Fallacies regarding free will.
Nor do I propose here to plumb into the analytic depths by spelling out in full detail just what it is that claims to free will involve. All that matters for my present purposes is that such freedom calls for an agent's being in control of what he does in ways that are at odds with the prospect that his thoughts and intentions could be bypassed in an adequate explanation of events.
The first fallacy to be considered relates to an objection that Daniel Dennett has formulated as follows:
If determinism is true, then our every deed and decision is the inexorable outcome, it seems, of the sum of physical forces acting at the moment; which in turn is the inexorable outcome of the forces acting an instant before, and so on to the beginning of time. Thus, if determinism is true, then our acts are the consequences of the laws of nature and events in the remote past. But it is not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore the consequences of these things (including our present acts) are not up to us. (1)
It is exactly in this italicized transit from "and so on" to "the beginning of time" that constitutes what I shall call the Zenonic Fallacy. And fallacy it is because it overlooks the prospect of backwards convergence, drawing ever closer to a fixed prior terminus, but never passing it. (2)
The failing at issue here is substantially that of Zeno's notorious paradox of Achilles and the Tortoise. Both alike involve a fallacy in overlooking the circumstance that, thanks to convergence, an infinity of steps can be taken in a finite distance provided merely that the steps get ongoingly shorter. Once it is granted that, even if a cause must precede its effect, nevertheless there is no specificative timespan, however small, by which it need do so, the causal regression argument against free will loses all of its traction.
With Zeno, Achilles never catches the tortoise because his progress must go on and on before the endpoint is reached. In the present reasoning, explanation will never reach an initiating choice-point because the regress goes on and on. However, in both cases alike the idea of a convergence that terminates the infinite process at issue after a finite timespan is simply ignored.
Such a perspective leaves the Principle of Causality wholly compatible with freedom because all those causal consequences of the act remain causally explicable.
A second fallacy runs as follows:
If all events are explicable in the order of natural causality, then so are all of those supposedly free decisions of agents. This means that the Law of Causality leaves no room for agent causality and thus no room for free will. (3)
To avoid this fallacy we must draw some rather subtle distinctions, and involve ourselves in a bit of process metaphysics to boot.
The first and most crucial distinction here is that between two sorts of occurrences, mere events and eventuations. Events are occurrences that form part of nature processuality. They are happenings on the world's spatio-temporal stage. So they transpire over time: they have a finite lifespan and their time of existence always occupies a two-sidedly open interval of time.
Eventuations, by contrast, are not parts of nature's processuality but terminating points within it. They are temporally punctiform and lack duration. They mark the beginnings and endings of events.
Now all human acts (all actions and activities) are event-like. They occupy time, but the junctures of resolution that mark the completion of a process or choice of decision are not events. Such completions are not actually processual doings, but rather are mere junctures of passage--transitions that mark the beginnings and endings of events. Looking for something is an activity, but actually finding it is not. (There is no present continuous here. One can be engaged in looking, but not in finding.) Listening to someone is an activity, but hearing what they say is not. Activities are events, terminations and completions are not. The running of a race is an event (as are its various parts, such as running the first half of the race). However, finishing the race is an eventuation. Such eventuations are endings or culminations. One can ask "How long did he take to run the race?" but not "How long did he take to start the race?" And even as the race ends when it is won (or lost), so the task ends exactly at the moment when it is completed (or abandoned). Finishing is thus an eventuation, and accordingly the finishing-point of a race, instead of being the last instant of the race, is the first instant at which the race is no longer in progress. And just this is the case with the decisions and choices that terminate a course of deliberations.
Eventuations, so understood, are not parts of nature's processual flow, since parts of processes are always processes themselves. Rather, eventuations--their beginnings and endings--belong to the machinery of conceptualization that minds impose on nature: instrumentalities of descriptive convenience that do not correspond to anything enjoying independent existence in the real world. Like the North Pole or the Equator they are not real items existing physically in nature, but rather thought instrumentalities projected into reality by minds proceeding in the interests of description and examination.
