Philosophy of Science

Values and Science
(note prompted by, but not exactly about, the Sokal affair)
Curtis Brown


Representation and Reality (Word and Object)

We need to distinguish between the real world and theories about the real world. Many works in science studies seem to play fast and loose with this distinction – for instance providing evidence that theories are social constructs and concluding that the world is a social construct. (Even Kuhn comes close to doing this when he argues that before and after a scientific revolution, scientists live in different worlds. But Kuhn stresses that although scientists live in different worlds, the world itself remains the same. More extreme is the case of Latour and Woolgar in Laboratory Life.)

Context of Discovery vs. Context of Justification

In very broad terms, there are two stages of scientific inquiry: constructing a hypothesis or theory, and testing it. (Or maybe three: asking a question, constructing a hypothesis, testing it.) Theory construction is clearly subject to all sorts of social forces; it is influenced by the ideas in currency in a culture at a time, by the concepts available to frame questions, by what questions are deemed worth pursuing and which are seen as a waste of time. This poses no problem for scientific objectivity. But the study of the influences on question formation and theory construction are certainly appropriate subjects for sociological inquiry. What makes science distinctive is not theory construction, but rather theory testing or justification. And it is with regard to justification that science attempts to minimize the role of social and ideological factors. To the extent that one’s views about the truth of a theory are influenced by one’s values or ideology, one is doing bad science. But this is precisely what many “science studies” advocates appear to believe is impossible.

Such advocates argue that the values of objectivity and rational debate are themselves just one more self- (or class- or gender- or whatever) serving ideology, no better or worse than any other at discovering the truth (if indeed there’s any such thing as truth to be had). I think this is extremely pernicious. I agree that it’s extremely important to discover the extent to which one’s ideology, preconceptions, prejudices, and so forth influence one’s views about the world – but it is important to be self-conscious and critical about this in order to devise techniques to prevent these ideological concerns from influencing one’s assessment of the evidence. Science as a social enterprise has many ways of doing this – replication of experiments, peer reviewing of journal submissions, blind or double-blind experiments, and so on.

(Aronowitz seems to me to fall into something like this trap when he suggests that the views of the Bell Curve are mistaken partly because they are incompatible with democratic ideals. This is to turn science into nothing more than ideological dispute; conservatives can argue that intelligence is highly heritable because that provides a potential explanation of social inequalities that doesn’t involve discrimination, while liberals can argue that intelligence can’t be highly heritable for the very same reason. And since ideologies are difficult or impossible to confirm or disconfirm, this turns science into nothing more than a shouting match. But it doesn’t need to be like this, and shouldn’t be like this. If Herrnstein and Murray are wrong, then further empirical work can reveal this, but accusing them of racism doesn’t prove anything one way or the other.)

Knowledge vs. Certainty

I have the impression that much of the literature on these topics simply confuses these two different notions (or perhaps assumes with Descartes and other early modern philosophers that knowledge requires certainty). Critics of science provide reasons to think we can’t be certain about the world, and emphasize that we need to be critical and questioning regarding claims to truth. So far so good; but they often conclude that knowledge is impossible. A view that has only two categories, knowledge or conjecture, and then interprets knowledge in such a way that it is impossible to attain, makes all theories equal: creation science becomes no different from evolutionary theory, astrology becomes no different from astronomy, and so on.

Ways Values May Affect Science

Things social factors may influence:

1. what questions get asked
2. what concepts we use to frame the questions
3. the hypotheses we construct
4. the extent to which we regard the hypotheses as confirmed or disconfirmed by the evidence

all these are worthy of study, but only the last threatens the objectivity of science. (Except for the important point Okruhlik makes, that 3 may indirectly affect 4: To the extent that theories are only evaluated relative to rival theories, we may overestimate the value of a given theory because the really good competitors never get considered.) My own view is that 4 does sometimes occur, but that when it does we are dealing with pathological science; when science is at its best it succeeds in screening ideological factors out of the assessment of the evidence. (Maybe we can’t be completely successful in this, but we should certainly keep it as an ideal.)

Last update: March 28, 2005
Curtis Brown | Philosophy of Science | Philosophy Department | Trinity University