The Evolution of Modern Metaphysics: Making Sense of Things (82 page)

What is striking – and this is the crucial point in the current context – is that both the work that Lewis takes his modal apparatus to perform and the damage that it briefly looked as though it might wreak are fundamentally
philosophical
. Descartes had a metaphysical system that was intended to subserve science; Spinoza, a metaphysical system that was intended to
subserve ethics; Hegel, a metaphysical system that was intended to subserve everything. And the damage that their systems were in danger of wreaking was similarly far-reaching. Even Leibniz, whose metaphysical system came closest to being intended as an end in itself (
Ch. 3
, §1), was in danger, partly indeed for that very reason, of fostering passive resignation in the face of the world’s evils. The positive or negative assessment of what Lewis offers, by contrast, seems very much a matter for the philosophical study.

For Lewis, metaphysics is intended to subserve the rest of philosophy. Thus even the example that I gave in §7 of the Introduction, of how Lewis combines metaphysics with science in addressing questions about the existence and nature of properties and universals, showed him putting metaphysics to work in the service, not of science, but of the philosophy of science, trying to make the best sense
of
science. Likewise, the example that we considered in §2 above, of how Lewis combines metaphysics with mathematics, was an example of his putting metaphysics to work, in the service, not of mathematics, but of the philosophy of mathematics, trying to make the best sense
of
mathematics.
²⁶
(This of course relates to the mathematical conservatism that we also saw him proclaim in that section.) Not that the metaphysics in which Lewis engages need be viewed as anything other than a maximally general attempt to make sense
of things
, of universals, for instance, as opposed to the concept of a universal, or of sets, as opposed to the concept of a set. It is just that its focus, as befits its place in the analytic tradition, is as much on the sense that is being made of these things as it is on the things themselves. In the offing are always philosophical questions about the more particular sense that is available to be made of these things, in the specific areas of human thought and experience in which they especially figure.

Let us return to the puzzle that I posed in §1. How is that, in the wake of Quine, and against a fundamentally naturalistic backcloth, philosophers have felt encouraged to revert to the kind of metaphysics that was practised in the early modern period, to reflect, in their armchairs, on substance, universals, and the rest? The story, I suggest, goes something like this. Quine, arch-naturalist that he was, believed that a crucial task for naturalism was ‘[to blur] the supposed boundary between speculative metaphysics and natural science’ (Quine (
1961b
), p. 20). He gave renewed respectability to the idea that metaphysicians had a legitimate contribution to make to the overall project of making sense of things, of a piece with the various contributions made by natural scientists, but at a much higher level of generality than most. Later, the boundary that he and many other analytic philosophers
took his naturalism to have blurred seemed, in the eyes of a large proportion of the analytic community, perhaps even the majority, to be as sharp as ever. (It had rarely seemed
razor
sharp, certainly not when ‘speculative metaphysics’ had been recognized as a legitimate activity at all.) Analytic philosophers became altogether less suspicious than they had been under the sway of Quine of the various associated distinctions, most notably the distinction between what is necessarily the case and what is contingently the case. At the same time the idea that philosophers in general, and metaphysicians in particular, could muscle in with natural scientists in discovering truths about reality (and not just promoting clarity of thought,
a là
Wittgenstein, or articulating linguistic frameworks for the expression of other people’s discoveries,
a là
Carnap) retained its allure. So philosophers, if they were to hold on to this idea, needed to conceive of the contribution that they had to make to the overall project of making sense of things as genuinely distinctive, not merely different in degree from the contribution that natural scientists had to make. The most obvious way for them to do this, if not the only way,
²⁷
was to see it as their business to reflect on how reality must be, as opposed to finding out empirically how it happens to be. And in this way, or in some related way, they were able to reattain to a broad conception of philosophy, and in particular of metaphysics, that had been dominant in the early modern period, but that had been consigned to oblivion throughout most of the analytic tradition.

Why then did metaphysics in the early modern period have so many more repercussions beyond philosophy than its analytic descendant, metaphysics of the kind that Lewis practises, or ‘naturalistic’ metaphysics, as we might call it?

Part of the answer to this question is illustrated in what I said above about Lewis himself. The focus, when naturalistic metaphysicians attempt to make maximally general sense of things, is characteristically as much on the sense as it is on the things – which indeed connects with the idea that the project is one of saying how the things must be, not simply how they are. But that is only part of the answer. Even if it explains the special connection between naturalistic metaphysics and the rest of philosophy, it does not explain the broader connections enjoyed by metaphysics in the early modern period. What did the early moderns bring to their endeavour that naturalistic metaphysicians do not?

The answer to
this
question – to continue in the same dangerously glib vein – is that they brought a due regard, not only for the things, and not only for the things together with the sense made of them, but for the making of that sense. Metaphysics in the early modern period had a distinctive kind of self-consciousness that naturalistic metaphysics lacks. (Recall my
complaint at the end of §1.) Every single one of my protagonists in
Part One
of this book tried to make maximally general sense of things by making sense, in particular, of making sense of things – as I tried to indicate in each of their respective chapters. This enabled metaphysics to serve, for each of them, as ‘a humanistic discipline’, to use Bernard Williams’ phrase (Williams (
2006m
)).
²⁸
And this in turn enabled it to enjoy the broader connections it did. It became inextricably bound up with the attempt to understand the place of humanity in the larger scheme of things: to understand, for example, how human beings can arrive at
scientia
(Descartes); or how they can achieve their highest virtue, by acquiring adequate knowledge of the essence of things (Spinoza); or how they can acknowledge this world as the best of all possible worlds, despite its appearing not to be (Leibniz); or so forth. This is not
yet
to say that metaphysics was in the service of anything other than philosophy. To substantiate that claim, some story needs to be told about how understanding the place of humanity in the larger scheme of things could in turn help humanity to live in that place. But each of my protagonists did have such a story to tell. And because of the distinctive self-consciousness that we find in early modern metaphysics, each such story became a story about the use to which metaphysics could be put beyond the philosophical study – provided, of course, that it was metaphysics of the sort practised then.

