Compatibilism holds that free will can coexist with determinism. But what exactly is determinism? The term is sometimes used casually and with somewhat fluctuating meanings. Different versions (or aspects) constrain free will theories to quite different degrees.
SOME DEFINITIONS
I was introduced to determinism during one Christmas break, when I read Will Durant’s The Story of Philosophy to get some background before my first college philosophy course. As I recall, Durant said determinism was based on a vision of the universe as a giant machine, cranking along as it inevitably must. The categories of actual and possible overlap 100%, which means that everything that happens is the only possible thing that could happen. My naive undergraduate mind didn’t see much room for free will in that scenario. If anything, getting rid of foofy notions like free will seemed to be precisely the point.
More recently, Al Mele defined determinism in the Encyclopedia of Consciousness as “the thesis that the combination of a complete statement of the laws of nature and a complete description of the condition of the entire universe at any point in time logically entails a complete description of the condition of the entire universe at any other point in time.” Adina Roskies defines it in TICS as belief that “the state of the universe is entirely a function of physical law and the initial conditions of the universe.” The Big Questions in Free Will Lexicon of Key Terms (http://www.freewillandscience.com/wp/?page_id=63) defines it as “The thesis that a complete statement of the laws of nature together with a complete description of the entire universe at any point in time logically entails a complete description of the entire universe at any other point in time.”
I see up to three elements to these definitions, and different people may embrace them to different degrees. That could cause confusion and arguing past each other.
The three are pan-causality, reductionism, and predestination. With a nod to a classic Western movie, I think of them as the good, the bad, and the ugly. Let me comment on each and what it does to free will.
THE GOOD…
The simplest aspect of deterministic thinking says that everything is caused. This “gentle” aspect of determinism is not hard to swallow and may even be compatible with some randomness. I suppose it depends on what is meant by “causing,” but as far as I can tell that is simply a description of how one event or state leads to another. (Actually, in practice, multiple events or circumstances come together to create another; single-cause explanations are probably almost never correct. Every causal principle known to psychology ceases to operate above 200 degrees Fahrenheit, for example.)
Full-blown determinism, as indicated by the definitions I gave earlier, generally involves more than just pan-causality. Still, I have come to realize that many eminent scientists (including perhaps my sometime debating opponent John Bargh) understand determinism as no more than pan-causality, and so it is no wonder they are upset when people question determinism. To them, science is constantly “proving” determinism insofar as it furnishes evidence of causality. Hence it is helpful to consider a watered-down, gentle form of determinism that insists only on causality.
There are many types of causes. The causal principles one would invoke to explain why an electron changes its orbit might well bear no resemblance at all to the principles one would invoke to explain the international economic collapse of 2008. Meanwhile, pan-causality even allows for causes that operate in large systems, for causes that operate by changing probabilities, and for multiple possible outcomes in the future. I’m not sure what all causes have in common, except that some things lead to other things. New kinds of causes could even emerge. Natural law has presumably been there since the Big Bang, but still, there is no reason new causes couldn’t emerge, is there? Meanwhile, cultural laws are certainly more recent. Next year there will be new laws, unless all governments go on strike.
Pan-causality by itself seems not that hard to reconcile with many definitions of free will. Free will could be simply another one of the many different kinds of causation. Unlike billiard balls, electromagnetic fields, the laws of probability, gravity, and other sorts of causation, free will operates with a single agent who conceptualizes multiple possible future outcomes and chooses among them based on reasons that may include symbolism and meaningful calculations.
THE BAD…
The formal definitions of determinism I quoted above refer specifically to natural laws. Cultural laws are not mentioned, even though they clearly have causal impact. (For example, traffic laws alter the trajectories of many molecules!) The only excuse I can imagine for failing to acknowledge cultural laws — and all other social, cultural, and symbolic causes of behavior — is that people may think that natural laws can fully explain them. This brings up the second element in notions of determinism, namely reductionism.
Reductionism is the belief that what happens on one level can be fully explained by events at lower levels, presumably all the way down to subatomic particles. To reductionists, your thoughts are entirely explainable as the results of chemical and electrical activity in your brain. Your brain is fully explainable as the result of biological causes, which in turn can be entirely accounted for by chemical reactions, which in turn are the results of molecular structure, and so on. Physics is the bedrock, the lowest level, to which everything else can be eventually reduced.
