Interpretation and Social Knowledge: On the Use of Theory in the Human Sciences

INTERPRETATION AND SOCIAL KNOWLEDGE

THE UNIVERSITY OF CHICAGO PRESS

Chapter One

I

The construction of social knowledge occurs at the intersection of two meaning-systems, one of theory and one of fact. Let us begin, briefly, with the latter.

The problems that surround what Robert Merton called "establishing the phenomenon" have long been the subject of methodological disputation in social research. Methodology is a reflection on the efficacy of our various techniques for establishing facts—survey data and in-depth interviewing, quantitative versus qualitative approaches to the historical archive, and so on. All of these methodologies (and the disputes about them) are, however, confronted by the problem that, in the case of human affairs, many of the most essential facts of the matter—the social facts—are not immediately observable. Rather, they are observable through what Émile Durkheim called their "individual manifestations."

Why is this the case? Certainly much of what human beings who live in societies do is observable, recordable, etc. And there are many behaviors that we do not observe or record directly that we can, nonetheless, be fairly sure happened. It is not our spatial or temporal distance from social facts that makes them a difficult category. It is that reporting on the carryings on of human beings requires a reference to the meaning of their actions, in a very minimal yet fundamental way. The telling of the facts of the matter, in human affairs, already involves a structure of meaning and intention, and, therefore, of inference on the part of the investigator toward aspects of life that are not visible, and never were nor will be visible. Social facts understood in this manner can never be fully stated in protocol sentences that are verifiable by literal observation, but must be inferred and understood in a dialogue about what is happening or has happened, at a certain time, in a certain space, in a given society.

For example, when one states that, in 1692, after examination by the village doctor, it was determined by a set of adults in Salem Village, Massachusetts, and its environs that the fits and screams of "afflicted" girls were due to their being under an "evil hand," one is stating a rather uncontroversial fact. We know this happened—that these adults made this determination. But understanding this fact already involves understanding the possible meanings for the people of seventeenth-century Massachusetts of the physically observable behavior of the girls, the meaning of the utterance "evil hand," the meaning of the term "doctor," and the expectations that other people had of doctors, and so on. It is in this way that social facts are "thick."

Dispute already reigns here. However, the "thickness" of human facts—that is, the way in which they already contain inferences to meanings that are, technically speaking, invisible—is but one aspect of a much larger problem of interpretation in social research. For, while establishing the phenomenon may be the most important and most difficult task a social researcher faces, it is neither the task that produces the most controversy in social science nor the final step in the production of social knowledge. For, as soon as we have established the phenomenon—or, some would say, thickly described it—then we ask the next question: how are we to understand it?

In other words, it is the responsibility of the social researcher not only to report the facts, but to propose a deeper or broader comprehension of them. When investigators attempt to do this, we reach for our theories. We do this because we need some way of comprehending what is, to speak colloquially, "underneath" the facts. We want to know what generates them, determines them, what their consequences are, how we should think about them politically, what their connection to the here and now is, and so on. To do this, it is very seldom enough to continue to gather more facts, no matter how thickly we comprehend them and present them to our colleagues. We need theory to help us explain and evaluate social life.

Our theories are, by their very nature, meaningful human constructions. They exist primarily in the heads of investigators and the pages of their books and journals. Sometimes they consist of a vast, abstract architecture of interrelated and highly consistent terms, sometimes they attempt to specify in the abstract a single mechanism, sometimes they propose a new way to think about something we all already think about, such as democracy. But the world of social theory is meaningful in the basic human sense of providing a coherent model for and model of the (social) world. The hope is that this meaningful world is also a useful one, so that our attempts to develop a deeper understanding of social phenomena are sometimes successful.

Our facts are thus a set of meanings, and our theories are a set of meanings. When we bring theories and facts together, then, we are bringing two meaning-full worlds, or meaning-systems, together. To a certain degree, of course, our theories may indeed influence our use of evidence to construct the facts, or our emphasis on different facts. But this happens less than we think, and at the level of bare social facts—the level of understanding required to report what happened—we quite often can achieve a good deal of consensus, despite our theoretical, or even epistemological, differences. (This is not always the case, and I will consider an example of a disputation over facts below.) Rather, the influence of theory on our knowledge claims most often comes in a much more conscious and controlled form—when we deliberately bring terms foreign to our subjects of study (e.g., "mode of production," "episteme," "ideological state apparatus," "habitus") to bear on our facts, in an effort to grasp some essential aspect of social life that is not given up easily by the facts.

