Critical Realism
Deena Higgs
http://www.philosophynow.org/issue42/42caldwell1.htm
[- Is Postmodernism finally on its deathbed? Roger Caldwell examines
the evidence and takes a look at its would-be successor: Critical
Realism. -]
For the last two decades of the twentieth century the dominant
cultural paradigm was that of postmodernism. But at the beginning of
the new millennium a new paradigm is on offer.
Postmodernism is dead. It is to be succeeded by the age of critical
realism. That at least is the promise that José López and Garry Potter
hold out as propagandists of the new movement (they edited a
collection of essays called After Postmodernism - An Introduction to
Critical Realism, published by Continuum in 2001). True, the two
movements have much in common in their sheer scope — offering an
overall view of science, social science and the arts, and all in the
interests of an emancipatory politics. However, although postmodernism
made an easy transition from academia to the media, critical realism
has shown to date no signs of doing so. From this, however, no adverse
inference should be drawn as to the quality of its thought.
The talk of paradigms recalls the term used by Thomas Kuhn in The
Structure of Scientific Revolutions: for him long periods of 'normal
science' were punctuated by crises leading to 'paradigm shifts'. For
Kuhn competing paradigms were incommensurable: they involved looking
at the world in radically different ways. Certainly, the world looked
at through the eyes of critical realism is vastly different from that
seen through the eyes of postmodernism — for a start, there is a
single world again — but there is more to the matter than an
irrational leap from one view to the other. For critical realism
begins with the awareness that the postmodernist project is fatally
flawed.
There is the danger of anachronism here. Roy Bhaskar may be regarded
as the founding father of critical realism, yet his first book, A
Realist Thought of Science appeared in 1975 when postmodernism was
still in its infancy.
Nevertheless, the central targets of the book, Kuhn and Paul
Feyerabend, were undoubtedly (and perhaps unwittingly) forerunners of
postmodernism in their questioning of scientific rationality. Of the
two it was Kuhn who was the closest to realism — he held that even
after a revolution at least part of the previously 'normal' science
proves to be permanent, and that science offers us our surest example
of sound knowledge. Indeed, it is hard to see in what way there could
be a growth of scientific knowledge except from a realist stance,
however finely nuanced that claim to realism may be.
Feyerabend, acting as a gadfly to all scientific pretensions, held
that there was no such thing as the scientific method and saw science
as an essentially anarchic enterprise in which 'anything goes'. The
one scarcely follows from the other, however. It is true that there is
no single method that marks out science from any other form of
rational enquiry but nonetheless there are a range of criteria — such
as explanatory scope, predictive power, experimental repeatability,
consistency with other well-established theory — that make it a
different sort of enterprise to, say, astrology or alchemy. Feyerabend
could scarcely have expected that his remark that "science is the myth
of today", intended no doubt as a provocation, would so soon become
orthodoxy, at least in the Humanities.
If philosophers outside science were led in an anti-realist direction
there were also developments within science itself — notably the
enigmas of quantum physics — that seemed to go against the normal
assumption that there is a single observer-independent reality. The
Copenhagen interpretation of quantum physics remains the most popular
one but it doesn't take us very far; it confirms that the equations
work but doesn't provide a physical model to account for their
success. Attempts to relate the collapse of the wave function in terms
of the 'real' world — such as Hugh Everett's many-worlds model — seem
unconvincingly extravagant. Much of science is counterintuitive, but
the notion that whole new universes are continually splitting off, for
all that it has eloquent defenders, would seem in need of firmer
foundations to be persuasive.
Here Christopher Norris (once a prolific writer on postmodernism and
now an avowed critical realist), rather than accepting the notion of a
universe which is dependent on human observers to exist, attempts to
bring quantum physics within the embrace of realism. There are several
points to be made here. If no realist model of quantum physics has yet
been agreed on, this may be because the science itself is incomplete,
or because no one has yet devised a suitable model, or because we have
yet to decide between competing models. (I understand that in the last
decade realistic models have been devised that don't demand the
extravagance of a 'many-worlds' interpretation). Also, even if no
agreement has been reached on an interpretation of quantum physics,
its capacity for precise physical prediction and the fact that it has
given rise to sophisticated technology potently suggest that it has
latched on to certain objective underlying features of physical
reality. It is further worth pointing out that the particular problems
of quantum physics don't carry over into the rest of physics or into
chemistry or biology, much less constitute any kind of general
scientific crisis. The existence of stars and planets, of DNA, of
human bodies and animal bodies is not thereby put into doubt, nor is
the validity of the considerable body of scientific knowledge we have
developed about these entities. Whatever problems there may be at the
subatomic level do not affect our ability to devise realistic theories
of the macroworld.
