Agents, Cognition and Free Will

For what it’s worth, my intuition :) is that our intuitions about free will are not reliable. Whenever I think about free will, I start by trying to define the agent, paring down nonessentials. I’m left with nothing. On other days, I add parts to the agent which play integral roles in influencing choices made. I end up with the whole shebang (i.e. a causal chain back to the Big Bang).

Most discussions of free will leave me unsatisfied because they fail to address who is supposedly making choices. Without that, the rest of the discussion makes no sense.

The more I think about it, free will seems like a human invention which doesn’t correspond to anything real (outside of our own intuitions, anyway). It also seems like people are so concerned about free will because their particular religious notions hinge on the idea of moral culpability. It seems like their religion prejudices the question for them.

Jonathan,

It also seems like people are so concerned about free will because their particular religious notions hinge on the idea of moral culpability. It seems like their religion prejudices the question for them.

If moral culpability were simply a religious commitment this would make a better argument. It seem, rather, that people are generally committed to moral culpability even when they are not religious.

I would expect us to personify more things. Maybe you will argue that this is just what people did more of in the past. What do you think?

I think it unarguably that people personify a lot of things – often inappropriately. The literature on cognitive science and religion give numerous examples. I’d suggest checking out Atran’s book if you haven’t yet. It’s one of the better ones on the subject. I’d discussed it a lot a few years ago on the old blog (such as here)

Jeff, whether we have reasons to believe ourselves an agent I think it undeniable that we do consider ourselves agents. Even folks who don’t think we’re really agents probably are engaging in what Peirce calls paper doubt. That is they write down that they don’t believe but probably function cognitively as if they did believe.

The problem with the robot example is, of course, that it’s debatable whether a robot could be so programmed. (The classic problem of AI)

We say it’s self-evident to oneself that one thinks. If one does seem to doubt that, one would rather seem to be doubting what words like “one” and “thinks” mean; but is there much doubt about the meanings of those words? One is oneself, which is inevitably a mysterious thing (no less than, for example, time, or even space, and there is no doubt that some things exist spatiotemporally) but is straightforwardly referred to, if anything is, surely; and you’ve clearly been thinking if you’ve read this far! So, if the brain is the medium for the mind, then we would expect it to contain systems for identifying agents, and what the mind would perceive (what a robot could not) is a posited subject, like oneself (in many ways… we could describe those ways, e.g. sentience and sapience, broadly, and so on more narrowly, but surely the complexities of such analyses (always obscure just beyond the cutting-edge) cannot undermine our trust in what gave the basic meaning to those words; no more than a robot’s ability to mimic sentience or sapience (or more likely some aspects of those at the cutting-edge of their analyses) could make it sentient or sapient)… is what I think (sort of); sorry if that doesn’t help (:s).

If moral culpability were simply a religious commitment this would make a better argument. It seem, rather, that people are generally committed to moral culpability even when they are not religious.

Point taken, though I wonder how much our secular ethics has been shaped by our religious history. Free will may not have been much of an ethical concern given a different religious history. I’m not saying, however, that religion is the sole source of our idea of free will.

I think there’s another part to it that comes from similar research: the mechanisms by which we detect our own agency. These mechanisms are designed to tease apart internal and external causes of events, and as a result, I think, we tend to see our own actions as entirely separate. That is to say, we see them as “free.” I’m thinking here of work by Dan Wegner, which we’ve talked about over at Mixing Memory a few times over the years.

Clark, I haven’t read Atran, thanks for the suggestion.

Blake Ostler gave a presentation on Spiritual Experiences as the Basis for Belief and Commitment which does touch on the heart of epistemology and the foundations of intuition.

“Now I ask again, can humans really know anything? Does the experience come from God, or do we merely interpret it to be experienced as coming from God? I’m going to deal with the strongest arguments that I know.”

“Now I’m going to make a suggestion. As counter-intuitive as it may seem, even our most rigorous reasoning rests, at bottom, upon human intuition. For reasoning, including formal logic and mathematics, cannot proceed without at least some basic axioms and assumptions: derivation procedures, formation transference rules, and rules of inference that cannot be proven by the same rigorous reasoning. These facets of reasoning can be justified only as basic givens which have the property of just seeming right or that lead to explanation and consistency of a range of experience and data.”

Clark, it seems to me the issue is not one of Free Will or Determinism but rather are we moral agents as shown by our moral behavior? And culpability depends on the ability to change. Our moral behavior is based on affective memory. How we differ from other animals is we can project alternative futures from our affectively established memory for the basis of future moral behavior. They cannot. This ability can be constrained based on the affective impact of particular memories. Fear established memory is almost impossible to extinguish. The challenge of that memory and its associated behavior will exist throughout that person’s life. This doesn’t mean the person is doomed to repeat that behavior. With enough desire (an affection) new workaround memories can be established and alternative future behaviors can be made available when it comes time to choose an alternative future.

Sociopaths, for example don’t seem to be able to do this. First of all they lack the ability to judge their own behavior in a social environment. This is simply another way of saying they lack empathy. Empathy is the ability to project other feelings as your own. Empathy is like a feedback loop which allows us to judge our actions in terms of what how we feel the other person feels. Mirror neurons seem to be at the center of this ability. Nevertheless, empathy is a trained ability. Given cold distant care givers, the ability to empathize never grows. This is called Associative Disorder, the extreme wing of which is Sociopathy. (All this is predicated upon a normal brain.)

Are we moral agents. We would have to say yes, under normal circumstances. That is, if we grow up in a nurturing environment with nurturing care givers and a normal brain so that we can establish new behavior through the development of new affectively based memories. With a cold non-nurturing environment it becomes more difficult to establish affectively based alternative futures, a prerequisite for change. By the way, reason, in this process, is the means by which we evaluate and choose alternative affectively based futures.

Rich

My question, which I didn’t state very well, is what question does Free Will (however you understand that term) and Determinism attempt to answer. I was under the impression that that it was the question of agency. I attempted to show in the case of moral agency we have no need of the concepts of Free Will and Determinism.

If what I sketched out is correct, why do we need the concepts of Free Will and Determinism. What question do they attempt to answer?

Rich

When I first started I was responding to the issue of free will. But as I thought more about the issue of free will, determinacy and agency I find myself answering the issue of agency without concepts such as free will and determinacy. For well over 2,000 years this argument has gone on with no real solution in sight. Maybe these concepts don’t have anything to contribute to the issue agency. So what I’m now asking is what issues do these concepts attempt to address.

Rich