In physics, reductionistic views once prevailed, and physicists all comfortably assumed that eventually their work would be able to explain everybody else’s. The turning point was Philip Anderson’s classic article “More is Different,” in which he argued (and the field came around to agreeing with his view) that different kinds of causes operate at different levels. In particular, new kinds of causes emerge at different levels or scales. Today I doubt you will find many professional physicists who think the banking crisis of 2008 could ever be explained in terms of the exchanges of photons between protons and electrons.
To appreciate the problem, imagine a book that explained the Civil War in lower-level terms, by describing only muscle movements, say, or (below that) nerve cell firings, or chemical reactions, or the movements of subatomic particles. Not only would this style be cumbersome and inconvenient. I’m pretty sure that even someone who read the whole book wouldn’t be able to recognize it as a history of the Civil War. It would completely miss the point of the war and, more profoundly, would fail to explain the true causes of all the action. Some of these causes existed only at the level of large systems of shared understandings among groups of people.
I said earlier that some scientists may believe that pan-causality is constantly being proven by findings that show causation. How someone could sustain a belief in reductionism is much harder to say, however. Reductionism has been “proven” wrong locally countless times. With regard to patterns of human behavior, for example, as research marches on, things almost always turn out to be more complicated than initially surmised. Over time, as new data come in, theories generally become more and more complex, not simpler as reductionists would have us believe.
In any case, if one does embrace reductionism, free will is still plausible but loses much (as compared with mere pan-causality). Ultimately free will would have to be fully explained by subatomic processes, which means you can completely account for it in terms of the handful of variables that physicists use, such as each particle’s mass, charge, velocity, and so forth. Responsibility and justice are just chemical reactions, so to speak. So is free will in a reductionist’s worldview.
AND THE UGLY
If causality is the good and reductionism is the bad, predestination is the ugly.
Predestination rules out multiple possibilities and insists that everything that happens is the only thing that possibly could. The future is just as fixed in stone as the past. Everything you will ever say or do has already been scripted.
In the Lexicon, predestination is essential to the definition of determinism, since pan-causality is specifically said to be not enough for full determinism. It has to be possible to deduce the full state of the universe at any time if you know the full state any other time plus have full account of the laws of nature (and presumably if you have one super kickass computer.)
In a sense, predestination undermines some concepts of choice and agency. The agent is just a throughput. The literal meaning of choice is to select among possible alternatives, but the predestinationist insists there is only one possible outcome rather than multiple possible alternatives, so the agent’s belief in their multiplicity is fundamentally mistaken. The agent is doing something and being part in the causal chain. But the agent is not doing what agents by definition are supposed to do, namely exert control in the sense of choosing and steering among multiple possibilities, only some of which come true.
The athlete’s belief that he or she can influence or help cause the outcome of the game is correct, but the belief that he or she can change the outcome is wrong, according to the predestinationist. The fundamental essence of a game as having multiple possible outcomes is an illusion too, if you accept predestination.
SIZING UP THE COMPATIBILIST BEDFELLOWS
Compatibilism means that free will can coexist with determinism. But the good, the bad, and the ugly aspects of determinism are not equally congenial bedfellows for free will. You can have a vigorous and fascinating free will compatible with the gentlest form of determinism (ie, pan-causality alone). My sense is that purists (including the Lexicon) agree that pan-causation alone is not full-blown determinism. However, it is probably a popular version with the (wo)man in the street and other philosophically challenged folks, scientists included, and of course it is far more likely to be true than the more demanding versions that bring in predestination and/or reductionism.
Once reductionism and/or predestinationism climb into the bed, however, the scope for free will narrows dramatically. As a trio, the good, the bad, and the ugly will be hard for free will to satisfy and hard to live with. A free will that is compatible with them all looks to me more like a puppet government, or even just a puppet.
Help me out here, dudes! Can a free will find compatibilist happiness together with all three bedfellows?
Also, is there something that all causes have in common?
Just to pinpoint:
"Predestination rules out multiple possibilities and insists that everything that happens is the only thing that possibly could."