This excursion into quantum physics is necessary because
postmodernists have drawn unwarranted conclusions about a general
epistemological crisis from, for example, Heisenberg's Uncertainty
Principle. In fact, as Sokal and Bricmont have shown in Intellectual
Impostures (1997) these conclusions are invariably based on a lack of
understanding of the relevant science. If postmodernism is indeed dead
— the announcement may yet prove premature — then Sokal and Bricmont
have surely been instrumental in hastening the death-throes. They show
that, scientifically speaking, the postmodernist gurus have feet of
clay.
Indeed, it is hard to give an overview of the major postmodernist
tenets without seeming to fall into parody. All knowledge, scientific
knowledge included, is held to be socially constructed through and
through. Science is therefore merely one story among others. The world
we know is one that is constructed by human discourses, giving us not
so much truths as 'truth-effects' which may or may not be
pragmatically useful. From this point of view, epistemologically
speaking, a scientific text is understood as being on a par with a
literary text. Further, given that for Derrida language is a
self-referential system, all communication is reduced to the model of
an avant-garde poem in which all meaning is indefinitely deferred.
So put this seems scarcely persuasive. (Indeed, as Garry Potter points
out, this is not even a plausible account of an avant-garde poem: if
there are no inherent meanings in the text it is not properly a text
at all but indistinguishable from an arbitrary jumble of words.) More
basically, a denial of realism can take two forms: the first is to
accept the possibility of there being an objective reality but to deny
that we are in a position to have knowledge of it; the second — more
typically postmodern — is to see reality as entirely composed of our
discourses about it. The effect of either form is that we no longer
are in a position to talk of reality or truth as such: rather, both
words are, as it were, to be put in inverted commas.
Clearly, if we adopt the latter form of anti-realism, we should have a
magical solution to all our problems. For example, as Ted Benton
points out, if nature were merely a cultural construct, all we would
need to solve our ecological problems would be to change the terms of
our discourse. That these theorists do not in fact take this step
suggests that they (for good reasons) fall shy of the consequences of
their own theories.
One may question whether it is even possible to state theories of this
kind without self-contradiction. If objective truth about reality is
impossible, then what is the logical status of the statement that
objective truth about reality is impossible, since it itself aspires
to objective truth? A similar problem arises, as Bricmont points out,
with regard to Richard Rorty's neopragmatism. If, as Rorty proposes,
we replace the notion of truth with that of usefulness, so that we
accept only those propositions which we find in general to be
'useful', then the question arises as to whether they are really
useful or not. That is, the very criteria by which we judge a
proposition to be useful involve the same recourse to a correspondence
with reality which the theory denies us in advance. We are left,
inescapably, with the conclusion that the theory is incoherent.
Critical realism, then, rescues us from the postmodernist nightmare
and restores us to reality. We cannot manage without a concept of
truth. There is (as most of us thought all along) a pre-existing
external reality about which it is the job of science to tell us.
True, we must be cautious about claims to objective reality, alert to
ideological distortions, and aware that the world is a messier, more
complicated place than the accounts of physicists would suggest. This
does not mean that such claims cannot plausibly be made. A central
plank of critical realism is that science can no longer be considered
as just another myth or story.
Ted Benton is concerned to restore the centrality of the concept of
nature to the social sciences. He notes that, among sociologists,
there is an ambiguous attitude to the natural sciences, debunking on
the one hand but envious of their success on the other. The notion of
nature, and for that matter human nature, tends to be seen as
essentially a social construct, which means that we can never speak of
nature as such but only of discourses about nature. The result of
this, combined with a suspicion of scientific thought as indissolubly
linked with political and social domination, is that sociologists are
powerless to contribute to debates about such important contemporary
issues as loss of biodiversity or ecological degradation, assessment
of which is crucially dependent on scientific analysis. If
sociologists deny the validity of a scientific account of nature to
begin with, dissolving 'nature' into so many discourses, they are left
with a hapless relativism, inadequate to deal with the 'real' problems
that clearly exist. This is not to deny that science may be put in the
service of political or social oppression, or indeed that
scientifically-based remedies may be inappropriately applied. The
answer to this is better political systems and more finely tuned
application of science. It does not constitute an argument against
scientific truth as a whole.
