Derek L. Phillips
Epistemology and the Sociology of Knowledge: The Contributions of Mannheim, Mills, and Merton
Theory and Society, 1(1)
Spring 1974, 59-88.
HHC: index added
In the natural and social sciences alike, there exists a rather rigid separation between those thinkers concerned with the practice of knowledge and those concerned with questions about the theory of knowledge. This is in contrast to the situation for the early Greeks and, much later, for such seventeenth century thinkers as Descartes and Locke, where there clearly existed an explicit concern with the connection between the theory and practice of knowledge. Just as clearly, the twentieth century has witnessed an obvious separation between the interest and practices of scientists and philosophers, and, consequently, between “science” and “epistemology”.
This distinction between the theory and practice of knowledge is heightened at present by the gulfs dividing different intellectual disciplines. Such a separation has been especially pronounced in sociology, where an emphasis on imitating certain methodological practices of the natural sciences seems to have reproduced the latter’s indifference to what are regarded as “philosophical” problems. It is perhaps partially because of their collective insecurity about the “genuine” scientific status of their discipline that sociologists have reacted with either indifference or antagonism to questions about the status of their knowledge.
There is, nonetheless, one branch of sociology where problems of knowing
I wish to thank my colleague, Alvin W. Gouldner, for his suggestions and comments - both substantive and editorial - on earlier versions of this essay. The essay itself is a product of our continuing dialogue over the past eighteen months, regarding questions of science and knowledge. I also owe a continuing debt to Ludwig Wittgenstein (Philosophical Investigations, New York: Macmillan, 1958; Lectures and Conservations, Berkeley: University of California Press, 1972; Remarks on the Foundation of Mathematics, Cambridge, Mass.: MIT Press, 1972), who, more than anyone else, recognized the fully social nature of science and knowledge.
and knowledge have been an explicit focus of attention: the sociology of knowledge. Here, as well as in the sociology of science which forms one of its subparts, there has been some interest in certain aspects of knowledge, including epistemological questions. As these fields have developed in recent years, however, their practitioners have tended to ignore epistemological questions in favor of questions concerning such matters as the origin of scientific ideas, their communication to other scientists, scientific productivity, the reward systems in science, and related matters. Indeed, the direction of these fields is, I believe, very much at odds with many of the earlier formulations of the sociology of knowledge and science.
Many sociologists of knowledge and science have been concerned with the origins of scientific ideas, and with the relation of these “discoveries” to social and cultural factors and contexts. Robert K. Merton (l970a), for example, systematically reviews a number of questions pertaining to the social origins of knowledge, while generally omitting questions about the validity or justification of the knowledge-claims involved. This distinction between the genesis of scientific ideas and their evaluation, between what Reichenbach (1938) termed the contexts of discovery and justification, is ignored by most sociologists. Indeed the division of labor between philosophy and sociology, authorized by sociologists, is maintained in the practice of sociology. This, of course, leaves epistemological matters to the philosopher and (more recently) to the historian of science.
By ignoring epistemological issues, sociologists have put themselves in the position of having very little to say about two problems of great concern to many contemporary thinkers: first, the problem of the theoretical and empirical foundations upon which authority rests in Western society; and, second, the problem of authority and competence in science.
The former problem has been dealt with at length in an excellent article by John Schaar (1970). Schaar argues that legitimate authority is declining in the modem state, and that (Schaar, 1970:279) “the crisis of legitimacy is a function of some of the basic, defining orientations of modernity itself; specifically, rationality, the cult of efficiency and power, ethical relativism, and equalitarianism”. Sociologists, I believe, by generally neglecting questions regarding their status as knowers and the status of their knowledge, have effectively cut themselves off from a concern with this issue of legitimate authority. Questions about what it is to “know” something and about who are to establish the criteria or standards for showing that one does know or that one group knows better than another, are simply ignored by most sociologists. If one shares with Schaar, as I do, the belief that the modern
condition is characterized by the shattering of authority, then one longs for (Schaar, 1970:292): “an account of reality, an explanation of why some acts are preferable to others, and a vision of a worthwhile future toward which men can aspire”. Sociologists have had very little to say about such matters.
An awareness of the absence of moral absolutes and certainties is, of course, widespread in contemporary society. In ethics, the notions of “right” and “wrong” have come to be recognized as culturally-dependent. But now there is a growing awareness that science, - which has been viewed by many, including sociologists, as the source of absolutes and certainty is a fully human enterprise, where truth is not something lying “out there” but, rather, a construction of scientific communities. Witness, for example, recent controversies in the philosophy and history of science, involving, among others, Popper, Kuhn, Lakatos, and Toulmin. Despite the enormous attention given these problems today, they are almost ignored in the sociological literature. This is somewhat surprising in that Kuhn, especially, has emphasized the sociological nature of his work, and the term “sociological” is utilized by Popper, Lakatos, and other critics of Kuhn, as a word of degradation.
It is not my intention to pursue this controversy here, but only to provide a very brief overview of one aspect of their discussion. Popper and Lakatos, who refer to themselves as “demarcationists”, believe that there exist universal criteria by which scientific theories can be compared and appraised, and by which science can be distinguished from pseudo-science. Feyerabend, on the other hand, holds that scientific theories occupy no privileged epistemological status as compared with other families of beliefs; no one belief-system is any more “correct” or “better” than another. Kuhn and Toulmin, like Feyerabend, reject the idea of universal criteria for comparing theories. But whereas the demarcationists lay down statute laws of rational appraisal, Kuhn and Toulmin (and Polanyi, as well) hold that science can only be judged by case law. That is, only the members of a specific scientific community are competent to judge about specific questions of scientific practice within that community. Lakatos refers to this as “sociologism”.
Both of the above problems - authority and legitimacy more generally, and within science specifically - are clearly major problems of our time. On the one hand, by accepting that each separate culture or group should decide by its own standards what properly counts as “scientific understanding” (or “equality”, or “justice”) we opt for relativism. On the other hand, by accepting the existence of universal, abstract definitions of “scientific understanding”, “equality”, and the like from outside, we land ourselves in absolutism. The question is whether we must choose between these, or whether
there exists a middle way which allows us to steer a course between the relativist and absolutist extremes. The major issue is, in short, what intellectual authority can be claimed - in principle - for one set of standards rather than another?
In the following pages, I will consider the views of three sociologists - Karl Mannheim, C. Wright Mills, and Robert K. Merton - as they touch on these matters and, especially, as they bear on some of the issues raised more recently by Thomas Kuhn and others. My intentions in considering these men are three. First, to remind sociologists of the enormous sensitivity of these earlier writers to the issues that were central to the so-called “revolutionary” ideas set forth by Kuhn in his influential book, The Structure of Scientific Revolutions. At the same time, I will note some of the “advances” attributable to Kuhn, as well as the similarities between Kuhn and the others. Second, to suggest some possible reasons as to why sociologists failed to take seriously the epistemological implications of these earlier views. Third, and finally, to argue that many sociologists of science and knowledge today are paradoxically less sociological than are their contemporaries in other fields - especially Kuhn, Feyerabend, and Toulmin.
It is useful to begin by briefly reviewing some of Kuhn’s more central themes. Time is necessary in order to gain a full appreciation of the extent to which these earlier thinkers, and Wright Mills, especially, anticipated many of the ideas which have today made Kuhn a center of scientific and philosophic debate. It is, I think, rather ironic that many sociologists who today show an enormous enthusiasm for Kuhn’s work (or, at least, for his notion of paradigm) should have forgotten or ignored much in the pioneering contributions of Mannheim, Mills, and Merton. In a profession which evidences an almost pathological tendency to claim various past luminaries as “sociologists” (for example, Marx, de Tocqueville), one would have expected a great outpouring of analyses showing the seminal contributions of these earlier sociologists to problems which are at issue in contemporary science and philosophy. Perhaps there is a reason for this failure that is worth noting.
Kuhn (1962: 10) argues that “particular coherent traditions of scientific research”, which he terms “normal science”, take their shape from paradigms. While he uses the notion of paradigms in a variety of ways, in the Preface he defines them (Kuhn, 1962: x) “as universally recognized scientific achievements that for a time provide model problems and solutions to a community of practitioners”. Paradigms include (Kuhn, 1962:10) “law, theory, applications, and instrumentation together” and “they are the source of the methods, problem-field, and standards of solution accepted by any mature
scientific community at any given time. As a result, the reception of a new paradigm often necessitates a redefinition of the corresponding science”. Such paradigms (Kuhn, 1962:108) “provide scientists not only with a map but also with some of the directions for map-making. In learning a paradigm the scientist acquires theory, methods, and standards together, usually in an inextricable mixture”. For Kuhn, then, a paradigm indicates the existence of a coherent, unified viewpoint, a kind of Weltanschauung, which determines the way a science’s practitioners view the world and practice their craft.
