evidence is thus usually rushed, and we often have to make forced decisions under a time limit which does not permit careful deliberation. As we hastily examine the evidence, some reason stands out as "good enough"(93) to justify stopping deliberation to decide. In parliamentary terms, the motion is made for nominations of potential reasons to cease. We determine that this reason or motive is of sufficient weight for us (or give it sufficient weight) such that we estimate that nothing is likely to outweigh it. This "good enough" reason becomes the winner, and we decide to pursue this course of action. Once we set out on a course, we are not easily dissuaded. Our minds are made up, and it may take overwhelming evidence or obstacles to change our minds. Deliberation is not an endless activity; it always ends in a decision (even if that decision is not to act right now).

        A decision is the willd choosing one course over another. As Nozick comments, while we cannot control earlier events, "we do control which ones cause current actions."(94) Decisions are not irrational, then, but are made for evidence that is sufficient and reasons which are compelling for the agent. These reasons provide rational justification for the agent, whether or not it seems irrational to an objective observer or not.(95) In the words of Robert Kane,

        agents who succumb to temptation do not do so without reasons . . . . When agents
        overcome temptation they are acting for the former [moral, long time goods, motivated by
        duty] reasons, when they succumb to temptation, they are acting for the latter [prudential,
        short-term goods, motivated by pragmatism] reasons. In either case, it is important to note
        that the agents had reasons of both conflicting kinds before they acted; it is just a question
        of which ones they will act upon . . . . [W]e might think of the choice as being rational from
        one or the other equally valid point of view, whichever way it went. The problem is that this
        supposition does not adequately describe the situation: the two sets of reasons or points of
        view are decidedly not on a par in the mind of the agent. Moral (or prudential) reasons are
        thought by the agent to be overriding, this is what creates a problem about moral or
        prudential weakness.(96)

Deciding involves determining not only an end, but at least a provisional plan of means to accomplish that end.

        (f) Execution of Intention. Once a decision is made, some interaction with the neurological system is initiated. The volitionalism here advocated is incompatible with physiological monism, but presupposes some variety of substance dualism. Interactionism seems the most promising option, since it affords something of a hermeneutical circle in which physical events and mental events can interact. Intention is executed in trying, which in turn is usually expressed in action. But in addition to the intended action, there are several other accompanying possibilities: a failure to achieve the end, a change of volition in light of unanticipated obstacles, and unanticipated consequences.

        (g) Moral Weakness. This view of freedom affords almost unlimited potential for evil choices. All people have ethical ideals which they violate and will not be realized because of they willd + r to do evil (sinfulness or immediate desires) rather than good. We live in an evil environment so pervasive that choosing righteousness consistently is extremely difficult, and in fact virtually impossible. Temptation is a powerful force in perverting and blinding humans to desire and to choose less than their ideals. But the temptation is resistible.

        The Apostle Paul's statement in Rom. 7:14-25 that he could not resist doing some evil and that he could not make himself do some good is cited by Hare and others to suggest that he could not do otherwise (in Hare's terms, "ought but can't"). But Paul's statement must be read alongside his affirmation in 1 Cor. 10:13 that no temptation is irresistible. Read in this light, Romans 7 should then be understood as affirming that Paul would like to be more moral, but he does not yet willc to do his willi. Standard accounts understand one primary meaning of Paul's frequently used term "flesh" as being virtually equated with his selfish will. His claim, then, was that his will was immoral, at least until God could recreate a more moral will within him. But while working out one's salvation with fear and trembling in sanctification, there will be an inevitable conflict between the immoral selfish will and the sanctified spiritual will. Indeed, Paul's remorse and regret are incomprehensible if he could not genuinely resist his sins, for then he would not be accountable for them. Only if his sins were resistible would regret and remorse be appropriate.(97)

        Sometimes repeated evil choices can result in a silenced conscience, a depraved mind, or sinful nature, which is expressed in what an external observer might describe as foolish, irrational, and self-destructive choices. Ordinarily, however, moral weakness involves (as Robert Kane has suggested) choosing short-term prudential reasons over long-term moral reasons:

        If an agent is capable of inner moral or prudential struggle at all, then it must be assumed that the agent consciously believes and would avow that moral (or prudential) reasons ought to prevail; in that sense they outweigh or override in the mind of the agent. But if the agent's conviction about this was unshakable and never was threatened by countervailing reasons (some perhaps unconscious) there would be no inner conflict or struggle.(98)

        Moral weakness is expressed in the dynamic between character traits and individual actions. Characteristically self-controlled persons can exhibit lack of control in certain areas, and characteristically incontinent persons can exhibit great self-control in certain areas. But whereas we ordinarily act consistent with our character, we can deceive ourselves even about our own character. Chekov's concept of moral slavery reminds us that we can engage in self-deception by thinking better of ourselves that we ought based on the fact that we refrain from mortal sins, ignoring our persistent practice of daily acts of selfishness, anger, lying, and disloyalty.(99)

        That we are morally weak, that we yield to temptation, that we fall short of our ideals, is indisputable. But this human moral fallibility is not merely because of ignorance, compulsion, desire, belief, or insufficient resolve. We are morally weak because we will to do so. As James reminds us, "For him who knows to do right and does it not, to him it is sin (Jas. 4:17).(100)

Sources Consulted

Adler, Mortimer. The Idea of Freedom. New York: Doubleday, 1961.

