Free will has been a tricky and lingering philosophical problem. Here's my solution.
|Emergence of Free Will
Many free will libertarians and Agent-Causation theorists want to put human agency on the same plane of existence as neurons themselves. I think this is as poor of a strategy for showing agency as pointing to an individual fish in a school and saying that that one represents the school. Some think that free will is an illusion that we should keep on believing or perhaps cannot help but believe in, but this may be missing the point altogether. If agency is something which emerges, as I will propose, calling it an illusion may be a useless and misleading claim.
In a simple but useful example, suppose there is a magic eye, an autostereogram, in which a three dimensional picture emerges from two dimensional designs. One may say that the three dimensional image is an illusion, insisting that it does not map to reality. He will say that there is no three dimensional image, just the illusion of one in two dimensional designs and the 2-D designs are all that exist. 3-D image does appear much more illusory than the 2-D one, but this is simply because we are not used to relating to the world in such a way that 3-D images emerge. We could say that the 2-D images are merely illusions of half-tone colored dots, and the colored dots really have no color, they just bounce light back at a certain frequency, and we could keep breaking it down everything into “illusion” like this. Does the “world-as-is” actually contain images or colors, or are these also perceptions and concepts we get from grouping together patterns of dots or wavelengths of light? Should we also say that the notion of a solid as opposed to a liquid is an illusion, since there are no solids or liquids in the world-as-is but simply different arrangements of molecules that with our mental faculties we come to recognize as solids or liquids? Reducing the world to its fundamental building blocks and calling everything else an illusion, just seems silly.
With this talk of illusion, an incompatabilist may ask for a demonstration of free will. But asking for demonstration or proof of free will is quite an odd claim. I think a lot of the literature on the issue thinks the question of free agency to be something of similar structure as a teapot orbiting around Jupiter. Libertarians actually have to demonstrate that their view matches with reality; like I would in claiming that a teapot orbits Jupiter; if they cannot demonstrate it, it is not necessarily false, but of course there is insufficient evidence to justify the belief in the teapot. The claim that humans are agents who act according to their intentions should not be anything like belief in a teapot orbiting Jupiter; it should be of an entirely different sort.
In “The Intentional Stance” by Dan Dennett he describes Martians for which the notions of agency or intentions never come to exist. Their knowledge and intelligence is so vast that that they never come to recognize patterns of behavior into emergent intentions and agency, but can still predict every human action in simply mechanistic terms. If we can make the conceptual leap for the sake of the example so that we could communicate with them, suppose we were to try to show them “evidence” of human agency. What would we point to? If they see things only in mechanistic terms there is nothing to which we can refer to demonstrate intention or agency. The only things we could actually point to would be lumps of matter and they would say, “What is special about this lump of matter? It behaves mechanistically like everything else.” In the way they relate to things, the notion of agency cannot be demonstrated and is incoherent.
Only by relating to someone at as an agent do I come to the understanding that he actually exists as an agent. You demonstrate X at the level of relation at which X is defined. But since the Martians don’t relate to the world in such a way that involves agency, I cannot demonstrate it to them. Still, this does not mean that it is irrational for us to relate to the world in a way that does. Do we need to demonstrate to the Martians that agents are “real” in order to justify our belief in them? No, all we need is to note that relating to the world on the level of agency is completely self-consistent, and insofar as we see it as an emergent relationship, it contradicts nothing we already know about the world. We are then justified to interact with the world in such a way, realizing that it is impossible to demonstrate our way of interacting to someone like a Martian who relates to the world on a lower level.
The incompatabilist may keep insisting for demonstration, but once something emerges, there is no further need for demonstration because it is simply there. That is why the idea of demonstration in this whole issue of agency is horribly misleading. It can be pointed out with a joke:
Skeptic: You can’t demonstrate agency because there is no such thing as agency, only neurons and chemicals.
Believer: Wait, who are you arguing with again?
Skeptic: I’m arguing with YOU! Pay attention!
Believer: Ah, thought so. Well there’s my argument.
Because the Skeptic is already on the level of relation that involves agency nothing needs to be demonstrated to him. I can demonstrate the patterns in the world, and the mental faculties of human beings which may give rise to the unification of things in a certain emergent relation, but demonstrating the relation itself makes no sense.