This is probably one of the most common confusions about the free will debate. It's good to see this confusion expressed by a psychology professor, since Thomas Nadelhoffer and Eddy Nahmias claim that folk will not be victim to this confusion - and then criticize Nichols et al. for using the phrase 'had to happen' in conveying determinism. My point is that if academics make this mistake, folks will surely make it...
Here is where I think you are confused, at least if we are concerned with 'possibility' in the sense that philosophers use it: The fact that universe has a single future at every point in time doesn't mean that everything that happens is the only POSSIBLE thing that COULD happen.
For instance, the laws of nature could have been different and yielded a different set of outcomes. The initial conditions of the universe could have been different.
Note that according to the notion of possibility at work here, it's even possible that one could go faster than the speed of light. (It's not physically possible; but logically, conceptually it is.)
Just to illustrate the sort of necessity at work, it's not possible that 1+1 not equal to 2 - or it's necessary that 1+1 equal 2.
You may wonder what this distinction has to do with anything in regards to free will.
I would suggest that (as an example) you check out the work of John Fischer and his theory of guidance control that heavily depends on counterfactuals in his arguments for the compatibility of determinism and moral responsibility (but not free will)...
Posted by: Cihan | 06/28/2010 at 09:00 PM
Cihan,
I think you can forgive Roy for this lack of precision in language. He articulated the gist of determinism regardless of whether he used "possible" in its philosophical sense. Your criticism really bears no relevance to his argument. After all, the crux of the argument regards free will, not moral responsibility, so what's the point in appealing to John Fischer's work?
Posted by: Noah Te Stroete | 06/28/2010 at 10:40 PM
Roy--if I may--a great post.
I'd ask first: does determinism as a concept apply to anything at all? And it seems it does. I'd say systems of events as descriptively constrained by criteria of adequate scope--talk about billiard ball collisions as against Compton electron and photon collisions--allows us also to talk about the determinism of billiard ball collisions as against the indeterminism of Compton scattering.
So it seems the question is: are mental/brain events related to FW questions ones best branded by billiard-ball talk or Compton talk?
Of course there are empirical questions about whether, as you say, reductivist concerns intercede at more subtle levels. Pool and electrons are one thing, as plausibly remote from one another; decisions and electrons another as much more plausibly related. There might be real blurring in the latter not at all descriptively relevant to the former. I concede that.
But here is where empirical neuroscience intrudes. What if we can show that neural processes as related to cognition/behavior are more like billards than Compton phenomena? Then there would be good reason to prefer a deterministic description of them than the alternative. Of course epistemic uncertainty about experimental data, etc. might dilute confidence here, but still, it seems conceptually possible that evidence of a certain sound sort might favor a real claim to determinism of mind as against any alternative. (In fact I assume that many neuroscientists have already laid claim to such determinism, though I would argue tentatively at best. Not all loops are closed here, as Roger Penrose has tediously maintained.) And if science can show that determinism is an apt description of mental processes, then isn't any concern about indeterministic FW thus cast into the role of hopeful speculation or unmitigated faith at the very best? This hasn't been done of course--but I think that indeterminists about FW might best prepare for it. Using margins of experimental error for the claim of determinism as a basis for the possibility of indeterministic influence has been (I woulod argue) a losing battle first waged by C.A. Campbell and pursued by libertarian minimalists like van Inwagen and (more empirically) Kane ever since.
Posted by: V. Alan White | 06/29/2010 at 12:55 AM
You write: "The simplest aspect of deterministic thinking says that everything is caused. This “gentle” aspect of determinism is not hard to swallow and may even be compatible with some randomness. I suppose it depends on what is meant by “causing,” but as far as I can tell that is simply a description of how one event or state leads to another. (Actually, in practice, multiple events or circumstances come together to create another; single-cause explanations are probably almost never correct."
The best way to understand the universal governance of causality is to consider the evolution of the universe from one state (literally, and as a singularity, which addresses misnomered references to multi-universes) to the next. As LaPlace noted, the state of the universe at one moment is COMPLETELY determined by the state of the universe during the previous moment. The ensuing causal chain is universal, absolute, and not subservient to even quantum interpretations of particle behavior as random. Because the universe is a singularity, this causal progression is not a "multi-event" phenomena. This universal, and by definition all- pervasive, aspect of determinism is anything but "gentle."
