No. 37, November 9, 2000

Book Review by Soshichi Uchii:

Ian Hacking, The Social Construction of What?, Harvard University Press, 1999.

Editor: Soshichi Uchii

Ian Hacking, The Social Construction of What?, Harvard University Press, 1999.

Ian Hacking, now one of the big names in the philosophy of science, talks about the "social constructionists", "culture wars", "science wars" and all that. Unlike Alan Sokal (we now have a Japanese translation of Sokal & Bricmont's Fashionable Nonsense, by two Japanese friends of the authors, from one of the most prestigeous publishers*), Steven Weinberg, or Richard Dawkins who are combatants of these "wars" in one capacity or another, Hacking tries to let us "hear from a foreign correspondent, not about the wars, but about an idea that has been cropping up all over the place". Thus, as a "foreign correspondent", he tries to give us fair pictures, cool analyses, sometimes sympathetic, critical at other times.

Topics are quite varied: from natural sciences (chapter 3), madness (chapter 4), child-abuse (chapter 5), weapons reseach (chapter 6), rocks (recent works on dolomite, chapter 7), and Captain James Cook (1728-1779; he made three big expeditions, and on the last, he was killed in Hawaii; chapter 8). But I shall be mostly concerned with the case of natural sciences.

*For this and related discussions, see: http://www.math.tohoku.ac.jp/Pkuroki/Sokal/
The URL mentioned in Hacking's book is not exact. Japanese readers may visit:

Science Studies and Constructionism

All right, what does he say? Maybe it is instructive to know, in advance, his general stance.

Science studies, sociology of scientific knowledge, science and technology studies: these are where the action has been in the philosophy of science over the past few years. I do not mean to belittle specialist studies of quantum mechanics, space and time, systematic bilogy, neurophilosophy, or questions of cause and effect, theory and experiment, probability and induction, and much else. But if we ask what has played the thoroughly lively role, in the republic of letters, of Thomas Kuhn, or Karl Popper, or John Stuart Mill, or Francis Bacon, in recent years, the answer must be science studies. (186)

Here are some samples of science studies: Laboratory Life: the social construction of scientific facts by Bruno Latour and Steve Woolgar, Constructing Quarks by Andrew Pickering. However, Hacking's sympathetic attitude to science studies does not mean he is also sympathetic to social constructionists.

I have seldom found it helpful to use the phrase "social construction" in my own work. When I have mentioned it I have done so in order to distance myself from it. It seemed to be both obscure and overused. (vii)

Yes, I agree! I myself was annoyed by some eminent (Japanese) philosophers and left-wing (or anti-establishment) scientists, when they speak of "social construction", in order to attack Darwinism, for instance, or in order to defend their "up-to-date" or "fashinable" views; what is this thing called "post-modernism" or "deconstruction" and all that? Hacking, however, is cool most of the time, and informs us of the point of the social constructionist.

So what are social constructions and what is social constructionism? With so many inflamed passions going the rounds, you might think that we first want a definition to clear the air. On the contrary, we first need to confront the point of social construction analyses. Don't ask for the meaning, ask what's the point.

... a primary use of "social construction" has been for raising consciousness. This is done in two distinct ways, one overarching, the other more localized. (5-6)

But what is "raising consciousness"? Hacking tells us that "social construction work is critical of the status quo". When X is said to be a social construction, what is claimed is that X need not have existed, it is not determined by the nature of things, and it is not inevitable. Often, it is further claimed that X is bad, and we would be much better off if X were done away (6).

He reminds us of Beauvoir's famous dictum, "one is not born, but rather becomes, a woman"; thus gender is also socially constructed, and it is not as it should be!

Later, another key word is introduced: unmasking. Actually, there are several grades of constructionist commitment, and "unmasking" is also a name for the third grade (among six). Hacking borrows this word from Karl Mannheim, and it is supposed to mean something like this: unmasking "does not seek to refute ideas but to undermine them by exposing the function they serve" (20). A professor's kindness to a female student is unmasked when it turns out this professor (beware, he or she) had merely sexual interest in her!

Anyway, when it comes to natural sciences, it is quite understandable why many scientists get angry with the social constructionists, because their point is, quite often, to denounce sciences, by calling them a "social construction". But be patient for a while, and let's see what are involved in the constructionist thesis for the natural sciences.


