This is the first in a series of postings I've commissioned by leading philosophers on "where the action is" in various subfields of the discipline. (I wrote about Nietzsche studies a couple of weeks ago.) Hopefully these short essays will be illuminating for both graduate students interested in the particular subfield and philosophers working in other areas who want to keep abreast of developments outside their specialty.
Today, John Martin Fischer at the University of California at Riverside writes about "where the action is" on free will and moral responsibility.
=====================
The topics of free will and moral responsibility continue to be the subjects of lively and active debates in contemporary analytic philosophy. Not surprisingly, no consensus about results has been reached (cf. the fallen angels in Milton's Paradise Lost), and, like much contemporary philosophy more broadly, there is a fairly wide range of methodologies. Additionally, some philosophers approach these topics with an emphasis on ethical issues (such as moral responsibility, blameworthiness/praiseworthiness, judgments such as "ought/ought not", "right and wrong," and so forth), whereas others approach the topics from the perspective of metaphysics and/or philosophy of mind.
There was a tremendous amount of work on the relationship between God's omnisicience and human freedom (and responsibility) about twenty-five years or so ago, but this discussion has been relatively less lively recently (with certain exceptions). The doctrine of Molinism has been a hot topic, especially in certain parts of Indiana. According to Molinism, God can know prior to any providential activity on his part a set of "counterfactuals of freedom"--subjunctive conditionals about what human beings would freely do in various circumstances. He combines this "middle knowledge" with his "natural knowledge" of the various possible initial situations to derive truths about what human beings would freely do, and he uses this information to guide his creation and providence. There has been much debate about whether there can be true "counterfactuals of freedom," with Robert Adams leading the charge against this possibility. The problem is that such conditionals would seem to need "metaphysical grounding," but if causal indeterminism is true, all that would seem to be grounded are "might-conditionals". (Molina and his followers believe that causal determinism is incompatible with the relevant sort of freedom.) Alfred Freddoso and Thomas Flint are contemporary exegetes and defenders of Molina. Whereas there is much of interest in these debates, I believe they are orthogonal to the problem about the relationship between God's foreknowledge and human freedom. In my view, Molinism provides a picture of how God could know about future actions of humans, and how he could use this knowledge in his providential activity. But it does not provide an answer to the problem about the relationship between God's foreknowledge and human free action; rather, it simply presupposes some answer to this problem. Indeed, Molina's favored solution to the latter problem was that of Aquinas, according to which God is outside of time (and thus has omniscience but not foreknowledge), and Flint's solution is Ockhamism. Some philosophers (for example, Robert Kane in his otherwise excellent introductory book on free will in Oxford University Press's Fundamentals of Philosophy series) seem to think that Molinism is an answer to the problem about omniscience (or foreknowledge) and freedom, but this is a mistake.
I think it might be interesting to explore whether any contemporary work on the nature of belief (in philosophy or perhaps the cognitive sciences, but probably not neuroscience!) might be employed to seek to make some progress on the traditional problem about the relationship between God's omniscience (foreknowledge) and human freedom. This is not where the action is, but it is one place where I'd like to see some (more) action. (For discussion, see the Prosblogion blog.)
There is some intriguing work on the nature of causal determinism. J. Howard Sobel's book, Puzzles of the Will, has a subtle discussion of the various forms of determinism and how they might be thought to threaten human freedom. He (in my humble and contentious opinion!) successfully shows that the Consequence Argument (or Basic Argument for Incompatibilism) can be formulated rigorously in various ways, not all of which require a modal principle such as the Principle of Transfer of Powerlessness (the principle that acts like a slingshot that can transmit powerlessness from one thing [the past] to another [the present] via the natural laws). Some philosophers, notably Peter Van Inwagen and Ted Warfield, maintain that all valid versions of the Consequence Argument rely explicitly or implicitly on such a modal principle; I have been a persistent (they might say pesky!) critic of their view. Carl Hoefer has a fascinating paper in which he points out that contemporary physics may not sit well with the basic intuitions that go into the Consequence Argument; in Hoefer's view, the intuitions that play a role in this argument (and buttress the Fixity of the Past and Natural Laws) come from common sense (and the A-Series picture of time), whereas physics presupposes the B-Series (without a moving now, as it were, and thus without past, present, and future).