Kuhn’s argument reminds us, of course, that science is a social enterprise, with an organized consensus of men determining what is and is not to be warranted as knowledge. Among other things, Kuhn’s views are at odds with those formulations of science which sharply differentiate facts from interpretations. He questions the belief that the world we know is a collection of individual observable “facts” which various sciences try to order so as to predict certain events on the basis of others. Kuhn argues that what is seen as a “problem”, a “fact”, a “solution”, and so on depends on presuppositions which constitute part of a paradigm.
Kuhn’s views seem to lead to a kind of relativism. By stressing the determinative influence of paradigms, as they affect the ways in which scientists view the world, including their very conception of what is or is not a fact, Kuhn apparently denies the possibility of comparing and making judgments about the choice of paradigms. That is, since there are no such things as “independent” facts, or any other independent features or standards, there can be no “good reasons” for choosing one paradigm over another. For, according to Kuhn, what constitutes a good reason is itself established by the paradigm. For instance, Kuhn (1962:147) states that “the competition between paradigms is not the sort of battle that can be resolved by proofs” and adds (Kuhn, 1962:150) that “in these matters neither proof nor error is at issue”. Furthermore, he asserts (Kuhn, 1962:119), “we may… have to relinquish the notion, explicit or implicit, that changes of paradigms carry scientists closer to the truth”. Kuhn’s arguments, then, lead to the brink of abandoning the traditional idea of objectivity and progress in science.
Let us now turn to the three sociologists of principle interest in this essay, beginning with Karl Mannheim.
In his Ideology and Utopia, first published in 1929 and available in an English translation in 1946, Mannheim sets forth two goals for the sociology of
knowledge (1972:237): “as theory it seeks to analyse the relationship between knowledge and existence; as historical-sociological research it seeks to trace the forms which this relationship has taken in the intellectual development of mankind”. Mannheim’s discussion firmly anticipates many of Kuhn’s central themes. Mannheim holds that not only does the individual speak the language of his group, but he also thinks in the manner in which his group thinks. He has at his disposal only certain words and their standardized meanings. These, to a large extent, govern his avenues of approach to the surrounding world. Individuals come to perceive the world and its objects in the way the group to which they belong does. Mannheim (1972: 243) notes that: “Every epoch has its fundamentally new approach and its characteristic point of view, and consequently sees the “same” object from a new perspective”. (We find echoes of this last observation in Kuhn’s (1962:121) statement that “When Aristotle and Galileo looked at swinging stones, the first saw constrained fall, the second a pendulum”.)
Mannheim emphasizes that “knowing” is a fundamentally collective enterprise, and that it (Mannheim, 1972:28) “presupposes a community of knowing which grows primarily out of a community of experiencing prepared for in the subconscious”. Rather than formulating knowing as an individual matter, he lays heavy emphasis on the social and communal character of knowing and of knowledge.
Much of what Mannheim says is suggestive of Kuhn’s notion of paradigm as an organizing Weltanschauung. Mannheim notes that every perception is ordered and organized into categories, and that the extent (Mannheim, 1972:77) “to which we can organize and express our experiences in such conceptual forms is, in turn, dependent upon the frames of reference which happen to be available at a given historical moment”. And he adds (Mannheim, 1972: 250) that: “the approach to a problem, the level on which the problem happens to be formulated, the stage of abstraction and the stage of concreteness that one hopes to attain, are all and in the same way bound up with social existence”.
What one finds in Mannheim, then, is an acute sensitivity to the paramount influence of social factors on the various modes of social thought and knowledge. But one sees further his recognition of the impossibility of considering any element of social life - whether language and meaning, perception, knowledge, truth - outside of a communal or social context. It is not surprising, then, that Mannheim (1972: 80) acknowledged that “every point of view is particular to a certain definite situation…” Then how does one distinguish true and false knowledge? In other words, how did Mannheim
deal with the “relativity” problem that Kuhn and others have wrestled with in recent years? As did Kuhn, more than thirty years later, Mannheim appears to reject the idea that there exist firm, unchanging, ultimate “truths”. The very notion of truth had a social character: “We see, therefore”, says Mannheim (1972: 262), “not merely that the notion of knowledge in general is dependent upon the concretely prevailing form of knowledge and modes of knowing expressed therein and accepted as ideal, but also that the concept of truth itself is dependent upon the already existing types of knowledge”. And, he adds further, “… we must reject the notion that there is a ‘sphere of truth in itself’ as a disruptive and unjustifiable hypothesis”.
Mannheim designates the standpoint of the sociology of knowledge as “relational”, which he contrasts with relativism. With relationalism, all intellectual phenomena are subjected to the question (Mannheim, 1972: 254): “In connection with what social structures did they arise and are they valid?” The point in relationalism is not that there are no criteria of rightness or wrongness in a discussion, but rather that such criteria can only be formulated in terms of the perspective of a given situation. This, he argues, is different than “philosophical relativism”, which he (Mannheim, 1972: 254) characterizes as denying the validity of any standards as well as the existence of order in the world. But Mannheim is unclear as to exactly what consequences relationalism has for establishing the validity (truth) of one or another assertion. Consider the ambiguity of the following statement (Mannheim, 1972: 256): “The function of the findings of the sociology of knowledge lies somewhere in a fashion hitherto not clearly understood, between irrelevance to the establishment of truth on the one hand, and entire adequacy for determining truth on the other”. Apparently, however, Mannheim believes that the relevance of the sociology of knowledge is in some way dependent on a comparison with the “facts”. Thus, he states that (Mannheim, 1972: 256) “the mere delineation of the perspectives is by no means a substitute for the immediate and direct discussion between the divergent points of view or the direct examination of the facts”. This statement of Mannheim’s is quite unexpected, as one would expect him to hold the view that whether something is, for example, “consistent” with the facts is itself dependent on what are regarded as facts, and as consistency, within different social contexts. That is to say, such matters as consistency, similarity, divergency, and the like, are, one would think, themselves matters of social conventions in different groups. This is, of course, Kuhn’s position. Here we see that Kuhn goes beyond Mannheim by arguing that even matters of “similarity” and “difference” are dependent on social conventions. Thus, Kuhn is more radically relativistic. 
1. As I have noted elsewhere, (in Derek Phillips, Abandoning Method, San Francisco and London: Jossey-Bass, 1973), however, Kuhn is not at all consistent or clear on this matter.
Mannheim has more to say about the issue of “facts” in one of the last sections of his book, where he directly confronts “the epistemological consequences of the sociology of knowledge”. Perhaps more than any other portion of his book, this has most relevance for the present essay. Mannheim (1972: 25 7) notes that the fact that the position of an observer influences the results of his thought, and the fact that “the partial validity of a given perspective is fairly exactly determinable, must sooner or later lead us to raise the question as to the significance of this problem for epistemology”. He begins by questioning the belief that the genesis of an assertion is irrelevant to its truth, arguing against an epistemology that holds this as an a priori premise. He goes on to argue that epistemology itself must be willing to alter its foundations as it encounters new modes of thinking: “Through the particularizing procedures of the sociology of knowledge, we discover that the older epistemology is a correlate of a particular mode of thought” (Mannheim, 1972: 260). We are, Mannheim says, “thus implicitly called upon to find an epistemological foundation appropriate to these more varied modes of thought”. He calls for a new kind of epistemology which will take into account the facts brought to light by the sociology of knowledge.
