Sunday, January 30, 2011

Once more on "evolutionary" psychology

This is a kind of follow-up on two recent posts (here and here) on "sociobiology" and "evolutionary" psychology.

It's obvious to all (and to scientists in particular) that human beings learn from science and act on the basis of scientific knowledge all the time. That is to say, we rationally regulate our conduct on the basis of beliefs grounded in scientific knowledge (e.g. I don't bring a plugged-in toaster in the shower with me because I have the belief that it would electrocute me, a belief that is based on accumulated scientific knowledge about electricity, etc.).

This illustrates the obvious point that we act on the basis of beliefs we hold. The set of beliefs I hold forms the framework within which I act. Beliefs may be considered separately from actions, but actions (in order to be such) cannot ever be made sense of apart from beliefs. For something to be an action at all, apart from a mere event (e.g. physical movement), it must express some belief we hold. It's worth saying a bit more about the distinction between a physical movement and an action in order to drive the point home. As Alasdair MacIntyre puts it:
The distinction is obvious if we consider the criteria of identity in each case. The same physical movements may constitute in different contexts quite different actions. So a man may go through the same physical movements involved in signaling his name and be concluding a treaty or paying a bill, which are quite different actions. But is not this man performing the same action in each case, namely signing his name? To this the answer is that writing one's name is never merely by itself an action; one is either signing a document or giving information or perhaps just doodling. All these are actions, but writing one's hand is not. Equally, the same action may be constituted by quite different physical movements. Writing on paper, passing coin, even saying the words may all constitute the same action of paying a bill.

When we talk about explaining human behavior, we sometimes blur this distinction. Because there is no human action that does not involve physical movement we may suppose that to explain the movement is to explain the action... My head nods and I am asked, "why did you nod your head?" If I answer by referring to a nervous tick, I point in the direction of a story about necessary and sufficient antecedent conditions, a story about nerves and muscles and possibly about conditioning in early childhood... But to explain the nod as a nervous tick is precisely to explain it not as something that I do, but as something that just happens. If on the other hand I explain the nod by saying that I had been asked a question and was answering "Yes", then I certain explain the nod as an action. I do so by pointing to the purpose which it serves. This reference to an agent's purpose always carries with it implicitly and often explicitly a reference to some antecedent event or condition... But if I explain my nod of the head as an answer to an antecedent question, I certainly do not refer to a generalization stating some constant conjunction which holds between questions and answers. From the fact that the question has been asked it never follows that it will be given some particular answer or indeed any answer. What makes a reference to antecedent questions part of an explanation of what one was doing is that one is explaining the role of one's nod in carrying through a socially established and recognized practice, that of asking and answering questions. That is, the background to my explanation is formed by the customarily recognized rules of a particular social order". (See MacIntyre (1969: p.56) in Laslett et al. Philosophy, Politics and Society, Oxford: Blackwell)
Something will not have been an action unless some agent undertook it and did so on the basis of a reason. Actions are those "distinct sorts of events for which we may appropriately demand reasons or justifications from subjects whom we take to be responsible for such events occurring."

As MacIntyre makes clear, it is quite obviously true that the beliefs that I am liable to hold are strongly influenced by the available concepts in the society into which I am thrown at birth. I couldn't reasonably have formed beliefs about electricity in the Bronze Age, say, since electricity presupposes all sorts of bundles of complex knowledge and concepts that did not exist at the time. If you want to understand and explain why people act the way that they do, you cannot but include in your explanation the concepts, dominant ideas, theories, culture, etc. that are available to the agent. Without taking such things seriously, you cannot explain human behavior at all. Again, as MacIntyre describes it:
Suppose that a team of Martian social scientists is observing human behavior. What they are watching we should describe as chess-playing, but unhappily they lack the concept of a game. Mars is gameless. They therefore do not discern the rule-governed character of the players' behavior, although they arrive at many statistical generalizations about the movement of small pieces of wood by human beings. What is it that they do no understand when they fail to understand these movements as a game of chess?

They fail of course first of all to grasp the players' action as distinct from their physical movements. It is not that they wrongly explain what is done; rather they fail to identify the actions which are to be explained.... The analogy with a social system is clear. To explain actions within it we have to identify the rules and their connexion with reasonable or unreasonable, true or false beliefs. Thus we cannot explain actions by means of beliefs and not raise questions of truth and falsity, reason or unreason... Thus in any society we shall only be able to identify what is going on if we have identified and assessed the established methods of reasoning and criticism in that society... Explaining actions is explaining choices, and explaining choices is exhibiting why certain criteria define rational behavior for a given society.
So, let us compare all of this with the model of explanation adopted by "evolutionary" psychologists (I use scare quotes because I doubt that such a research program has any serious scientific grounding in the theory of evolution itself).

For these folks, human behavior is explained in mechanistic terms. The lines between beliefs, purposes, reasons on the one hand, and action on the other, are cut. Actions are understood in the same way that mere physical events are understood.