Deliberations, so regarded, are seen as events--as processes that occur over open-ended intervals of time and culminate in decisions as eventuations. Here the situation is as per Display 1:
As this display shows, there will always be an interval of time between a decision and any subsequent action--an interval able to accommodate intervening events to serve as causal explainers of that decision-consequent action. Since there is no such thing as a next time subsequent to a point of decision, there will always be room for squeezing in further events before any particular decision-subsequent event. The prospect of determination by events is thus ever-present. Analogously, there is no first decision-succeeding event that excludes the prospect of an occurrence-explaining prior event. And just this is critical for the present positions regarding the causal explainability of actions. Freedom of decision accordingly does not impede causal explicability. However, what one has in the wake of a free decision is a phenomenon that might be characterized as causal compression. Every event that ensues from that decision can be accounted for causally--but only with reference to occurrences during the immediately preceding but decision-subsequent time span, the duration of which converges to zero as the point of decision is approached.
In sum: once we duly distinguish events from eventuations we can regard all action (as events) to be causally explicable in terms of what precedes. Free will becomes reconciled to the causal explicability of actions.
A free decision inaugurates a series of events each of which is fully explicable and determinate on the order of natural causality. But this is something that is true of all those decision-subsequent events, and does not hold for that free decision itself.
A third fallacy runs as follows:
Since predetermination is incompatible with free will, so is the determination of a decision's outcome by the agent's own decision-engendering deliberations.
This objection overlooks an important distinction, namely that between predetermination and what might be called precedence determination. The former calls for predictability as of some antecedent time; the latter involves no such thing. This crucial difference is illustrated in Display 2.
With predetermination what happens at [t.sub.0] is determined, that is, law-deducible, from that which happens at some earlier time t. Already at this earlier time the decision becomes settled: a foregone conclusion that is reached in advance of the fact. Some earlier state of affairs renders what occurs at the time causally inevitable. With precedence determination, by contrast, what happens at [t.sub.0] is also determined by what goes before--but only by everything that happens from some earlier time t up to but not including [t.sub.0]. (4) Both are modes of determination by earlier history, but unlike the former, the latter requires an infinite amount of input-information which is of course never available. What we thus have in this latter case is a mode of antecedence determination that does not give rise to predictability and is in fact incompatible with it.
Just exactly such precedence determination can and should be contemplated in relation to free decisions and choices: a determination by the concluding phase of the course of the agent's deliberation that issues in the decision or choice at issue.
Predetermination means that the outcome becomes a foregone conclusion at some antecedent time. The events that constitute a course of deliberation antecedent to a decision or choice so function as to determine the outcome, but it is only the end-game, the final, concluding phase that is decisive. The entire matter becomes settled in advance of the fact. This is indeed incompatible with free will because it deprives the agent of the power to change his mind. There is some time in advance of the point of decision when the whole matter becomes settled.
Precedence determination, by contrast, means that the final phase of the deliberation is decisive. Only the entire course of the agent's thinking from some earlier point up to but not including the point of decision suffices to settle the issue. The outcome is never settled in advance; it isn't over "until the fat lady sings." And it should be clear that this sort of antecedent determination geared to the unfolding course of deliberation in its final phase is nowise at odds with freedom of the will.
The situation of a free choice among alternatives is thus associated with the following sort of picture regarding the situation at issue. Consider, by way of example, a course of deliberation for deciding among three alternatives A, B, and C with the decision ultimately arrived at in favor of A at time [t.sub.0], the "point of decision". At every time t before [t.sub.0] there are three possible outcomes A, B, and C, the probabilities of which (at any given time prior to [t.sub.0]) make up a band of width 1 overall, as per Display 3.
Throughout the course of deliberation these probabilities may wobble across the probability band but in the end they must converge in a way which at [t.sub.01] gives the whole probability to one outcome alone. But at any time prior to [t.sub.0] there is a nonzero probability that any of the three outcomes will result--at no anterior time is the outcome a foregone conclusion. The endgame is never definitively settled before the end is reached: only at the very end (at [t.sub.0]) is there a probability collapse into 1 and O's. Until the issue is fully decided there is a nonzero probability of the agent's making a choice different from the one that ultimately eventuated. As the point of decision is reached it becomes more and more likely how the issue will resolve itself. But there are no guarantees. At no time before that point of decision is there a point of no return where the resolution becomes a foregone conclusion.
So once again a distinction comes upon the scene to save the day. The objection in view is fallacious because it overlooks the crucial distinction between the two very different modes of "determination by what precedes" represented respectively by predetermination and precedence-determination of the sort just described.
One other point is important here. The first question to ask of any mode of determinism is determination by what? By matters outside the agent's range of motivation is one thing, but by the agent's own deliberations--by the manifold of inclination that encompasses his wants, wishes, aims, and choices--is something else again. Determination of decision outcomes by the agent's thoughts is surely a requisite of free will rather than an obstacle to it.