Mannheim discusses two directions taken by epistemology: one stressing comprehensiveness; the other, emphasizing the neutralizing function. With regard to the first direction, Mannheim (1972: 271) states that “here preeminence is given to that perspective which gives evidence of the greater comprehensiveness and the greatest fruitfulness in dealing with empirical materials”. In this statement, Mannheim again reveals his lack of full commitment to his own general thesis concerning the influence of social position. As Kuhn has more recently pointed out, matters such as “comprehensiveness” and “fruitfulness” are decided by invoking various communal standards. And these human standards may be in conflict in the same way as are the “points of view” discussed by Mannheim. In fact, Mannheim’s position is far from clear about such matters. For instance, he rejects the idea that the sociology of knowledge is relativistic, because assertions are relativistic, he says, only when judged from the standpoint of (Mannheim, 1972: 270) “external, unperspectivistic truths independent of the subjective experience of the observer”. And Mannheim does not accept this older, static ideal of eternal truths. Thus, deciding which of two or more points of view is the best cannot rely on a comparison of some independent measure - for there is none. On the other hand, Mannheim allows that decisions about the best point of view may be made on the basis of the greater comprehensiveness of one viewpoint over another. He treats comprehensiveness as if it were a fixed, stable, standard, instead of recognizing that - like truth - it is a matter of communal judgment as to which of several points of view has the greatest comprehensive-
ness. Thus, for Mannheim, there are “outside” independent standards. He fails to see that just as points of view may appear differently among people in different social positions, so may comprehensiveness or fruitfulness also appear differently.
With the second direction that can be taken by epistemology, Mannheim suggests that, rather than absolutizing the concept of “situational determination”, it may be possible, by discovering the element of situational determination in various views, to thereby “neutralize” it. This neutralization then creates a wider and more comprehensive basis of vision. By being fully aware of situational determination, it is possible to harness it and “use” it in moving toward a more formal and abstract level of analysis. But Mannheim (1972: 274) notes that “we are not yet in a position today to decide the question as to which of the two above-mentioned alternatives the nature of the empirical data will force a scientific theory of knowledge to follow”. Again, it can be seen that in some unspecified sense “empirical data” are considered as if they were free from the influence of situational determination. Clearly, then, there is evidence of a kind of lingering positivism in Mannheim’s position.
In summary, Mannheim is enormously sensitive to the influence of people’s social positions on what they can perceive, what they define and accept as knowledge and truth, as well as their views, opinions, goals, and values. But he seems to think the “facts” are something existing external to human actors which can be used as a reference point for checking the influence of various social determinants. Furthermore, and despite his criticisms of the prevailing conceptions of science, he believed that the natural sciences were immune from the influence of social factors. In his words, natural science (Mannheim, 1972:261) “is largely detachable from the historical-social perspective of the investigator…” This view no doubt served to encourage the development of a sociology modelling itself on the natural sciences. But Kuhn’s work, as we know, has shown the extent to which the natural and biological sciences are fully social activities, and therefore, always subject to the influence of social factors.
It is disturbing, being neither just nor scholarly, that C. Wright Mills, one of the best-known American sociologists of the twentieth century, should be so thoroughly ignored when it comes to issues concerning the sociology of knowledge.  While his involvement with this was never evidenced in a major
2. For example, none of Mills’ work is included in James Curtis and John Petras, eds., The Sociology of Knowledge, New York: Praeger, 1970.
book-length monograph, nonetheless, one would have expected his seminal articles to have been a source of continuing interest for sociologists. In the following, I will consider two of Mills’ articles that deal with issues crucial to knowledge and methodology. These were published in 1939 and 1940, shortly after the appearance of Mannheim’s Ideology and Utopia in English. They show the influence of Mannheim’s thinking but, in my view, go considerably beyond Mannheim’s formulations.
Like Mannheim, Mills had as his major concern in these articles the social determination of ideas and mentality. In his earlier article, he points out the need for a concept of mind which would allow for explicit linkages between mind and other social factors. What is needed, he stresses, is fuller understanding of the social-psychological processes by which the connection between the mind and social-historical influences can be established. Mills is emphatic in inserting the social-psychological dimension into the sociology of knowledge. Whereas George Herbert Mead had conceived of the “generalized other” as incorporating the “whole society”, Mills lays stress on the “selected societal segments” to which different individuals orient themselves at different times. In either case, a pattern of internal conversation (thinking) between the thinker and his selected audience constitutes the structure of mentality. It is in such a manner that ideas are “logically tested”. As Mills (reprinted in 1963: 427) puts it: “One operates logically (applies standardized critiques) upon propositions and arguments (his own included) from the standpoint of the generalized other. It is from this socially constituted viewpoint that one approves or disapproves of given arguments as logical or illogical, valid or invalid”.
At this point, Mills makes an important observation about the nature of logic. Rather than regarding the rules of logic as an innate expression of the human mind, or as having a timeless and unchanging character, he recognizes that they are human and conventional. “No individual can be logical”, Mills (1963:427) points out, “unless there be agreement among the members of his universe of discourse as to the validity of some general conceptions of good reasoning”. What we term “illogicality” is very much like immorality; both are deviations from social norms. Correspondingly, the criteria of logicality may be different at other times and in other groups. He also emphasizes that not only what are accepted as valid arguments in the discourse within a particular social group but also what constitutes the elements of reasoning and analysis within a given individual are the result of social conventions. That is to say, in general, the acceptance and diffusion of ideas is dependent on conformity to what counts as following logical rules within a given group. 
3. Mills’ views on the fully social nature of logic and reasoning clearly anticipate current controversies surrounding these issues. See, for example, Thomas Kuhn, The Structure of Scientific Revolutions, Chicago: University of Chicago Press, 1962 and 2nd ed., Chicago, 1970, as well as his paper “Reflections on my critics”, in Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge, Cambridge: Cambridge University Press, 1970; Imre Lakatos, “Falsification and the methodology of scientific research programmes”, ibid.; and Bryan Wilson, ed., Rationality, New York: Harper and Row, 1970.
HHC: footnote 3 displayed on page 69 of original
Mills also adds a new emphasis in the sociology of knowledge: the fundamental role of language in thought. As Mills (1963: 433) notes: “Our behavior and perception, our logic and thought, come within the control of a system of language”. Mills recognizes that language precedes any given actor. A socially sustained system of meanings has a priority over any given individual. In Mills’ (1963: 434) words: “Meaning is antecedently given; it is a collective ‘creation’.” Thus Mills recognizes the extent to which an individual thinker - including the individual scientist, although Mills says nothing about this - is circumscribed by an audience. In order to communicate, to be understood, by this or that audience (his family, peer group, or scientific colleagues) he must use language in such a manner that it evokes the same response in them as it does in himself. This is much of what Kuhn is talking about when he views a paradigm as establishing what constitutes “similarity”, “new facts”, and the like. For the individual thinker, as Mills (1963: 435) observes: “The process of ‘externalizing’ his thought in language is thus, by virtue of the commonness essential to meaning, under the control of the audience”.
In a second important article, published one year later, Mills evidences a developing originality and insight concerning the relationship between the sociology of knowledge, epistemology, and methodology. While his focus is on the social sciences, many of his ideas and conclusions have direct relevance for problems of all scientific disciplines. He (Mills, 1963: 453) begins by criticizing the views of those - for example, Hans Speier, Talcott Parsons, Robert MacIver, and Robert K. Merton - who hold “that the sociology of knowledge has no relevance for epistemology; that sociological investigations or inquiries have no consequences for norms of ‘truth and validity’.”
It is true, Mills says, that one cannot deduce the truth or falsity of an individual’s statements by virtue of knowledge of his social position. He argues, however that the matter is considerably more complicated than that, and he sets out to provide answers to two broad sets of questions (Mills, 1963: 454): “(1) What is the generic character, derivation, and function of epistemological forms, criteria of truth, or verificatory models? (2) Exactly wherein, at what junctures, and in what types of inquiry may social factors enter as determinants of knowledge?
Mills had earlier emphasized that “truth” and “objectivity” have meaning only with reference to some accepted system of verification. That is, truth and objectivity are a matter of communal definition and, therefore, may differ within different social groups and under different social conditions. Mills (1963: 454) rather cryptically suggests that: “He who asserts the irrelevance of social conditions to the truthfulness of propositions ought to state the conditions upon which he conceives truthfulness actually to depend; he ought to specify exactly what it is in thinking that sociological factors cannot explain and upon which truth and validity do rest”. Mills points out that what had once constituted validation and truth in the official “paradigm” of medieval scholasticism, for example, was certainly influenced by a number of social factors. And, he argues, the fact that the truthfulness of propositions is dependent on criteria of validity and truth with are themselves subject to social-historical relativization means that the truth or falsity of various statements or propositions is influenced by social factors. Mills (1963: 455) asserts that “Criteria, or observational and verificatory models, are not transcendental”. He observes that, for the most part, individual thinkers and scientists do not consciously select a verificatory model, clearly anticipating Kuhn, who later stresses that the very criteria for verification at different times and in different scientific communities are dependent on the world-views or paradigms in which the criteria are located.