No such story is relevant to metaphysics of the sort practised now, by those of Lewis’ ilk, that is to say naturalistic metaphysics.
²⁹
For reasons that I tried to make clear in the previous chapter, it is precisely this distinctive self-consciousness that the naturalism at work in such metaphysics prevents it from replicating – at least when that naturalism assumes anything like the extreme form that it assumes in Quine’s case.
³⁰
Such naturalism badly
misconstrues, or at the very least fails to make adequate provision for, making sense of making sense of things. In Lewis, and in other metaphysicians of his stripe, we find insight, invention, and illumination. But we also find evidence, it seems to me, of the debilitating power of their naturalism, which, by forcing
their metaphysics into an inappropriately scientific mould, seriously restricts its impact.

1
David Lewis, the subject of the present chapter, accepts an (unsharp) analytic/synthetic distinction. But he also works hard at rebutting Quinean objections to what he accepts: see
Convention
, pp. 200–202 and Conclusion.

Note: throughout this chapter I use the following abbreviations for Lewis’ works: ‘Abstract’ for Lewis (
1986b
); ‘Anselm’ for Lewis (
1983c
); ‘Attitudes’ for Lewis (
1983f
); ‘Ayer on Meaning’ for Lewis (
1998b
);
Convention
for Lewis (
2002
);
Counterfactuals
for Lewis (
1986d
); ‘Counterpart Theory’ for Lewis (
1983d
); 1st Introduction for Lewis (
1983b
); 2nd Introduction for Lewis (
1986a
); ‘Knowledge’ for Lewis (
1999d
);
Metaphysics and Epistemology
for Lewis (
1999a
); ‘Observation’ for Lewis (
1998a
);
Papers I
for Lewis (
1983a
);
Parts of Classes
for Lewis (
1991
);
Plurality
for Lewis (
1986c
); and ‘Reduction’ for Lewis (
1999c
).

2
See esp. the essays in
Papers I
and
Metaphysics and Epistemology
for Lewis’ own treatment of these topics. For samples of other classic treatments of them within the analytic tradition, from the second half of the twentieth century, see Strawson (
1959
); Armstrong (
1968
); Kripke (
1981
); van Inwagen (
1981
); Nagel (
1986
); McDowell (
1996
); Mellor (
1998
); and Wiggins (
2002
). Note: I talk about the resurgence of metaphysics as practised in the early modern period, though what Lewis epitomizes is also highly reminiscent of scholasticism, not least in its close attention to language. Were this comparison to be pushed, then we might further say that science plays the role that was played in scholasticism by theology, and Frege’s varied legacy the role that was played in scholasticism by Aristotle’s.

3
Here and throughout this chapter I use the term ‘naturalism’ in more or less the Quinean way defined in §2 of the previous chapter. (Contemporary analytic philosophy exhibits other, less demanding uses of the term; see n. 8 of that chapter.)

4
Cf. Introduction, §5. See further §4.

5
There are some notable exceptions. See in particular Davidson (
1984b
) and (
2005a
), Williamson (
2007
), and Papineau (
2009
) – the last of which is especially robust in its commitment to naturalism. I should also mention the many excellent essays in Chalmers et al. (
2009
). Note, however, that these essays are concerned specifically with questions in ontology. They do not exhibit the broader self-consciousness about metaphysics that I have in mind.

6
Here the analogy with post-Kantian German philosophy certainly breaks down!

7
In ‘Observation’ and ‘Ayer on Meaning’ he considers in detail certain problems that afflict the formulation of semantic empiricism and provides a solution to these problems. Even so, he remains non-committal as far as the
truth
of semantic empiricism is concerned.

8
Later he writes, ‘I’m moved to laughter at the thought at how
presumptuous
it would be to reject mathematics for philosophical reasons’ (ibid., p. 59, emphasis in original), and goes on, very wittily, to present a catalogue of philosophy’s more outrageous pronouncements as evidence of its credentials. (Here as elsewhere there is a curious convergence between Lewis’ attitude and that of the later Wittgenstein: see
Ch. 10
, §5, where I cited Wittgenstein (
1967a
), Pt I, §124.) That Lewis’ attitude to the presumptions of philosophy with respect to mathematics provides a clue to his attitude to naturalism is also proposed by Daniel Nolan in Nolan (
2005
), pp. 11–12. But Nolan is more circumspect in his proposal than I am inclined to be; see in particular p. 12.

9
This passage evokes Neurath’s image; see §3 of the previous chapter.

10
‘The world’ here – to anticipate the next section – stands elliptically for the actual world.

11
Notice how, in the terms of the previous chapter, Lewis, like Quine, subscribes to both provisional physicalism and regulative physicalism. (But note that Lewis himself calls his thesis ‘materialism’ rather than ‘physicalism’, in combined acknowledgement of its ancestry and deference to earlier usage. (Fichte, for one, used the label ‘materialism’: see
Ch. 6
, §2.) ‘Physicalism’ is the better label, however. This is for a reason that Lewis himself gives: ‘our best physics acknowledges other bearers of fundamental properties [than matter]: parts of pervasive fields, parts of causally active spacetime’ (ibid., p. 293).)