So, by bringing our theoretical terms to bear on what happened at the Salem Witch Trials, we might come to understand that rather horrendous set of actions as an expression of the economic transformation of early America and the politico-economic interests of the parties involved. Or we might grasp it as one of the last violent episodes in the vast formation of early modern European patriarchy, in which the inner resentments and fears of men found their grisly public resolution. Or we might understand Salem as an early expression of American populism, a willingness of some actors, some of the time, to speak outside of the legal structures established by elites, whatever the risks—and thus as a story that should be recuperated by those interested in the establishment of a more democratic United States today. Each of these proposals gains in power what it loses in obviousness, and each incorporates into its interpretation the basic reports of what happened at Salem. The results of this incorporation are very different, however, and that is what makes them exciting and valuable.

II

To say that theory and fact are both meaning-systems is not to say that they are meaning-systems that work in exactly the same manner. Indeed, the intellectual disciplines dedicated to the study of meaning—hermeneutics and semiotics—are flush with typologies, dichotomies, and elaborate theoretical artifices all designed to work out the different ways in which language—or, more generally, signification—can work in its various social contexts. And needless to say there are surely many ways of gathering evidence and thus producing factual reports on what happened in social life, and many genres of theoretical exploration and imagination. But let us stick to the basics, at least at first. What is the difference between theory and fact as meaning-systems?

I think the central difference is that in the meaning-system of fact, we expect evidence to function referentially or indexically when indicating what happened, and in the meaning-system of theory, we expect theoretical terms to function relationally or conceptually. Some evidence is directly indexical—we think of it as a trace of a physical act that happened at a certain point in time and space. But, as discussed above, most of the reality we are busy studying in the human sciences is not reducible to its biophysical supports. Thus it is safer to say that we expect the gathering, organizing, and presenting of evidence, as an active and dynamic meaning-system, to serve a primarily referential function—even if what it references is not (or is not only) a material object or biological person. In the language game of fact, there are myriad evidential signs—sentences, photographs, quotations, assertions, graphs, tables, charts—and we expect these signifiers to express a certain content that is or was in the social world. This means that meaningful facts result from the connection of evidential signs to a ground that emerges, from research, as the object of investigation—the selected set of social actions that happened. Evidential signs, colligated together, connect the sociological investigator, and the people who read her text, to a set of social actions that are the ground of factual signification (see fig. 1).

The English locutions "evidence for" and "theory of" hint at how differently theoretical meaning works. In theory, meaning develops to a great degree by the ways in which contrasts between expressions—e.g., "forces of production" and "relations of production"—create conceptual contrasts in the minds of researchers. This is not to say that those concepts cannot, in turn, reference something else, perhaps something in the world. But if they do (and this book is, in part, an effort to figure out what, if anything, theory references) it is surely reasonable to point out that we do not expect theory to reference the social world in the same concrete manner that we expect evidence to reference the social world. Indeed, the whole point of theory is to be abstract and conceptual. The necessary result of this is that what theory "references," first and foremost, is not really a referent at all in the concrete sense of meaningful social actions that actually happened. Rather, the immediate reference of theoretical expressions is, as far as I can tell, (1) other theoretical expressions and (2) imagined societies, social actions, and social relations whose primary existence is in researchers' heads.

This may seem a strange thing to say, but anyone who has recently had the experience of being initiated into a group of theoretically informed social researchers (or, even worse, into a group of social theorists) probably has an intuitive sense of its truth. The disputations of the early and the late Wittgenstein aside, if there is one language you cannot learn by asking, at each turn in the conversation, "can you point to what that word refers to?" it is the language of social theory. This does not, however, make theory irrelevant to or useless for understanding social reality. Quite the opposite is the case. It is precisely because theory is abstract that it enables analysis of facts, and, ultimately, the construction of knowledge. Furthermore, this distinction between theory and evidence is relative; there are relational aspects of evidence as a language, and referential aspects of theory as a language. As an example of the former, consider how gender history must, even at its most empirical and evidential, use the binary of male/female as source of meaning. As an example of the latter, think of the referential importance of Speenhamland for Polanyi's abstract theories of economy and society. Here reference to a set of quite actual processes enabled the development of highly abstract definitions and elaborations of the historical prerequisites for the institutionalization of a "market society"—a theoretical term understood in relation to the abstractions of Marx, Weber, and others.