Thus, Mills refutes those writers who view the sociology of knowledge as having no consequences for the validity of statements or propositions, in a specific way: by affirming the historicity of models of verification. Certain models of verification may be the “accepted” models at different times or in different groups because of the power and influence of certain social or scientific elites.
Mills mentions two additional aspects of knowledge and truth that may be open to social-historical influences. First, the “categories” used by different groups of individuals are dependent on social conditions and influences, Mills (1963: 459) notes that “What is taken as problematic and what concepts are available and used may be interlinked in certain inquiries”. Secondly, there is the influence of social factors on perception. It is worth quoting Mills (1963: 459-460) at length here, because I want to contrast his observations with more recent remarks by Kuhn:
In acquiring a technical vocabulary with its terms and classifications, the thinker is acquiring, as it were, a set of colored spectacles. He sees a world of objects that are technically tinted and patternized. A specialized language constitutes a veritable a priori form of perception and cognition,
which are certainly relevant to the results of inquiry… Different technical elites possess different perceptual capacities.
Compare this with Kuhn’s remarks about paradigms and paradigm-changes. He argues (Kuhn, 1962: 111) that: “Paradigm changes… cause scientists to see the world of their research-engagement differently. In so far as their only recourse to that world is through what they see and do, we may want to say that after a [scientific] revolution scientists are responding to a different world”, Galileo’s work provides Kuhn with an example for his thesis (Kuhn, 1962:117-118):
Since remote antiquity most people have seen one or another heavy body swinging back and forth on a string or chain until it finally comes to rest. To the Aristotelians, who believed that a heavy body is moved by its own nature from a higher position to a state of natural rest at a lower one, the swinging body was simply falling with difficulty. Constrained by the chain, it could achieve rest at its low point only after a tortuous motion and a considerable time. Galileo, on the other hand, looking at the swinging body, saw a pendulum, a body that almost succeeded in repeating the same motion over and over again ad infinitum.
Making use of the relatively new theory of motion - the “impetus theory” - allowed Galileo to “see” what the Aristotelians could not. Until the impetus theory (Kuhn, 1962: 1 20) “was invented, there were not pendulums, but only swinging stones, for the scientist to see”. What Kuhn, the historian of science, has done, then, is to document Mills’ theses regarding men’s conceptual language and their perceptions as affected by social-historical conditions.
A few words about Mills’ brief consideration of “relativism”, which, like Mannheim’s, I find unconvincing. Mills attempts to deny the charge of relativism by calling on Mannheim’s distinction between (Mannheim, 1972: 254) “philosophical relativism which denies the validity of any standards and of the existence of order in the world” and “relationalism”, where all intellectual phenomena are subjected to the question (Mannheim, 1972:254): “In connection with what social structures did they arise and are they valid”? But both Mannheim and Mills ignore the question as to how we establish knowledge of “social structures”; after all, this is done from some socially-sedimented cognitive standpoint.
It is not enough to answer, as Mills does (1963: 46 1), that: “The imputations of the sociologist of knowledge may be tested with reference to the verificatory model generated, e.g., by Pierce and Dewey. Their truthfulness is then in
terms of this model”. Why would this not be considered as relativism? After all, the choice of one over another verificatory model is made from some standpoint within a particular epoch and culture. Thus, the manner in which the sociologist of knowledge approached such problems as the choice of a verificatory model is - consistent with the central canons of the sociology of knowledge - conditioned by his particular standpoint. If this is not relativism, what is?
Mills (1963: 461) argues that: “The assertions of the sociologist of knowledge escape the ‘absolutist’s dilemma’ because they can refer to a degree of truth and because they may include the conditions under which they are true. Only conditional assertions are translated from one perspective to another”. But this begs the question of the “relativist’s dilemma”; that either the relativist’s own assertions are themselves relative, and, therefore, lacking truth value; or his argument is unconditionally true, and, consequently, relativism is self-contradictory.
“Relationalism” supposedly avoids having to choose between these two (logical) possibilities. Instead, however, it ignores the problem by failing to recognize the full epistemological implications of the sociology of knowledge. For relationalism is “relativistic” in that it cannot provide an answer to such questions as: from what standpoint do Mills and Mannheim decide what should count, as, for example, the “conditions” under which this or that is true? If it is from within a designated culture and epoch (and it could not be otherwise), then it is certainly “relativistic”. If, on the other hand, it is argued that such judgments are made from some “absolutist” standpoint, then why call it “relationalistic”? The fact that Mills and Mannheim were unable to provide a satisfactory solution to this problem - that is, a “solution” that allows them to conduct investigations in the sociology of knowledge while, at the same time, dealing with the question of how their own position is to be formulated vis-à-vis the implications of the theoretical position which they espouse - is not surprising. For even now, thirty-five years later, this problem is a source of debate among an increasing number of thinkers (for example, Kuhn, 1970a, 1970b; Lakatos, 1970; Bennett, 1964; Feyerabend, 1970; Popper, 1963, 1970; Winch, 1958; Lukes, 1967; Jarvie, 1972). What could have been very important for the development of sociology as an intellectual discipline is Mills’ and Mannheim’s recognition of the need to try to consider knowledge from a uniquely sociological standpoint. With regard specifically to Mills, he exhibited a theoretical awareness - as he did in so many other areas of sociological and intellectual inquiry - that placed him years in advance of the dominant tendencies and directions of the sociology of his day.
Robert K. Merton’s contributions to the sociology of knowledge and science, like Mannheim’s, have been widely recognized. They began as early as his 1938 doctoral dissertation, re-published in 1970, in which Merton showed a concern with problems central to these fields. Like Mannheim, and especially, Mills, Merton also recognized that science is a social activity; (Merton, 1970b: 225) “the verification of scientific conceptions is itself a fundamentally social process”. He held that matters of scientific knowledge and truth are dependent on the scientific community to which the individual scientist directs his truth-claims (Merton, 1970b: 219):
Science is public and not private knowledge; and although the idea of “other persons” is not employed explicitly in science, it is always tacitly involved. In order to prove a generalization, which for the individual scientist, on the basis of his own private experience, may have attained the status of a valid law which requires no further confirmation, the investigator is compelled to set up critical experiments which will satisfy the other scientists engaged in the same cooperative activity. This pressure for so working out a problem that the solution will satisfy not only the scientist’s own criteria of validity and adequacy, but also the criteria of the group with whom he is actually or symbolically in contact, constitutes a powerful social impetus for cogent, rigorous investigation.
While not going quite so far as to argue that scientific truth and knowledge exist solely by virtue of being warranted by the relevant scientific communities, Merton states that the “discoveries” of one or another scientist are only (Merton, 1970b: 220) “imbued with significance through contact with other scientists”. Speaking of scientific theories, he points out that (Merton, l970b: 220) “...long after the theory has been found acceptable by the individual scientist on the basis of his private experience he must continue to devise a proof or demonstration in terms of the approved canons of scientific verification present in his culture”. Merton also clearly recognises that the scientific standards which the investigator must meet may differ in different cultures.
Merton’s early use of a distinction between the contexts of discovery and justification has not, however, always been evidenced in his more recent work (or in the work of his students) in the sociology of science. In a paper first published in 1945, he again emphasized the critical distinction between discovery and justification. Criticizing Sorokin’s emphasis on intuition in scientific work, Merton (1970a:357) observes: “[Sorokin] indicates that ‘intui-
tion’ plays an important role as a source of scientific discovery. But does this meet the issue? The question is not one of the psychological sources of valid conclusions, but of the criteria and methods of validation”. Still, in one of his most recent publications, Merton (1972) appears to drop the distinction between the sources of knowledge and the scientific community’s verification of knowledge claims. And when he does touch on the distinction, he totally ignores his own earlier observations concerning the criteria of verification in different cultures. Let us, then, consider Merton’s views as expressed in this article.
Merton (1972: 11) is concerned with recently emerging claims to group based truth: “Insider truths that counter Outsider untruths and Outsider truths that counter Insider untruths”. Speaking of the insider doctrine that you can only understand blacks, then only white scholars can understand whites”. But specific claim, it would appear to follow that if only black scholars can understand blacks, then only white scholars can understand whites”. But Merton fails to see here that claims to understanding, like claims to knowledge, are a matter of meeting public (communal) criteria. Further, he fails to consider the consequences of this for the problem he is considering.
What the individual black scholar, for example, may say is of no special consequence as regards “understanding” blacks, unless the relevant scientific community warrants the correctness of his claims. If the black sociologist formulates a sociological explanation concerning blacks, it becomes a “sociological truth” about blacks only by being warranted as such by a sociological community largely composed of whites.
