RichardKennaway comments on Open thread, Dec. 22 - Dec. 28, 2014 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (218)
As a person with a scientific background who suddenly has come into academic philosophy, I have been puzzled by some of the aspects of its methodology. I have been particularly bothered with the reluctance of some people to give precise definitions of the concepts that they are discussing about. But lately, as a result of several discussions with certain member of the Faculty, I have come to understand why this occurs (if not in the whole of philosophy, at least in this particular trend in academic philosophy).
I have seen that philosophers (I am talking about several of them published in top-ranked, peer-reviewed journals, the kinds of articles I read, study and discuss) who discuss about a concept which tries to capture "x" have, on one hand, an intuitive idea of this concept, imprecise, vague, partial and maybe even self-contradictory. On the other hand, they have several "approaches" to "x", corresponding to several philosophical trends that have a more precise characterisation of "x" in terms of other ideas that are more clear i.e. in terms of the composites "y1", "y2", "y3", ... The major issue at stake in the discussion seems to be whether "x" is really "y1" or "y2" or "y3" or something else (note that sometimes an "yi" is a reduction to other terms, sometimes "yi" is a more accurate characterisation that keeps the words used to speak of "x", that does not matter).
What is puzzling is this: how come all of them agree they are taking about "x" while actually, each is proposing a different approach? Indeed, those who say that "x" is "y1" are actually saying that we should adopt "y1" in our thought, and by "x" they understand "y1". Others understand "y2" in "x". Why don't they realise they are talking past each other, that each of them is proposing a different concept and the problem comes just because they want all to call it like they call "x"? Why don't they make sub-indices for "x", therefore managing to keep the word they so desperately want, but without confusing each of its possible meanings?
The answer I have come up with is this: they all believe that there is a unique, best sense to which they refer when they speak about "x", even if it they don't know which is it. They agree that they have an intuitive grasp of something and that something is "x", but they disagree about how to better refine that ("y1"? "y2"? "y3"?). Instead, I used to focus only on "y1" "y2" and "y3" and assess them according to whether they are self-consistent or not, simple or not, useful or not, etc. "x" had no clear definition, it barely meant anything to me, and therefore I decided I should banish it from my thought.
But I have come to the conclusion that it is useful to keep this loose idea about "x" in mind and believe that there is something to that intuition, because only in the contemplation of this intuition you seem to have access to knowledge that you have not been able to formalise, and hence, the intuition is a source of new knowledge. Therefore, philosophers are quite right in keeping vague, loose and perhaps self-contradictory concepts about "x", because this is an important source from where they draw in order to create and refine approaches "y1" "y2" and "y3", hoping that one of them might get "x" right. ((At this point, one might claim that I am simply saying that it is useful to have the illusion that the concept of "x" really means something, even though it actually means nothing, simply because having the illusion is a source of inspiration. But doesn't precisely the fact that it is a source of inspiration suggest that it is more than a simple illusion? There seems to be a sense in which a bad approach to "x" is still ABOUT "x"))
I would be grateful if I got your thoughts on this.
P.S. A more daring hypothesis is that when philosophers get "x" right in "y", this approach "y" becomes a scientific paradigm. This also suggests that for those "x" where little progress has been made in millennia, the debate is not necessarily misguided, but what happens is that the intuition is pointing towards something very, very complicated, and no one has been able to give a formal accout of the things it refers to.
It might be useful to look at what happens in mathematics. What, for example, is a "number"? In antiquity, there were the whole numbers and fractions of everyday experience. You can count apples, and cut an apple in half. (BTW, I recently discovered that among the ancient Greeks, there was some dispute about whether 1 was a number. No, some said, 1 was the unit with which other things were measured. 2, 3, 4, and so on were numbers, but not 1.)
Then irrationals were discovered, and negative numbers, and the real line, and complex numbers, and octonions, and Cayley numbers, and p-adic numbers, and perhaps there are even more things that mathematicians call numbers. And there are other ways that the ways that "numbers" behave have been generalised to define such things as fields, vector spaces, rings, and many more, the elements of which are generally not called numbers. But unlike philosophers, mathematicians do not dispute which of these is the "right" concept of "number". All of the concepts have their uses, and many of them are called "numbers", but "number" has never been given a formal definition, and does not need one.
For another example, consider "integration". The idea of dividing an arbitrary shape into pieces of known area and summing their areas goes back at least to Archimedes' "method of exhaustion". When real numbers and functions became better understood it was formalised as Riemann integration. That was later generalised to Lebesgue integration, and then to Haar measure. Stochastic processes brought in Itô integration and several other forms.
Again, no-one as far as I know has ever troubled with the question, "but what is integration, really?" There is a general, intuitive idea of "measuring the size of things", which has been given various precise formulations in various contexts. In some of those contexts it may make sense to speak of the "right" concept of integration, when there is one that subsumes all of the others and appears to be the most general possible (e.g. Lebesgue integration on Euclidean spaces), but in other contexts there may be multiple incomparable concepts, each with its own uses (e.g. Itô and Stratonovich integration for stochastic processes).
But in philosophy, there are no theorems by which to judge the usefulness of a precisely defined concept.
I think this is a very good contrast, indeed. I agree with your view of the matter, and I think I will use "number" as a particular example next time I recount the thoughts which brought me to write the post. Thank you.