Intellectually, we are not at home in the
world.  To tell a likely tale, once upon
a time we recognized not only a natural world of sticks and stones, cats and
dogs, sun and moon, but in addition peopled it with gods, spirits, nymphs,
djinns or angels, from which realm derived moral demands, the necessities and
impossibilities of natural things, even the hard edges of mathematics and
logic, if we turned our minds to them. 
Our minds were a spark of this divine realm, somehow linking us and
it.  The great advance was when we
reconceived all this, some bits more easily than others, as our contribution, nomos not physis, shifting the non-natural from a factitious supernatural to
an all-too-human conventional status. 
But the dialectic does not stand still. 
The non-natural remains stubbornly non-natural.  It, and the minds that engender it, stick
out like a sore thumb in a mechanistic or post-mechanistic world-view, and
beget any number of ingenious attempts to make themselves disappear.  It is the hope of Joseph Rouse’s new book to
lay this ghost once and for all, to offer a fully naturalistic account of
language, scientific thought, and action, but one which requires us "to
reconceive scientific inquiry, the world it discloses, and the relation of both
to philosophical reflection about scientific understanding" (p. 5).

Rouse has then great ambitions.  He has an enviable familiarity with styles
of philosophy apparently alien to the analytic tradition in which he mostly
works.  How pleasant to be able to drop
a remark like "Unified science was conceived as a historically situated
task oriented toward making predictions; authentically resolute Dasein was a
thrown project of being-toward-death" (p. 72) in the course of commenting
on Neurath’s and Heidegger’s disagreements with Carnap and Husserl
respectively.  But it has its dangers.  For
one who gives such prominence to the practical turn, to the appeal to human
practice as the touchstone and ne plus
ultra, Rouse remains aloof from actual examples of the human action he so
much lauds.  I suspect the aptness of
Mackie’s comment on Dummett’s style of philosophizing: "he cannot simply
go on appealing to a ‘principle’ which may sound plausible when it is stated
only in general terms, but which becomes utterly implausible when implemented
in particular cases" (Logic and
Knowledge, p. 243).  At any rate,
born an Aristotelian, I need the help illustrations may provide.  Rouse’s abstract, Platonic argumentation too
often leaves me uncomprehending.  

But what are the bare bones of Rouse’s project?  He starts by setting out two conditions on
an acceptable naturalism: it must not require any arbitrary impositions upon
the sciences; it must not appeal to anything mysterious or supernatural (this
is mainly cashed in terms of rejecting notions of representation or meaning
supposedly destroyed by celebrated Quinean or Wittgensteinian
considerations).  He recognises that
implementing these requirements is itself a historically conditioned and
contentious activity — we don’t know exactly what an arbitrary imposition may
be and we aren’t always sure what is or isn’t mysterious, or what is and what
isn’t science.  But we will continue to
argue about it, and that seems enough to satisfy him.  Rouse thinks that most attempts to characterize science and the
world it reveals have in fact rendered their satisfaction impossible:

 The "natural
world" to be understood scientifically and the normativity of the
thoughts, utterances, and actions in which such understanding could be realized
have been preconceived in ways that prevent their reconciliation…. Nature so
conceived could not have become manifest through scientific practices that are
themselves a part of nature so conceived. (pp. 11-12)

Rouse’s
constructive response centres on practice and a way of construing practices,
not as rule-governed regularities, but as "identifiable by their normative
accountability" (p. 12). 
Normativity is indeed the focus of Rouse’s discussion, but he says
remarkably little to display what aspects of human life he is thinking of.  He has some discussion also of necessities,
and towards the end he makes play with a notion of "real
possibility", but the normativity "that is constitutive [another key unexplained notion] of meaning and
knowledge" (p. 31) is his prime target.  
To illustrate my bafflement, one could invoke Rouse’s early rejection of
Frege’s idea of the normativity of logic. 
Whatever the mysteries of Frege’s full-blown metaphysics, it seems
pretty clear what he is doing in rejecting psychologism and insisting that
logic is normative.  In empirical
investigations, I have found, for instance, that many people are inclined to
think arguments of the form No A are B,
no B are C, so no A are C are valid. 
(They are, of course, patently invalid, so much so that the fallacy
doesn’t even have a traditional name.) 
So it’s a fact that people will argue in that way; it’s equally true
that the form is invalid — we ought not to. 
But I don’t yet see why we shouldn’t treat this in Quine’s way as a
technological, hypothetical ought: if we want to preserve truth in our
inferences, don’t use that form.  We
don’t need to suppose Frege is relying on some sort of objective prescription
about inference.  Rouse seems not to
descend to even refute such simple-minded positions.

 Rouse rejects the common assumption that
"social relations and natural objects form relatively autonomous
components of the world" (p. 12) and repudiates the general suspicion of
causality among philosophers.  He admits
that many will think his task impossible to achieve, but once an alternative is
seen to be viable, we can naturalise the normative, not by tying it to
something non-normative, but by recognising normativity all the way down:

there is no causal structure, no determinacy to the components of a causally
intra-active system, apart from its involvement with discursive normativity (p.
290). Without a pattern of utterances responsive to one another and their
circumstances in semantically significant ways, the world would indeed have no
semantic structure, and hence no boundaries or properties (p. 291).

Neither sentences nor things can be what they are
without the other (p. 292).

These last quotations, beside intimating a kind of
idealism he officially repudiates, reflect Rouse’s plausible view that almost
everyone else has mistakenly taken a world of objects, with their criteria of
identity and essential natures, for granted, when enlightenment comes from
"the opposite direction, beginning with causally intra-active phenomena
and asking how objects and natural kinds are understandable through their
practical-inferential role expressing the connections between various
phenomena" (p. 24).  Later he uses
this notion to make an interesting criticism of Kripke’s notion of rigid
reference (p. 345).  There are, indeed,
many comments and criticisms of contemporary writers throughout the book:
extended discussions of Brandom and Haugeland, briefer engagements with Quine,
Davidson, Rorty, Cartwright, Heidegger, Foucault, and many others.  Nor does Rouse omit discussions of various
approaches to science studies, in particular the contributions of feminist
theorists (feminist criticism of actual science seems not to be a case of
placing arbitrary impositions on the sciences).  But too much of it is opaque. 
To take a sentence at random: "What is authoritative over and
constitutive of human agency and meaning, I will argue, is not the independent
objective natures of things, but the emergent configuration of a situation as
having something at stake in its outcome" (p. 257).  The surrounding paragraph does not make it
any easier to discern what that amounts to.

The issues Rouse confronts are important. While it
may contain valuable insights, I regret that I cannot in the end recommend his
book to the non-professional reader. 
Professionals may well be bound by the constraints of their ongoing
conversations to find something to recover here; after all, they "can only
become what [they] are in concert with other agencies and circumstances, and
hence [their] constitutive normativity involves responsibility … to
coordinated, intra-active performances, issues, and stakes" (p. 359),
which I presume includes having to read the unreadable.

© 2003 Ed Brandon

Ed Brandon is, by
training, a philosopher, and now is working in a policy position in the
University of the West Indies at its Cave Hill Campus in Barbados.

Categories: Philosophical