Strong and weak theories in the social sciences

There is something funny about how the word “theory” is used in the social sciences, and especially the humanities.

One hears of attempts to “theorize” gender, or power, or postcoloniality, or the erotic, and anything else one can think of. Shelves full of college books hold the label “theory.” One often reads of Person X’s attempt to theorize Y. As in, Lacan’s or Girard’s theories of desire, Foucault’s theory of power; Saussure’s theory of the sign; Geertz’s theory of culture; Lukacs’s theory of commodity fetishism; Butler’s theory of gender; Said’s theory of orientalism, and so on.

But in each of these cases, the word “theory” can be replaced with “idea” or “description” or “account” without loss, as far as I can see.

It would appear that the term “theory” is used merely to grant arbitrary prestige to the idea in question. The conclusion is that anyone in an institutionally powerful position (access to publication, institutional prestige) who writes a few sentences about topic X instantly becomes X’s “theorist.” All ideas, good or bad, true or false, demonstrable or merely speculative are instantly described as “theories.”

Fine then, if we want to use the term “theory” for “ideas about something,” let’s at least recognize the distinction between what is traditionally described as a theory in the academy.

The normal requirements for a set of ideas to be called a “theory,” as in Darwin’s theory of evolution or Einstein’s theory of relativity are, at minimum, descriptive and explanatory adequacy, with a search to provide predictive power.

When a descriptive account, a model, adequately accounts for all available facts on the topic, it achieves explanatory adequacy. When it can make predictions whose accuracy can then be confirmed, it achieves predictive power.

Hence, the model for planetary motion can be shown to accurately describe the known data on planetary movement within our solar system, and can successfully predict the position of the moon next Wednesday, and the rotation of the earth for the next year. Our models for weather are less successful, but they have a high enough success at predictive power that we still watch the news each morning to decide whether to bring an umbrella. The predictions delivered are not always as precise as we would like, but they are certainly more accurate than anyone could have made a century ago. So here are two examples of strong theories that achieve some level of explanatory adequacy and predictive power, and are thus viable “theories.”

When it comes to the social sciences, there is nothing even approaching this. Theory of democracy? Theory of power? Theory of sexuality? Predictions of events or behavior? An increase in accuracy over time? Are we verifiably better at describing desire, power, gender, or democracy than we were a century ago, or a millenium ago?

There is no rigorous description that comes close to being able to account in precise terms for the known phenomena in a given area. This is compensated for, justifiably, by using vaguer terms, since precise ones (in the sense that a biologist would understand precision) don’t get us anywhere. But then we have nothing like a “theory” of anything, we have only a vague sense of something. As one can see on the pages of Nature or Science, the leading science journals, we can’t even describe insect behavior accurately in any real theoretical sense. Human behavior, infinitely more complex, is not even a pipe dream for the current state of knowledge.

Being unable to provide even a descriptive model, a predictive one is far beyond our hopes. It’s hard to imagine what that would even look like. A theory of power would attempt to rigorously summarize all human interactions in which people are being controlled against their will or according to the will of others. Not one of these terms (power, control, will) is nearly precise enough to go beyond this vague formulation. There is nothing like a model where input X produces output Y. Or output Y is fully accounted for by mechanisms A, B, and C.

So we can certainly hope to provide adequate formulations of human phenomena in this general sense. But this is nothing approaching what one would call a “theory” of power, and the same is true for other topics.

If the social sciences and humanities want to continue to describe their activities as the ongoing “theorization” of human affairs, we can call them “weak theories,” in the technical sense meaning “ideas discussed in the social sciences,” reserving the notion of “strong theories,” beyond the purview of the social sciences, for the technical sense of “rigorous descriptive and predictive models discussed in the hard sciences.”

This entry was posted in Social sciences--critique of. Bookmark the permalink.

7 Responses to Strong and weak theories in the social sciences

  1. andreakente says:

    I agree that it would be useful to have some kind of differentiation between the various senses of theory, especially as that term is used by both the “hard” and “soft” sciences to mean different things. I don’t necessarily see why the term should be more properly be used to refer to the former (accepted, predictive) than the latter (speculative, conjectural). Webster’s dictionary lists six definitions, the second of which is speculation, the fifth of which is an accepted explanation for scientific phenomena. Not that this is authoritative per se, I just wonder why you assume the one use is more primary or valuable than the other (if, indeed, I am understanding you correctly).

  2. beautype says:

    My guess, though I won’t take the time to prove it for now, is that social sciences are self-consciously modeled on the hard sciences in a number of respects, including this one, and that notion of “theory” and “theorization” are miming the hard sciences for the prestige they incur. Newton has a theory, and Darwin, and Einstein, so why not Freud, Foucault, and Jameson too?

    What one would want to look into is the historical use of the term. My guess is that the social sciences use of “theory” to describe their activities is pretty recent.

  3. jmac says:

    You obviously have never read a social science text on theory building. Since I have, let me give just a few words of input. In all the sciences, theory is always EXPLANATION and only sometimes PREDICTION. This reality is quite evident, dating from the earliest days of debate about what is science and what is theory, beginning with science philosophers long ago.

    In the human world, too, prediction is not quite so simple as in chemistry or astronomy. Prediction is not the holy grail you assume in this post.

    If you just substitute the word ‘explanation’ for ‘theory’ in all the examples used at the start of this thread, you can see what social scientists are up to: explanation-building.

    If you want to learn more, read The Structure of Scientific Theories (Suppe), Varieties of Social Explanation (Little), Explication (Chaffee), and How to Build Social Science Theories (Shoemaker). Suppe’s text gives the long historical view; even reading his table of contents will debunk most of this post author’s confusion.

    • beautype says:

      Thanks for the ideas on this topic. Appreciate the feedback, and take your point about the limits of prediction within the human world. You certainly have a point when you say that “in the human world … prediction is not quite so simple as in chemistry or astronomy.”

      I disagree with the gist of your assertions, however. You state that in “all the sciences,” theory is an explanation that only occasionally has predictive power. This is incorrect, I think. The very criterion for a theory’s prestige, and its acceptance as a real theory, is precisely its predictive power. That is, if it cannot explain the real world adequately enough to predict it in some quantitative, precise sense, then the so-called theory is intellectually useless, no more than words. It may sound right, and it may actually be right, but we have no way of verifying it, so there’s no reason to take it seriously. You cannot know that a theory or explanation is correct except insofar as it predicts something precisely measurable with consistent accuracy. This notion can be described as the predictive power of a theory. Atomic theory and evolutionary theory are two examples. A far less powerful, but still relevant notion is that of explanatory power. Insofar as known facts are adequately explained by a model or proto-theory, one has achieved explanatory adequacy, which is intellectually substantive, meaningful but not yet powerful as an explanation.

      The point you make about the human world being inherently resistant to prediction is very pertinent. That is why there are few, if any, “theories” about the human world that are worth anything–they are usually made up of vague assertions, unfalsifiable statements, and which claims that fail basic tests of explanatory adequacy.

      Thanks for the recommended titles, I’ll check them out.

  4. Tomas says:

    Wow! Very well put. Easy to follow your line of thought. Supremely enjoyable. Thank you.

  5. timbrazia allen says:

    That was very very helpful thank you so much for that and I don’t like science but know I do just because of this.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s