Still, the point holds, especially when push comes to shove: the most effective way to criticize another researcher's evidence is to suggest that it fails to accurately represent the phenomenon it claims to represent—that it is referentially incorrect; theory, when it is taken to task on its own, separate from any evidence, can be devastated if the critic can show conceptual incoherence. However, perhaps more frequent than either of these criticisms is the critique of a social knowledge claim as suffering from a disjunction between theory and evidence: the evidence does not bear out the theory, the theory does not fit the evidence, this is the wrong theory to use to interpret this evidence, etc. This sort of response, by an audience, to a failed knowledge claim gives us a clue as to what happens when theory and evidence do come together successfully.

III

In most of its contacts with actual happenings in the social world (represented by evidence), theory is metacommentary. It proposes to rethink, reframe, and recast facts that have already been established; it proposes to set up a research question to be investigated; it hypothesizes about a cause whose traces can be either measured quantitatively, confirmed comparatively, or perhaps verified via testimony and interview; it creates the conditions for critique by denaturalizing the inevitable, reopening the possible, and exhorting for the radically democratic. All of these functions are invaluable, but they are also supervenient upon the existence of referential evidence in a well-colligated meaning-system of fact. In this format, the sign-system of theory combines with the sign-system of fact in an obvious way: facts provide an "example of" a theory, theory provides "a new way to view" the facts.

The importance of these functions of theory should not be underestimated. However, they refer to a kind of discourse in which the difference between theory and fact remains obvious because the two meaning-systems remain to some degree in disjunction. The stakes are higher when the meanings of theory and the meanings of fact mix in a more extensive and effective manner to produce the social knowledge claims that I will call maximal interpretations. In maximal interpretations, theory and fact articulate in such a way that the referential functions of evidence and the relational functions of theory are subsumed under a deeper understanding. No longer is evidence used merely to shore up a factual "example" of a theoretical expression. Rather, the signs of evidence become themselves intertwined with the signs of theory, such that both come to express a deeper social force, a longstanding democratic imperative, or an underlying discursive formation. They become part of a maximal interpretation.

In other words, while "establishing the phenomenon" requires interpretation at the level of evidence and method—the arrangement of evidential signs—such work tends toward one end of a spectrum that runs from minimal to maximal interpretation. At the minimal end of the spectrum, the frequency of theoretical terms is slight (or ... minimal), and the claims tend to be less controversial—though they can, on rare occasions, be startlingly new. The maximal end of the spectrum involves statements that mix, in a consistent and deep way, theoretical and evidential signification, in an effort to produce a powerful comprehension of the matter at hand. Here is a minimal interpretation: 'On the night of August 4, 1789, feudal privileges in France were abolished.' Here is a maximal interpretation: ' The French Revolution was a social revolution with political consequences.'

To believe the first statement to be true, one has to understand the meaning of certain basic terms (most notably 'feudal privileges'), and one has to agree that a series of actions took place at a specific time and place that can be adequately described as the 'abolition' of feudal privileges (and that that place can be adequately described as something called 'France'). This can, of course, be disputed in a variety of ways, including looking empirically into what demands aristocrats still made on the peasants who worked on their land after August 4, 1789, and whether these demands were fulfilled. Still, this statement can inspire a certain degree of agreement, primarily because of its status as a relatively minimal interpretation—there is nothing in the statement to suggest why feudal privileges were abolished, or what it meant to the future of France that they were abolished. No social research can exist without minimal interpretations, but very few works in the human sciences limit themselves to them, and even those have to choose some statements over others. Minimal interpretation is necessary but not sufficient for powerful social research.

(Continues...)

---

Excerpted from INTERPRETATION AND SOCIAL KNOWLEDGE by ISAAC ARIAIL REED Copyright © 2011 by The University of Chicago. Excerpted by permission of THE UNIVERSITY OF CHICAGO PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---