And this brings us to a fourth fallacy which runs as follows:
An act can be free only if its productive source is located in the thoughts and deliberations of the agent. But this is never the case because the tight linkage of mind-activity to brain-activity means that the thoughts and deliberations of the agent's mind are always rooted in and explicable through the processes at work in the agent's brain.
To see what is amiss here consider the classic freshman physics experiment of a gas-containing cylindrical chamber closed off by a piston at one end. The temperature inside the chamber is lock-step coordinate to the distance of the piston-wall from the fixed wall. When the piston moves the temperature changes correspondingly, and conversely when temperature-changes are induced the piston moves correspondingly. But this condition of functional lock-step correlation leaves the issue of iniative wholly open: one may either be changing the temperature by moving the piston, or moving the piston by changing the temperature. Thus lock-step coordination as such does not settle the question of the direction of determination of which of those coordinated variables is free and which is dependent. The fact that two parameters are lock-step coordinated does not settle--or even address--the issue of processual initiative.
For the sake of illustration consider two small boys on a teeter-totter or, alternatively, a pulley wheel with weights suspended on either side. Here the up-or-down motion of the one side is inseparably tied to the corresponding motion of the other. This illustrates the larger point: however tight and rigid the functional coordination between two operative agencies may be, the issue of initiative and change-inauguration is something that yet remains entirely open and unaddressed. Mark Twain's tendentious question "When the body is drunk, does the mind stay sober?" is perfectly appropriate, but then the inverse question "When the mind panics does the body remain calm?" is no less telling.
All of those myriad illustrations of a condition between thought and brain activity are simply immaterial to the issue of who is in charge. For what is involved cannot settle the question of whether mind responds passively to brain-states changes or whether it actively uses the brain to its own ends.
For the determinist, to be sure, agents are productively inert. What they do is always the product of what happens to them: they simply provide the stage on which the causality of nature performs its drama. The voluntarist, by contrast, sees intelligent agents as productively active participants in the drama of the world's physical processuality. The reality of it is that mind-brain correlation cannot effectively be used against him. It is simply fallacious to think that the intimate linkage between brain activity and thought puts the brain in charge of the mind.
Yet, if mind as well as matter can seize the initiative with respect to human action so that we can act in the mode of agent causality, while nevertheless all human actions can be explained on the side of natural causality, then we confront Kant's paradox of reconciling the two modes of causality. (5)
On such an approach, the brain/mind is seen as an emergently evolved dual-aspect organization whose two interlinked domains permit the impetus to change, lying sometimes on the one side and sometimes on the other. For the direction of determination so far remains open. Given these interlocked variables, the question of the dependent vs. independent status is wholly open and the question of initiative unresolved. The fact that mind and brain sail in the same boat, is no reason why mind cannot occasionally seize the tiller. What is at issue is a partnership of coordination not a state of inflexible master-servant subordination. In particular situations, the initiative can lie on one side or the other, all depending.
But all depending on what? How does it get decided where the initiative lies? Well, think again of the pulley situation. When the weight rises, is this because someone is pushing up on it or because a bird has alighted on the wheel? The system itself taken in isolation will not answer this for you, but the wider context--the overall causality synoptic and dynamic context--will decide where the initiative lies. It is all a matter of where the activity starts and what stands at the end of the causal line. The free will situation is much the same. When I read, the mind responds to the body; when I write, the body responds to the mind.
Consider the following objection:
If the acts of an agent are anywise determined--if they are somehow, that is, anywise necessitated--then they cannot possibly qualify as free.
Both Aristotle and the Stoics sought to reconcile the volitional freedom they deemed requisite for morality with the determinism they saw operative in the circumstance that character dictates decisions. To accomplish this without adopting the Platonic myth of character selection, they maintained that what would impede freedom is not determination as such but only exogenous determination rooted in factors outside the agent's self-produced motivations. The crux of freedom, so viewed, is not indetermination but auto-determination--determination effected by the agent's agency itself--sua sponite as the medievals put it.
On such a compatibilist view, the crux of the matter is not whether or not there is determinism; it is conceded that there indeed is, albeit of the agent-internal variety. The crux is whether there is an agent external determinism, a determinism where all reference to the agent and his motivations can be out of consideration in matters of explanation. The crux of freedom does not lie in the "that" of determination, but in its "how," its procedural mechanisms. For as long as those deliberative factors play a role the basis for freedom is secured.