Recently philosophers have distinguished at least two importantly different ways in which causal determinism might threaten moral responsibility. One way stems from the Consequence Argument. If causal determinism rules out the sort of freedom that involves alternative possibilities, and this freedom is required for moral responsibility, then causal determinism is incompatible with moral responsibility. Some compatibilists deny that the Consequence Argument successfully rules out the sort of freedom that involves genuine alternative possibilities. Others contend that moral responsibility does not require this sort of freedom. There are at least two different versions of this latter strategy. Harry Frankfurt and I (and others) have contended that "Frankfurt-style" examples (first invented by John Locke) show that moral responsibility does not require alternative possibilities. The debate about the Frankfurt-style examples still rages; a good recent collection is edited by David Widerker and Michael McKenna: Moral Responsibility and Alternative Possibilities: Essays on the Importance of Alternative Possibilities (Ashgate). David Widerker has been an important and influential critique of the strategy based on the Frankfurt-examples; there are rumors that he has a new paper in the works on Frankfurt-examples that might surprise many. (Is it too much to hope that Widerker will defect to the side of the good guys?) Another strategy builds on the seminal work by Peter Strawson; there are important recent developments of Strawsonian approaches in Paul Russell's Freedom and Moral Sentiment and R. Jay Wallace's Responsibility and the Moral Sentiments.
Another way in which causal determinism might threaten moral responsibility is directly and not via threatening to expunge the sort of freedom that involves alternative possibilities. A whole range of different options are available here. "Source incompatibilism" is the term typically used to refer to this family of worries; this view has it that if causal determinism is true, then an individual would not be the "source" of his choices and behavior in the way required by moral responsibility. Important developments of this view are found in Robert Kane's The Significance of Choice and Derk Pereboom's Living Without Free Will. Although there is still much interest in the Frankfurt-style examples and Strawsonian approaches, the "action" seems to have shifted somewhat in the direction of Source Incompatibilism recently. (I cannot help noting--please excuse this--that in my 1982 Journal of Philosophy article, "Responsibility and Control," I identified Source Incompatibilism as a significant view with which compatibilists must come to terms; but of course in subsequent years I have sought to do just this, whereas such philosophers as Kane and Pereboom have developed and defended the position with considerable detail and skill.)
Timothy O'Connor's book, Persons and Causes, seeks to develop an agent-causal view. It is the most sophisticated and comprehensive development of such a view of which I know. In an interesting and important recent book, Libertarian Accounts of Free Action, Randolph Clarke surprisingly backs off from his former defense of agent-causal libertarianism; there is still hope that he will come all the way over to the compatibilist's side, although he remains unconvinced.
A major problem for any version of libertarianism--event-causal (Van Inwagen, Kane, Ekstrom) or agent-causal (O'Connor) is the problem of arbitrariness or luck. Much recent work has addressed this set of problems, and Alfred Mele has written a book on this subject forthcoming in early 2006 with Oxford University Press. I hope he solves the problem, because I believe that moral responsibility should not "hang on a thread"--should not depend on whether the conditionals that link past to present in physics have associated with them 100% probabilities or merely 99.9% probabilities! Since I believe our most basic views about ourselves as free and responsible should not hang on a thread (at least a thread of this sort), I believe that these views should be resilient with respect to a discovery that causal determinism were true; also, they should be equally resilient with respect to a discovery that causal indeterminism (of certain sorts) were true. So good luck, Al--no pun intended!
Various important challenges have been raised to our views of ourselves as robustly free and responsible, including Galen Strawson's Freedom and Belief, Saul Smilansky's Freedom and Illusion, and Derk Pereboom's Living Without Free Will. (One can imagine a whole series of Living Without X books: Living Without Ethics, Living Without Knowledge, Living Without Aesthetics, and so forth…)
There is much intriguing and suggestive work employing somewhat nontraditional methodologies. Neurophilosophers such as the Churchlands have written recently about freedom and responsibility. Also, there is a booming industry in "experimental philosophy" which seeks (among other things) to clarify and establish what people actually believe about the range of issues pertaining to freedom and responsibility. I think that progress in the future on free will and moral responsibility will most likely be made through a broad-minded and eclectic approach whereby one takes insights from both traditional methodologies and also the developing informational sciences, psychology, and neuroscience. I think it is misguided and not fruitful to brush aside centuries of thought on these ideas as my friend Daniel Dennett does in his intelligent, lively but in my view disappointing book, Freedom Evolves. Similarly, we need to be open to the insights potentially available through the work of neuroscientists, psychologists, and philosophers employing these intellectual tools.
I shall end this (too long but also obviously too short) post with my puzzlement over a "splinter" in the work on free will/moral responsibility. Some theories are called "normative" accounts (Watson, Wolf, Wallace, Benson, et. al.), whereas others are called "metaphysical" (Van Inwagen, Ginet, Fischer and Ravizza, O'Connor, Clarke, et. al.) It is striking to me that the "normative" group is highly miscellaneous; the specific ways in which they are "normative" are very different, and it would be useful to get a clearer understanding of the distinction. I doubt that it is as sharp or important as some suppose. The difference may be as much a matter of "conversational networks" (as Manuel Vargas has pointed out to me) than content.
I should note that much of interest had to be omitted from this discussion--I'm sorry. I welcome you all to the discussions at "The Garden of Forking Paths."
(Please note that comments may take awhile to appear.)