Hacking kindly reminds us of the distinction between ideas and objects, and warns not to confuse them (28-31). Confusing them, and arguing (consciously or uncnsciously) on the basis of this confusion is nothing but an equivocation, something any analytically-minded philosophers should avoid. Atoms spoken in physics and chemistry are objects, but the idea of atom has changed historically, and presumably the current idea of atom is "socially constructed", but not atoms themselves (if they do exist). To make things worse, there is another category: what Hacking calls elevator words. "Fact", "truth", or "knowledge" are typical, in that they are often made to work at a different level than those words for objects or words for ideas; and they are also said to be constructed. So let us be careful about them too.

Fair enough! But Hacking (committing this fallacy himself in one of his previous books) diagnoses that concepts, practices, and people interact with each other, and such interaction is often the very point of talk of social construction (29), especially when it comes to social phenomena like child-abuse; that may easily lead to the confusion. Thus we have to ask "what is constructed?" Another distinction (also quite familiar) is that between process and product, which applies to the very word of "construction"; the social constructionist argues that the product is not inevitable by showing how it came into being (historical process), and by pointing out the "purely contingent historical determinants" of that process (38).

One more item for the preliminaries. Hacking points out that Kant was the great pioneer of construction.

Construction brings with it one or another critical idea, be it the criticism of the Critique of Pure Reason or the cultural criticisms advanced by constructionists of various stripes. We have logical constructions, constructivism in mathematics, and, following Kant, numerous strains of constructionism in ethical theory, including those of John Rawls and Michel Foucault. (41)

Thus, by alluding to Kant, Hacking is free to bring such distinguished, tough-minded philosophers as Carnap, Russell, and Nelson Goodman into his picture of social constructionism. I am not quite sure whether this liberal attitude of Hacking helps; does the logical construction by these philosophers share the same point with the current social constructionists?; does van Fraassen's constructive empiricism share the same point? I think not, because these philosophers' criticisms are logical or epistemic criticisms, not social, moral or political (or whatever) which the social constructionists primarily intend.

Sticking Points

Now Hacking touches on natural sciences in relation to constructionism. What's the point of speaking of natural sciences in terms of social constructs? One author claims that the science wars are about "the image of selfless scientific objectivity and authority"; one party tries to undermine it, and the other party wants to recover it. Against this "political" picture, another author tries a peace-making, by finding common ground in the battle of opposition and rage. But Hacking sees three "sticking points"---the points of fundamental disagreement about natural sciences--- in this battle, and he claims that "my sticking points emphasize philosophical barriers, real issues on which clear and honorable thinkers may eternally disagree" (68). Let's listen to what Hacking says, referring to Pickering's Constructing Quarks.

A social construction thesis for the natural sciences would hold that, in a thoroughly nontrivial sense, a successful science did not have to develop in the way it did, but could have had different successes evolving in other ways that do not converge on the route that was in fact taken. Neither a prior set of bench marks nor the world itself determines what will be the next set of bench marks in high-energy physics or any other field of inquiry. (32-33)

Hacking confesses that it is hard for him to state this idea, let alone to believe. But here is the first sticking point, contingency or the denial of inevitability. Here, each side of the combat sticks to his/her own position.

The constructionist maintains a contingency thesis. In the case of physics, (a) physics (theoretical, experimental, material) could have developed in, for example, a nonquarky way, and, by the detailed standards that would have evolved with this alternative physics, could have been as successful as recent physics has been by its detailed standards. Moreover, (b) there is no sense in which this imagined alternative physics would be equivalent to present physics. The physicist denies that. Physicists are inclined to say, put up or shut up. (79)

As Hacking understands, the physicist thinks that if successful physics took place, then it would inevitably have happened in something like our way.

The second sticking point is metaphysical; it is nominalism, or name-ism, which asserts that there is nothing peculiar to the objects picked out by a common name, save the fact that they are called by the same name. The social constructionist endorses one version or another of nominalism, saying that what we call "structure" is not in the world but in our representations of the world. The opponent denies this, and claims that the world itself is inherently structured, and the point of science is to discover this structure (83-84).