In one sense, Merton does acknowledge the relevance of public criteria for settling claims to scientific understanding. To see this, it is necessary to quote Merton (1972: 42) at length here:
It is the character of an intellectual discipline that its evolving rules of evidence are adopted before they are used in assessing a particular inquiry. These criteria of good and bad intellectual work may turn up to differing extent among Insiders and Outsiders as an artifact of immediate circumstances, and that is in itself a difficult problem for investigation. But the margin of autonomy in the culture and institution of science means that the intellectual criteria, as distinct from the social ones, for judging the validity and worth of that work transcend extraneous group allegiances. The acceptance of criteria of craftsmanship and integrity in science and learning cuts across differences in the social affiliations and loyalties of scientists and scholars. Commitment to the intellectual values dampens
group-induced pressures to advance the interests of groups at the expense of these values and of the intellectual product.
In this affirmation of the transcendental standards of scientific institutions, Merton seems to forget that all the standards of science are humanly established. He forgets his own earlier observation that scientific standards (“criteria of good and bad intellectual work”, “criteria of craftsmanship and integrity”, and the like) may differ in different epochs and cultures. He fails to see that the standards of a particular scientific discipline may have arisen from, and may be supported by certain powerful elites; and that, therefore, the standards of the group dominant in one or another scientific community may be in conflict with the standards held by other (minority) groups within the discipline. The conflicts and controversies surrounding the view of a Galileo, a Darwin, or a Lysenko, make this clear. Of course, sometimes there is the involvement of what are easily identified as non-scientists or political authorities, as in the case of Russia in the 1950s where the authorities supported Lysenko’s position against the neo-Darwinists. But, for the most part, it is not at all easy to locate or establish permanent and universal criteria that allow for a clear demarcation between “scientific” and “non-scientific” considerations (or between what Merton terms “intellectual” and “social” criteria).
I come now to the second point of interest in this essay: the question as to why the epistemological implications of the theses advanced by these early writers in the sociology of knowledge have been ignored, or at least not taken as a topic for sociological inquiry. I do not deny that the epistemological issue, in the guise of relativism, was recognized by sociologists. The real question is: why did they fear relativism? Although I have been unable to find any explicit reactions to Mills’ two articles, sociologists were quick to see the implications of Mannheim’s views. Stung by critics’ assertions that his standpoint led to total relativism and nihilism, he came to argue in terms of a pragmatic theory of adjustment to the specific requirements of particular historical situations and, later, to stress the position of the “socially unattached intelligentsia”. By emphasizing pragmatism and the unattached intelligentsia, he sought to escape the charges of relativism. After the Nazis seized power in Germany, he emigrated to England, where his intellectual interests underwent an enormous change. As Coser (1971: 447) notes: “one might say that while Mannheim’s German work stood under the shadow of Hegel and Marx, his British work stood under the shadow of Durkheim”. Not only did Mannheim himself resist the epistemological consequences of his
earlier work but, by abandoning a concern with the sociology of knowledge, he helped assure that epistemological matters did not become a central concern to sociologists.
Despite Mannhejm’s efforts to save his assertions from the charge of relativism, Ideology and Utopia was severely criticized. In reviews appearing shortly after the book’s publication in English, von Schelting (1936) and Becker (1939) raised questions about the epistemological status of the sociology of knowledge. The tenor of their criticisms was echoed in Merton’s (1957: 503) observations, originally published in 1941, that Mannheim’s view “leads at once, it would seem, to radical relativism with its familiar vicious circle in which the very propositions asserting such relativism are ipso facto invalid”. Noting Mannheim’s remarks about men speaking in categories which are inappropriate, Merton (1957: 503) points out: “Moreover, determination of the ‘appropriateness’ or ‘inappropriateness’ of categories presupposes the very criteria of validity which Mannheim wishes to discard”.
Especially at a time when the German universities were undergoing a racialist purge, it was understandable that there would be a great resistance to any work that even suggested that science (natural and social) is necessarily affected by social factors. Merton noted in 1938 (reprinted in Merton, 1973: 260) that we must resist the idea that “Scientific findings are held to be merely the expression of race or class or nation”. The extent of this resistance is revealed by Merton in that same article where he states (Merton, 1973: 260): “It is of considerable interest that totalitarian theorists have adopted the radical relativistic doctrines of Wissenssoziologie as a political expedient for discrediting ‘liberal’ or ‘bourgeois’ or ‘non-Aryan’ science . Politically effective variations of the ‘relationalism’ of Karl Mannheim (for example, Ideology and Utopia) have been used for propagandistic purposes by such Nazi theorists as Walter Frank, Krieck, Rust, and Rosenberg.” Thus, one reason for the rejection of the epistemological issues raised by Mannheim was undoubtedly the political struggle against Nazism prior to and during World War II.
But there were other reasons as well. Among these was a social climate favoring pragmatism and empiricism as opposed to the European emphasis on theorizing and speculation; thus there was an increasing stress on the development of empirical sociology in the United States. Certainly during the war years, the use of sociology for war purposes (for example, The American Soldier) laid heavy emphasis on empirical, as contrasted with theoretical, inquiry. And at Columbia University the struggle for control between the department’s more speculative wing and its more empirically oriented coun-
terpart was resolved largely in favor of the latter (Jay, 1973:218). Perhaps partially as a result of this increased emphasis on empirical work and partially as a result of the kinds of problems facing American society (and sociology), Wright Mills was to generally ignore the kinds of issues which preoccupied him in his early work, discussed in the first part of this essay. In any case, at a time when sociology was only beginning to become respectable in American academic circles, and when its practitioners themselves were striving to become a “real” science, it is not surprising that there was no great enthusiasm for viewpoints that tended, if taken seriously, to throw into question the very cognitive stabilities of sociology itself.
But this is not to say that sociologists were unaware of the epistemological issues raised by Mannheim and Mills. Indeed, I believe that it was precisely because they were aware of these issues that sociologists concerned with the sociology of knowledge chose to ignore or dismiss the epistemological problems raised by Mannheim and Mills. The fact that Mannheim and Mills themselves failed to follow through with further inquiries into these problems, of course, made it even more unlikely that epistemological issues would concern American sociologists. Given the insecure status of sociology in the 1940s, it could, in a sense, ill afford to entertain questions about the grounding of its own knowledge. To have faced these epistemological questions squarely would have forced sociologists to consider the existence, or lack of same, of a dividing line between sociology and ideology, the very difference that sociology had been intent on affirming from its very beginnings. Whatever the social, cultural, and professional conditions conducive to the dropping of epistemological questions in sociology, sociologists of knowledge, or those utilizing certain aspects of that general perspective, came to focus on issues of ideology, on issues concerning the importance of understanding the social context in which ideas develop, and related matters. Indeed, the sociology of knowledge was cryptic ideology, genteel ideology, prudent ideology: ideology-critique academicized.
Furthermore, there developed a unique area of specialization within sociology - the sociology of science - which was, in a way, predicated on the rejection of epistemological questions. After all, the natural sciences - which are the main focus of concern for sociologists of science - are, at least by Mannheim’s account, immune from the influence of social factors. And while Merton (1957: 635) noted some fifteen years ago that there were few sociologists who “could bring themselves, in their work, to treat science as one of the great social institutions of the world”, that has surely changed. Today there exists a considerable literature in the sociology of science, but, with the exception of a small number of British sociologists,  those working in the
4. See, for example, Michael Mulkay, “Some aspects of cultural growth in the natural sciences”, Social Research 36, 22-52, 1969; S.B. Barnes and R.G. Dolby, “The scientific ethos: a deviant viewpoint”, European Journal of Sociology 11, 3-25, 1970; M.D. King, “Reason, tradition, and the progressiveness of science”, History and Theory 10, 3-32, 1971; and Richard Whitley, “Black boxism and the sociology of science: a discussion of the major developments in the field”, The SociologicaiReview Monograph 18, 61-91, 1972.
HHC: footnote 4 displayed on page 78 of original
sociology of science have rejected epistemological concerns. As Whitley (1972: 61) points out: “Ignoring the cognitive aspects of scientists’ activities, they restrict sociology to discussion of social relations and processes”. Ignoring epistemological questions, they exclude questions pertaining to the social nature of science (including sociology) itself. Thus, the maturity of sociology - its own self-awareness - is what is ultimately at issue here.