Thus, we have to distinguish between endogenous (agent-internal) and exogenous (agent-external) determination. Clearly if a determination is effected without reference to the agent by forces and factors above and beyond his control by thought, then we can hardly characterize that agent as free. But if those determinative factors are agent-internal, if they are a matter of the agent's own plans and projects, his own wishes, desires, and purposes, then the deliberation of the values of decisions and choices nowise stands in the way of the agent's freedom. Quite on the contrary, a choice or decision that was not the natural and inevitable outcome of the agent's motivations could hardly qualify as a free decision of his.
And so freedom of the will is nowise at odds with the Principle of Causality as long as the locus of causal determination is found in the thought process of the agent, that is, as long as causal determination is canalized through the mediation of the choices and decisions emergent from his deliberations. There is consequently no opposition between freedom and causal determination as long as that determination is effected by what transpires within the endogenous principle of the agent's activity and the matter is one of agent-causality. (6)
In sum, to set free will at odds with determinacy is fundamentally fallacious because to do so is to ride roughshod over the crucial distinction--that between the agent-external causality of impersonal events and the agent-internal causality of deliberative thought.
Yet another fallacy inheres in the following reasoning:
Free will is mysterious and supra-natural. For it requires a suspension of disbelief regarding the standard view of natural occurrence subject to the Principle of Causality.
Along these lines one recent writer complains:
Agent causation is a frankly mysterious doctrine, positing something unparalleled by anything we discover in the causal processes of chemical reactionism, nuclear fission and fusion, magnetic attraction, hurricanes, volcanoes, or such biological processes as metabolism, growth, immune reactions, and photosynthesis. Is there such a thing? When libertarians insist that there must be, they [build upon sand]. (7)
But this sort of complaint is deeply problematic.
Free will, properly regarded, hinges on the capacity of the mind to seize the initiative in effecting changes in the developmental course of mind-brain coordinated occurrence. Need this, or should it be seen as something mysterious and supra-natural?
With the development of minds upon the world stage in the course of evolution, various capacities and capabilities come upon the scene emergently, adding new sorts of operations to the repertoire of mammalian operations--remembering past occurrences, for example, or imagining future ones. One of these developmental innovations is the capacity of the mind to take the initiative in effecting change in the setting of mind from coordinate developments.
Now the explanatory rationale for this innovation is substantially the same as that for any other sort of evolution-emergent capability, namely that it contributes profitability to the business of natural selection. There is nothing mysterious or supra-natural about it.
Thus, this present fallacy rests on a failure of imagination. It is predicated on an inability to realize that with the evolution of intelligent agents there arises the prospect of intelligence-guided agency determined through the deliberations of these intelligent agents.
Down the corridors of time have echoed the words of Spinoza:
Men believe that they are free, precisely because they are conscious of their volitions and decision, and think not in the slightest about the causes that dispose them to those appetites and volitions, since they are unknown to them. (8)
And here Spinoza was echoed by Charles Darwin when soon after the voyage of the Beagle, he wrote:
The general delusion about free will is obvious because man has the power of action but he can seldom analyze his motives (originally mostly instinctive, and therefore now [requiring] great effort of reason to discover them.) (9)
Apparently Darwin thought (with Spinoza and perhaps Freud) that action is only genuinely free when it is activated entirely by recognized and rationally evaluated and approved motives, but this simply confounds free with rational agency. As long as the agent acts on his own motives--without external duress or manipulation--his action is free in the standard (rather than rationalistically reconfigured) sense of the term. Motivation as such does not impede freedom--be it rationally grounded or not. Our motives, however inappropriate and ill-advised they may be and however little understood in terms of their psychogenesis, nevertheless do not constrain our will externally from without our self, but are the very core of its expression.
A will that is responsive to an agent's motivation is thereby free and it matters not how compelling that motive may be in relation to the resolution at issue. (10) After all, a person's nature is manifested in his decisions and finds its overt expression realized in them. His decisions are nothing but the overt manifestation of his inner motivational nature. It is through his decisions and consequent actions that a person displays himself as what he actually is.
Consider this situation. I ask someone to pick a number from 1 to 6. He selects 6. I suspected as much: his past behavior indicates that he has a preference for larger numbers over smaller and for evens over odds. So his choice was not entirely random. Does that make it unfree? Not at all! It was nowise forced or constrained. Those number preferences of his were not external pressures that restricted his freedom: on the contrary they paved that way to self-expression. It would be folly to see freedom as antithetical to motivation. Quite to the contrary! Volitional freedom just exactly is freedom to indulge one's motivations.