Audi, Robert. "Weakness of Will and Rational Action." Australasian Journal of Philosophy 68,
        no. 3 (September 1990), 270-281.

Aune, Bruce, and Keith Lehrer. "Cans and Ifs: An Exchange," in Free Will, ed. Gary Watson.
        New York: Oxford University Press, 1982, 36-45.

Bigelow, John, Susan M. Dodds, and Robert Pargetter. "Temptation and the Will." American
        Philosophical Quarterly 27, no. 1 (January 1990), 39-49.

Bratman, Michael. "Practical Reasoning and Weakness of the Will." Nous 13 (1979), 153-171.

Charles, David. "Rationality and Irrationality." Proceedings of the Aristotelian Society 83 (1983),          191-212.

Chisholm, Roderick M. "Agents, Causes, and Events: The Problem of Free Will," in Agents,
        Causes, and Events, ed. Timothy O'Connor. New York: Oxford University Press, 1995,

Davidson, Donald. Essays on Actions and Events. Oxford: Oxford University Press, 1980.

_______. "How Is Weakness of the Will Possible?" in Moral Concepts, ed. Joel Feinberg.
        Oxford: Oxford University Press, 1970, 93-113.

_______. "Paradoxes of Irrationality," in Philosophical Essays on Freud, ed. R. Wollheim
        and J. Hopkins. Cambridge: Cambridge University Press, 1982, 289-305.

Day, J. P. "Temptation." American Philosophical Quarterly 30, no. 2 (April 1993), 175-181.

Dennett, Daniel. "On Giving Libertarians What They Say They Want," in Agents, Causes, and
        Events, ed. Timothy O'Connor. New York: Oxford University Press,1995, 43-55.

Double, Richard. "Libertarianism and Rationality," in Agents, Causes, and Events, ed. Timothy
        O'Connor. New York: Oxford University Press, 1995, 57-65.

Evans, Donald. "Moral Weakness," Philosophy 50 (1975), 295-310.

Frankena, William. "Hare on Moral Weakness and the Definition of Morality." Ethics 98 (July
        1988), 779-792.

Frankfurt, Harry. "Freedom of Will," in Free Will, ed. Gary Watson. New York: Oxford University
        Press, 1982, 82-83.

Ginet, Carl. "Reasons Explanation of Action: An Incompatibilist Account," in Agents, Causes, and
        Events, ed. Timothy O'Connor. New York: Oxford University Press, 1995, 69-93.

Hare, R. M. Moral Thinking: Its Levels, Method, and Point. Oxford: Clarendon, 1981.

_______. Freedom and Reason. Oxford: Oxford University Press, 1963.

Hill, Thomas E., Jr. "Weakness of Will and Character." Philosophical Topics 14, no. 2 (Fall 1986),

Hogarth, Robin M. "Beyond Discrete Biases: Functional and Dysfunctional Aspects of Judgmental
        Heuristics." Psychological Bulletin 90, no. 2 (1991), 199-211.

Hurley, Paul. "How Weakness of Will Is Possible." Mind 101, issue 401 (January 1992), 85-87.

Jackson, Frank. "Weakness of Will." Mind 93 (1984), 1-18.

Kane, Robert. Free Will and Values. Albany: SUNY Press, 1985.

_______. "Two Kinds of Incompatibilism," in Agents, Causes, and Events, ed. Timothy O'Connor.
        New York: Oxford University Press, 1995, 115-150.

Kennett, Jeanette. "Decision Theory and Weakness of Will." Pacific Philosophical Quarterly 72
        (1991), 113-130.

Korsgaard, Christine. "Skepticism about Practical Reason." Journal of Philosophy 83, no.1
        (January 1986), 5-25.

Leibniz, Gottfried Wilhelm. Die philosophische Schriften von Gottfried Wilhelm Leibniz, ed. C. I.
        Gerhardt. Hildesheim: G. Olms, 1961.

_______. Theodicy. LaSalle: Open Court, 1985.

McCann, Hugh J. "Intrinsic Intentionality." Theory and Decision 20 (1986), 247-273.

_______. "Rationality and the Range of Intention." Midwest Studies in Philosophy 10 (1986),

_______. "Settled Objectives and Rational Constraints." American Philosophical Quarterly 28, no.
        1 (January 1991), 25-36.

_______. "Volition and Basic Action." The Philosophical Review 83 (October 1974), 451-473.

Marshall, Graeme. "Action on the Rationality Principle." Australasian Journal of Philosophy 59
        (1981), 54-67.

Matthews, G. "Weakness of Will." Mind 75 (July 1966), 405-419.

Mele, Alfred R. "Akratic Feelings." Philosophy and Phenomenological Research 1, no. 2
        (December 1989), 277-288.

Nagel, Thomas. "The Problem of Autonomy," in Agents, Causes, and Events, ed. Timothy
        O'Connor. New York: Oxford University Press, 1995, 33-42.

Nozick, Robert. "Choice and Indeterminism," in Agents, Causes, and Events, ed. Timothy
        O'Connor. New York: Oxford University Press, 1995, 101-114.

O'Connor, Timothy, ed. Agents, Causes, and Events. New York: Oxford University Press, 1995.

_______. "Agent Causation," in Agents, Causes, and Events, ed. Timothy O'Connor. New York:
        Oxford University Press, 1995, 173-200.

Paskow, Alan. "Moral Denial, Moral Weakness, and the Complicity of the Self."   International
        Philosophical Quarterly 28, no. 1, issue no. 109 (March 1988), 65-78.

