By Schubert Ogden
...
Metaphysics may be reasonably thought to have begun with Parmenides, who held that the philosopher (qua metaphysician) is like the natural scientist in giving an account of the universe. But, unlike the scientist, the metaphysician bases her or his account, not on observation and experiment, but, primarily, on an analysis of concepts.
Aristotle thought of metaphysics, or first philosophy, more as explaining things we already know to be true than as giving reasons for the assumptions we make in the sciences and everyday life, so as thereby to provide the underpinings of science and common sense.
Kant offers a threefold analysis of "metaphysics" as:
...
(3) a "human propensity."
For logical positivism, a statement is "metaphysical" if it purports to make a statement of fact but fails to do so, thereby also failing to have a meaning, because no observations count as evidences for or against it. Accordingly, a metaphysical statement is a pseudoinformative statement that is really meaningless.
For Peirce, metaphysics is "an observational science whose job is 'to study the most general features of reality and real objects.'"
For Walsh, metaphysicians historically have wanted to say both that their propositions possess a peculiar certainty and that they are significant as a purely analytic proposition is not. So metaphysical propositions pretend to the status -– in Kantian terminology -– of synthetic a priori truths. The principle of causality, for example, is not a very wide empirical truth mysteriously known in some nonempirical way, but rather the expression of a of a rule of procedure that serves to tell us, not what properties things have, but how to interpret them. But, then, Walsh reasons, there's a possibility of decoupling Kant's principles of the understanding from their exclusive connection with our sense knowledge of the world underlying science and thinking of them as also providing prescriptive principles for interpreting the rest of human experience. Thus metaphysics comes to be viewed as a set of principles of the understanding that, when applied, yield a unitary account of things, and the metaphysician as "a man with a vision of the scheme of things entire;' ," who then proceeds to develop his vision into a theory. The metaphysician, accordingly, is someone concerned to advocate, articulate, and apply a set of basic interpretative principles, "categorical principles;' ," that promise to make sense of all empirical data.
That this view of metaphysics may well blunt "Hume's fork" to the extent of declining to accept his simplistic two-part analysis of meaningful propositions seems clear enough. But I take it to be equally clear that it is completely shattered on the rocks of truth and argumentation. What good does it do to hold that one metaphysics may be more "illuminating," or "enlightening," than another if the judgment as to what is illuminating is itself a purely "persona!personal," as distinct from, in any way, a "public" judgment? Unless I'm mistaken, Walsh's view of metaphysics saves it only by destroying it -– or, if you will, only by completely changing the subject.
On the other hand, I find his response to the Wittgensteinian appeal to language games, "this game is played," and so on, interesting. I entirely agree with him that there is a certain priority to be claimed for "the language game in which we say how things are" and that, if we could but find a way of speaking that would enable us to express the true nature of the world (!) , we would have a yardstick by which to measure the ultimate testability as opposed to the immediate use of particular language games.
Since Kant, metaphysics has been variously understood:
...
(3) (also popularly) as the spiritual, the religious, and even the occult;
(4) (in modern philosophy) as concerned with questions about the kinds of being there are and their modes; and
(5) (especially in the 18th and 19th centuries) broadly as including a concern with questions about the reality of the external world; the existence of other minds; the possibility of a priori knowledge; the nature of sensation, memory, abstraction, etc., all of which are now included in epistemology.