Monday, September 5, 2016

Are brains machines?

Recently I was somewhat amused to see an online discussion on the merits of philosophy as a research field turn to the question whether our brain can properly be called a machine. One of the reasons why I found this amusing was that I had not too long ago been involved in the very same discussion at a completely different venue.

What I think is going on here is less that the people arguing about the word actually disagree substantially, but rather that the word is used in two completely different ways, with different meanings.

If we think about it, it is really interesting how we use analogies to describe concepts. There are, of course, words that have pretty much only one proper meaning, where a word is a clear, unambiguous name for a concept that you have to learn directly without the help of an analogy. In my area, for example, one might think of words like phylogeny or clade. Yes, those words are derived from Ancient Greek terms that actually meant something else originally; klados is apparently a branch, and phylon a “race, tribe, kind”. But the point is, few English speakers today will hear the word clade and think branch. They will either have no idea what it means or they are biologists who have learned the technical meaning without first having learned Ancient Greek.

A phylogenetic “tree”, on the other hand, uses an analogy to capture the idea of a branching structure with no reticulations. This analogy works fairly well indeed, but there are stranger ones. Take “identification key”, for example; where does the metal object for unlocking doors come into this picture?

Anyway, back to the brain. Does it make sense to discuss the brain in terms of a machine? And if we do so, what do we mean by that? Is it a helpful analogy?

Again, the problem that I perceive is that there are two groups of people using the same word for very different analogies. There are many different definitions of “machine” that we can find in dictionaries, but the two relevant ones appear to be:

1. A technological device built by humans to achieve a purpose, and

2. A complex structure consisting of interrelated parts with separate (sub)functions.

Note that in the first definition the point is about a machine as a human artefact, really any human artefact beyond a certain level of complexity. In contrast, the second definition is considerably more general and can also apply to natural structures that can be seen as having a function, even in the context of evolutionary biology.

And this, I think, is the problem. There are those coming from a computer science background or people skeptical of supernatural claims about our minds who want to stress that the brain is purely physical and “merely” a very complicated mechanism for information processing. And they are correct (in my eyes). And then there are those who hear “the brain is just a machine”, are skeptical about hyperbolic claims around the feasibility of mind uploading or brain simulations and respond, no, there is something fundamentally different about machines and the human brain. And they are also correct. Obviously there is a massive difference between a purpose-built car or even a supercomputer on one side and the messy, squishy, improvised end product of billions of years of natural selection and drift on the other.

The picture that one side wants to draw with the analogy does not work for the other side, because they understand the word in a different manner, because to them this specific analogy doesn't work. It is possible that this confusion underlies a few other discussions in areas that are hard to talk about without using analogies.

No comments:

Post a Comment