Pears, David. Motivated Irrationality. Oxford: Clarendon, 1984.

Rappaport, Elizabeth. "Explaining Moral Weakness." Philosophical Studies 24 (1973),

Reilly, Richard. "Moral Weakness." International Philosophical Quarterly 17 (June1977), 167-177.

Robinson, Kirk. "Reason, Desire, and Weakness of Will." American Philosophical Quarterly 29,
        no. 4 (October 1991), 287-298.

Rowe, William. "Two Concepts of Freedom," in Agents, Causes, and Events, ed. Timothy
        O'Connor. New York: Oxford University Press, 1995, 151-171.

Scaltsas, Theodore. "Weakness of Will in Aristotle's Ethics." Southern Journal of Philosophy 24,
        no. 3 (1986), 375-382.

Schiffer, Stephen. "A Paradox of Desire." American Philosophical Quarterly 13 (1976),

Simon, Herbert. The Sciences of the Artificial. Cambridge: MIT Press, 1969.

Strawson, Galen. "Libertarianism, Action, and Self-Determination," in Agents, Causes, and Events,
        ed. Timothy O'Connor. New York: Oxford University Press, 1995, 13-31.

Strawson, Peter. "Freedom and Resentment," in Free Will, ed. Gary Watson. New York: Oxford
        University Press, 1982, 59-80.

Swanton, Christine. "Weakness of Will as a Species of Executive Cowardice." Canadian Journal
        of Philosophy 21, no. 2 (June 1991), 123-140.

Taylor, Charles, ed. Human Agency and Language: Philosophical Perspectives 1, Cambridge:
        Cambridge University Press, 1985.

Vailati, Ezio. "Leibniz on Locke on Weakness of Will." Journal of the History of Philosophy 28,
        no. 2 (April 1990), 213-238.

van Hooft, Stan. "Weakness of Will." Southern Journal of Philosophy 26, no. 3 (1988), 403-421.

van Inwagen, Peter. "When Is the Will Free," in Agents, Causes, and Events, ed. Timothy
        O'Connor. New York: Oxford University Press, 1995, 219-238.

Watson, Gary. "Free Agency." The Journal of Philosophy 75 (1975), 205-220, reprinted in
        Free Will, ed. Gary Watson. New York: Oxford University Press, 1982, 96-110.

_______, ed. Free Will. New York: Oxford University Press, 1982.

_______. "Skepticism about Weakness of Will." Philosophical Review 86 (1977), 316-339.

Wiggins, David. "Weakness of Will, Commensurability, and the Objects of Deliberation and
        Desire," in Essays on Aristotle's Ethics, ed. A.D. Rorty. London: University of California
        Press, 1980, 241-256.

Woods, Michael. "Reasons for Action and Desires." Proceedings of the Aristotelian Society 46
        supp. (1972), 181-191.


1 The vocabulary problem is compounded by the many meanings of "will." One might distinguish at          least five senses of "will":

wf = the faculty of will--the mental faculty uniquely possessed by the psychosomatic unity we
        might describe as an "agent" or "person".
wd = a decision at a time--an expression of volition after deliberation (be it lengthy or
        virtually instantaneous) which is influenced but not determined by antecedent events,
        states, beliefs, or desires; a punctiliar act of intention formation which may be reversed
        by later decisions.
wr = will as resolve--the resolve and commitment to carry out an intention over time.
wi = the ideal will--choosing to do what one judges to be the perfect, ideal, morally best thing to
        do in ideal circumstances.
wc = the circumstantial will--choosing to do in an actual circumstance what one wills to do in this
        particular situation, usually guided by prudential or pragmatic considerations.

2In particular, Augustine's views as are voiced in books 1 and 2 of On Free Choice of the Will.

3Socrates argues, ". . . then the argument is absurd which affirms that a man often does evil knowingly, when he might abstain, because he is seduced and overpowered by pleasure; or again, when you say that a man knowingly refuses to do what is good because he is overcome at the moment by pleasure." Protagoras, 355-256. All quotations from Socrates and Plato are from the translation by B. Jowett, ed., The Dialogues of Plato (New York: Random House, 1937).

4"Then, I said, no man voluntarily pursues evil, or that which he thinks to be evil. To prefer evil to good is not in human nature; and when a man in compelled to choose one of two evils, no one will choose the greater when he may have the less." Protagoras, 358-359.

5The principal passages in the Socratic dialogues are in Protagoras 352-359, Meno 77-79, and Gorgias 468-469. For this purposes of this paper, the hermeneutical issue of how much of Plato's views are written into Socrates' views will be bracketed. For further discussion of Socrates' denial of akratic behavior, see William Charlton, Weakness of Will (New York: Basil Blackwell, 1988), 13-33; Justin Gosling, Weakness of Will (New York: Routledge, 1990), 7-20; Richard E. Hughen, "Some Arguments in Support of the Socratic Thesis that There Is No Such Thing as Weakness of Will," Journal of Thought 17 (Spring 1982):85-93; Ronald D. Milo, Aristotle on Practical Knowledge and Weakness of Will (Paris: Mouton, 1966), 71-83; Gerasimos Santas, "Plato's Protagoras and Explanation of Weakness," Philosophical Review 75 (January 1966):3-33; and James Walsh, Aristotle's Conception of Moral Weakness (New York: Columbia University Press, 1963), 4-27.

