Post-truth, or Post-Academic?: The Transformation of Science and the Obsolescence of “Reality”Abstract
Discussions of post-truth typically appeal to a notion of science as objective and therefore non-ideological. Objectivity has been attacked as itself an ideology, and a large literature has developed which claims that there are alternative “epistemologies” which reveal facts which this ideology excludes. The focus on the question what is a fact leads to “post-truth: this question is in turn treated as ideological. But this is misleading in two ways. One is that the kinds of facts that are supposed to be added by these alternative epistemologies have been part of sociology all along. The disagreement is not about facts, but about affirming the validity of viewpoints. Another is that sociology problematizes ideology itself, without affirming any ideology. This approach provides its own discipline: we come to understand what something like “oppression” means, and means for the people involved, in terms of the difference between our assumptions and practices, from our living life, and theirs.
Keywords:
post-truth; objectivity; sociological explanation; Mauss; practices
Resumo
As discussões sobre a pós-verdade apelam geralmente para uma noção de ciência como objetiva e, portanto, não ideológica. A objetividade foi, contudo, atacada como sendo, em si, uma ideologia, desenvolvendo-se nesse contexto uma ampla literatura sobre a existência de “epistemologias” alternativas capazes de revelar fatos que essa ideologia excluiria. O foco na questão do que é um fato leva à “pós-verdade”: essa questão, por sua vez, é tratada como ideológica. Mas isso é enganoso por duas razões. Por um lado, os tipos de fatos que supostamente seriam acrescentados por essas epistemologias alternativas fizeram sempre parte da sociologia. O desacordo não é, nesse sentido, sobre fatos, senão sobre a afirmação da validade dos pontos de vista. Por outro lado, a sociologia problematiza a ideologia como tal, sem afirmar nenhuma ideologia. Neste contexto, a abordagem aqui proposta oferece seu próprio enquadramento: passamos a compreender o que algo como “opressão” pode vir a significar - e o que significa para as pessoas envolvidas -, em termos da diferença entre nossas suposições e práticas, enraizadas na nossa forma de vida, e as delas.
Palavras-chave:
pós-verdade; objetividade; explicação sociológica; Mauss; práticas
It is common for discussions of post-truth to focus on the no longer fashionable fashion of post-modernism, and affirm the importance and value of science as the alternative to, and refutation of, the denial of “truth.” But this response overlooks the changes in science itself, and in its notions of truth-changes that fundamentally alter the epistemic status of science and its claims to authority. John Ziman was a pioneer in recognizing the changes. He saw that rather than pursuing the older, self-selected goals of a comprehensive understanding of reality, science by the 1980’s had become an activity which provided “reliable knowledge” in response to the needs of funders for this knowledge. Ziman was writing before the transformation of science by the three trillion dollars of funding for biomedical research since 1995. This period produced a new stage in science: of “reliable enough knowledge.” The new demand was for knowledge that met bureaucratic and regulatory standards for knowledge, especially for drug efficacy, but also for policy. One sign of this change was the rise of the Randomized Controlled Trial as the “gold standard” of research. This was a conventional means of showing effects. But it was a low threshold for knowledge, and produced “knowledge” without understanding. And it - together with a rise in statistical methods, modelling, stimulation, and imaging - created a new kind of “science” that was prone to error, fraud, replication issues, and corruption. But most significantly, the concerns of this kind of science were not with a comprehensive understanding of reality, but with the effects of intervention in complex systems that were themselves poorly understood. This much weaker notion of science is itself a form of post-truth.
Sociology arrived at this weaker notion of science first, but, so to speak, from a different direction. From the start sociology has been faced with demands for utility, which is to say for good enough knowledge to deal immediately with particular social problems. Because reform was typically connected to social movements, sociology was also pressed to be public, in the sense of being produced in a form which could be part of public discussion or background learning. These constraints, together with constraints on funding which follow from them, have always defined sociology. To be sure there has also been “theoretical” sociology and pure sociology that was closer to a philosophy of the social, but these efforts have normally involved simplifications of complex processes that make them more closely akin to ideology; indeed, social theories typically have been denounced as ideologies because of their ideological implications, intended or not. So at the theoretical level sociology has never had much coherence, and consists, as empirical sociology does, of competing models. These go by various names, but they are in the end what Weber called ideal-types. And Weber recognized very early that this is all sociology as a conceptual enterprise could consist of: ideal-types which were imperfect but clear simplifications of social reality (Weber, [1904] 2012).