Bhaskar himself tends to argue on an ontological level (he asks what
kinds of entities — natural and social — exist) rather than on an
epistemological one (that is, asking what different ways there are of
arriving at knowledge). There are good reasons for this. If scientific
method does not differ essentially from other ways of determining the
probable truth of a state of affairs, then it is hard to see how there
can be competing epistemologies. Think, for example, of a murder
enquiry: X has been shot, and the evidence available suggests that it
was Y who did it. He was known to have a grudge against X, he had
previously threatened to shoot him, there is good DNA evidence, and he
was seen standing over the body with a smoking gun. Then, in the
absence of evidence to the contrary, it would surely be rational of
anybody to conclude that X was murdered by Y.
Now, this judgement may be wrong: later there may come conclusive
evidence that it was Z who did it, having cleverly incriminated Y. In
which case our conclusion will be revised accordingly. But any
investigator considering the available evidence should reach the same
conclusion regardless of their age, gender, race or sexual persuasion.
That is what we mean by objective truth. If this is true of a murder
enquiry it is surely true of how progress is made in the physical
sciences: just as X was murdered either by Y or Z or somebody else, so
the speed of light is either one value or another. It cannot be the
case that the speed of light has one value for one theorist and
another for another: either one or the other (or both) are mistaken.
It is considerations of this sort that make nonsense of Luce
Irigaray's notorious question; "Is E = Mc2 a sexed equation?"
Equations cannot be sexed like humans or chickens: the equation in
question is true or false, regardless of who discovered it. It just
happens that it was Einstein: it could easily have been someone else.
Clearly, critical realism is by now a diffuse and interdisciplinary
movement, covering a wide spectrum of opinions. The question is: how
broad a church can critical realism be if it is to remain both
critical and realist? Most of the contributors to López and Potter's
anthology clearly accept scientific objectivity: it is far from clear
that the contributors to the section entitled 'Ways of Knowing' are
similarly committed. Jenneth Parker, invoking Lyotard, Feyerabend and
feminist epistemology, explicitly argues that the 'reductionism' of
Western science derives from the economic and political organisation
in which it is embedded. This reductionism has allegedly led to the
loss or marginalization of less privileged knowledge-systems. This may
be so, but the term 'knowledge-systems' rather rigs the question in
advance. If instead we talk of belief-systems — which say cover, for
example, witchcraft, Christianity, astrology, not to say science
itself — we can then ask the crucial question: are they true? For only
then can they become knowledge-systems proper. And I'm not clear on
what basis Parker could decide this.
She argues that Western science should not be privileged over, say,
acupuncture; that to include both is likely to lead to a better
understanding of the human body. Again this may be so: acupuncture is
clearly widely-practiced and may have beneficial effects on health.
But how many ways of understanding how a human body works can there
be? If behind acupuncture there lies genuine knowledge about the human
body so far unrecognized by science then the only rational procedure
for scientists is to modify their theories so as to take this new
knowledge into account. If this is thought to privilege the hegemony
of science then I make the alternative proposal: that if there is
genuine knowledge in Western science about the human body not
previously taken account of by acupuncture (and,of course, relevant to
healing) then it is only rational of acupuncturists to incorporate
that knowledge into their practice, if it is possible to do so. The
result is, in the first case, that science remains science, but better
science. The result is, in the second case, that acupuncture becomes
more scientific.
Parker is arguing for pluralism. However, whilst there can be, and
obviously is, pluralism in regard to the values which particular
societies endorse, it is unclear in what way there can be a pluralism
in regard to truth. Obviously, there is a pluralism of ways of looking
at the human body — an artist, a sexual partner, a surgeon will all
look at it from very different perspectives. But it seems to me that
only the biologist is in the business of explaining how the human body
functions. It is theoretically possible that at any one time there may
a number of competing biological theories, but only one (or none) of
them is likely to be correct. A plurality of ways of looking does not
translate into a plurality of ways of knowing.
Alison Assiter, writing on Descartes, adopts similarly dubious
tactics. She argues that Descartes' philosophical project foundered on
its failure to take other people and their beliefs into account, and
on Descartes' own assumption that he could isolate himself from his
particular values and beliefs to produce knowledge. She further argues
that Descartes' ultimate reliance on God is a result of his having
severed any dependency on anything else, and that, from the standpoint
of feminist epistemology, there is no 'project of pure enquiry' but
that all enquiries are dependent on a social context. One may well
agree that Descartes failed in his project, though scarcely for the
reasons she gives. Assiter is here falling back into positions that
are closer to postmodernism than to that of critical realism.