6For further discussion of Plato's views, see Charlton, 13-33; Gosling, 20-24; and Glenn Lesses, "Weakness, Reason, and the Divided Soul in Plato's Republic," History of Philosophy Quarterly 4, no. 2 (April 1987), 147-161.

7Republic, 4. 440a-b.

8I am following Lesses in this interpretation. He bases this reading of Plato primarily on the Republic 4. 439-444, 505, 580-581, and 602-604.

9NE 7. 1. 1145b22-1146a9; NE 7.3. 1147b6-17. All Aristotelian quotations are from the translations in Jonathan Barnes, ed., The Complete Works of Aristotle, 2 vols. (Princeton: Princeton University Press, 1984). NE will be used as an abbreviation for Nicomachean Ethics. Helpful discussions about Aristotle's view of akrasia include Steve Buckler, "Moral Weakness and Citizenship in Aristotle," Polis 10 (1991):65-95; T. D. J. Chappell, Aristotle and Augustine on Freedom: Two Theories of Freedom, Voluntary Action, and Akrasia (London: St. Martin's, 1995); Charlton, 35-58; Norman O. Dahl, Practical Reason, Aristotle, and Weakness of Will (Minneapolis: University of Minnesota Press, 1984); Gosling, 25-47; Fred D. Miller, Jr., "Aristotle on Practical Knowledge and Moral Weakness," in The Georgetown Symposium on Ethics (Lanham, MD: University Press of America, 1984), 131-144; Milo, 9-113; Richard Reilly, "Weakness and Blameworthiness: The Aristotelian Predicament," Philosophical Studies 24 (1976):148-165; Theodore Scaltsas, "Weakness of Will in Aristotle's Ethics," Southern Journal of Philosophy 24, no. 3 (1986), 375-382; and Walsh, 60-188.

10Robert Audi, "Weakness of Will and Rational Action," Australasian Journal of Philosophy 68, no. 3 (September 1990), 273-278, and Theodore Scaltsas, "Weakness of Will in Aristotle's Ethics," Southern Journal of Philosophy 24, no. 3 (1986), 375-382. The primary passages in Aristotle are in the Nicomachean Ethics, 1109-1152.

11NE 7. 1. 1145b12.

12NE 1. 13. 1102b15-21.

13NE 7. 10. 1152a19-24.

14NE 7. 3. 1147a24-36.

15Anthony Kenny and Norman Dahl have a sustained attack on the notion that Aristotle actually accepted the Socratic paradigm, especially the rationality requirement. Their ingenious arguments are based on a close reading of Aristotle, giving significant attention to issues in textual criticism such as textual variants and alternative translations. Despite their interesting approach, which to refute would take more than the present effort can afford, this paper will follow the majority of scholars in a more straightforward reading and traditional interpretation of Aristotle's doctrine. See Dahl, 156-226; and Anthony Kenny, "The Practical Syllogism and Incontinence," Phronesis 11 (1966):162-189. To be fully enkratic, the agent must also grasp correctly the particulars that apply in this situation, which become the minor premise(s) in the practical syllogism, but not to do so is not necessarily akratic.

16 In On the Movement of Animals, 701a10-25, for example, Aristotle suggests three very different practical syllogisms:

1--(a) Every man ought to walk, (b) I am a man, therefore (c) I ought to walk;
2--(a) I ought to create a good, (b) A house is a good, thus (c) I ought to create a house; and
3--(a) I need a covering, (b) a cloak is a covering, therefore (c) I need a cloak and thus should         build it.

See also Aristotle's dry food syllogism in NE 7. 3. 1147a5-10; the chicken example in NE 6. 8. 1141b15-25; and the sweets example in NE 7. 3. 1147b25-31. Milo describes Aristotle's practical syllogisms as "to say the least, very peculiar." Milo, 46.

17The lower animals cannot be incontinent because they cannot comprehend the universal principle. NE 6. 8. 1142a25-35; NE 7. 3. 1147b1-5.

18NE 7. 4. 1148a9-11; NE 7. 6. 1149a24-1149b26.

19In common usage, this word meant "crazy" or "hotheaded." In ancient medical science it was thought to have been produced by the heating of black bile. Later, Galen used it to mean "sanguine" or "choleric." A contemporary analogue might be "anxious." See John Burnet, The Ethics of Aristotle (London: Methuen, 1900), 322.

20NE 7. 7. 1150b15-29; NE 7. 10. 1152a15-31.

21Scaltsas, 375-382.

22NE 3.5. 1113b23-1114a4.

23The agent is responsible for voluntarily developing good habits and character. Persons are not responsible for their innate characteristics or nature, but they are responsible for the character they develop over time. Aristotle believed it is possible, however, for a habit or character trait to become so ingrained over time that, like a thrown stone, it cannot be changed. There was a time when it was in the power of the agent to change it, but that time has passed. NE 3. 5. 10-31; NE 7. 3. 1147a20-24.