Here's what I mean. Think of the way that a physicist explains events. They observe a lot of events, note the constant conjunction of what appear to be cause and effect, and then, by way of an inductive inference, hypothesize about general laws in light of which such events are to be understood. We have no reason to think that the conditions in which such laws are valid are subject to change anytime soon.

Now, what the sociobiologist does is assume that human actions are just like physical events. They conflate physical movement with action. Thus they can say that there are general laws, like those governing the behavior of gases, that can explain human behavior. They assume that actions are mere events which can be explained by reference to general laws inferred from the constant conjunction of certain variables.

So, for instance, they see, say, men in the United States behave in a certain way (e.g. hit on women at bars) and then say: "might there not be some invariant general law that explains this behavior?". They assume that human beings are mere objects that are caused ("programmed", "hardwired") to act in this way: the only question then becomes what the natural law is that appropriately generalizes from this cause and effect relationship. Finally, they go all-in for the view that the content of such general laws are wholly (or largely) determined by an individual's genetic makeup. The line of causation from genetic makeup to behavior, they think, is one that is opaque to us from the first-person perspective. There's nothing internal here at all: the causal story here has nothing to do with what it's like to be a human agent from the inside. It is purely external, in the same way that we externally explain and observe the behavior of gases.

So there are at least three components to their view: (1) the mechanistic view of human action which says that there are general laws that explain it (analogous to ideal gas laws), and (2) the view that the general laws boil down to facts about genes, and (3) the view that the causal mechanism from genes to behavior is non-rational and external to first-person reasons, intentions, etc.

Let me say why I think (1) is false. First of all, it is absurd to say that we can understand human actions without reference to belief. We cannot even identify an action as such without at least in principle presupposing that some belief is being expressed by it (the MacIntyre examples above make this point clear). The social sciences (or the "human sciences") must therefore study beliefs in order to study human action. But, and this brings me to my second point, studying the beliefs of an agent means studying the available concepts, theories, socially recognized practices, norms etc. in the society in which the agent is located.

But what is a social scientific theory? It is a collection of concepts, beliefs about humans, etc. Sophisticated social theories are continuous with our "naive" beliefs about the society we live in that guide our actions. So this means that social science is part of its own object of study, it is a self-reflexive undertaking. This is not so with physical science. It is not the case that our beliefs, concepts, scientific methods, norms and justification practices are part of the object of study of physical science itself. Such things are external to the object of study of physical science: physical science does not study it's own methods and practices as physical objects: it is not the case that physical science itself is part of the object of study of physical science.

So, physical science is not self-reflexive in the way that human sciences are, and that's fine given its object of study. Raymond Geuss is excellent on this point:
"To treat a belief as if it were about a mere object in nature implies both something about how one can and should investigate it, and about how one can use it to argue, evaluate, and guide action. An inert external object can sometimes be manipulated if I have sufficient knowledge of it. If I discover that the unwieldy sofa has a handle on the other side, I can perhaps grip and shift it, whereas I could not before. In doing this, of course, I need take no account of the beliefs or preferences of the object- it has none... Objectifying thought is especially prone, however, to false universalization for the obvious reason that it will encourage investigators to construe local phenomena as universal... If I think it will rain tomorrow, my belief and the weather are two distinct states of affairs. The weather is what it is regardless of what I might think. My belief itself is completely "external" to it and would not appropriately figure as part of the meteorological report. It is perfectly natural for us to adopt an objectifying attitude towards the nonhuman world of nature. The question is whether it is also appropriate to adopt such an attitude toward a society of which I myself am actually a member... But the beliefs and attitudes people in the society have about themselves and their society are themselves an integral part of the society. If everyone in a society, say, early twenty-first-century Britain thinks that people are universally selfish, then that belief is reflective in a way that my belief about the weather is not. Since it is a belief about people in general, it includes the members of the society in question, and holding it will have an effect on that society."
This is all just to say that beliefs are part of the object of study of the human sciences, and therefore the social/human sciences are self-reflexive. Social theories are continuous with their object of study, and, interestingly, they can therefore play a role in their own falsification and confirmation. If propagating a certain theory causes people to have different beliefs and thus to act in radically different ways, then the data which could confirm or disconfirm this very theory have themselves changed radically. This is not true of the physical sciences. So much for the assumption that human behavior could be studied mechanistically, on the model of the physical sciences. So much for the assumption that human behavior could be explained by reference to invariant physical laws. Human behavior is bound up with beliefs, and the possible beliefs an agent could hold must be explained by reference to a certain social order. And as everyone knows, social orders change and develop over time. Thus, human behavior must be explained and understood in historical context. Throw in the premise that history is political (all history is the history of class struggle), and you have thus reached the conclusion that human behavior is dynamic, plastic and must be understood relative to the configuration of power in that society, the level of material development, etc. This does not mean that human behavior is determined by social/political causes rather than physical ones. It only means that behavior is dynamic, liable to change, and develops in relation to a determinate context.