If the latter involves, as she says, "a socio-historical situating of
knowledge", there is a singular failure in her essay to locate
Descartes' own philosophical project socio-historically. The main
purpose of what we now think of as Descartes' philosophical works was
to establish a certain foundation for his physical science which he
hoped, nervously aware of the fate of Galileo, would be acceptable to
the Catholic Church. In this, as we know, he failed: for all his
efforts to placate the Church, his works were placed on the Index of
Prohibited Books. In his scientific work, like Galileo, he attempts to
provide an account of the natural world in the light of human reason
and independent of theology. If this is to be acceptable to the Church
then he must find a way of showing that human reason is somehow
guaranteed by God, that God is not a malignant trickster.
If we approach Descartes' project historically in this way it is easy
to see that Assiter's charges are misconceived. Descartes' dependency
on God, in the context of his period, is scarcely a pathological
matter, requiring a Freudian reading. The method of hyperbolic doubt
is a heuristic device for a particular end, not a universal
prescription: it is, I agree with Assiter, not something that school
teachers should recommend to their charges, but Descartes would not
have recommended it either. She finds it strange that Descartes, as a
practising scientist, should not have emulated the procedures of the
sciences in seeking help from others. This is anachronistic with a
vengeance: there were in Descartes' time no scientific institutions in
our sense. Science was necessarily carried out by individuals in
isolation. Indeed, contra Assiter, individualism in this sense has had
a rather successful track-record in science.
Whether or not there is 'a project of pure enquiry' one only has to
think of the achievements of Newton, Darwin, Mendel, Einstein to doubt
Assiter's recommendation that truth is best validated in collectives.
What Assiter is mainly concerned to do, however — and here Descartes
is only a convenient whipping-boy — is to advocate "the more
collective, cooperative, self-reflective" approaches advocated by
feminist methodology as exemplifying scienticity. It is hard to see,
however, in what this methodology consists, or what defects in
non-feminist methodology it seeks to remedy. Assiter invokes the
insights of Sandra Harding for whom feminism requires us "to reinvent
science and theorizing". The achievements of 'feminist science',
however, as Susan Haack reminds us, have been unimpressive. Harding
tells us that, thanks to feminist scientists, "we now know that
menstruation, pregnancy, and menopause aren't diseases." We may wonder
what other great discoveries are to follow.
Haack argues in her book Manifesto of a Passionate Moderate (1998)
that the profusion of incompatible themes offered as feminist
epistemology itself speaks against the idea of a distinctively female
cognitive style. Besides, if there are insights available to women
that are not available to men, it is hard to see how men could even
come to understand what feminist science is saying. The same, of
course, applies to women attempting to understand masculinist science.
But if we are to posit different epistemologies for men and women,
what logical reason have we to stop there? Are there perhaps 'gay'
ways of knowing as opposed to 'straight' ones? Are there perhaps black
as opposed to white ones, urban as opposed to rural ones, childrens'
as opposed to adults'? And so on. If so we are each of us a confused
site of many ways of knowing. The question is: what difference does it
make to our ability to arrive at objective knowledge? I fail to see
that it makes any difference at all. Assiter may well argue that it is
impossible for us, Descartes-style, to strip ourselves of our social
context, of our assumptions and values. In one sense it is. But if
your purpose is, say, to discover the structure of DNA, they are not
going to be of much use to you. As we know, Watson and Crick made the
discovery; it could have been — and, as we know now, nearly was —
Rosalind Franklin.
This was not a triumph of a male cognitive style over a female one. If
Franklin had made the discovery it would not have been the triumph of
a female cognitive style over a male one. In either case the structure
of DNA is a double helix.
To postulate the existence of competing epistemologies in the way that
Parker and Assiter do is surely regressive — it involves a fracturing
of knowledge and, by implication, leads to the relativistic impasses
that are characteristic of postmodernism. It invites the suspicion
that not all of those who now choose to operate under the banner of
critical realism have the right to do so — they have changed the label
but not the brew. It perhaps illustrates too that to go over from one
paradigm to another is a messy business, and takes time. To the degree
that critical realism has broken free of its successor it is surely to
be welcomed — we have reality once again, and we have the possibility
of progress in knowledge. We have (potentially) a social science that
operates on the basis of a realistic conception of the natural
sciences. There is at last light on the horizon. On the fringes of the
movement there may be a few dubious practitioners who wish to return
us to the postmodernist night in which all cows are black. But the
centre seems firm enough, and we can only hope that it will hold.
© Roger Caldwell 2003
© 2003 Philosophy Now.
~~~~~~~~~~~~~~~~~~~~
Not for commercial use. Solely to be fairly used for the educational
purposes of research and open discussion. Content and/or links do not
necessarily reflect the views of the poster.
~~~~~~~~~~~~~~~~~~~~