In this wide-ranging book, LaPorte takes on the orthodoxy in
philosophy of science and metaphysics regarding a number of central topics. LaPorte
argues that the causal theory of reference does not guarantee stability of
reference to natural kinds across theory changes; that scientists do not discover
that sentences like "whales are not fish" are true; that mainstream
proponents of the causal theory of reference are committed to the analytic/synthetic
distinction, and that twin earth thought experiments do not have the moral they
are usually taken to have.  As anyone acquainted with mainstream analytic
philosophy of the past three decades will instantly recognize, these claims
amount to something like heresy. Nevertheless, the position LaPorte defends is
no irrationalist or relativist rejection of science and scientific progress.
Instead, it is a kind of moderate Kuhnianism, affirming the existence of
incommensurability but denying that it undermines scientific progress.

LaPorte’s central claim is that scientists do not discover
the essence of natural kinds. The orthodox view is that natural kind terms have
their reference fixed by a formal or informal dubbing ceremony: ‘By "gold"
(or "mammal" or "water", or whatever) I mean that stuff there
and everything else which has the same essence (or everything else that shares
its essence with the majority of that stuff, should the sample prove to be
impure)’. After such a ceremony, our natural kind terms refer, but we don’t
quite know what it is they refer to until scientific investigation uncovers the
essence of the natural kind (the essence is usually supposed to be its
microstructure: its atomic weight, or chemical composition or genetic code).
Once we know what the essence of our natural kind is, we often discover that we
have been applying the term to some objects erroneously. Once we discover what
mammals are, and what fish are, we discover that whales are mammals, not fish.

LaPorte argues, against this view, that our natural kind
terms are vague before science investigates them. Vagueness is
discovered, but the essence of natural kinds is not. It is not true, for
instance, that scientists discovered that monotremes (egg-laying marsupials)
are mammals. Instead, the discovery of monotremes exposed an unsuspected
vagueness in the natural kind term. We might have decided that monotremes were
not mammals (either because they lay eggs, or because of their evolutionary history
– the principal criterion of taxon membership according to most biologists – or
because of their genetic relatedness to paradigm mammals) just as correctly.  We
might even have decided that "mammal" does not name a natural kind at
all, as scientists have done with "algae". Whichever way we went, we
had to make a decision, and the decision would represent a refinement of the
natural kind term. We do not illuminate what speakers always meant by "mammal",
we refine the meaning.

This is not to say that classification is arbitrary. Far
from it. Scientists do discover the facts that underlie our classification
systems. It is just that given these facts, there are many possible, equally illuminating
, classifications. This does not alter the fact that many possible
classifications would be plain wrong. We could decide that whales were, or were
not, fish, or that "fish" did not name a natural kind. We could not rightly
decide that whales were rodents or primates.

LaPorte runs the same line of argument against Putnam’s twin
earth thought experiment. Despite what most philosophers have concluded, we
might just as easily decide that XYZ is water, or even that H₂O is
not water, as that H₂O alone is water.  There is no need to think that
miscrostructure necessarily trumps descriptions. LaPorte buttresses this
claim with a fascinating discussion of the term "jade". Putnam
himself talks about jade, but (LaPorte claims) he gets the history wrong. "Jade"
is a real life example of a twin earth-type case.

According to Putnam, Chinese speakers have always used the
word "jade" (in Chinese) to refer to two minerals, with different
microstructures. Since the original dubbing ceremony more-or-less explicitly was
aimed at samples of both, "jade" refers to both. But since we only
had Earth water to point to when we dubbed it "water", only H₂O
is water. LaPorte argues this is false. In fact, "jade" (in Chinese)
referred to only one mineral, nephrite. It was only after thousands of years of
working with this stone that they encountered a new mineral, jadeite, with very
similar properties. Chinese craftsmen knew nephrite’s properties intimately,
and immediately realized that jadeite was a different mineral.

Thus, the Chinese were in a strictly analogous situation to
the visitor to Twin Earth. They were confronted with a similar entity, but one
with a different miscrostructure. If Putnam is right, they should have declined
to call the mineral "jade". But, after a period of debate and
uncertainty, the Chinese chose otherwise. They extended the term "jade"
to cover jadeite as well as nephrite. Resolving the vagueness of "water"
in the way Putnam argues we ought is perfectly permissible, but so is the opposite
choice made by the Chinese. Each represents a refinement of the natural kind
term, and neither is wrong.