All of this should make clear how absurd it would therefore be to suppose that extant social norms, practices, and human actions could be explained by a collection of facts about our genes. At best, our genes are but one variable in a very complex field. There is no scientific warrant for singling out one such variable, genetic makeup, and supposing that it trumps all others.

The reasons to reject (3) are largely supplied above by the MacIntyre quote. It is patently absurd to say that we could explain actions without reference to beliefs and reasons. That is to say, it's patently absurd to say that we could explain human action without reference to what it's like to be a human agent from the inside. We are the kinds of beings that have identity crises, who are not sure what we should do with our lives, who struggle with hard decisions as to what the right thing to do is. That is because we act on the basis of reasons. We think that the question "what should I be when I grow up?" is a real question, and a genuinely hard one at that. Nobody thinks that "do what will best pass on your genes" is a plausible answer to that hard question. No parent would suggest such a thing to their child. From the inside, internally, we take ourselves to act for all sorts of reasons, the vast majority of which are not explainable in terms of the external function of "passing on one's genes". What the fuck do I care about my own genes anyway? I care about my life going well. I care about love, relationships and people. I care about justice, about fighting oppression, about human emancipation and freedom. It's highly possible that caring about these things puts me at odds with the goal of passing on my genes most effectively. I care about all kinds of things that don't seem at all to me like mere epiphenomena, mere window-dressing for the "unconscious desire" to pass my genes on.

When someone unfairly wrongs me, or breaches norms governing our relationship, I don't think it would be acceptable for them to say "but I'm just passing on my genes" or "I'm just allowing my genetic makeup to manifest itself behaviorally unbenknownst to me". I would hold such a person responsible for what she did. I would resent them in a way that I would not resent a mere object.

I think we should be more than a little suspicious of a theory of human action which is so radically at odds with everything we, as human beings, think about human action. At the very least, the burden of proof on such a theory should be extremely high, given that it should cause us to doubt and revise the vast majority of what it is we think that we're doing when we act.


-sf said...

Stuff like this geeks me out, but will likely "cause us to doubt and revise the vast majority of what it is we think that we're doing when we act."

My favorite part was when experimenters used magnetic pulses to "turn down" a specific part of the brain and found that they could temporarily alter people's moral appraisal of others' actions. Since I'm a proponent of honest debate, I will say it is possible that their manipulations could simply divorce a person's beliefs with the physical movement of pressing a button, but there are easy controls for that kind of thing, and its likely they did them.

A few take-away quotes from Neil DeGrasse Tyson's concluding statements:
"We need the tools and methods of science to shield us from bias, blunder, and delusion... What are thoughts but electrical impulses among brain cells? What are ideas but novel firings of those cells?"

If the way we view and interpret our world with its stimuli, ideas, and beliefs is through our brain, which is simply an electrical circuit subject to physical laws, why is it so hard to believe that perhaps our conceptions of the world around us is also governed by physical laws?

t said...

Let's take the following claim that you mention: "What are thoughts but electrical impulses among brain cells? What are ideas but novel firings of those cells?"

Perhaps this is true, but it is far from obvious that this isn't just a category mistake (a category mistake is philosophical example of such a mistake would be to ask "what color is three?").

Now, everybody agrees that thoughts require brain cells, electrical impulses, and all the rest of it. But what's not obvious is that thoughts are simply reducible to, and nothing but, electrical impulses, etc.

For instance, I may require certain physical materials to produce a painting (e.g. oil paints, a canvas, brushes, etc.) but it's not obvious that the meaning of the painting is well-explained by an analysis of the materials required to produce it. Simply because one thing is necessary to produce something else, it hardly follows that this necessary component exhausts and throughly explains this something else.

Also, it is not clear that we are actually explaining consciousness itself when we reduce it to talk about brain cells, neurons, etc. To be sure, the talk about brain cells, neurons, etc. is important and indeed such things are conditions of the possibility of being conscious at all (i.e. no brain, no consciousness). But it's not clear that consciousness itself is captured by such analyses. Some have wanted to say that there's a qualitative "what it's like" character to being a conscious agent that isn't captured by analysis of the physical (or, if you like, chemical/biological) composition of our brains, though it is enabled by it. The example is the qualitative first-person, felt experience of being a bat. The claim is that we could learn everything there is to learn about the neural structure of bat's brains, and this is important... but such facts do not capture what it's actually like to be a bat. That's something else, and the study of it requires different concepts than those required for the study of the physical composition of the brain.

There's plenty more to say here, and I hope I haven't given the impression that philosophers all agree here (some are naturalists and are sympathetic to what Tyson is saying, though I think they are making serious errors in doing so).

Anonymous said...

t is a modern day Lysenko