Social Construction

People have, justly, gotten extremely defensive about what various social constructivists are said to have said. The same thing happened to Kuhn, though there, perhaps, there is a stronger case to made that he said some stuff he shouldn't have. In his defense he was the first one to walk into this, and he didn't realize he needed to be so careful.

Of course scientific theories are in some sense reflecting discoveries about the objective world, and of course the job of the scientist is different from that of the novelist: scientists not only don't think that all that they are trying to do is produce good stories, but that isn't all that they are doing. So far as I can tell, everyone (even LaTour and Woolgar) agree to that.

BUT there has a been a tradition according to which what scientists do is channelling reality. The Baconian picture of science has done a lot of harm in that respect: the story is that Bacon thought that if only you looked at reality hard enough, it would reveal its secrets to you directly, no creativity required. Bacon himself never thought anything like that, though he did sometimes sound like he did, but there is a long tradition of attributing such a view to him and, worse, of actually adopting such an attitude about science. The logical positivists "Egypticism" and empiricism betrayed a similar attitude: if one just proceeded according to the scientific method, Nature would reveal all secrets. One important reason Popper (falsificationist) was popular among scientists was that he emphasized that science isn't mechanical and automatic, creativity is required.

Kuhn, Bloor, LaTour , Woolgar and the rest emphasized and got excited about a feature of science that had been neglected: it is done by scientists, who are human and lead lives outside (and sometimes even inside) the laboratory. Scientific work is influenced by economic concerns (will it get me tenure, a raise, a grant, ...), political concerns (does this make my friends look good, so I can hire them), and social and political concerns of the larger society. These are facts that had been hidden from official discourse about science. They weren't supposed to have anything to do with the scientific method, and so forth. Those who have been concerned to emphasize that science is done by people have not been concerned to emphasize that science explores the natural (as opposed to human) world. Everyone already knows that. They are frequently read as not knowing that, as claiming that science is a purely social game or system that makes no contact with anything objective.

The such attacks we find, the more disclaimers we see in the work of social constructivists. Somehow, it never seems to be enough.

Suppose you are a sociologist studying the beliefs of a society and the way that society maintains and modifies its system of beliefs. That is what Bloor takes himself to be doing. How do you do that? You figure out who counts as an authority, about what, how authority is acquired, how authoritative beliefs are disseminated, what background beliefs and methods play an important role, and so forth. You might talk about shamans, secret teachings, apprenticeships, whatever. One thing you don't talk about is whether the beliefs are correct, the methods good ones, and anything else of a similarly evaluative sort. Why not? That isn't the task at hand. A sociologist is trying to get an accurate picture of how a system works, not to figure out whether to adopt it, try to change, or anything else of that sort. Those kinds of questions may be important, but it is a good idea to figure out what a system is before evaluating it. That means that, in trying to study the sociology of knowledge systems, the symmetry principle, the principle that the explanation of what is believed should be independent of whether what is believed is true or false. That does not mean that it isn't either true or false, or that it is unimportant whether it is true or false, only that that consideration is inappropriate for the task at hand.

Bloor had the temerity to apply these insights to science, considered as a sociological knowledge system. He talked about authority instead of truth and pissed off a lot of scientists. They, or their philosophical defenders, said Bloor was ignoring the most important part: that the scientists were right. He was doing his damndest to ignor e that, that was the whole point. That didn't mean that he suspected or thought that that wasn't also an interesting fact about science. It wasn't what he was interested in. To the extent that one can learn interesting things about science using sociological methods, perhaps one doesn't need other explanations of those things, and so, perhaps, one might take his work as an implicit criticism of bothering to do philosophy of science. That reaction is often misguided. Example: Peer review is important to evaluation in science, more important than whether work is really really true. That does not mean that truth doesn't matter, it does mean that it might be worth thinking about what peer review has to do with determining truth.

