What the Free Will Problem is About

In the effort to increase chatter on this blog, I will continue talking to myself. For a change of pace, this time I will be talking in a widely speculative and impressionistic fashion that is only dimly supported by facts.

The folks at The Garden of Forking Paths ask, "What is the Free Will Problem About?": http://gfp.typepad.com/the_garden_of_forking_pat/2005/01/what_is_the_fre.html.

Unsurprisingly, the default choices were all surrounding whether something called "Libertarian Free Will" exists (my non-philosophers friends should note: this is a metaphysical idea; it has nothing directly to do with Libertarian Party politics). I'm no scholar of the free will problem, but I doubt anyone could give a satisfactory definition of "Libertarian Free Will," which, as I will argue, is part of the problem. The expression does, however, undoubtedly signify a cluster of vaguely related ideas about locating the ultimate origination and responsibility for an agent's actions within its Deep Self--i.e., the idea that each time I act freely it is a metaphysical fact that I--in a sense that needs to be spelled out--am the definitive causal source of my actions and the only thing that is morally responsible for them.

This notion of Libertarian Free Will is supposed to be incompatible with determinism because if every event, in including any of my actions, is preceded by a set of events in the past that can serve--in conjunction with the laws of nature--as a sufficient cause of my action, then I cannot be the ultimate originator of these actions. Because I cannot be the ultimate originator of these actions, the Incompatibilist picture says that I cannot be ultimately responsible for them. The Hard Determinist infers from this either that I am not ultimately responsible for these actions or that I have no free will. The Libertarian infers from this that determinism is false. The Compatibilist tries to tweak our understanding of the facts so that I am still perfectly responsible, despite lacking ultimate origination.

The entirety of the free will "problem" is, in my opinion, a result of the fact that this notion of ultimate origination and the picture of the Deep Self that lies behind it are deeply confused and artifacts of the way our culture has tried to construct notions of political and moral autonomy.

The free will problem should not be a question about deep metaphysical facts such as, "am I the ultimate causal source and locus of responsibility for my actions and self?" The free will problem should rather be a set of normative questions followed by a corresponding set of practical questions. The set of normative questions would be something like this: "how should I go about taking responsibility for this self?--how should I go about justifying my actions to myself and to others and to the world?--how should I go about endorsing my desires and their fruits in action?--what should the structure of my will be?--what are the desires I should desire to have?" The set of corresponding practical questions would simply be questions about what set of social institutions, laws, and moral duties best describes a program for implementing the answers to these normative questions.

This shift in perspective would make the free will problem a branch of normative and practical ethics, which is fine by me. The result is simply classic extistentialism, filtered through Harry Frankfurt's notion of a hierarchical will from "Freedom of the Will and the Concept of a Person" and some thoughts of Charles Taylor's in "Responsibility for Self" (Professor Dennett also seems to have something like this in mind towards the end of Freedom Evolves, so too do Professor White in his essays that touch on free will and Richard Moran at Harvard in Authority and Estrangement: An Essay on Self-Knowledge).

People are often unsatisfied by a focus on the normative and practical questions because they seem to leave out important factual issues that are necessary for free will. For example, we want to be assured that our selves are actually sources of causal creativity in the world--that is, we want to know that our choices make a difference and that the future is really a place of open possibilities, such that our choices decide between these possibilities and our actions serve to realize the results of our choices. We also want to be assured that we are in at least some sense the originators of our actions--that is, that we are the locus of doings in the world rather than just happenings. Determinism seems to threaten this picture of the self by reducing us to moving points in space through which wider causal stories flow, rather than allowing us to be the productive sources of these causal stories. It seems that, if these gloomy facts about the world are true because of determinsm, then the normative and practical questions raised above cannot even get off the ground: we are left with an empty phenomenology of free will that does not touch down in the reality of the world.

Now, as has been pointed out by naturalistic approaches to the free will problem (especially by Professor Dennett in Ch. 3 of FE, "Thinking About Determinism"), what we are imagining here about the threat that determinism poses to the self's causal creativity is simply contrary to fact. The parts of our brain that instantiate the decision-making faculties of our minds are of such a complexity that our actions are inherently not-maximally-predictable to cognizers in the outside world and even to ourselves prior to the formation of an intention-in-action (this is especially true of the gut-wrenching decisions made under uncertainty and emotional ambivalence, which seem to stir up something like chaos in the brain).

When we imagine a Laplacean God taking in the spectacle of our choices and smiling at our naivete, we are probably not imagining something coherent. It is still a very open question as to how to interpret Libet's experiments on the brain--i.e., an open question as to what kind of information about our decisions the observable spikes in Readiness Potential encode. Any Laplacean God--**who cannot read my thoughts**--is going to have to picture my actions as events under a certain description: if the dimensionality of the description is too low, then the Laplacean God runs into amplified quantum indeterminism; if the dimensionality is too high, then the Laplacean God runs into the fact that I eat Cheerios on Monday and Corn Flakes on Tuesday, despite the fact that the decisions are caused by the same equally foul mood each morning (i.e., a mood which keeps me from taking the time to cook some eggs). Some Libertarians, such as Robert Kane, are willing to grant all of this, but they then say that Libertarian Free Will demands not only in-principle epistemic unpredictability, but also in-fact metaphysical indeterminism. I say that a difference that can make no difference to us makes no difference: if we can't count whether the number of stars in the universe is odd or even because of deep physical limitations in our cognitive faculties that could never be remedied by technological advance, then there is simply no fact of the matter as to whether there is an odd or even number of stars in the universe (I get this example from Professor White).

Now, what this inherent unpredictability does not get us is "Ultimate Responsibility," because we are still being situated within a network of wider causal forces, all of which impact our decisions, and some of which we cannot even plausibly be held metaphysically responsible for (even if we must normatively take responsibility for them). What inherent unpredictability does get us, however, is individuality and openness to the future. The plasticity of the brain is such that I cannot be seriously taken as a 'robot' in the colloquial sense of a being that is inexorably implementing the program that nature has designed for its life. There is no such program, because at each instant my mental program is being rewritten in objectively undpredictable fashion, partly as a consequence of the decisions I make. Moreover, if the future is inherently unpredictable, then the fatalistic itch cannot get scratched (and some interesting metaphysical issues in the logic of tensed statements open up).

However, the flip side of all this is that "who I am" is not something that "I am . . ."--in any deep metaphysical sense--". . . responsible for," but it is something that "I must take responsibility for." The key practical question of free will becomes: how do I build self I am willing to take responsibility for? Here, the 'I' can be taken somewhat loosely, because in giving up Ultimate Responsibility, I am no longer as concerned with issues of autonomy: I can use the social and physical world as active tools in order to get the self I want. The issue becomes less one of personal power than "the experience of freedom," which is something like one's picture of the world under a certain value-laden description being realized.

The main challenge that remains to this picture of free will becomes dealing with the sci-fi extremes: is being a brain in a vat that is pumped with pleasant sensations a kind of self worth wanting, if all that matters is getting the "experience" of having one's values realized (Nozick's question in the "Experience Machine" thought-experiment from Anarchy, State, and Utopia)? Clearly, what we want is for the world to be a certain way as well. I have no response to this dilemma, other than to say that the normative question of free will is not just a question about what kind of self I should want to take responsibility for, but also what kind of society and world I should want to create.