Over the course of my philosophy career, I’ve come to realize I sometimes use technical philosophical terms without an exceedingly determinate conception of their precise meaning. But I do, however, know how to use them in a particular philosophical context that will make sense to an interlocutor–reader, discussant, student–who has a background similar to mine. (Perhaps this is all that is required with just about any word? What more could be required after all? But I digress.) Thus, I muddle through, talking about philosophy, writing on it, teaching it, debating it. Heck, I’ve made a career out of it.
A classic example of an ambiguous, yet useful and widely used term is ‘humanism.’ I made heavy use of it in the first paper I wrote in graduate school, in a paper on Marx and Feuerbach‘s views on religion. I described Marx and Feuerbach (and possibly Hegel) as humanists, referred to the Young Marx as an arch-humanist in distinguishing him from the Later ‘Das Kapital‘ Marx, and so on. Over the years though, I’ve come to sense that I don’t have a real handle on the term other than to say it refers to ‘human-centered philosophies.’ When asked to explicate that term, I launch into various examples: early Marxism, existentialism, secularism–stress its affinities–philosophical naturalism, for instance–and point to other schools of thought that employ the term, like, say, renaissance humanism. Within the context of these examples, I am then able to try to clarify what is meant by ‘human-centered.’ This past fall, when introducing students to existentialism via Sartre–besides the obvious import of the slogan that ‘(human) existence precedes (human) essence’–I stressed his claim that Sartrean existentialism is humanism because it emphasizes, centrally, the human freedom and ability to make choices. And as I’ve mentioned affinities above, it is worth mentioning humanism’s affinities with pragmatism. In particular, William James, who took ‘humanism’ to describe his pragmatism, offers us some wonderful characterizations of it:
[I]t is impossible to strip the human element out from even our most abstract theorizing
[T]o an unascertainable extent our truths are man-made products.
The ambiguity of philosophical terms should not be too shocking: many philosophical terms have been employed in a wide variety of disciplinary contexts; they have extensive histories of usage and thus resist precise definition (as Nietzsche usefully pointed out a long time ago); they are used to clarify, extend, and resolve philosophical debates in more than one arena of disputation; sometimes, they are drawn from different languages and then encountered in translation; they often enjoy extensive deployment in non-philosophical contexts, and thus create ambiguities between antecedent and current usage. Furthermore, philosophical traditions that stress conceptual analysis can sometimes exacerbate the confusion: by emphasizing necessary and sufficient conditions for usage, they risk smoothing out, by force and fiat, the rough, serrated edges of meaning that make the term as useful and ubiquitous as it has been.
A philosophical term that is all too easily defined should make us just a little suspicious about its usefulness.
Samir Chopra 
I think Richard Rorty, who I have been reading lately, would also agree with you. As a follower of James, he seemed to extend the idea that definitions, like truths, are linguistic contingencies.
Tom,
You must read Rorty’s ‘Contingency, Irony and Solidarity’ – the first chapter on language is fascinating.
I have. Twice. It’s brilliant. I even tried to have one of my characters espouse some of his ideas in my novel. Thanks also for the James recommendations.