Friday, January 25, 2013

More on Churchland

Continuing from the previous post, in this article she also discusses free will in the context of self control. It is an adaptive mechanism not limited to humans. She said:

“When the very abstract question of free will is put in this context, I am no longer sure exactly what the question is. If it means can we have self-control, then obviously the answer is yes. If it means can we create a choice with no causal antedecent, in all probability the answer is no.”

I’m guessing there are those who make such a claim for the second type of free will, but I am not one of them. Hence I’m referring to it as a means of self control and executive or top-down decision making. She further notes that executive control is also a key part of social mores.

In this article on conscious and unconscious control she reiterates what Eagleman said in Incognito, that quite a bit of control is handled unconsciously. Like him she gave the example of skills once learned, how they then operate automatically. What she didn’t discuss was that those skills required conscious control in the learning process. The same goes for learning social mores or learning anything, for that matter. And that when we must learn new data the conscious control aspect again comes to the fore.

In the last section of the article (p. 346) she said:

“To be clear, we are not advancing the radical thesis that there is no such thing as consciousness or conscious control. The main point of this article is rather that although consciousness – for instance of goals and what the neo-Kantian would call ‘reasons’ – does sometimes have an important role in control, it is not required for  control. Nonconscious control can be – and frequently is – exercised, and this control can be
every bit as genuine as the conscious variety.”

Of course it can, but she doesn’t focus on the conscious aspects of control because it is not part of her agenda. She then admits that research to date is scant on the interaction of conscious and unconscious control, and conscious control is not well understood. Well duh, no wonder if neuroscience researchers are agog with unconsciousness and too busy debunking consciousness. The good news is that there is an entire field called decision neuroscience that is examining the basis of conscious control. Yet we don’t hear that much about it, if at all, from the eliminative reductionist circles.

As to the issue of emergence, in this article she describes it as follows:

By emergent property, I do not mean anything spooky or metaphysical. I merely mean that the property is a function of both the intrinsic properties of neurons in the network and the dynamics of their interactions. I mean it is a network property. The network provides the neural mechanism whereby the phenomenon is produced” (108).

Which is how I’ve been using it term, not as some disembodied Platonic ideal. Anyone who has read this blog at all knows I’ve railed against such idealism from the get go. Earlier in the article she notes that morality itself emerges from earlier structures but was an advance into network properties beyond the earlier structures (96).


  1. I don't disagree much with what Churchland says. But note, that if your definition of free will involves top down causation, than that definition is pretty watered-down. Even machines exert top down causation, let alone, as Churchland notes, non-human animals (and as I noted in another comment, cells). Do machines have free will? Do cells? If your answer is yes, then by that definition of free will, humans do, too. But if machines do not have free will, nor cells, nor primitive organisms, I'm not sure how you make the argument that humans do.

    Human behavior is more complex than that of any of these other entities, but it still involves causal chains. These causal chains are frequently too complex to predict outcomes, and it may even be that some outcomes are inherently unpredictable, or non-deterministic. But that does not mean they are the product of free will. Top down actions, fine, but again, what is it about human actions that is different in kind from that of a machine? According to the current scientific worldview, which Churchland supports, nothing.

    I don't have a problem with Churchland's point that we need a sense of responsibility in order to maintain social stability. But what this really boils down to is a standard of acceptable behavior, free or not. If someone with a brain tumor repeatedly commits murder, we might say that he can't help himself, we might say he can't take responsibility for his actions, but we still would lock him up. Indeed, most people believe that wild animals can't help attacking humans in certain situations, but we don't have a problem killing them when they threaten humans. So the question of individual responsibility really does not preclude punishment for certain forms of behavior.

    Let's also not forget that another rationale underlying punishment is deterrent. We punish certain forms of behavior to make them less likely. The notion of a deterrent makes much more sense if behavior is not free. if an individual could chooses his behavior freely, why would he be swayed one way or another by a deterrent? He could choose to be or not to be. But if our behavior is the outcome of a large number of factors, one of which is the fear of consequences, then deterrents make good sense.

    But despite all these philosophical maneuvers, at the end of the day, we have to ask ourselves, does the person we call "I"--in other words, the brutal reality of our everyday lives--have free choice? I don't see how anyone who observes his behavior carefully can possibly believe this. I have learned again and again and again that i do not have free choice, that everything I do is the outcome of certain factors that "I" have no control over.

