I'm at Madison now, ready to officially start my immersive Arabic program tomorrow. Or rather, not ready, but being pushed off the cliff anyhow. So I have a couple ideas that I want to get out before I am unable to use English or be around large amounts of English-speakers for two months (mercifully I have Friday nights off, at least, but I don't plan on writing philosophy then). First, I've been thinking that philosophy (and truth-seeking in general) is a lot like computer search algorithms, and that the analogy can help us to understand the place of the history of philosophy.
One can use simple search algorithms, called hill-climbing algorithms. The point is simple: you look everywhere around your current position, and you find the direction that takes you higher (where what counts as "higher" depends on what you are searching for; in this case, perhaps it is what is more rational/coherent/explanatory/pragmatic/all of the above). You take this step, and you repeat the procedure until you find a maximum.
It's nice, it's simple, it gets some results, but the problem with the hill-climbing algorithm is the same as with many thinkers: if you only look from within your own position, you are as likely as not only going to find a local maximum. That is, you found your own little hill in the search-space, but it is quite possible that you are surrounded by the Himalayas. Or plains, for that matter. The fact that you found a local maximum has absolutely nothing to do with the surrounding terrain. That's the problem with "faith seeking understanding": if you start with some ground that you refuse to question, you may get interesting results, but that says absolutely nothing whatsoever about what is ultimately true.
One solution to the problem with the basic hill-climbing algorithm is to start searching at multiple points. There's still no guarantee of finding a global maximum, the absolutely highest point present on the search-space, but one increases ones chances greatly. Even if one doesn't find the highest point, one at least has multiple local maxima to compare. One has a better chance of finding oneself to be in the Himalayas, even if one doesn't find Mt Everest.
Similarly, in philosophy, studying multiple styles of philosophy and assorted thinkers is necessary, not to find the global maximum (the absolute truth), but at least to have different local maxima at hand. Such an endeavor must be done sympathetically, though; if one can't actually see the world through the eyes of different thinkers, then one has not actually searched through the world with them.
One type of search algorithm that enables one to search multiple points is the genetic algorithm. One starts a pool of strings encoded with information. They are tested via a fitness function, and the fittest are allowed to pass on to the next generation. Further, the fittest strings are also mixed up to create new strings with information taken from multiple old strings, thereby enabling change to occur so that new areas on the search space can be explored.
This is how I think that the history of philosophy can work. There is (1) progress, which (2) needs what was done in the past and (3) which doesn't necessarily lead to "the truth" (since we can't necessarily know when we've actually hit a global maximum). We take the "memes" given by other philosophers, ancient and contemporary, and we remix them in our own persons. Some combinations will prove fitter than others, and there can be overall progress. Nevertheless, we never get rid of the past, since it is part of our own intellectual makeup and is a constant source of ideas to stimulate our own searches. Also, the diversity of philosophies is what enables the search to continue and possibly hit better and better local maxima. It isn't about getting it right once and for all, but rather it is about participating in the communal effort for truth as it gradually improves over time.
2 comments:
Very cool. The analogy is apt, and makes good sense of a purpose (but not necessarily "the" purpose, if indeed there is such a thing) or at least motive behind studying the history of philosophy. The only thing I'm a bit hesitant to assent to is your claim that this search for truth gradually improves over time. It sounds like some sort of Hegelian Myth of Progress to me.
I don't really have something against progress in the general sense. I think that overall we know more about the world than the ancient Greeks did, or at least have access to more memes about the world which can be put together to create an overall better picture (i.e. there can now be the possibility of people putting together modern natural science with Neoplatonic metaphysics, something Plotinus couldn't do).
The advantage of the genetic algorithm analogy contra Hegel is that this progress is incredibly messy and can't really be predicted ahead of time. A string from one time isn't necessarily better than a string from another, and perhaps the overall fitness of strings at a later time is worse than that of the overall fitness of strings at an earlier time, though the later time has more useful variety and so a greater potential for fitness than the earlier time.
If you prefer, though, then in the analogy we can say that the fitness function also changes over time. Thus, the strings are not becoming more fit in the same environment, but are constantly trying to stay fit in a changing environment. There is then no progress, but there is still a need for continual re-thinking.
Post a Comment