Such universal, singular, and inviolable state causality is why free will is absolutely impossible. Any and all logical/rational refutations clearly fall under the weight of the singular, all-encompassing causality demonstrated by the evolution from one state of the universe to the next.
Posted by: George Ortega | 06/29/2010 at 03:48 AM
Hi Roy, this is a useful post, since it indicates how easy it is to confuse determinism, as used in the philosophical debates about the compatibility question, with other theses, such as reductionism and fatalism. Once we make these distinctions, we can see that determinism is not the problem many people take it to be, including many "ordinary folk" and many scientists.
I humbly suggest that people might find this old post of mine at the Garden useful:
http://gfp.typepad.com/the_garden_of_forking_pat/2009/05/defining-determinism-and-such.html
Posted by: Eddy Nahmias | 06/29/2010 at 09:08 AM
Eddy,
If the universe as a singularity is completely deterministic, as is clear by considering it's state-to-state evolution, every specific event, including the actions of human will, would need to be completely constrained by that same state-universe causality. Humans freely willing actions that somehow violate determinism would pose a huge problem in that one would then have to explain how the universe was presumably completely deterministic pre-humans, and with the first human suddenly became a curious combination of acausality in the human-specific existing alongside complete and absolute causality in the state universe-general. The above context naturally subsumes both fatalist and reductionist perspectives.
To better clarify "determinism," since absolutely precise prediction is impossible at BOTH macro "determinist" and quantum levels, determinism is essentially analogous to the causality expressed in state-universe evolution. As such, it poses a huge, indeed universal, problem to free will.
Posted by: George Ortega | 06/29/2010 at 10:44 AM
George Ortega's comments bring up another interesting point. He insists that the universe is one-state singularity. I gather this means that all events are interconnected. The universe moves from one state to the next as a whole, a unity. Is there any evidence that that is true?
If anything, the opposite seems more correct, or at least worth considering. After the Big Bang there may have been plenty of independent events. Causes here and there do their thing without reference to what happens far away, on other continents or planets.
Indeed, the intersecting of independent causal chains (eg, whether a falling piano hits the passerby or just misses) seems like one potential source of unpredictability that would the thwart the determinist's super-computer trying to predict the exact future.
Posted by: Roy Baumeister | 06/29/2010 at 01:58 PM
Evolution of free will: clarifying one point in George O's comment.
George says "Humans freely willing actions that somehow violate determinism would pose a huge problem in that one would then have to explain how the universe was presumably completely deterministic pre-humans, and with the first human suddenly became a curious combination of acausality in the human-specific existing alongside complete and absolute causality in the state universe-general."
Those of us trying to understand the evolution of freedom would not believe things to be sudden, (nor do we look at free will as acausal, but that's another issue).
As Tomasello's research and others make clear, humans differ from other animals in understanding the world in terms of invisible causes. This use of inferential thought is a collective achievement, as is accessing symbolic meaning. No single person could invent a language on his or her own. It is the use of meaningful thought that adds a different dimension to human decision making, absent (or minimal) in the animal world. This would not have been sudden at all. Probably it took tens of thousands of years for humans to build up culture, including language and the associated understandings.
Hence learning to guide one's behavior based on symbolic ideas and socially transmitted meanings (which are not physical things: they lack the properties of physical objects, such as mass, velocity, molecular structure, physical location, etc) would have been a gradual process, dependent on cultural progress based on accumulation of knowledge. That to me is the likeliest candidate for how freedom evolved. It does then allow nonphysical realities to enter into the stream of physical events. Nothing that happens violates natural or physical law, but it goes beyond it in that some causes are not physical things.
Posted by: Roy Baumeister | 06/29/2010 at 02:10 PM
Roy,
Regarding your 06/29/10, 1:58pm comments;
I didn’t say that the universe is a “one-state” singularity. I said it was simply a singularity in that there is only one universe, which, of course, encompasses the entirety of reality. You asked for evidence of this, and the short answer is that there is no direct empirical evidence. To provide empirical evidence one would have to be able to perceive, or measure, the entire universe. Such evidence is unattainable to subjective beings such as humans.
