1.  The essays in this
collection cover various topics — classification in the human sciences,
radical interpretation, ‘styles of reasoning’, Wittgenstein’s meta-philosophy,
Foucault’s ethics and Descartes’ dreams.  In the introduction, Ian Hacking claims
that each is an exercise in the ‘historical ontology’ of the title.  But what
is ‘historical ontology’?

When
we speak of ‘history’, we sometimes speak of the study of the past,
and other times of the past itself.  Something similar is
true when we speak of ‘ontology’ — sometimes we mean what there is, and
other times we mean the study of what there is.  As a study, ontology is
characterized by its high level of generality and abstraction.  It is concerned
with such things as objects, events, substances and facts — but not with particular
instances of these things (particular objects such as my spectacles or
particular facts such as ‘House prices are rising’), rather with objects and
events in general.  History, by contrast — although also in a sense the
study of ‘what there is’ — is usually thought of as studying this in all its
concrete and localized specificity.  In addition, whereas the stuff of ontology
is usually thought to be timeless, the stuff of history necessarily rises and
falls with the passing of time.  How can there be a kind of inquiry that brings
these two distinct disciplines together?  

Hacking’s
answer is that the objects of our knowledge come into being over time.  So ontology
— in the sense of what there is — is historical.  And historical ontology —
in the sense of the study of what there is — describes the processes by which these
objects come into (and go out of) being.  Such a study is ontological because
it is concerned with kinds of objects and not their particular instances;
and it is historical because it is concerned with the various specific ways in
which these kinds come and go.

2.  This project already
seems to have a somewhat Kantian flavor — and this appearance is strengthened
by Hacking’s numerous favorable references to the work of Michel Foucault and
his project of ‘historicizing Kant’.  Hacking celebrates Foucault political
radicalism (especially in the area of prison reform), but likes to downplay his
philosophical radicalism, presenting Foucault — and by implication himself —
as a philosophical conservative.  I find this somewhat disingenuous, because
there is obviously something radical in the project of ‘historicizing Kant’ —
not least because there is something radical about Kantianism itself.

As
is well known, Kant took it for granted that we possess objective knowledge of
the world, and then asked how that knowledge is possible.  On at least one
reading of Kant, he concluded that the objects of our knowledge are not given
in experience, but constituted out of the manifold of sensory intuitions by the
synthesizing activity of our minds.  On the assumption that ‘to constitute
something’ is to bring it into being, Kant’s view seems to be that the objects
of our knowledge and experience are brought into being by the activity of our
minds.  

There
is a problem with this idea — a problem that applies to all versions of
Kantianism (at least on the reading currently in play).  The problem is this: activity
is not ‘pure output’ or ‘mere behavior’; it is action, and action is
distinguished from ‘pure output’ by the fact that it is individuated (in part) by
the contentful mental states of the agent (or agents) whose action it is.  These
mental states will most commonly be intentions, and so will have contents that
we could express by saying "that I do such-and-such".  For instance, Jane
might intend that she picks up that glass of water.  It seems,
therefore, that if there is activity then there are determinate contents. 
However, determinate contents are necessarily of determinate objects of
experience.  In the foregoing example, the content of the intention is of
both Jane and a glass of water.  Consequently, it seems that if there is activity
then there are determinate objects of experience.  However, the Kantian appeal
to activity is supposed to explain how these determinate objects come into
being in the first place.  The problem is that it cannot do that, for that
which it appeals to — namely, activity — presupposes such objects.  The
erroneous assumption that activity does not have this presupposition is simply
a myth.  Susan Hurley has a useful name for it — she calls it the myth of
the giving.  

We
see this myth in the following passage:

The
knowledge about [for instance] a particular tripeptide produced by the hypothalmus
of mammals or the established hypothesis of quarks is best described not
in terms of discovery, but in terms of social construction.  This
is not to say that these authors think that the products of the construction is
not "really" a fact (now) — only that the unthought world does not
come in facts.  The factisation of the world is a human activity. (p65)

The
general problem here is that the very idea of ‘human activity’ presupposes the
idea of thought (or ‘content’, as I have been calling it).  Consequently, we
cannot appeal to activity to explain how a world without thought gets ‘factized’,
for a world that contains activity must already be a ‘thought’ world.  The
moral is this: take activity for granted and you take the idea of thought for
granted — and thus the idea of determinate objects of experience/ knowledge.

3.  It is unclear whether or
not Hacking is committed to any kind of ‘Kantian’ story about the constitution
of objects as such.  So it is unclear whether or not he falls prey to
the myth of the giving.  Sometimes it seems that he does  — when he writes in
support of Foucault, and when (in his essay ‘Language, Truth and Reason’) he defends
the relativist ‘idea’ that a sentence’s content is relative to a framework constituted
by various practices (or ‘styles of reasoning’, as he calls them).  But at
other times he seems to draw a line.  Below the line are those objects that
come — as it were — ‘pre-constituted’.  They are part of what Daniel Dennett
calls ‘the basic furniture’ of the universe.  And above the line are objects
that are not part of this furniture, and hence must be brought into being by us
in our everyday (and usually scientific) activity.  As he says, "many
categories come from nature, not from the human mind" (p.106) — he
mentions horses and planets.  However, he insists that such categories as gloves
and multiple personality come not (or not entirely) ‘from nature’, but (at least
in part) ‘from’ our practices.  In this more modest picture, only those objects
that belong to the latter category have a historical ontology.

            Hacking’s
argument for why we should draw this line is not convincing.  In perhaps the
collection’s most famous paper, ‘Making Up People’, he tries to argue for the
claim that there is a line and some items are ‘above’ it by making use of Elizabeth
Anscombe’s idea that intentional actions are always ‘under a description’.  Anscombe
had a clear aim in mind when she introduced this idea.  She wanted to counter
the suggestion that when I — say — greet my friend by waving my hand and
(therefore) moving my body, there are not three different actions going on (my
moving my body, my waving my hand and my greeting my friend), but simply one
action described in three different ways.  However, Hacking mis-reads Anscombe
as saying that "if a description is not there, then intentional actions
under that description are not there either: that, apparently, is a fact of
logic" (p.108).  The conclusion he draws from this is that actions, and
everything that involves them, must be located ‘above the line’.  The (very
vague) account of action that Hacking attributes to Anscombe seems to be this: there
must be practices of describing certain items of behavior as actions in order
for these items to be actions in the first place.  So actions — and all that involve
them — belong ‘above the line’, for they are constitutively dependent on our
practices of describing.  But ‘practices of describing’ are actions.  So
the account is flagrantly circular (even more so than those accounts of content
that fall prey to the myth of the giving).  Consequently, I do not think we
have been given grounds for placing actions — and all that involve them — ‘above
the line’. 

This
is an interesting book.  Each of the essays contains much to stimulate the mind,
and each is written in Hacking’s usual crisp style.  There are excellent footnotes
and bibliography, and Harvard’s overall presentation of the book is as careful
and as pleasing on the eye as you would expect.  Hacking’s views are intriguing
— but they are not always as clear as they might be, and they rarely
convince. 

© 2003 Adrian Haddock

Adrian
Haddock, Ph.D., Department of Sociology, Exeter University, UK

Categories: Philosophical