24NE 3. 1. 1110b17-24. Aristotle drew a distinction between acting out of ignorance and in ignorance. In ignorance refers to someone like a drunk who is not conscious of her actions but nonetheless is held responsible for being in that state. The person who acts out of ignorance, however, is truly unaware of the circumstances at hand was not acting voluntarily and thus cannot be held responsible. That which is done in or of ignorance is simply non-voluntary. One must also be aware of the distinction Aristotle draws between non-voluntary (ouk ekousion ) and involuntary (skousion). All acts done in ignorance are non-voluntary, but only those which the agent regrets after she becomes cognizant of the fact count as involuntary. Aristotle thus did not count acts done in ignorance, but without regret when made aware of them, as truly involuntary, because the agent would have done the act voluntarily if she had been aware of it. For example, the person who runs over her neighbor's cat which she did not see while backing her car out of her driveway would be involuntary if she were sorry for the accident, but non-voluntary if she were not. NE 3. 1. 1110b17-28.

25Examples of true compulsion, for Aristotle, are when the moving principle is entirely outside the agent, such as someone being carried away by the wind or by other people. In other cases, when an external mover is exerting extreme pressure on the agent, such as in extortion, threat of life, or other extreme circumstances, Aristotle described the action as "mixed," but nearer voluntary than involuntary because the agent still has a choice, however unhappy it might be. Only when the agent has no active role at all may the action be exempted from being akratic according to Aristotle's definition. But genuine instances of ignorance do not count as akratic behavior. NE 3. 1. 1109b30-1110ba15; NE 3. 1. 1113b23-27.

26NE 7. 10. 115a28-31.

27Not unlike Socrates, Aristotle appears to believe that rationality requires action. For example, note the imperatives in the practical syllogism relating to sweets:

"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 affirm the conclusion, while in the case of opinions concerned with production it must immediately act (e.g. if everything sweet ought to be tasted, and this is sweet, in the sense of being one of the particular sweet things, the man who can act and is not constrained must at the same time act accordingly)." NE 7. 3. 1147a25-31, italics mine. As Ronald Milo comments, "Aristotle seems to suggest that it is psychologically impossible for one who accepts the premises not to act in the required manner (unless, of course, he is prevented from acting by other external physical forces). . ." Milo, 50. An akratic person, then, is psychologically driven to apply the universal truth.

28Both excess of pleasure and loss of desire for pleasure are unfortunate extremes, for Aristotle. The desirable action is something of a Golden Mean between them--the relative standard of what most people do. A person should thus judge his alcohol consumption, violence, and resolve against what the average person normally does. If an agent finds herself being excessive in comparison with others, she should curb her activity and get closer to the norm. If, despite recognizing she is excessive compared with others she continues this behavior, she is akratic. NE 7. 7. 1150b7-18; NE 7. 8. 1151a1-5; NE 7. 10. 25-27. In a different context, (NE 2. 6. 1106a29-1106b6), Aristotle is prepared to make the mean relative to an individual, but he does not offer this option in discussing akrasia.

29NE 7. 4. 1147b20-1148a18.

30NE 7. 3. 1146b7-1147a24; NE 7. 3. 1147b6-17.

31Chappell, 105-106; Milo, 82-84.

32Donald Davidson, "How Is Weakness of the Will Possible?" in Moral Concepts, ed. Joel Feinberg (Oxford: Oxford University Press, 1970), 93-113, reprinted in Donald Davidson, Essays on Actions and Events (Oxford: Oxford University Press, 1980), 21-42; see also "Actions, Reasons and Causes," 3-19; and "Intending," 83-102; in the same volume; and "Paradoxes of Irrationality," in Philosophical Essays on Freud, ed. R. Wollheim and J. Hopkins (Cambridge: Cambridge University Press, 1982), 289-305.

33Michael Bratman, "Practical Reasoning and Weakness of the Will," Nous 13 (1979), 153-171.

34Graeme Marshall, "Action on the Rationality Principle," Australasian Journal of Philosophy 59 (1981), 54-67.

35The problem with this wording is a lack of clarity and definition about what qualifies as "best" or "good." What makes something the better or best judgment? Better in whose judgment? Best in what sense? Good in what way? Since weakness of will is framed by classical thinkers, it presupposes an objectivity which most modern commentators cannot accept. But unless one accepts objective truth and morality (as Plato and Aristotle certainly would have), what problem would there be in going against qualified subjectivist or relativist human projections of morality or truth?

36Davidson, "How Is Weakness of the Will Possible?" in Moral Concepts, 95.

37Thomas E. Hill, Jr., "Weakness of Will and Character," Philosophical Topics 14, no. 2 (Fall 1986), 94, 103; Jeanette Kennett, "Decision Theory and Weakness of Will," Pacific Philosophical Quarterly 72 (1991), 113-115; Richard Reilly, "Moral Weakness," International Philosophical Quarterly 17 (June 1977), 167-169; Christine Swanton, "Weakness of Will as a Species of Executive Cowardice," Canadian Journal of Philosophy 21, no. 2 (June 1991), 124-129; and Ezio Vailati, "Leibniz on Locke on Weakness of Will," Journal of the History of Philosophy 28, no. 2 (April 1990), 221.

38Stan van Hooft, "Weakness of Will," Southern Journal of Philosophy 26, no. 3 (1988), 403-404.

39R. M. Hare, Freedom and Reason (Oxford: Oxford University Press, 1963), 67-85; Moral Thinking: Its Levels, Method, and Point (Oxford: Clarendon, 1981).

40David Wiggins, "Weakness of Will, Commensurability, and the Objects of Deliberation and Desire," in Essays on Aristotle's Ethics, ed. A.D. Rorty (London: University of California Press, 1980), 241-256.

41David Charles, "Rationality and Irrationality," Proceedings of the Aristotelian Society 83 (1983), 191-212.