However, we can hold this state-to-state, moment-by-moment, evolution of our universe as axiomatic if we accept the following two premises:
1) There exists an “arrow of time,” meaning that time always moves from past to present to future, and never in reverse.
2) Our universe does not exist outside of time.
Thus, if we consider that the universe in its entirely is the only TRULY closed system in nature (all other systems being open to outside influences), it MUST evolve as a unity from state to state via a singular causal progression. To suggest otherwise invites the notion of more than one universe, which, of course, by definition, would be quite incoherent.
When the universe is considered in its entirety as a singularity, relativity no longer applies, (there being no other object the universe could be relative to), and no manner of interdependent causal chains, (a closed system singularity permitting only one causal chain).
LaPlace was, of course, completely right in asserting that if one were to know completely the entire state of the universe at one moment, one could accurately predict the state of the universe at any future moment. Of course the subjectivity that prohibits the simultaneous measurement of particle momentum and position, and the lack of a complete description of any universal state, render that prospect potentially possible only to a universe with complete self-knowledge (as certain definitions of God would allow).
Posted by: George Ortega | 06/29/2010 at 10:51 PM
Roy,
Regarding your 06/29/10 2:10pm comments;
Whether humans could develop the ability to violate state-universe causality suddenly as opposed to over an indefinite period of time is not the salient consideration. Our problem manifests when acknowledging that for the above scenario there must exist a moment of initial causality violation. To assert otherwise would suggest that causality could be violated in stages, a prospect fraught with conceptual impossibilities.
Regarding Tomasello’s research, it is one matter to be able to understand invisible causes and another matter altogether to by such means be able to violate the causal laws of nature.
Being a materialist, I cannot accept the premise that “symbolic ideas and socially transmitted meanings” are not physical. Because we have not been able to, through our state-of-the-science technology, detect such realities does not mean they are not physical. You seem inadvertently to defend the physicality of all reality with your statement “Nothing that happens violates natural or physical law…” If everything is, in fact, governed by physical law, everything must, by definition, be physical.
Posted by: George Ortega | 06/29/2010 at 11:31 PM
George,
I respect your position. But here's how I see it. When you say that symbolic ideas and socially accepted meanings are physical but we can’t detect them, I must part company. First, we detect them all the time, in that we talk about them and use them. We just don’t detect them with physical measures, because they aren’t physical things.
For example, physical things can be weighed, because they have an objective mass. But our discussion here cannot be weighed, because it is not a physical thing. (Though no doubt it should be heavy, ha ha.) It is a symbolic communication of ideas based on shared understandings (and maybe a few misunderstandings).
That brings us to the second and more important point. We know well the properties of physical things, and the list is not that long: mass, precise location, velocity, acceleration, molecular structure, chemical composition, and so on. Every physical thing has these properties.
But what is the chemical formula of justice? What is the mass of the French language? What is the molecular structure of the new health care reform act? What is the precise location of the national debt? These things are real and have physical consequences, but they are not themselves physical things. They have essentially none of the properties of physical things, and to be physical things they would have to have all of those properties.
Those are social realities. They exist not in individual brains but in the nonphysical 'space' of sharedness of understandings among them, using mostly pre-existing worlds of ideas that can only be accessed by large groups of intelligent creatures working together to accumulate knowledge across generations. That is what humans evolved to do.
And, to go a step further, working together to accumulate knowledge is why our action control system is so much more complex and advanced than what other animals do. That is the real phenomenon, which is I think what people call free will. One can argue whether it deserves that name or not, depending on definitional issues. But that’s the real thing, whatever you call it.
Posted by: Roy Baumeister | 07/02/2010 at 10:01 PM
Roy,
I didn’t say that we can’t detect ideas and socially transmitted meanings, per se; I said that we can’t detect them “through our state-of-the-science technology.” Our minds are, of course, conscious of, and can detect, them.
Our discussion digressed to the question of dualism vs. universal physicality, and while I maintain my stated position, we can return to the determined vs. free will question from your understanding that some things are not physical.