---

Latour and Woolgar studied a laboratory and gave a sociological report about it that viewed it as a system (machine) that takes certain inputs (rats, blank paper, money) and produces facts as outputs. They viewed the whole process as being an arcane and slightly bizarre system of rituals much as a disapproving outsider might describe predicting the outcome of a battle by reading entrails. They talk about the process as one of separating things from their origins and hiding that they were the result of a process, in Hacking's term, of Egypticizing them.

In a simple case, consider the difference between,

It seems to me that whenever I put a pair of things someplace and then another pair, I, at least usually, wind up with a foursome of them

and

2+2=4.

The two items have, in a clear intuitive sense, something in common, but the second has been promoted to an "objective" fact. That is what the they view the scientific process as doing, but they completely omit considering whether each step toward facthood is well motivated, a good idea, anything like that. The ways in which they go further then Bloor include, at least, omitting the information that might be relevant to seeing whether it is a good idea and declaring that the process they describe is all there is to "manufacturing" facts.

On traditional accounts of science, facts are discovered. They are aspects of the world there for uncovering. As we have seen, science isn't as clean as that. Not only nature, but we play a role in what we take to be facts. Facts are in part, manufactured. Latour and Woolgar, however, seem to view scientific facts as nothing more than social artifacts. They do say, at least occasionally, that manufacture is a process constrained by what will work, but they minimize attention to that detail. They deny that nature or reality impose any constraints on the construction of facts because reality and nature are themselves constructed, but in order for an artifact to become a fact it must "stabilize," and there are pragmatic considerations in what will stabilize.

-- ShaughanLavine - 23 Feb 2006

Latour and Woolgar: "Facts and artefacts do not correspond respectively to true and false statements. Rather, statements lie along a continuum according to the extent to which they refer to the conditions of their construction." 252

If one misunderstands the claims Latour and Woolgar make about the social construction of facts as claims about truth and falsehood, it looks like they have omitted any objective component from what they take to be facts. But what they are actually discussing is the transition from artifact to fact. Both artifacts and facts are "real" in the sense that they are actual events in a laboratory. It is not up to us what results when, say, you mix two solutions. The result either blows up or it doesn't. Latour and Woolgar never deny that. But, as we'll see in detail for a lot of the rest of the semester, there is in a lab frequently a question about whether what happened is a result of an error, an impurity, a mistake, a misconception, and so forth, and so a lot of laboratory work is devoted to checking whether what happened is worth reporting as fact. It isn't worth reporting that, say, when Francis mixed the stuff in the yellow jar on the table with the stuff in the cabinet, it blew up. What one wants to report is that if nitric acid and glycerin are mixed, the resulting liquid is explosive. In this case, probably, most of what is relevant had already been established.

That is not true in the case of TRF(H). The relevant substances had to all be established in the lab. That is, one had to find a way to make them, characterize them, and so forth. To say "make them" suggests that there were substances in advance, but what substances were being made? The characterization of what would count as the right substance was developed at the same time as the procedures for producing it. In that very real sense, the substance glycerine, for example, is clearly a social construct: whether you can make it only from animal fat, or vegetable fat will do will depend on whether those differ in explosive characteristics.

A big part of what goes on in laboratories is standardizing processes, tests, and substances so that one can describe manufacture and assay in a reproducible way. Nancy Cartwright talks about the laws of physics as being obeyed by "nomological machines": you need the lab equipment to get the laws obeyed.

Once you have a good set of standards and procedures, one that is "stable" you can forget about their origins and detach their results from the procedures. Once you have figured out what to count as glycerine, there is just this stuff, glycerine, now thought of as independent of its original characterization and manufacture. It is nonetheless not helpful in explaining the original introduction of glycerine to describe it as the "discovery" of something that was antecedently there. Why not? That is a part of what was in doubt during the discovery, and so such a description misses out on what it was like deciding that there was such stuff in the first place. That kind of triumphalist after-the-fact description makes excellent textbook history: It is a good way of introducing people into the current theory, practices, devices, and so forth, so they can use them. But it is a lousy way of seeing what people saw, thought they were doing, at the time.