42Gary Watson, "Free Agency," The Journal of Philosophy 75 (1975), 205-220, reprinted in Free Will, ed. Gary Watson (New York: Oxford University Press, 1982), 96-110; and "Skepticism about Weakness of Will," Philosophical Review 86 (1977), 316-339.

43David Pears, Motivated Irrationality (Oxford: Clarendon, 1984).

44Stephen Schiffer, "A Paradox of Desire," American Philosophical Quarterly 13 (1976), 159-203.

45Swanton, 124. See also Galen Strawson, "Libertarianism, Action, and Self-Determination," in Agents, Causes, and Events, ed. Timothy O'Connor (New York: Oxford University Press, 1995), 20.

46Audi, 271; Hill, 102; and Kirk Robinson, "Reason, Desire, and Weakness of Will," American Philosophical Quarterly 29, no. 4 (October 1991), 293-295, 298.

47Hugh McCann warns against the reductionism of causally deterministic accounts of action that "do not allow for independent mental states of intending, but rather seek to reduce intention to other states, often a combination of desire and belief which, when they cause behavior of an appropriate kind, are held to issue in an intentional action." Hugh J. McCann, "Intrinsic Intentionality," Theory and Decision 20 (1986), 247.

48"Intentions do frequently accord with strongest desires, and the frequently accord with judgments of what is best. The problem is, however, that strongest desire and best judgment may be out of accord with each other, and then intention can go either way." McCann, "Intrinsic Intentionality," 252. See also John Bigelow, Susan M. Dodds, and Robert Pargetter, "Temptation and the Will," American Philosophical Quarterly 27, no. 1 (January 1990), 39-41; Hugh J. McCann, "Rationality and the Range of Intention," Midwest Studies in Philosophy 10 (1986), 192-194; Alfred R. Mele, "Akratic Feelings," Philosophy and Phenomenological Research 1, no. 2 (December 1989), 284; Robinson, 293-298; Scaltsas, 381; van Hooft, 411-415; and Wiggins, 246.

49G. E. M. Anscombe, Intention (London: Oxford University Press, 1957), 64. She humorously suggests that the universal premise that all people should eat sweets would obtain only if everyone had maniacal sweet tooths.

50Milo, 52-53.

51Dahl, 160.

52Anscombe, 64.

53Milo, 50.

54Hare, Freedom and Reason,68, 75-77; see also Donald Evans, "Moral Weakness," Philosophy 50 (1975), 299.

55Reilly, "Moral Weakness," 170, 173.

56Thus, for Frank Jackson, "it seems that causing agents to act contrary to better judgement is neither necessary nor sufficient for weak-willed action." Frank Jackson,"Weakness of Will." Mind 93 (1984), 3.

57Dahl, 163.

58Jackson, 4.

59Watson, "Skepticism about Weakness of Will," 325.

60Watson, "Skepticism about Weakness of Will," 330-332.

61Daniel Dennett, "On Giving Libertarians What They Say They Want," in Agents, Causes, and Events, ed. O'Connor, 49-50.

62Audi, 275.

63Milo, 80.

64Thomas Aquinas, Summa Theologica, 2.1.6. While all movement proceeds from God, I believe, God does not directly cause all things. But through providential care, God always upholds the universe (general providence), often involves himself personally in human lives (specific providence), and sometimes intervenes in dramatic ways to change the usual course of events (miraculous providence). One of the arguments used frequently by determinists against agency theory is that it leads to an infinite regress. While agency theorists have provided adequate defenses against this argument, it is somewhat ironic that atheistic determinists would raise this issue, since a non-theistic, literal reading of the universal law of causation (upon which their view depends) itself requires an infinite regress of causes. Theists can offer a more satisfying account of origins, much more compatible with the big bang cosmology that has dominated physics for the last half century than a non-creationist account.

65McCann, "Intrinsic Intentionality," 247-252, "Rationality and the Range of Intention," 192-194; and Bigelow, Dodds, and Pargetter, 40.

66Robert Kane, Free Will and Values (Albany: SUNY Press, 1985), 46. See also Richard Double, "Libertarianism and Rationality," in Agents, Causes, and Events, ed. O'Connor, 59.

67Harry Frankfurt, "Freedom of Will," in Free Will, ed. Watson, 82-83; and Bigelow, Dodds, and Pargetter, 48. See also Kennett, 126. Frankfurt's second-order desires involve not merely desiring, but to desire to shape one's desires to become a better person. I share the view that animals are capable of teleological choices, although they likely lack the second-order desires that Frankfurt describes.

68Scaltsas, 412-413. We want not just shelter but a beautiful home, not just food but gourmet food, not just warmth but attractive clothing. Or, as Scaltsas comments, these drives are understood "adverbially," such that they satisfy not just biological needs but also the three distinctively human basic drives Kant lists in his Anthropology--wealth, power, and esteem.

69Unless otherwise indicated, in this paper "freedom" will refer to creaturely freedom, not divine freedom, which does not suffer from creaturely limitations.

70Rowe (following Reid) suggests that being tortured on a rack in great pain would "cause directly the volition to reveal the important secret." Rowe, "Two Concepts of Freedom," in Agents, Causes, and Events, ed. O'Connor, 160,

71van Inwagen allows only Buridan's ass/ice cream flavor cases, duty versus inclination cases, and incommensurable values cases to count as genuine freedom. Peter van Inwagen, "When Is the Will Free," in Agents, Causes, and Events, ed. O'Connor, 219-238.