  2. I think discussion of Libet is a red herring. Both supporters and critics of his findings—or of the newer research—are missing the point. Consciousness, according to the scientific view, is part of a causal network, so its temporal relationship to some behavioral act is irrelevant.

    Suppose someone shows that the conscious intent to make a movement precedes the movement, precedes activation of the neural pathway that results in movement. Does this show that consciousness caused the movement? One could say that, but then one has to ask, what caused the conscious intent? And the answer, according to science, is other neural activity in the brain. That activity, in turn, was probably caused by events in the outer world. The idea that human consciousness is isolated from the rest of the world is of course long out-dated.

    This is why the whole idea of free will—in any meaningful sense—is incoherent. If everything is cause and effect, how can we say that free will exists in one cause and not in another? Even the highest level of some system is subject to forces from other systems outside of it.

    I think Churchland understands this. She wants to address the problem of individual responsibility, because she thinks without it, society is likely to collapse. I think she’s probably right about that, at least at this point in our social evolution. We have to act as if we, and those around us, are responsible for our actions. But this is just a useful fiction. We have lots of them in our lives. Physics tells us that matter is not solid, but we treat it as though it were. Physics also tells us that the entire world around us is moving tens of thousands of miles per hour in space, but we act as though it’s stable. Science has taught us that many aspects of our existence are not as they superficially appear to us, and one of those is our sense that we are in control.

    What we can say is that as life evolves, it becomes increasingly free from lower level causality (Dennett’s Freedom Evolves). The molecules in our body, unlike free molecules in the air, do not diffuse away from each other. They do not start moving faster in response to heat, and so on. So, yes, we are free to some extent from the physical world. But that just means we are subject to higher order forms of causality.

    You could interpret this, as some have, as saying free will is not either-or, but a matter of degree. But if you want to take that approach, you have to say that even machines have some degree of free will, and very low forms of life even more. As I’ve said before, I think “freedom from” is a much better way to describe this than “freedom to”.

  3. Forget the expression "free will," as it has too much baggage from idealists and I don't subscribe to that. Let's stick with Churchland's recontextualization of control, or top-down causation as I'm calling it. Indeed, other entities express it, even chemicals to some degree. But as you admit, with humans it's on an entirely different, i.e., more complex level, so the degree of choice is much higher.

    Hence control is of course not "isolated from the rest of the world" and I never made that claim. Again, to the contrary, this is why I cite Lakoff's work in embodied realism. But emergence does indeed allow the human brain-mind to exert much more executive control over lower functions. Yes, it is a causal chain, but one that is sometimes (not always) top-down and these causes are not only not just the result of lower causes but could never arise just from the lower causes. It is the network, or the interactions of the lower causes as Churchland describes, that cause the top-down control. And you agree with that, that there are higher orders of causality.

    Even Churchland above said she is not arguing against conscious control, including goals and reasons. Which is pretty much all I'm saying. And that we can to some extent control our lower natures via methods like meditation or biofeedback or even rationality. And that these emergences, while being physically constrained by lower causes, are not completely determined by them. Again, as you admit, we are "increasingly free from lower level causality." So I can accept your "freedom from" definition.

    It comes down to what can we control, to what degree, and does it matter? I say it matters quite a bit. And it makes a huge difference in whether our species, and many others, survive climate change. We need to decide to fix it and do what we can within our power or control or its game over. Maybe not for some very low forms of life like viruses, or their close cousins Republicans, but accepting what control we have brings with it a responsibility, one we must accept and get busy with fixing the monster we created in climate change.

  4. As but one example of that control, and another reason I cite Lakoff, is that language framing is a very important means of shaping public opinion to garner enough support to enact policy agendas like climate change. He is using the neuroscience that tells us we can control our lesser urges with hour higher angels, to some degree, via the use of framing, a conscious method that acts on unconscious processes. Obama's campaign took his advice and won re-election. And overall these 'liberal' (progressive)values won the day in opinion polls, hence we get regressives (my word for Republicans)caving on increasing taxes on the rich, or no longer holding the debt ceiling hostage, etc. Framing is a very conscious method of top-down control with real-world effects.


Note: Only a member of this blog may post a comment.