Addiction is a Choice

Today in Philosophical Foundations of Cognitive Science we had an interesting discussion about Dennett's response to Functionalism - and Stich's response to Dennett. I won't bore you with the details just yet (it's coming, though) - but I did want to complain a bit about a side argument that I had with two others about the nature of addiction.

Stitch's response (ok, well, but it's relevant - only the barest skeleton, I promise!) to Dennett relies in part on questioning whether people really have intentionality - that is, do they really act rationally based on internal motivations like beliefs and desires, or is this just a convenient shorthand for some poorly understood brain mechanisms. (It's important to understand here that Dennett's response isn't that beliefs are ultimately a product of neurons firing - the Funcationalists believe that too. His response is to question whether "belief" is not a kind of illusion - at the very least an inaccurate description of the brain process that actually goes on.) The basis for Stich's question is that we frequently observe people doing things which seem to be irrational. Stich uses the example of eating chocolate - which seems to offer small payoff in terms of pleasure (which is momentary, fleeting) weighed against the negative effect it has on one's health. And indeed, from some purely rational point of view, it doesn't make much sense to eat chocolate.

But I think this is a misunderstanding of what it means to make a conscious, desire-driven choice. Kant put it better. Kant's formulation of this situation (admittedly from an ethical rather than a philosophy of mind point of view) is summed up in the Hypothetical Imperative. Put informally, the idea is that actions have meanings and should be evaluated in accordance with those meanings. It's a kind of ethical decision whether or not to eat chocolate, but this decision is left up to the individual - you and I have no grounds for coercing the person to behave one way or the other. (This is as distinct from situations involving the Categorical Imperative, where we can require that other people do what we want - for example, we can tell people that they may not murder anyone.) To use the chocolate example - what the Hypothetical Imperative says is that eating chocolate means in some sense that the person who eats it values the pleasure it gives him more than he values health and long life.

It seems to me that there is no real cogent argument against this. Now, one can well say that we sometimes don't think things through when we act - but Kant would say (and I agree with him) that that isn't really at issue. People should think things through - and they are at least capable of doing so, whether they choose to or not.

So, that's more or less the position I argued from. Where Stich gets it wrong isn't in saying that people do things that aren't, strictly speaking, rational. What's wrong about it is his assumption that this means that "decision" is a bad word for what they do. In fact, there is a decision involved - it just doesn't seem (from an outsider's point of view) to be the proper one.

I brought up the example of a smoker (which I think is a better one). Let's say someone is highly addicted to cigarettes to the point where he believes it is impossible for him to quit. It seems to me that even a heavily-addicted smoker has a choice about whether or not he smokes - just that his choice is heavily weighted by physical factors toward smoking. He gets cravings, etc. that I don't have, and that makes it more comfortable for him to smoke than not. For those of us not addicted, it's easier to decide not to smoke because we don't have any strong physical motivation to do so. There are no pains that can be relieved by doing it - only (minor) pleasure to be gained (and indeed, some people hate the feeling of inhaling smoke so much that it's a no-brainer for them to choose health).

So there is a choice, just a heavily weighted one.

These two other students in the class, however, kept wanting to say that an addiction is something else - that it is, by definition, an involuntary thing. But I don't see how this could be. I mean, if nicotine addiction means that one involuntarily obtains nicotine, then how do we explain the series of conscious choices that leads them to go to a convenience store, ask for cigarettes, pay for them, open the wrapper, pull out a smoke, light it up, and inhale? In an aboslute sense, there's a lot of avoidable effort involved in getting your fix! Now, to be fair, I guess we could say that people frequently are at the store anyway - but they still have to go to the counter and ask for a pack, pay for it, open it, etc. Clearly, there are choices involved. I do not think anyone seriously believes that such people are in a zombie-like state when they perform these actions.

Now one of the two students made the point that he has often managed to buy food at Subway while semi-conscious on alcohol. And that's a decent comeback, I guess, but I think it has two problems. First of all, although it's true that alcohol affects your judgment, it's really the associated memory-loss/dampening that leads him to believe he went to Subway in a zombie-like state. He knew what he was doing at the time (albeit in a severely diminished capacity), he just doesn't clearly remember having done it now. The other problem is that be that as it may - even if we grant that people under the influence of alcohol can't reason properly (and I know for a fact that they can't - I have had rather a LOT to drink over the course of my life and have done some pretty embarassing things under the influence!), I don't think anyone believes that this is true of people who buy cigarettes. No one believes that the smoker who goes into the convenience store is in an addiction-induced stupor when he does this, so the example is ultimately irrelevant.

No - it's a choice. It has to be. What makes it an "addiction" is the sheer weight of the physical motivation to make the choice. I agree that smokers are not thinking clearly when they decide to indulge their addictions - but it isn't that they aren't (at least capable of) reasoning about what they're doing.

And so, in some sense, it is indeed an "intensional" and "rational" decision when a smoker gives into a nicfit. What he's saying, in terms of Kant's hypothetical imperative, is "making this unpleasant feeling go away is more important to me than avoiding possible health consequences 20 years down the road." It may not be an advisable or easy decision, but a decision it nonetheless is.