The third sticking point is on explanations of stability (84). The scientist's favorite examples of stability are the Second Law of Thermodynamics and Maxwell's Equations of electro-magnetism. Are not they in the world, regulating the course of events in the world? But they involve ambiguity: they may well work, either as an object (a law is in the world, and hence like an object), or as an idea (a law was given its form by so-and-so, thus his/her idea is responsible for that), or as an elevator word (isn't law a general fact, or a truth?). How should we explain such stability? The social constructionist opts for external explanations (that is, external to the content of a science; e.g., social interests, institutionalization, etc., are external factors), whereas the physicist regards external factors irrelevant.

Who sticks to What?

Now I have finished Hacking's list of the sticking points, but I am sure few can see how they work. So let us see Hacking's analysis of Weinberg's statement on Maxwell's Equations (written in connection with Sokal's hoax), in order to illustrate Hacking's point. Here are two statements (of Weinberg's, condensed and divided into two):

[A] Maxwell's Equations in the present form are valid, despite for a limited context; they have survived for a century and may be expected to survive indefinitely.

[B] Maxwell's Equations are an example of the law of physics corresponding to "something as real as anything else we know". While physicists like Sokal and Weinberg believe this, many social constructionists deny the objective nature of scientific knowledge.

On Hacking's diagnosis, [A] is uncontroversial. But [B], which is supposed to be learned from [A], contains elevator words "real", "objective", and "knowledge"; and by their means "we are seamlessly moved up by the elevator" (88). Notice that these words were not contained in [A]. Moreover, [B] involves all of the three sticking points, in one way or another. First, what does this phrase mean: "as real as anything else we know"? Doesn't Weinberg mean that if anyone is to think about electro-magnetism, he or she is bound to end up here, Maxwell's Equations? Hacking confirms this reading by another statement from Weinberg. But, then, this is an inevitability thesis, a denial of contingency, sticking point one! Secondly, what sort of conviction does Weinberg have when he says this? Presumably none other than his belief that Maxwell's Equations express (or correspond to) part of the inherent structure of the world. Then this comes down to sticking point two, nominalism or its denial.

Thirdly, when Weinberg emphasized the difference between physicists and social constructionists (in particular, he mentioned Latour), at least part of what he had in mind was the way the two parties explain the stability of Maxwell's Equations. For Weinberg, Latour and other constructionists deny "the objective nature of scientific knowledge", despite their lip service to "reality" and "facts", because they think that external factors (network of events and agents that lies behind an item of knowledge) are highly relevant to the stability, while Weinberg (and Sokal also) denies this. Thus sticking point three is also involved.

Up to this point, Hacking's analysis in terms of sticking points looks helpful. And he adds an amusing check list, for measuring your degree of "social constructionism"; for each of the three sticking points, there are five grades 1 to 5, and 5 means you stick on the constructionist side, 1 the opposite side. Hacking's own list is:

(1) Contingency: 2
(2) Nominalism: 4
(3) External explanations of stability: 3

However, I find sticking point three still ambiguous. For there are a variety of "external explanations" among the constructionists, and likewise there are a variety of the denial of them. Moreover, even among the sympathyzers of science, "non-external" explanations of stability may be ambiguous. For instance, one may ask "Stability at what level?" (Recall Maxwell's Equations are now different from James Clerk Maxwell's originals). The realist would understand Maxwell's Equations literally, and would attribute stability to the whole structure of this rendering, whereas the non-realist (like Mach or van Fraassen) may divide the structure into two parts, the "observational" and the rest, and would prefer to attribute stability to the former part only. Of course, this move changes "external/nonexternal" division accordingly, and somehow go hand in hand with sticking point two. In a word, both "explanation" and "external" are ambiguous, so that the sticking point three is doubly ambiguous. That may be one of the reasons why Hacking gave "the ambivalent score of 3" (234), although he gave a quite different reason (apparently, at least) for that.

Be that as it may, this book is written in a fair and cool manner, and it would be useful for both combatants and noncombatants in the "science wars" and in the "culture wars" (I have omitted chapters related to the latter) as well. In addition to being a good "primer" in this field, it contains original and valuable analyses, and as is usual with Hacking's books, it is elegantly written.

Last modified Nov. 30, 2008. (c) Soshichi Uchii