I think all of that is right, I think Latour and Woolgar played an important role in showing that it is right, and I think that failure to realize it has been the source of a lot of bad history and philosophy of science. In particular, we've talked a lot this semester about experiment and observation, but always as if it were a simple, one-or-two stage process: you set things up and you see what happens. It is that simplicity, in part, that makes experiment a candidate for deciding between theories. But the simplicity is an illusion.

Latour and Woolgar do not say that science is a social construct unconstrained by any facts. They think that there genuinely are facts, and that science is constrained by them. It is just that the facts are socially constructed.

---

Having said all that, lots of social constructivists, probably including Latour and Woolgar, have said lots of stupid, incautious stuff that makes it sound as if they think that science is just a certain social game. I'm pressing what I'm pressing because they have been so soundly reviled that there is real danger on missing what they have to say that is useful and important.

---

James Brown

He criticizes social constructivism as nonsense and the root of much nonsense. He views criticizing social constructivism as the intellectual equivalent of shooting fish in a barrel, and so he, time after time, fails to take careful aim.

He claims Latour and Woolgar believe

TRF(H) exists if and only if bioassay B is accepted 261

Later on the page he admits that "perhaps this is an uncharitable reading" since "the exact claim is 'Without a bioassay a substance could not be said to exist.'" That is not the same thing at all. Misquoting is not usually a good way of establishing your academic superiority to your opponents. Worse, he misinterprets the correct quote. He takes it to mean that without a bioassay we have no grounds to assert the existence. (paraphrasing) But what they actually mean is that without a bioassay we have no way of characterizing what we are looking for in trying to see if there is such a substance as TRF(H).

On p. 262, Brown says "the crucial thing is that they are independent reasons for thinking that the particular bioassay adopted is the right one ... we do not have a little circle consisting of only two propositions ... Once we allow that people have to work ... with facts which are _not constructed by them, we might as well admit the possibility that sometimes people have to work with facts which are _not constructed, simpliciter.

---

Boghossian

He accuses "the postmodernists" of an elementary mistake in claiming that the Zuni origin myth is "just as valid" as prehistory based on archeology.

It is sometimes suggested that the intended sense in which the Zuni myth is "just as valid" has nothing to do with truth or justification, but rather with the different purposes that the myth subserves ...The trouble with this as a reading of "just as valid" is not so much that it's false, but that it's irrelevant ... If I say the earth is flat, and you make no assertion at all, but instead tell me an interesting story ... 271

That totally ignores the plausible claim that the myth cannot serve its purpose unless it is taken to be true, not "just a story."

I think that many of the claims opponents of postmodernism and social constructivism have made are true and important. However, the opponents have tended to argue against claims that either postmodernists and constructructivist have never actually made, or only argue against the most extravagant claims made in incautious moments, skipping over many interesting claims similar to those that get dismissed. Worse, in polemics against postmodernists and social constructivists they argue badly and unfairly.

---

If anyone thinks that scientific facts are in no way constrained by anything other than social factors, they are wrong. On the other hand, if anyone thinks that scientific facts are unconstrained by social factors, but only by facts of objective reality, they are equally wrong. Moreover, appeals to objective facts are nearly always misleading in accounts of how those facts were established. That becomes clear when one takes seriously that we are sometimes wrong, that what we take to be established facts, established in the usual way sometimes aren't.

Here is an example of a scientific fact: at fixed temperatures and low pressures, pressure is inversely proportional to the volume of a gas. That is a fact about pressure, volume, and gasses. The fact is independent of human intervention once one has identified the relevant kinds: gasses, volume, pressure and put in place procedures for preventing and detecting leaks. The point is that what we take to be relevant kinds is socially constructed. But the fact is only a fact about the kinds.

-- ShaughanLavine - 02 Mar 2006