Thus sameness of microstructure is not necessary for
sameness of reference. Nor is it sufficient. LaPorte illuminatingly contrasts
the cases of "topaz" and "ruby". Originally "topaz"
was applied only to a yellow stone. When blue stones were discovered with
basically the same microstructure, the word was extended to them. However, "ruby"
is restricted to red stones, and microstructurally identical stones of a
different color were excluded from the extension of the term. We decided that
topaz would be a natural kind, while ruby became an artificial kind.

One of the attractions of the causal theory of reference was
supposed to be that it neutralized the danger that incommensurability
apparently poses to scientific progress. Suppose Feyerabend and Kuhn (at least
on some interpretations of his work) are right, and science is subject to
occasional revolutions, after which scientists no longer confront the same
world as did their predecessors. Given that pre- and post-revolutionary scientists
approach different problems with a different vocabulary, it seems difficult to
argue that the revolution represents progress. Progress requires a common
standard against which to measure it. The causal theory of reference was
supposed to underpin the common standard. Since natural kind terms were supposed
to refer to precisely the same entities before and after the revolution, they ensure
that incommensurability is, at most, partial.

Despite the failure of the causal theory to ensure stability
of reference, LaPorte argues that scientific progress is not under threat. In
fact, meaning refinement is a consequence of progress, not a threat to
it. The meaning of, say, "species" changes because scientists
discover that it is vague. They then know more about both the older use of the
term, and the new. They are able to substitute statements that are clearly true
for vague statements that were neither clearly true nor clearly false. That, it
seems, is plausibly taken to be progress.

Finally, LaPorte takes on the orthodox view that the analytic/synthetic
distinction is dead. In fact, LaPorte argues, anyone who is committed to a posteriori
necessity of the kind made familiar by Kripke is committed to the
analytic/synthetic distinction as well. If Hesperus = Phosphorus, this is
because Hesperus is a synonym for Phosphorus. But, as Quine saw, synonymity
was sufficient to save the notion of analyticity. If Hesperus = Phosphorus is necessarily
true, then (contra Quine) there are analytic statement which are not open to
revision by theory change alone. The only way to bring it about that Hesperus =
Phosphorus is false is by changing the meaning of the terms. Naturally, this view
requires that we be able to distinguish meaning change from theory change. LaPorte
argues that we can, and that those who accept the causal theory are committed
to this as well. If meaning change and theory change were really
indistinguishable, then it would not have been possible to discover that
Hesperus = Phosphorus; discovery (as Kripke uses it) is theory change.

As this summary of the main themes makes clear, this is an
ambitious and wide-ranging book. In just 173 argument-filled pages, it sketches
heretical views on some of the most interesting and difficult issues in
contemporary metaphysics. Inevitably, it moves too fast and covers too much
ground to be entirely convincing. In particular, it seems that there are
possible lines of reply to the contention that we can as easily decide to
identify a natural kind with observable properties as with microstructure. LaPorte’s
"jade" example is underdeveloped: we are left wanting to know how
important the distinction between nephrite and jadeite is to the Chinese. He
tells us they mark it, at least sometimes. If that is the case, the fact that
they call both "jade" seems no more to refute the Kripke-Putnam
thesis then does the fact that sometimes we call Koalas "bears". This
is not meant to be a rejection of LaPorte’s claims. Instead, it is the
expression of the hope that he will develop these themes, and that his claims
will receive the critical attention they deserve from other philosophers. I
wholeheartedly recommend this book to all serious students of philosophy of
biology, philosophy of science, and metaphysics. Whether they agree with it or
not, they will find it extremely stimulating.

© 2004 Neil Levy

Dr Neil Levy
is a fellow of the Centre
for Applied Philosophy and Public Ethics at Charles
Sturt University, Australia. He is the author of two mongraphs and over a
dozen articles and book chapters on Continental philosophy, ethics and
political philosophy. He is currently writing a book on moral relativism.

Categories: Philosophical