by Ian Hacking
Harvard University Press, 1999
Review by Christian Perring, Ph.D. on Nov 22nd 2001
Of the eight chapters here, only the first five will be of much interest
to most philosophers and psychologists. The last three are on weapons research,
geology, and different versions of the death of Captain Cook. Most of these
chapters are derived from previously published articles and book reviews.
Only the first and third chapters are completely new, and parts of the
first chapter also appeared in a rather different form in the magazine
Lingua Franca, in an article titled "Are You A Social Constructionist?"
(May/June 1999 issue).
Hacking's approach in these chapters is rather unlike most contemporary
philosophers -- he writes with a casual style much of the time, and often
with clear feeling behind his words, rather than the obscure terminology,
painstaking setting out of initial and then one or more revised definitions,
and relentless consideration of possible counterobjections that one finds
in philosophy journals and academic books published by university presses.
He speaks plainly, not leaving readers to guess what his real view is.
In short, when one is interested in his subject matter, his work is a pleasure
The pleasure in reading Hacking's approach is enhanced for me in that
his views and arguments strike me as right, at least for the most part.
In setting out the culture wars and controversies over social construction,
Hacking steers a middle course that makes a great deal of sense. Given
the controversies about the social construction of gender, race, sexuality,
illness, subatomic particles, emotions, science, and reality, it is good
to have thoughtful, careful, and moderate guide to help us think through
One of the central points that Hacking makes is that the claim that
X is socially constructed is generally meant to undermine our previous
belief in the inevitability of X. He distinguishes between different
grades of undermining and attitudes towards X: historical, ironic,
reformist, unmasking, rebellious and revolutionary. He also makes a strong
distinction between claims about the social construction of things and
the social construction of ideas or classification schemes. As a rule,
it is far more plausible to suppose that ideas and classification schemes
are socially constructed than that objects in the world are. We are more
able accept that our ideas about race, gender, illness, science and reality
are socially constructed if we understand that this is not meant to deny
that there are objective truths about these things which are so independently
of our existence. As many philosophers have pointed out, we cannot talk
about brute reality without having ideas of reality, and Hacking is quite
willing to accept that a great deal of our understanding of the world is
contingent and revisable, rather than necessary and fixed.
Hacking is somewhat impatient with the carelessness with which many
social constructionists have set out their ideas, and indeed, he says that
he rarely finds the phrase "social construction" useful. He sets out a
wide variety of ways in which people have argued that the way we think
about the world could have come out differently. In his discussion in chapter
three of the natural sciences, sets out three "sticking points": contingency,
nominalism, and explanations of stability.
The contingency of science is a claim about how science could be different.
Of course, science could have been very different - we might not have had
any science at all, and could still be in a preindustrial age. It is tempting
to say that contingency is about how science could explain the same set
of phenomena as our current science, but differently. This would, however,
be misleading. First, as philosophers of science such as Hanson, Kuhn and
Feyerabend insisted, we can't talk of "phenomena" independent of our theories,
since our understanding of phenomena tends to depend on certain understandings
of the world and assumptions in our methodology, and these depend to some
extent on what theories we accept. The idea is roughly that science could
have progressed and delivered results using an utterly different set of
theories and descriptions of data. Kuhnians might put the point by saying
we could have had a different paradigm than the one we did. Applied to
subatomic physics, this would say we didn't have to describe the world
in terms of protons, neutrons, electrons. We could have had a different,
non-equivalent theory that would have worked according the standards of
success that we used to assess it. Hacking himself is not particularly
sympathetic towards this claim.
Nominalism is about how we could have used a different set of names
and concepts to describe the world. It does not deny that there are facts
about the world, but insists that there are different possible ways of
framing those facts, and that none is fundamentally more accurate than
any other. Hacking says that nominalism, as he defines it, denies that
the world comes with inherent structure that can only be classified one
way. Although Hacking does not say much about this aspect of social constructionism,
he is very sympathetic to it.
The issue of the stability of scientific belief comes down to asking
how it is that scientists come to agree on some theories. Constructionists
argue that this is at least partly explained by purely social and political
factors independent of the nature of the world and the scientific experiments
that have been performed. Scientists who argue against constructionists
on this point insist that scientists come to agree on theories purely on
the basis of evidence, facts, and logic. Constructionists don't deny that
these can play a part, but insist that social factors can play a role too.
Hacking is somewhere in the middle in this argument.
This leads us to the most interesting chapter for those interesting
in the philosophy of psychiatry, on whether mental illness is in some sense
socially constructed. Hacking has elsewhere discussed multiple personality
disorder and other forms of dissociation, and in many ways those are soft
cases, obviously influenced by fashions in the mental health profession.
Here he takes on the harder cases of mental retardation, childhood autism,
and schizophrenia. Hacking focuses on the issue of how we classify these
disorders, and whether the classification changes how people experience
themselves and behave, thus enforcing the classification: if so, then the
people in the classification are part of what he calls an "interactive
Hacking argues that indeed, the label of mental retardation does affect
how people are treated and how they think of themselves. Similarly, autistic
people also come to see themselves differently as a result of the diagnosis
they are given. And again for schizophrenia, Hacking conjectures that there
has been variability to what it is definitional of schizophrenia, and it
has not always been required that it involves hallucinations. In none of
these cases does Hacking go into much detail. He is more concerned with
the philosophical issues here than the empirical. He aims to show how the
classification scheme of psychiatry is contingent and self-reinforcing,
as opposed to simply being utterly objective and inevitable, even when
it comes to the most real of mental disorders.
It is important to be clear that Hacking is not denying that there are
strong neurological elements to these mental disorders. There may be an
underlying pathology, which does is not affected by human classification.
Nevertheless, the classification may affect the experience and self-experience
of patients. For autism, retardation, and schizophrenia, the situation
is still more complicated, because there is quite likely to be not simply
one fundamental pathology, but rather a spectrum of pathologies of different
sorts. Where we draw the lines of our criteria across these spectra of
pathologies is not a matter of scientific investigation, but of our decision,
based on non-scientific factors. Hacking suggests that rather than frame
our investigation as a search for the essence of mental disorders, we can
find out about what is going on at the neurological level of people with
mental disorders and also pay attention to how our classification schemes
affect people, and indeed, ultimately how this affects their neurology.
The suggestions Hacking makes here are programmatic. It will be up to
future researchers to follow them through and see to what extent they are
viable. The ideas that our classification systems are in flux and are subject
to a number of social pressures, and that these classifications affect
both how we treat people and how they experience themselves, are by no
means new. The value of Hacking's work to philosophy of psychiatry is to
provide a more sophisticated way of understanding the issues in an area
dominated by impoverished paradigms -- be they antipsychiatry, psychodynamic
theory, or biological reductionism.
© 2001 Christian Perring. First Serial Rights.
, Ph.D., is Chair of the Philosophy Department at Dowling College, Long
Island. He is editor of Metapsychology Online Review. His main research
is on philosophical issues in psychiatry. He is especially interested in
exploring how philosophers can play a greater role in public life. He is
available to give talks on many philosophical or controversial issues in