In what follows I will explain how the problem of post-truth arises, and why it is an artificial and misleading way of thinking of sociological knowledge, though in some ways an understandable error. I will then turn to the real problems of sociology, and why sociology produces such highly variable results and claims. These are serious enough, but the variability in question is not the dead end variability of post-truth.
The General Problem
Sociological topics do not arise simply in the framework of sociology as a “science.” A sociological topic is a difference of some sort, for example, between the different rates of occurrence of some outcome in two groups or categories of person or institution which can be given a sociological explanation, or of different practices or values, or of some anomalous event, such as a riot or revolution, that reveals difference between social categories. There is in some sense a generic “sociological” interest in these facts: if we stumble on, or find in the course of a systematic survey, some difference of this kind between social categories, it is a puzzle. But our interest in these puzzles is largely conditioned by our non-scientific or valuative interest in the outcomes in question: crime, unemployment, poverty, conflict, and so forth. And the comparisons may be to hypothetical alternatives: states of peace, prosperity, co-operation, and well-being that do not exist but which someone thinks could exist.
Often these alternative states exist only in the form of ideology, that is to say ideal possibilities. Rousseau said, in the opening sentence of The Social Contract ([1762] 2012), that “man is born free, but is everywhere in chains.” His detractors commented that “sheep are born carnivorous, but everywhere eat grass” (Hardy 2018, 191). Is there a sociological question here? In this form, a pair of claims about human nature-one affirming natural freedom, one pointing out the absurdity of the analogous reasoning as applied to sheep-there can be no resolution. Claims about human nature and human possibility can only hope to be resolved if there is some common ground of evidence or some shared form of reasoning that both sides accept. This is the beginning of the problem of post-truth.
Normally the problems of sociology are ideological, but we need to be careful with this highly contested concept. It is customary, if problematic, to divide ideologies into “total” ideologies, meaning the largely tacit ideational background to a way of life, and conscious ideological constructions, which are opposed to present conscious and unconscious ideologies. The terminology here is problematic. A simpler way of putting it is provided by Fred Brownlee, who distinguishes “the opinions that men discuss” from “the ideologies that men fight about because they cannot discuss them” (Brownlee, 1943, ix), which we may regard as the tacit and emotional substrate of belief.
Sociology not only examines ideologies, its content depends on, and its problems arise from, ideological materials. The social world is partly the product of the ideational background to action and belief. When we encounter something different-a different pattern of social action, or different results which appear to be based on a different tacit background than our own-we are faced with a puzzle. Similarly, if we believe that people are born free and are everywhere in chains, we are faced with the puzzle of why? Each case involves some standard which is not being met, whether this is the implicit standard of normalcy in the sociologist’s home setting or society, or an ideological construction, such as the society free of alienation, exploitation, oppression, or inequality. These puzzles, whether they are generated by the difference between our tacit background and theirs, or between an ideal standard and reality, are the things that sociology claims to solve.
There is a trap concealed in this basic feature of the construction of social things to explain. The reason the failure of our expectations about the behavior or beliefs of others compels us to look for an explanation is normally connected to the moral salience of our expectations for us. We are not sine ira e studio in relation to our expectations, the expectations that lead to the intellectual puzzle in the first place. So there is a temptation, which the reformers and moralists regularly succumbed to, to simply condemn what we find in other settings or traditions or social groups as “failure” or moral culpability. And this implied moral superiority. When this was done in the context of race, it had the effect of stigmatizing, stereotyping, and ignoring individual differences. It is not surprising that the reaction to this kind of condemnatory (and non-explanatory) moralism was to problematize the viewpoint which was the source of the expectations, and to reverse the direction of the condemnation and claim moral superiority over these moralists. Thus attacks on “whiteness” pathologize rather than explain “white” behavior sociologically. This is indeed the path to post-truth. And the strategy of turning the tables on the moralists has a long history in sociology.