72For example, in the following taxonomy of various nonequidistant degrees along a continuum of exertion of will, I affirm that in each of these cases, w1-w7, some exertion of will is made:

w1 -- Default Setting Choice--A choice so unopposed by alternatives that it seems almost
        reflexive, requiring a relatively strong exertion of will not to do. (For example, not to blink
        when someone claps their hands in front of your face, not to scratch an itch, not to jump
        out of the way of an oncoming car, or to risk your life in an emergency to save your own
        child's life).
w2 -- Customary Choice (or standing intention)--A course of action ordinarily done, but seeming
        less reflexive less emotion, and slightly more deliberation than in w1. (For example, turning on
        the light switch while entering a room in your house, or choosing your favorite soft drink from
        a drink machine).
w3 -- No Brainer Choice--One obvious choice among several choices, but one which is more
        novel than w1 or w2, and hence requires more reflection and exertion of will. (For example,
        to accept a gift of a million dollars, or to accept a job offer for your dream job).
w4 -- Decision among Equals/Featherweight Choice--A decision between two options which
        seem to have virtually the same weight of preference, and can be decided by the slightest
        leaning one way or another. (For example, to choose between two keyboard keys with the
        exact same function, to choose between two equidistant routes to drive home, to make a
        choice in Buridan's ass cases or vanilla/chocolate cases, to choose which box of corn flakes
        to pick up from the grocery shelf, to choose where to go to lunch between five of your
        favorite restaurants, to choose a place for a vacation between three equally preferred
        destinations, or a judge's choice of the best of show award at a dog show).
w5 -- Incommensurate Values, Duty vs. Inclination, and "Weak Will" Choices--There are
strong reasons for and against an action, but some factor weights one choice above others.
        (For example, to enroll in school or get a job, or a diabetic choosing to eat or withhold from
        eating his favorite ice cream).
w6 -- Decision against the Stream/the Road less Travelled Choice--A choice which may seem
        odd to one's contemporaries, and against which there are considerable reasons which would
        argue against it. (For example, to join the Peace Corps or the French Foreign Legion, to
        become a priest or nun, or to quit a safe, well-compensated career for a less secure dream
w7 -- Decision against Interest/Great Resolve--The circumstances or so against the choice that it
        requires a great act of resolve to do it, resolutely acting in a way that overrides one's own
        vested interest, doing only by a strong will against strong reasons. (For example, to jump on
        a grenade to save others, or to march into battle against a fortified line of defense).

73Mortimer Adler defines "freedom of self-determination" means that "the individual's . . . freedom of choice . . . rests with his power of self-determination, through its causal indeterminacy, is able to give dominance to one motive or one set of influences rather than another. Far from motives or other influences determining which of several decisions is made, it is the other way around. . . the self determines which motive or set of influences shall be decisive." Mortimer Adler, The Idea of Freedom (New York: Doubleday, 1961), 2:232, quoted in Robert Nozick, "Choice and Indeterminism," in Agents, Causes, and Events, ed. O'Connor, 112. Self-determination need not denote the view of radical freedom advocated by Sartre and other existentialists.

74Indeterminism can be understood negatively to mean simply that determinism is incorrect, or positively to affirm a view of events (based on quantum physics) as chaotic, random, or accidental.

75Carl Ginet prefers "adeterminism" to "indeterminism" to avoid the suggestion of randomness in the indeterminism of quantum physics. Carl Ginet, "Reasons Explanation of Action: An Incompatibilist Account," in Agents, Causes, and Events, ed. O'Connor, 91.

76Ginet defines "an undetermined event" as "one that was not nomically necessitated by the antecedent state in the world," and "a determined event" as "one that was nomically necessitated by its antecedents." Ginet, 69. As Nozick reminds us, "It is neither necessary nor appropriate . . . to say the person's action is uncaused . . . . It is undetermined which act he will do," keeping in mind the "distinction between an action's being caused, and its being causally determined," because "'uncaused' does not entail 'random.'" Nozick, 102, 104-105. It was mildly amusing to me for Peter Strawson (in an otherwise excellent article) to be casting about for a definition of determinism when, if determinism were true, one definition would be predetermined for him. Certainly, if determinism were true, it would not be a matter of choice in choosing one definition or another! Peter Strawson, "Freedom and Resentment," in Free Will, ed. Watson, 59-80.

77Reductionists are always with us, from Thomas Hobbes's reduction of emotions as glandular activity to that of present-day geneticists such as Candace Pert, who told Bill Moyers in a recent PBS special report that emotions are "nothing but" neuropeptides. Such extreme reductionism is naive and dangerous. Indeed, if their view were correct, whatever they said would be merely genetic, and thus totally beyond verification. But as Thomas Nagel notes, there is "no room for agency in a world of neural impulses and chemical reactions." Thomas Nagel, "The Problem of Autonomy," in Agents, Causes, and Events, ed. O'Connor, 34.

78"What special power then does indeterminism confer on free agents? Agents in the undetermined world have the power to make choices for which they have ultimate responsibility. They have the power to be the ultimate producers of their own ends . . . . That is to say, they have the power which can only be explained in terms of their own wills (i.e., character, motives, and efforts of will). No one can have this power in a determined world. If we must give this power a name, let us call it 'free will'." Robert Kane, "Two Kinds of Incompatibilism," in Agents, Causes, and Events, ed. O'Connor, 146.

