You said: "I am always surprised that aligning an ontology with how we speak about the world rather than with how the world is is something we seem to seek to do, but I am an engineer, not a linguist."
Seeing "how the world is" rather than just "how we speak about the world" is a program that drove Philosophy from the time of Plato into the early 20th century. In Continental Philosophy, Husserl's bracketing was an attempt to do just that, and his greatest student, Heidegger, repudiated it. In Analytic Philosophy, the latter Wittgenstein repudiated the picture theory of meaning which was part of his Tractatus logical atomism, and no one has picked it up since. Richard Rorty wrote extensively about out inability to obtain a "God's eye" view of things, and Wilfred Sellars described what he called "the myth of the given". Quine's semantic holism, arising out of his demolishment of the analytic/synthetic distinction, was another nail in the coffin of the belief that we can see things as they really are, apart from our conceptualizations. And, of course, one could readily argue that this whole theme began with Kant, and that the "semantic ascent" to language of 20th century analytic philosophy replaced Kant's talk of "faculties" with a more tractable way of analyzing the contribution of the subject to his knowledge of what there is.
[MW>] I beg to differ. If the point of metaphysics is not to understand how the world is, then there is no point to it at all. That people find it hard, and frequently get it wrong does not mean it is not worthwhile. The best we can do is more useful than the worst we can do, whether or not we have reached a true understanding – which we probably could not be certain of even if we had.
All this can be ignored when we are engaged in everyday talk about everyday things, but not when we push beyond "mid-range physical objects" (Quine's term) into what we, as upper-level ontologists, talk about.
I believe I detected something of the same attitude you expressed in Pat Hayer's recent response to me (and I say this with all due respect to the extensive and important work Pat has done) -- that we down-to-earth ontologists / knowledge engineers (as opposed to up in the clouds (pace Aristophenes) philosophers who can never get down to anything real and solid) build things, real things, and get real work done.
I suggest that ontologists / knowledge engineers who share that belief are simply philosophers who aren't aware that they are doing philosophy. They are those who continue in the mistaken belief that, in matters upper ontological, there really is a "given" -- or at least a few well-defined alternative givens.
[MW>] Well I’m happy to concede that there are an infinite number of possible theories of how things are. I have seen a huge number of theories of bits of the world that break when they are taken out of the very specific context in which they were developed. In fact this is how I got into data modelling and ontology, Shell was concerned in the early nineties about why it was when Shell Group companies were developing systems for particular purposes, they were unusable by other companies. I was part of the team that was assigned to find out why and what we could do about it. We discovered that the reason was that the systems had built in constraints that might be valid in the particular context the system was developed in, but were not necessarily true outside that context.
So yes there a lot of possible theories you can adopt, but most of them do not hold up to any critical analysis. As I said before, it comes back to the commitments you make. You can’t choose not to make them, and there are not too many (reasonable) options available for each choice you have to make.
I will also suggest, on the other hand, that while we upper-level ontologists are doing ontology, i.e. metaphysics, i.e. philosophy, there is an important difference between the way we do it and the way it has been done in traditional Philosophy. We formalize our speculations, from which we derive the benefits of being better able to see where important ambiguities still reside, or down the deductive road implications that we may or may not be comfortable with.
[MW>] Indeed. The main purpose I see for an upper ontology is as an analysis framework that helps you to ask the right questions to sort out these ambiguities.
Regards
Matthew West
http://www.matthew-west.org.uk
+44 750 338 5279