Writers like Robert and Merrill Lynd did this in Middletown (1929), attacking the moral superiority of what the moralists of the time embraced as “rugged individualism,” concluding that the unwillingness on the part of the subjects to pursue collective action and their moralistic attitude of blaming the individual for their failures rather than blaming collective arrangements was the real failure. Attacks on individualism of the same kind have been a staple of American sociology, through to more recent texts such as Reisman’s Invidualism Reconsidered (1954), Bellah’s Habits of the Heart (1985), and Putnam’s Bowling Alone (2000). The problematization of “Whiteness” works in the same way: it is about condemnation and pathologizing rather than explanation. “Psychological” facts do get explained sociologically: the differences between Asian and western children with respect to attributions of responsibility and individuality, for example, are clearly psychological facts, that have social consequences, but are also the result of the different social environments that produce them. To pathologize one or the other as a psychological failing is to substitute moralizing for explanation.
Facts
To make something into an empirical question, we need to agree on what “empirical” means, or more simply what kinds of facts count as facts for us. If we are agreed ideologically, we may still disagree about this question. But if we disagree ideologically, we may also agree about this. They are separate issues. If there was a conventional rule about facts to which we all adhered, this would not be an issue between parties on the opposite ideological sides, as long as they accepted the conventions. But even if they did, there still might be differences of interpreting the agreed-on facts, because more than one interpretation may be consistent with the facts. The advocates of post-truth epistemologies think that ideologies and facts line up: that there is an affinity between whiteness and certain epistemologies that sustain it, which reject the alternative epistemologies of the oppressed. But is this true? Is this why there are disagreements? Or is something else going on? The matter turns out to be complicated.
To get a start on understanding what goes wrong in these controversies, where we fail to agree, we can begin with the well-developed conventions about what sorts of statistical constructions of data - operationalizations-- count as facts. There are standard conventions for ethnographies of simpler societies, as well. Henrika Kuklick tells of writing on the history of anthropology, and being told by the historians that the only real evidence is archival, and by the anthropologists that the claims made about people are not really valid unless you have encountered the person directly, as an ethnographer does (Kucklick, 2014). In part, these conventions simply reflect the evidence that is available. One cannot do ethnography on the dead, so archives, inscriptions, numismatics, and monuments may be all the historian has. The anthropologist has no archives, but has the possibility of interaction, questioning, observation, and dialogue with the living. It is tempting to ground these conventions in something more - to provide some sort of general justification for them in terms of universal principles of scientific methodology - or to equate the conventions with the pursuit of the discipline itself.
To fetishize these conventions is to forget their origins and the purposes they originally served. In the case of ethnography, the rise of systematic ethnographic observation designed to enable comparisons with other simpler societies was an improvement on the travelers’ stories that preceded them, which concentrated on whatever struck the traveler as odd or marvelous. In the case of history, archival research was an improvement on the reporting of stories that had been passed down orally, and been transformed in the telling, modified to suit later purposes, and subjected to the political or social biases of previous generations. And each of these conventions came with corrective mechanisms. One must, as a historian, be critical of sources-source criticism is a basic adjunct to any use of the written past. Ethnographers can be fooled by informants as well, as Margaret Mead was accused of being (Freeman, 1983).
But in the end these are conventions which rest on common sense. Turning them into objectivity-guaranteeing machines occludes the reasons they were established in the first place, and their limitations. Travelers’ stories left out context. Archives compelled their users to assume a great deal about the practices and intentions of the original authors of the written material. These assumptions were not universalizable, and could not be warranted on universal grounds: they made sense of particular archival sources in particular contexts. One had to know in each case something that was not universal knowledge. To make sense of the object required local knowledge or equivalent assumptions about how the object was used and why, or how the archival remains were produced and selected to survive. The same kind of non-universal background knowledge was needed in ethnography. This is an important point to which we will return in the next section.