79Aristotle, Physics, 256a.

80If an event were causally determined, Nozick comments, "under exactly the same conditions repeated, exactly the same thing would have (again) to happen." Nozick, 103. Such nomic repeatability happens in laboratories, but not in human events.

81One can distinguish several senses of "causation":

cause1 = In the narrow sense, the universal law of causation; a repeatable, nomic Universal
        Law of Causation, which refers to cause-effect relationships among unconscious
        contingent physical things.
cause2 = Statistical explanation.
cause3 = A broader concept of explanation, which includes agent causation.
It is tempting for the agency theorist to describe reasons for acting in terms of cause3, but this course risks confusion with cause1 and cause2.

82Roderick Chisolm, "Agents, Causes, and Events: The Problem of Free Will," in Agents, Causes, and Events, ed. O'Connor, 97. Chisholm distinguishes two types of necessity--laws of logic (in every possible world, if a then b), and the laws of nature.

83"Laws of nature" is an unfortunate nomenclature because it is used in at least three senses:

LN1 = Strictly, a small set of natural laws, particularly identified with Newtonian physics.
LN2 = Generally, any observable, repeatable, nomic pattern which is inductively inferred (as a         covering law, statistical pattern, or a similar model) to be generally or universally true.
LN3 = A very broad, ill-defined sense, reflecting any pattern of regularity.
        "Laws of nature" has a common currency in contemporary philosophical discourse, usually referring to LN1 or LN2. The problem with LN1 is that it is difficult to conceive of how these few laws, which Peter van Inwagen ["The Incompatibility of Free Will and Determinism," in Free Will, ed. Watson, 48] calls more properly the "laws of physics," would impact human decision making at all. But if the determinist falls back to LN2 or LN3, there is no true or absolute causal determination.

84"What is inappropriate is to demand that a free choice be explained in a way that shows its is unfree." Nozick, 110. When the question is posed by the determinist, "Are choices causally determined or are they random accidents?" the indeterminist cannot honestly affirm either option.

85Using the word "indeterminism" leaves open the option of an understanding such as Richard Double's that indeterminism in action theory amounts to an indeterministic emission of a beta particle from an atom in the brain. Double, 62, 65. Determinists such as A. J. Ayer, J. J. C. Smart, and John S. Mill would likewise understand anything less than determinism as "accidents." Ginet, 74.

86The McCannian model is outlined most explicitly in "Formation of Intention," awaiting publication, and "Intrinsic Intentionality." See also "Rationality and the Range of Intention," "Settled Objectives and Rational Constraints," "Trying, Paralysis, and Volition," and "Volition and Basic Action," The Philosophical Review 83 (October 1974), 451-473. Scaltsas offers an abbreviated version of this pattern in four steps: evaluating alternative routes of action, concluding what is the best course of action, choosing the best route of action, and acting. Scaltsas's understanding of Aristotle's view of incontinence is that the weakly akratic person doesn't get beyond the second step, while the strongly akratic person completes all four steps. Scaltsas, 376-379.

87G. W. Leibniz, Theodicy, sec. 336, in Schriften, 6.314. A similar phrase is in a letter to Clarke, letter 5, sec. 8, in Schriften, 6.390.

88Nagel, 37. Described "from the inside," we experience autonomy as "when we act, alternative possibilities seem to lie open before us: to turn right or left, to order this dish or that, to vote for one candidate or the other--and one of the possibilities is made actual by what we do. Nagel, 36. Norman O. Dahl has proposed that "weakness of the will involves a failure to achieve autonomy . . . " [Audi, 273], but I would argue that moral weakness is precisely an expression of autonomy.

89Nozick, 102, 105.

90This is an updated and expanded account that parallels Leibniz's analogy (which he
considered "ill-conceived") of a queen who considers the input of her minister of state (representing the understanding), and her courtiers or favorite ladies (representing the passions). G. W. Leibniz, Theodicy (LaSalle: Open Court, 1985), 421; quoted in Timothy O'Connor, "Agent Causation," in Agents, Causes, and Events, ed. O'Connor, 188.

91Evans, 307; and Nagel, 42.

92Nozick, 103.

93Dennett, 49-50; and Nozick, 106. Both authors utilize Herbert Simon's term "satisfice" to describe this "good enough" judgment. See Herbert Simon, The Sciences of the Artificial. (Cambridge: MIT Press, 1969).

94Nozick, 112.

95Reilly, "Moral Weakness," 171, 173. Even those suffering from mental illness experience their reasons as being justifiable, although we might find their reasons irrational due to mental defect.

96Kane, "Two Kinds of Incompatibilism," in Agents, Causes, and Events, ed. O'Connor, 133.

97 For discussions of resistibility, see Hare, Freedom and Reason, 78-79; J. P. Day, "Temptation," American Philosophical Quarterly 30, no. 2 (April 1993), 177-180; Evans, 304-306; William Frankena, "Hare on Moral Weakness and the Definition of Morality," Ethics 98 (July 1988), 790; G. Matthews, "Weakness of Will," Mind 75 (July 1966), 414-415; Nagel, 42; and Reilly, "Moral Weakness,"169-173.

98Kane, 133.

99Evans, 303.

100. This paper was read for the March 1997 Southwest regional meeting of the Evangelical Theological Society, meeting at Dallas Theological Seminary. The author expresses his appreciation to the participants for their comments and suggestions.

Return to Ethics papers