To "free" the will from obeisance to the agent's aims and motives, needs and wants, desires and goals, likes and values, personality and disposition is not to liberate it, but to make it into something that is not just useless but even counterproductive. What rational agent would want to be harnessed to the decision-effecting instrumentality that left his motivations by the wayside? A will detached from the agent's motives would surely not qualify as his! It is a rogue will, not a personal one.
Our final piece of fallacious thinking is noted in the objection:
The very idea of free will is antithetical to science. Free will is something occult that cannot possibly be naturalized.
It is--or should be--hard to work up much sympathy to this objection. For if free will exists--if Homo sapiens can indeed make free choices and decisions--then this of course has to be part of the natural order of things. So if we indeed are free then this has to be so for roughly the same reason that we are intelligent--that is, because evolution works things out that way.
What lies at the heart and core of free will is up-to-the-last-moment thought control by a rational agent of his deliberation-produced choices and decisions in the light of his ongoingly updated information and evaluation. To see that such a capacity is of advantage in matters of survival is not a matter of rocket-science.
The objection at issue is thus fallacious in that it rests in the inappropriate presupposition that free will has to be something super or preternatural. If there is free will, it is an aspect of how materially evolved beings operate on nature's stage.
We have now examined some eight fallacious arguments against freedom of the will, and the list could easily be continued. But the overall lesson should already be clear. In each case the misconception at issue can be overcome by drawing appropriate distinctions, the heeding of which makes for a more viable construal of how freedom of the will--if such there is--should be taken to work. So at each stage there is some further clarification of what free will involves. There gradually emerges from the fog an increasingly clear view that what is at issue here is the capacity of intelligent beings to resolve matters of choice and decision through a process of deliberation on the basis of their beliefs and desires, a process that allows for ongoing updates and up-to-the-bitter-end revisability.
Properly understood, freedom of the will should not be at odds with our knowledge about how things work in the world. A Viable theory of free will should--nay, must--proceed on a naturalistic basis. And the idea that this is infeasible appears to be, by all the available indications, based on an incorrect and fallacious view of what freedom of the will is all about.
University of Pittsburgh
(1) Daniel Clement Dennett, Elbow Room: The Varieties of Free Will Worth Wanting (Cambridge: MIT Press, 1984), 99 (my italics). Compare also ibid., 134 and Daniel Clement Dennett, Freedom Evolves (New York: Viking, 2003), 16.
(2) To illustrate backward convergence, think of retreating to point 1 from point 2, and consider doing so by successive halfway steps, first to the midpoint between 1 and 2, then to the midpoint from there to 2, and so on With each step one draws closer to point 2, but will never succeed in reaching it.
(3) The formulation of this fallacy and of subsequent ones are my own.
(4) Note that while predetermination entails precedence-determination, the converse is not the case: precedence determination does not entail predetermination.
(5) See Immanuel Kant, Critique of Pure Reason, trans. N.K. Smith (New York: St. Marks Press, 1929), A803/B831.
(6) "A proper Principle of Causality nowise impedes free will" (Das wahre Causalprincip steht ... der Freiheit nicht im Wege); Hermann Lotze, Microcosmus: An Essay Concerning Man and His Relation to the World (Edinburgh: T. & T. Clark, 1885), 16.
(7) Dennett, Freedom Evolves, 120.
(8) Spinoza, Ethics, Book 1, Appendix.
(9) Charles Darwin, Charles Darwin's Notebooks, 1836-1844: Geology, Transmutation of Species, Metaphysical Enquiries, ed. Paul H. Barrett, Peter Jack Gautrey, Sandra Herbert, David Kohn, and Sydney Smith (Ithaca, N.Y.: Cornell University Press, 1987), 187.
(10) Lotze, Microcosmus, 287.
Correspondence to: 1012 Cathedral of Learning, Pittsburgh, PA 15260
|Printer friendly Cite/link Email Feedback|
|Publication:||The Review of Metaphysics|
|Article Type:||Critical essay|
|Date:||Mar 1, 2009|
|Previous Article:||The grounds of right and obligation in Leibniz and Hobbes.|
|Next Article:||An idea is not something mute like a picture on a pad.|