When Kuhn’s work began to appear, with its enormous impact on philosophy, the history of science, and elsewhere, there were heated reactions among philosophers and practicing scientists. But these were nothing as compared to what would have been likely had the same analysis been focused directly on the social sciences. After all, physics, chemistry, and biology, are generally seen to “work”. So that however deep Kuhn’s criticisms might go, they can in no way undermine the practices of natural scientists. With sociology, on the other hand - and this would have been even more true at the time when Mannheim’s and Mills’ work first appeared - the existence and continuance of a discipline of sociology would be seriously threatened by taking full cognizance of the social determination of all scientific views and standpoints. Mills’ suggestion that epistemology itself was relativistic would, if faced head on, have been highly threatening to those busy building a positivistic sociology. The same attitude seems prevalent today among many of those working in the sociology of science; they still refuse to see that the distinction between science and ideology is problematic at best, and that, from one point of view, science as ideology is an important topic for sociological inquiry. Thus, the internal cognitive nature and form of science are considered off-limits.
This brings me to the third theme in this essay: that the epistemological issues raised by Mannheim and Mills (and later ignored by both) are being pursued today by non-sociologists, that is, by non-card-carrying members of the profession. Despite the fact that Thomas Kuhn is cited with some frequency by sociologists, they often fail to understand the full implications of his views. They often concern themselves with parochial questions as to whether
or not sociology has a fully-developed paradigm, and, if not, the importance of acquiring one. Consider, for instance, a recent statement by Ben-David (1972: 4): “The existence of subconscious assumptions is not an important question at all… In science one obtains interpersonally valid knowledge through the subjection of personal ideas and explanations of reality to public test by logic, experiment or empirical observation. Thus personal biases and mistakes are corrected, and gradually eliminated”. To Ben-David, then, the content of science is apparently immune to social influences. Perhaps this is not surprising given his view in another recent work that (Ben-David, 1971: 1): “Sociologists study structures and processes of social behavior. Science, however, is not behavior but knowledge that can be written down, forgotten, and learned again, with its form or content remaining unchanged”. Consistent with this view of science, he asserts further (Ben-David, 1971: 13-14) that “the possibilities for either an interactional or institutional sociology of the conceptual and theoretical contents of science are extremely limited”. But in reaching this conclusion, Ben-David totally ignores the work of Kuhn (1962, 1970a, l970b) whom he cites in another context, as well as Hanson (1958), Feyerabend (1962, 1970a, 1970b), and Toulmin (1961) - all of whom are deeply concerned with understanding the contents of science. These men have as a central concern the ideological commitments which scientists must share in order for the scientific enterprise to succeed. They emphasize that the social nature of science is relevant to the validity of scientific theories (the content of science). What could be more in keeping with the aims of sociological inquiry than, for example, Kuhn’s (1970b: 240) statement that “the type of question I ask has... been: how will a particular constellation of beliefs, values, and imperatives affect group behaviour?” In fact, Kuhn’s “sociological” analyses have been thoroughly derided by his critics - especially Lakatos (1970), Shapere (1964), Scheffler (1967), and Popper (1970) - partially on the grounds that they are sociological.
Sociologists of knowledge and science, especially in the United States, might have made their own contributions to the post-positivist critique of knowledge and science had they more closely followed the leads of Mannheim and Mills. Despite the ambiguities of their views - in that they often seem to be providing a critique of positivistic science, while, at other times, holding that there exists a reality which is fully independent of the human observer (“independent facts” and “regularities” in nature, for example) - they recognized, as most sociologists do not, the social nature of language, perception, concept-formation, verificatory models, truth, and knowledge. It ironic that sociologists, with all their pretensions to high science and their frequent excuse that sociology is only a “young science”, should have failed to follow the leads provided by these two men. Instead it has been scholars
from outside the sociological community - men like Kuhn, Feyerabend, Toulmin, and Winch - who have been the most highly critical of the dominant positivist views of science.
Kuhn, for example, raises questions about the notion of theory-independent observation by pointing out that a theory (1962: 102) is a “conceptual network through which scientists view the world”. He asserts that (Kuhn, 1970a: 192) “People do not see stimuli: our knowledge of them is highly theoretical and abstract”. And Feyerabend (1962: 29) notes that: “Introducing a new theory involves changes of outlook both with respect to the observable and with respect to the unobservable features of the world… Scientific theories are ways of looking at the world; and their adoption affects our general beliefs and expectations, and thereby also our experiences and conceptions of reality”. What these men emphasize is that what counts as an observation of this or that, as well as the meaning of this or that, is theory-dependent. There are, then, no raw data, no brute facts, but only (Feyerabend, l962: 50-51) data “analysed, modelled and manufactured according to some theory”.
Among sociologists, Mannheim, Mills, and Merton, to varying extents, do recognize that truth and knowledge exist only by virtue of the relevant scientific audience warranting the truth- and knowledge-claims of individual thinkers. That is to say, they see that in every science, investigators must use various procedural rules for deciding whether propositions or statements are to be judged “factual” and, therefore, to be admitted to the corpus of scientific knowledge. They were able to recognize, to an extent that most contemporary sociologists do not, the distinction between the contexts of discovery and of justification - the first, having to do with the genesis of the inquirer’s ideas; the second, with his way of presenting the results of his inquiries. They see that it is in the context of justification that scientific truth and knowledge are established.
Popper has been extremely critical of the directions taken by early studies in the sociology of knowledge, but I think he misses the full implications of these inquiries. For example, he (1963: 216-217) asserts that the sociology of knowledge “shows an astonishing failure to understand precisely its main subject, the social aspects of knowledge, or rather of scientific method. It looks upon science or knowledge as a process in the mind or ‘consciousness’ of the individual scientist or perhaps as the product of such a process”. What Popper is calling attention to is the context of justification, with its necessary
reliance on procedural rules or, what he refers to as, “scientific method”. Popper shows his position most clearly in contrasting his view with, what he regards as, the sociological view (Popper, 1963: 216-217):
If scientific objectivity were founded, as the sociologistic theory of knowledge naively assumes, upon the individual scientist’s impartiality or objectivity, then we should have to say good-bye to it... No, what we usually mean by the term rests on different grounds. It is a matter of scientific method... Scientific objectivity can be described as the inter-subjectivity of scientific method. But this social aspect of science is almost entirely neglected by those who call themselves sociologists of knowledge.
While many sociologists of knowledge do neglect the social aspects of science, this accusation is not correct for the writers being considered here. Merton (l970b: 220) points out that a variety of scientific observations in seventeenth century England “were imbued with significance through contact with other scientists”. That is, (Merton, 1970b: 2l9), “the investigator is compelled to set up critical experiments which will satisfy the other scientists engaged in the same cooperative activity”. And Mannheim and Mills stress this throughout the work being examined here, as was seen in earlier pages of this essay. Furthermore, all three went beyond Popper, to emphasize, in a way that he does not, the full extent to which the “social aspects of knowledge” are relevant to knowledge-claims and, at the same time, subject to the influence of social-historical conditions. Only a hint of this is found in Merton’s early dissertation, where his focus was on other matters. But he does firmly indicate (Merton, 1970b: 220) that the individual scientist must “devise a proof or demonstration in terms of the approved canons of scientific verification present in his culture”. This statement certainly suggests an awareness that these canons of verification may be different in other cultures and at other times. Speaking of the verification of knowledge, Mannheim (1972:2 59) points out that the “very principles, in the light of which knowledge is to be criticized, are themselves found to be socially and historically conditioned”.
It was C. Wright Mills, however, who was the most explicit as to the social influences on the criteria and standards involved in justifying various claims to knowledge and truth. For example, he considers the “official and monopolistic paradigm of validation and truth accepted by medieval scholasticism”, and goes on to observe that (Mills, 1963: 455) “There have been and are diverse canons and criteria of validity and truth, and these criteria, upon which determinations of the truthfulness of propositions at any time depend, are themselves, in their persistence and change, legitimately open to social-
historical relativization”. Speaking of the current “scientific” thought-model, he notes that this model distinguishes between the genesis of an inquiry and the truth of its results (Mills, 1963: 458):
For this paradigm demands that assertions be verified by certain operations which do not depend upon the motives or social position of the assertor. Social position does not directly affect the truthfulness of propositions tested by this verificatory model. But social positions may well affect whether or not it or some other model is used by types of thinkers today and in other periods. By no means have all thinkers in all times employed this particular verificatory model.
