Persons, animals, and machines.(Humans, Animals, and Machines)

Animals and Machines

Animals aren't machines. Well, they are. But they aren't just machines.

That is to say, if you insist on thinking about an animal as a mere

machine, then you won't do justice to animals.

When you ask how an animal produces a certain kind of behavioral repertoire, you might think that all that needs to be said can be said by referring to the operation of its inner mechanism. And, indeed, treating an animal simply as a complex mechanism can seem, and historically has seemed, very attractive.(1) But the approach runs into trouble.

Take the case of an animal such as a frog. The mechanistic mode of explanation will give us an account of how it is that this entity responds to certain features in the world in appropriate ways. It will allow us to trace the causal links between a passing fly, the frog's visual apparatus, its motor apparatus, and the eventually successful fly-snatch. Moreover, we can be very confident about this mode of explanation. What we know about physics and chemistry--the hard sciences--convinces us that there will be no occasions on which the frog does something that we will not be able to explain using this mechanistic vocabulary.

There is a problem, however. The problem is that it is not at ail clear that what we would be explaining by this method would be the behavior of the frog. What do I mean here? Well, I mean simply that behavior cannot be equated with mere movement. Of course, if something is to behave, then it will need either to move or to refrain from moving. So I would not want to suggest that movement and behavior have nothing to do with one another. But nonetheless I want to claim that one cannot explain behavior by talking merely about the movements an organism makes and the causes of those movements.

The reason I want to make the distinction is that behavior, unlike movement, is a purposive term. Behavior is aimed at doing something. Sometimes behavior succeeds and sometimes it fails. But when it fails, it is no less the case that the behavior was aimed at success. We might talk of an animal searching for food and discuss the mechanism that it uses. And we might think that it is fair to describe that mechanism as a food-searching mechanism. But that the mechanism is there to search for food is not something that is apparent at the merely mechanistic level.(2)

Here is a different way of putting the same point. I am told that "flies... when given a choice between nutritionally worthless sugar fructose and some nutritive substance like sorbitol, will invariably choose the nutritionally worthless substance and starve to death" (Dretske 1988, 58). In such circumstances, the mechanism is operating just as it always has done. And if you were a physics monitor, making sure that the universe behaved itself, you would not see anything awry. All the physics is doing just what it should. But if you are a biology monitor, then you would notice that something has gone wrong. The fly is trying to feed and, in fact, failing. The physics monitor, at least at first shot, is unable to express the idea of the fly feeding or failing to feed.(3)

Now, I, as do many philosophers, think there is no straightforward way to bridge the gap between the mechanistic and the purposive. Not everyone agrees. Some people, reductionists, say that you can express what it is to be a fly, and what it is to have fly successes and failures in purely mechanistic terms. Although they acknowledge that fly talk inevitably involves distinguishing cases where the fly's behavior is successful and unsuccessful, they think that such talk can be paraphrased in mechanistic terms. Success equals one set of mechanistic conditions, and failure means another. The key move, however, is that these conditions are concerned only with the states and structure of the animal's body, considered independently of its environment and of its history.(4) Such an approach sees the animal as no more and no less than a complex mechanical device, and, moreover, one that can be comprehensively understood by the application of mechanistic principles to its intrinsic nature.

A more elaborate, though still reductionist, view will appeal to environmental and historical factors, such as a history of selection within a given environment, a history of learning within a given environment, or a propensity for success within a given environment.(5) In the case of such elaborate reductions, the strategy will be to continue to use the language of purpose but to deny that the language bears its face value. In this context, the reductionist says, purpose does not represent a genuinely distinctive normative standard and the use of normative language is no more than a useful shorthand for complex claims involving intrinsic and extrinsic properties (i.e., relations to the future, to the past, and to the environment. I doubt, however, that such strategies can ever succeed in reducing the purposive to the nonpurposive altogether. Rather, they seem to show how the complex purposes of different species and individuals can be seen as flowing from the very basic aims of surviving and reproducing shared by all biological organisms. But, if we dispense with these purposes, then, I suspect, we will have dispensed with the resources for picking out biological entities altogether.

My view, an antireductionist view, claims that in recognizing the entity before us as an animal, we are endorsing a certain standard of success and failure, a standard that we use to make sense of the animal's activity, to distinguish, for example, successful and unsuccessful behavior, to distinguish what the organism does (or tries to do) from what merely happens to it. Were we determined, we could, of course, refuse to recognize an animal, and say of the entity before us(6) that it is merely a complex machine, a machine to which no standards of success and failure apply.

The elaborate reductionist must agree with my view that we can take two attitudes: we can recognize the entity before us as an animal or we can view the entity before us(7) merely as a complex mechanism. But she claims that recognizing the entity as an animal only appears …