Before we understood the notion of emergence, the only way we could talk intelligibly about persons as free agents to come up with this notion of a “soul.” Libertarians mostly stick with this non-emergent concept but the language simply becomes more scientific; “soul” becomes an indeterministic agent which directly causes things. Otherwise, we insist, all there would be is a mesh of neurophysiological reactions; we need something person-like, something extra to account for it, and so we come up with the notion of dualistic souls. The fact that people are emergent entities that exist on a different level of relation as bodies, rather than being some supernatural entity on the same level of relation as bodies does nothing to devalue the notion that people exist. The realization of the notion of agents is more fundamental, and we try to seek out agency on a level in which it is not a fundamental concept and so we try to make it a fundamental concept on this level of relating; this leads to panicky metaphysics of libertarianism. The existence of agency on a higher level of relation is behind all thought of indeterministic agents.
The problem of free will libertarians is that they are looking for evidence, and in looking for evidence, they are looking for evidence on a lower level of relation; to find evidence of here is impossible.
Still, it seems there may be a concern about which type of explanation is primary; one in terms of mechanism, and one in terms of desires and intentions. If the latter arises out of the former, then it seems that intentions and desires are not the most fundamental type of explanations, and are in fact caused by mechanistic ones. In one way, this is the case, but not in the way that is relevant to freedom. In talk of a lower level process causing a higher one, it is not that the higher level explanation is now less valid nor does it mean that we should consider the lower one as primary while relating on a level that the higher explanation fits to. In Conway’s Game of Life, it does not make sense to talk about behavior on the lowest level when describing a scene. We can talk about gliders on one level, or on even a higher level we can talk about a glider-gun-breeder as a singular thing moving beyond individual glider, using the behavior of these things as causal explanations and not merely the simplest rules of the game.
Even though agent causal explanations arise out of a lower level that lacks agency, it does not mean that the explanations in terms of agents lose any validity. If we are relating to the world in terms of agents, it only makes sense to use agent-causal explanations. An analogy that can make this explicit is explanations in terms of life, which emerges from non-living things. If one was to ask “why did Abraham Lincoln’s presidential term end?” the proper causal explanation for the level of the question would be in terms of life, not in terms of non-living atoms and molecules. We would say that his term ended because his life ended. Sure, the notions of life and death arise from matter that is ultimately all non-living, but this does not mean that the “real” explanation for the cause of the end of Lincoln’s presidency would be in terms of non-living things. In the way of relating in which the question is asked, it only makes sense to talk in terms of life and death.
Similarly, when asked why I am choosing to see a certain movie over a play, my causal explanation will be in terms of intentions beliefs and desires, which emerge at a certain level of conscious life. Lower level explanations will miss something significant and not do the question justice. If my explanation is in terms of atoms and molecules I will have missed the point of the question entirely. Perhaps I will have given sufficient explanation for why the series of events takes place as it does, but it is a rather meaningless explanation in the context of the question asked. The same can be said, for an explanation in terms of life, but not agency. Saying “the brain of the organism, because of its makeup responds this way to this set of stimuli” does not answer the question properly either. But if I say, “I am choosing to see the movie, because I have been very excited to see it since the trailer came out, and frankly, plays usually boor me” my explanation gives the correct causal account for the question asked.
Perhaps if the question is asked in a lab by a physicist or a neuroscientist, we could get different lower-level explanations, but since agency is already a fundamental component to our way of interacting with the world, most of our everyday explanations for why people do things will be in terms of agents because that is the context in which a question will be posed.
Often in thinking of free will, philosophers will respond to a question in the wrong framework which gives a misleading answer. When asked if one freely performed an action, he may respond, “of course not, I cannot freely perform an action at all; there is only neurons firing and chemical reactions.” If one accepts this type of response, perhaps he should also consider John Wilkes Booth’s response: “Of course, I didn’t kill Lincoln! Lincoln was never alive, because there is no such thing as life, only non-living atoms, quarks, protons and neutrons. So how could I possibly kill him?” This is a misleading trick, an equivocation fallacy of sorts, of jumping down to a lower level of relation in response to a question that is on a higher level. It gives a seemingly real answer to the question which, as the Lincoln example shows, is not a real answer at all.
In asking questions of free will, we want real answers, ones that remain in the same level of relation as the question is asked. Such real answers to questions may conform to a module such as an account of free will based on values, but answers which eliminate agency entirely in their response do not actually answer the question as it is asked. They seem to, but they do no such thing.