Mannheim, Mills, and Merton, then, all give considerable attention to the context of justification and to the social conditions influencing what criteria and standards are viewed as relevant to the processes of validation of knowledge-claims. They also, however, consider the relevance of the genesis of ideas and statements for their truth-value or validity. Mannheim argues that the genesis of an idea may be relevant to its validity, while Mills and Merton (at least in his earlier writings) maintain that the validity of an idea is not dependent upon its genesis. They both argue that the motives or social position of an inquirer are irrelevant to the truth of his assertions, because the warrantability of his assertions is done by the scientific or other community to which he directs his assertions. Mills, though, points out that social positions are important in the sense that they may affect which verificatory models are used by different communities or audiences at different times. All three of these men are in general agreement, however, that truth-claims are settled in the scientific or intellectual community. It is by meeting various public criteria which satisfy other scientists or thinkers that truth is established.
Still, Mannheim and Merton, especially in his more recent writings, often talk as if there were some one “correct” position from which phenomena are to be viewed. This is clear in Merton’s (1972) article on “Insiders and Outsiders” and in Mannheim’s (1972: 80) assertion that: “What is needed... is a continual readiness to recognize that every point of view is particular to a certain definite situation and to find out through analysis of what this particularity consists”. Mannheim, as we know, believed there was one social group which was able to free itself from the influence of such particularisms: the unattached intelligentsia. I have criticized this view at length elsewhere (Phillips, 1973), and will not repeat my criticisms here. Rather, what I want to argue now is that there is a sense in which the genesis of ideas is relevant to their truth, and that, further, such a viewpoint does not assume the epistemology-
cally-privileged position of Mannheim’s unattached intelligentsia. Instead, it follows directly from the basic canons of the sociology of knowledge as formulated by Mannheim and Mills.
Simply stated, my thesis is as follows. Since it is the scientific community (and here I speak of it as a monolithic whole, although obviously it is not) which produces scientific truths and knowledge, and since further, the scientific community does consider the genesis of ideas as relevant to their truth, then indeed genesis does affect the truth of a scientist’s assertions. That is to say, those who accredit the truth- and knowledge-claims of the individual scientist may give close attention to his social position as serving to establish what Gouldner terms the scientist’s credibility.
In a sense, then, Mannheim and others who emphasize the importance of the social position of the thinker as relevant to the truth of his assertions are right. But this is not, as Mannheim seemed to believe, because some persons are in a better position than others to see or discover the truth. That is, we need not accept Mannheim’s argument as regards, for example, the unattached intelligentsia occupying an epistemologically privileged position by which they acquire a kind of “purified” mind allowing them access to undistorted reality, which they can then compare with the distorted images held by others. No, the social position of the thinker is important to the truth of various assertions because it is one of the factors considered as relevant by the scientific communities which produce truth and knowledge.
What I have been trying to emphasize here is that the communal nature of the context of justification does not preclude considerations of the genesis of the thinker’s ideas. And just as attributions of credibility may be dependent on the social position of the thinkers, so may they be dependent on such factors as his “motives”. If a scientific audience responds to a man’s publication by arguing, that “Of course, he’d say that; he’s just trying to get even with Y” or something similar, this means that the (attributed) motives of the thinker do play a part in the process by which his assertions are or are not accepted as “true” by the scientific community. If he is seen as having low credibility, there is less likelihood of there being communal attributions of truth to his work than if his credibility is seen as high.
One of the difficulties with Mannheim’s and Mills’ treatment of genesis is that they formulate the relationship between the genesis of an idea and its validity or scientific truth as if it were a private matter. They view the solitary individual thinker as setting forth ideas which may or may not be affected by his motives, social position, and the social conditions of his inquiry. Then, the
thinker’s ideas or assertions are verified by (Mills, 1963:458) “certain operations which do not depend upon the motives or social position of the assertor”. True, Mills does point out that there may be other verificatory models than the one which is dominant today. But he fails to see the inherent contradiction of regarding the assertor’s motives and social position as if they were fully independent of the process of verification. This may, indeed, be the model which scientists say they follow — where the contexts of discovery and justification are separate and independent — but in the actual practice of science it is otherwise.
The problem for the assertor is to convince the scientific community in which he shares membership to warrant the truth or validity of his assertions. In the actual process of verification, they may or may not attribute certain motives to him, they may or may not view his social position as having affected his scientific assertions. Of course, they, like some sociologists of knowledge, will regard these not as attributions but as “discoveries”. But it is they who - in the final analysis - provide whatever linkages are said to exist between the assertor’s motives, for instance, and his assertions. Putting it another way, if the scientific community decides that an individual thinker’s motives are relevant to the truth of his assertions, then they are relevant. It simply makes no sense to argue, for example, that his motives are “really” irrelevant but that this is unknown to the scientific community. For only they can decide matters of what is and is not relevant for scientific truth. After all, they decide what is to count as a “motive” or as “relevant” in such matters. There is no higher court of appeal, no superior vantage point from which such matters can be surveyed or settled.
Finally, with regard to the sociological studies of Kuhn, I wish to offer a few remarks as to the issues of relativism and rationality raised by Mannheim and Mills more than thirty years ago. The same difficulties ensuing from the standpoints of these two earlier thinkers are recognized as major problems in science and intellectual life today. But whereas they were a reason for rejecting many aspects of a sociology of knowledge orientation at that time, today they are taken as topics for serious contemplation and discussion.
One of the consequences of Kuhn’s The Structure of Scientific Revolutions, then, is to raise anew important questions about the relativism of scientific standards and intellectual viewpoints. Since Kuhn regards paradigms as sovereign, as providing alternative world-views, this means that those scientists working within one paradigm share no theoretical concepts with
scientists working under its rivals or predecessors. Lacking a common vocabulary, they may be unable to communicate with one another, and are, consequently, unable to even formulate topics for discussion or disagreement. From Kuhn’s standpoint, there is simply no vocabulary for comparing and contrasting the respective theoretical positions of men operating under different scientific paradigms. Although Kuhn has revised his views several times since the initial publication of his book in 1962, the problem raised by Kuhn remains. Of course, Kuhn did not really “raise” this problem, as it is one that has long plagued serious thinkers (see, for example, Collingwood, 1940). Whatever its origins, however, the problem remains.
If, as Kuhn, and in an inconsistent manner, Mannheim and Mills, argue, the concepts and standards accepted as authoritative in different milieux lead scientists to define the world in different ways, how can one find an impartial standpoint of rationality and thus escape the throes of relativism? How can one, for instance, compare scientific theories and decide which is the best? From what viewpoint can this be done? Kuhn (1970: 264) responds to accusations of relativism by asserting that “one scientific theory is not as good as another for doing what scientists normally do”. But this statement is highly ambiguous. Does it mean what scientists usually (ordinarily) do, or what they ideally (properly, normatively) do? If he means what they “usually” do, then there is no basis for criticizing the actual practices of a scientific community. If he means what they “ideally” do, then apparently he is an absolutist, holding that there are abstract, timeless, criteria of rationality. Should the latter be Kuhn’s meaning, then he is abandoning his original thesis.
Whatever the ambiguities in Kuhn’s position, he has been responsible for forcefully reminding us of the problem. Furthermore, he has stressed the need for a more historical and sociological approach to science. And I think that certain aspects of Mannheim’s and Mills’ writings give rise to the same concerns. All three writers argue that men think in terms of the intellectual and social “frames of reference”, “universes of discourse”, “technical languages”, “social categories”, and “presuppositions” available to them in their own culture or group. These determine what they can see, what they regard as evidence, as compelling, as consistent, and so on. Since men’s standards and preferences vary between different cultures and historical milieux, what intellectual or social authority can be claimed for one set of standards or preferences rather than another? The thorough-going relativist concedes final authority to the standards current in a particular milieu, at the same time denying that those standards have any relevance or authority outside that milieu. This is almost precisely the position taken by Mannheim and Mills with what they call “relationalism”, where they argue that intellectual
criteria can only be formulated in terms of the perspective of a given situation. As I noted earlier, this strikes me as fully relativistic.