Since it is human volition that is at issue, we can limit our exploration to decisions. If we agree that all decisions, regardless of their physicality or lack thereof, are events that occupy a specific and precise location within time, and that these decisions are caused, then whether or not the decisions are “physical” becomes inconsequential. The decisions are causality determined, and therefore cannot be freely willed. In other words, it is their existence as EVENTS along the timeline that renders them subject to causality.
I previously argued against your assertion that because there may be multitudinous causes of an effect, an opening for free will somehow arises by offering an holistic description of the state-to-state evolution of our universe. However, the causal evidence against free will is equally effective via the multitudinous causes perspective.
Whether there are two or two trillion separate causes that together constitute the necessary and sufficient determinants of a decision, every one of those individual causes has its own causal history that ultimately stretches back in time before, and thus beyond, a decider’s will, or control. Finally, addressing your distinction between other animals and humans, it is for the above reason that neither less complex life forms nor we can have a free will.
Posted by: George Ortega | 07/03/2010 at 09:16 PM
"Whether there are two or two trillion separate causes that together constitute the necessary and sufficient determinants of a decision, every one of those individual causes has its own causal history that ultimately stretches back in time before, and thus beyond, a decider’s will, or control."
George,
You are assuming that human decisions have sufficient causes. This seems to me to be begging the question.
It seems that you and Roy are at what John Fischer would call a "Dialectical Stalemate". Neither side can use their arguments in their current forms without first begging the question. You are assuming sufficient causes and Roy is assuming libertarian free will. Or am I mistaken?
It seems at the moment without any decisive proof of whether or not there are sufficient causes for human decision, belief in free will or determinism lies more heavily on values than evidence. That is why I remain an agnostic on the issue (at least until I encounter some convincing evidence to sway me in either direction).
Posted by: Noah Te Stroete | 07/06/2010 at 02:55 AM
Noah,
Causality is THE basic physical process, without which the universe would be COMPLETELY static, still, and without ANY manner of change. Time could not exist without causality, and causality could not exist without time. A decision is an event in time. It is, therefore, also an effect in time. All effects, which would clearly include human decisions, must, by definition, have a sufficient cause, or causes. The preceding statement is a deduction based on the reasons and evidence presented above. As such, it does not beg the question as you suggest. Consider also that the only possible alternative presenting is that human decisions are not caused, in which case they could not be the result of, or caused by, a free will.
Posted by: George Ortega | 07/07/2010 at 03:24 AM
"Consider also that the only possible alternative presenting is that human decisions are not caused, in which case they could not be the result of, or caused by, a free will."
George,
I may be mistaken but I think that libertarian free will means that human decisions have no sufficient causes. That's what free will means in this sense. "Free will" is not something such as a force that causes. It is a description of the (disputed) fact that decisions have no sufficient causes.
Also, I could dispute your premise that the universe would be completely static and without any manner of change without sufficient causes. The world may be populated with agents. Just to be clear, I think we should come up with a precise definition of "cause". Then and only then can we stop talking past each other and agree or agree to disagree.
Is a cause an invisible force or is it just a contiguous event that precedes another event?
Regards,
Noah
Posted by: Noah Te Stroete | 07/07/2010 at 12:26 PM
On the other hand, if free will is a force, then free will could be THE sufficient cause of decision. What could be meant that human decision has no sufficient causes is that it has no sufficient causes outside of the will. So, in this sense the will could be the only sufficient cause.
Posted by: Noah Te Stroete | 07/07/2010 at 01:05 PM
Noah,
Libertarian free will argues for an “agent” causing the choice in a way that is completely free of, rather than being caused by, the causal history predating the agent.
The perceived value of a free will is in claiming credit for the good we do, and credit for the wrong we refrain from doing. An agent acting without reason, or cause, would be of no use to the Libertarian imperative for accountability. You might see from this why Libertarian free will is a self-defeating, and thus incoherent, construct.
For a precise definition of cause, you might, again, consider it THE process that allows for change, as I explained in my earlier post. Causes may be invisible to subjective beings such as humans, or to our most advanced technology, but that is inconsequential. They are not merely contiguous events to their effects, or events; they are the reason (or reasons) for those effects, or events.
Posted by: George Ortega | 07/07/2010 at 01:58 PM