Friday, February 1, 2008

doubt on knowledge

some posit that "to have knowledge is to know something, and that something is true." i have some problems with that. for trivial statements such as, "it is snowing outside," while i don't have a well formed argument against it, i am somewhat uncomfortable with saying that it is true. however, for less trivial statements, such as "there is gravity," i think it is pretty easy to form an argument. the statement "there is gravity" can only be a summary for stating

"there is a large body of empirical observations that are all expected on the basis of positing the existence of some inexplicable force pulling all objects together. in particular, various physicists have concocted a precise formula that quantitatively characterizes this force for many cases."

i'm inclined to acknowledge that the first sentence is approximately true. however, the second sentence implies that the current model of gravity is imperfect, in that certain cases are not accurately characterized by even our best gravitational models. in that sense, the gravity that we posit to exist in correct, and so the statement "there is gravity" is also incorrect. furthermore, it is unlikely that we will ever have a comprehensive model of gravity, so the statement, "there is gravity" will never be strictly true.

mathematically, one could imagine that gravity is an "ill-posed problem", meaning there is no unique solution. thus, even if one could construct a comprehensive model of gravity, it would by no means by the only possible model; rather, it would be the only one we've thought of so far. aside from trivial changes in the model (like renaming something), in general, i see no reason to believe that all the observations relating to gravitational forces should only be explicable in a single way. therefore, there could be many equally good models of gravity. are they all correct? maybe. alternately, they may all be useful descriptions of the same phenomena. probably, some descriptions would be more useful than others (at the least, in some cases this should be true). in that case, we would prefer certain models at certain times. then, talking about whether a model is true becomes less important, and talking about whether it is useful is more important. so the question of truth vanishes in favor of the question of utility. i guess this makes me some what of a pragmatist.

this argument holds generally for any model of any feature(s) of the world. if one posits that all descriptions are really just linguistic models of the world, then the appropriate question to ask is which are useful. those models/descriptions that are useful are tho ones that should be kept and refined, others should be discarded. since we probably will never get a completely true model, i think this is the only reasonable desirate: utility.

1 comment:

joshyv said...

from joel: i think you want something by the way from the concept of knowledge very different from what philosophers want. very very briefly: they want to understand how it could even be possible for an observation to obtain such as to figure as a single variable within a spectrum of experimental results (for example with respect to gravity). your point on gravity relates to a particular model that explains more or less a set of observable features of the world or the universe. that these facts are observable at all would be enough to give one the paradigm case for knowledge...that we have a theoretical model able to explain the unity of all those observations seems to me a separate problem...i think there are many holes in this suggestion...but at the moment, my thoughts would head in that direction...also: it is probably a mistake to think of concepts or words as nothing more than models, as though they externally track the world, rather than being co-constitutive of it.