Central to Kuhn’s work and underlying the position of Mannheim and Mills then, is the necessity for philosophers of science and sociologists of knowledge to recognize the choice between the relativist approach, where the particular conceptual and theoretical systems current in one’s own scientific milieu are treated as locally sovereign; and the absolutist approach, where certain abstract, ideal, universal standards are imposed on all milieux alike. If one accepts the basic canons of Mannheim and Mills and the conclusions of Kuhn’s work, then one must choose the relativist position. Choosing the absolutist position, on the other hand, involves rejection of the basic tenets of the sociology of knowledge and of recent studies, like Kuhn’s, in the history of science. The decisive question, of course, is: Can one maintain the relativist viewpoint, and, at the same time, defend one’s own standpoint as rational? Toulmin (1972) has dealt at length with this question as to whether there is a middle ground between the absolutist and relativist extremes. While his arguments concerning this problem are intriguing, I do not feel that he has provided a satisfactory alternative to the absolutist/relativist dilemma.
But there remains a problem. If, indeed, people like Kuhn and Toulmin believe that some theories are better than others - so that they prefer their theories to those of Popper and Lakatos - how do they decide? Since they reject the existence of universal demarcation criteria which distinguish good from bad theories, what criteria do they and their audiences share that allow them to claim, and understand one another when they do, that “this” way of looking at science is preferable to “that” way? Of course, Kuhn claims that consensus concerning scientific knowledge is rather quickly arrived at in scientific communities. But this is certainly not the case in the philosophy or history of science, and most assuredly not in contemporary sociology. Given conflicting theories, how are some able to survive while others are not? Whereas Toulmin (1972) suggests a kind of survival of the fittest, this is not a terribly comfortable position to accept. Nor, on the other hand, can one be comfortable with the view that those theories survive whose advocates are the strongest. That is to say, while there are powerful elites in science as elsewhere, it is not, I believe, the case that “might makes right”. In short, if we reject the taken-for-granted belief in the rationality of science held by most sociologists - as I think we must - what are the full implications of this for the practice of science and for the way we individual scientists must live our lives? All of this is, of course, to raise questions for which neither I nor others concerned with these problems have ready answers. And, consistent with the line of inquiry followed here, we must face the question as to what is necessary for an “answer” to count as an answer.
And with all of these questions, the sociologist must ask “How do you know”? and “Why should we believe you”? As a beginning, I suggest that sociologists - especially those concerned with the sociology of knowledge and science - try to provide answers to a provocative pair of questions posed by Kuhn (1963: 395). He begins by observing that: “It is not, after all, the individual who decides whether his discoveries or theoretical inventions shall become part of the body of established science. Rather it is his professional community, a community which has and sometimes exercises the privilege of declaring him a deviant”. Kuhn then goes on to raise two questions that go to the heart of scientific and intellectual life: “Who are they to bear such responsibility? And on what ground should we trust their judgment?” The viability and health of the intellectual life of our time may be dependent on our ability to confront and answer these questions.
Barnes, S.B. and R.G. Dolby, “The scientific ethos: a deviant viewpoint”, European Journal of Sociology 11, 3-25, 1970.
Becker, Howard, “Review of Ideology and Utopia”, American Sociological Review 3, 260-262, 1939.
Ben-David, Joseph, The Scientist’s Role in Society, Englewood Cliffs, N.J.: Prentice-Hall, 1971.
Ben-David, Joseph, “Reflections on the state of sociological theory and the sociological community”. Paper read at Conference of the Research Committee on the Sociology of Science of the ISA, London, England, 1972.
Bennett, Jonathan, Rationality, London: Routledge and Kegan Paul, 1964.
Collingwood, R,G., An Essay on Metaphysics, Oxford: Oxford University Press, 1940.
Coser, Lewis A., Masters of Sociological Thought, New York: Harcourt Brace Jovanovich, 1971.
Curtis, James E., and John W. Petras, eds., The Sociology of Knowledge, New York: Praeger, 1970.
Feyerabend, Paul F., “Explanation, reduction, and empiricism”. In H. Feigi and G. Maxwell, eds., Minnesota Studies in the Philosophy of Science, vol. III, Minneapolis: University of Minnesota Press, 1962.
Feyerabend, Paul F., “Against method: outline in an anarchistic theory of knowledge”, Minnesota Studies in the Philosophy of Science 4, 17-130, 1970a.
Feyerabend, Paul F., “Consolations for the specialist”. In Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge, Cambridge: University Press, 1970b.
Hanson, Norwood Russell, Patterns of Discovery, Cambridge: Cambridge University Press, 1958.
Jarvie, I.C., Concepts and Society, London: Routledge and Kegan Paul, 1972.
Jay, Martin, The Dialectical Imagination, London: Heinemann Educational Books, 1973.
King, M.D., “Reason, tradition, and the progressiveness of science”, History and Theory 10, 3-32, 1971.
Kuhn, Thomas S., The Structure of Scientific Revolutions, Chicago: University of Chicago Press, 1962.
Kuhn, Thomas S., “Response to critics”. In A.C. Crombie, ed., Scientific Change, London: Heinemann, 1963.
Kuhn, Thomas S., The Structure of Scientific Revolutions, 2nd ed., Chicago: University of Chicago Press, 1970a.
Kuhn, Thomas S., “Reflections on my critics”. In Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge, Cambridge: Cambridge University Press, 1970b.
Lakatos, Imre, “Falsification and the methodology of scientific research programmes”. In Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge, Cambridge: Cambridge University Press, 1970.
Lakatos, Imre and Alan Musgrave, eds., Criticism and the Growth of Knowledge, Cambridge: Cambridge University Press, 1970.
Lukes, Steven, “Some problems about rationality”, Archives Européennes de Sociologie VIII, 247-264, 1967.
Mannheim, Karl, Ideology and Utopia, London: Routledge and Kegan Paul, 1972.
Merton, Robert, “Paradigm for the sociology of knowledge”, pp. 342-37 3. In James E. Curtis and John W. Petras, eds., The Sociology of Knowledge, New York: Praeger, 1970a.
Merton, Robert, Science, Technology and Society in Seventeenth Century England, New York: Harper Torchbooks, 1970b.
Merton, Robert, “Insiders and outsiders: a chapter in the sociology of knowledge”, American Journal of Sociology 78, 9-47, 1972.
Merton, Robert, The Sociology of Science, Chicago: University of Chicago Press, 1973.
Mills, C. Wright, “Language, logic, and culture”, American Sociological Review IV, 670-680, 1939.
Mills, C. Wright, “Methodological consequences of the sociology of knowledge”,American Journal of Sociology XVI: 316-330, 1940.
Mills, C. Wright, Power, Politics, and People, Ed., Irving Louis Horowitz, New York: Ballantine Books, 1963.
Mulkay, Michael, “Some aspects of cultural growth in the natural sciences”, Social Research 36, 22-52, 1969.
Phillips, Derek L.,Abandoning Method, San Francisco and London: Jossey-Bass, 1973.
Polanyi, Michael, “The republic of science”, pp. 1-20. In Edward Shils, ed., Criteria for Scientific Development, Cambridge, Mass.: MIT Press, 1962.
Popper, Karl R., The Open Society and Its Enemies, vol. 2., New York: Harper Torchbooks, 1963.
Popper, Karl R., “Normal science and its dangers”, pp. 51-58. In Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge, Cambridge: Cambridge University Press, 1970.
Reichenbach, Hans, Experience and Prediction, Chicago: University of Chicago Press, 1938.
Schaar, John H., “Legitimacy in the modern state”, pp. 276-327. In Philip Green and Sanford Levinson, eds., Power and Community, New York: Random House, 1970.
Scheffler, Israel, Science and Subjectivity, Indianapolis, Ind.: Bobbs-Merrill, 1967.
Shapere, Dudley, “The structure of scientific revolutions”, Philosophical Review 73, 383-394, 1964.
Toulmin, Stephen, Foresight and Understanding, New York: Harper Torchbooks, 1961.
Toulmin, Stephen, Human Understanding, vol. I, Oxford: Oxford University Press, 1972.
von Shelting, Alexander, “Review of Ideology and Utopia”, American Sociological Review 1, 664-674, 1936.
Whitley, Richard D., “Black boxism and the sociology of science: a discussion of the major developments in the field”, The Sociological Review Monograph 18, 61-91, 1972.
Winch, Peter, The Idea of a Social Science, New York: Humanities Press, 1958.
Wilson, Bryan, ed., Rationality, New York: Harper and Row, 1970.
Wittgenstein, Ludwig, Philosophical Investigations, New York: Macmillan, 1958.
Wittgenstein, Ludwig, Lectures and Conversations, Berkeley: University of California Press, 1972a.
Wittgenstein, Ludwig, Remarks on the Foundation of Mathematics, Cambridge, Mass.: MIT Press, 1972b.