These ancillary features of evidence are easily forgotten when an apparently universalistic standard is applied. In anthropology, there was a quest for universal generalizations, which required data in universal form. The Human Relations Area Files were designed to provide just this: a standard set of facts that could be analyzed quantitatively. What was the result? Not much. The essayistic descriptions in the files were more informative. And the entire effort was followed by a revolution in anthropology under the heading of “thick description,” an implied contrast to the thin description characteristic of the HRAF (Geertz, 1973). But this was itself followed by a kind of collapse: thick descriptions were no longer comparable in a way that lent itself to generalization. The lack of opportunities to study “primitive” societies turned anthropologists’ interests to accessible topics and to literary analysis (Boon, 1982) and critiques of older ethnographies. The new topics came with political baggage: ideological conflicts internal to the cases themselves, as well as new conflicts imposed on the cases by people with agendas taken from their own societies, motivated criticisms and alternative interpretations, and critiques, of the same data.
With these new agendas, and in light of the literary analysis of past ethnographies and ethnographers private diaries (e.g.Wax, 1972), the older idea that ethnographic conventions produced objective knowledge was discredited. Then a new element was added: claims about the epistemic validity of the approaches to the subject associated with the external and internal ideological agendas themselves. These claims were made not in the name of objectivity but of morally superior truth. The Imperial gaze is not just false, it is not even a competitor to the post-colonial, subaltern gaze. The male gaze, similarly, is an expression of power that distorts; the point of view of the subordinated is not merely different, but better (see Harding 2003). But significantly, these different perspectives are not universalizable. They belong to different people or groups. One cannot fully enter into the perspective intellectually: it requires having the appropriate group experience. There is no unengaged point of view, no such thing as a free-floating intellectual.
What does this have to do with sociology? Can’t one still rely on its methodological conventions? A great deal of advocacy, in all fields of social science, does rely on these conventions. The point of advocacy research is to persuade: conventional standards of data construction or at least the appearance of adherence to these standards is a powerful means of facilitating persuasion. The data is not enough, however: it must be interpreted and analyzed. And bias often slips into the presentation of the data itself.
In the case of qualitative data this takes the form of selectivity and loaded description. In older forms of reporting, such as works like W.E.B. DuBois’ The Philadelphia Negro (DuBois, [1899]1967), there was no attempt to hide this: value judgements were made in almost every description, so they were easy to identify. In other cases they are concealed, and concealed from the authors themselves, whose bias flows from their own experiences and tacit background. In quantitative work there is a great deal of “assumption” that needs to go into statistical analysis. Sophisticated users can manipulate apparently innocuous assumptions to change outcomes. But because the goal of advocacy research is to persuade, the presentation of the results-the data and the model―follows conventional standards, or appears to.
The more radical kind of attack on objectivity involves the claim that particular methods, even the concept of objectivity itself, are ideological― part of “White” sociology or whiteness, and therefore a form of racism (e.g. Rizvi, 2022a, 2022b). This is where post-truth entered: with the idea that every fact was the product of a viewpoint, and that the only way of deciding between them or validating them is a non-truth one. Variants of this kind of claim, which serve to problematize any grounds for resolving conflicts between perspectives, are of course not new. Feminists made similar claims, and there is a long history of denying the ideal of pure science as a “bourgeois” ideological construction. But often the same kind of claim was concealed under different language, with the same effect. One variant of this argument is that facts, such as the facts of ethnography, are not “complete” unless they are framed in terms of the role of capitalism in producing them. To be “incomplete” in this sense is to be ideological: it is a kind of omission (Wacquant 2002). This appears to be a neutral demand for more truth, but of course it assumes as a matter of fact that capitalism produced the realities that are being described by the ethnographer. Because the alternative of no capitalism is not a comparison to anything real, but to an ideal whose realizability is open to question, we cannot make the kinds of comparisons that arise through the discovery of differences between societies or social settings. We can only imagine, and our results would be equally imaginary. Similarly for the study of “inequality.” We can measure inequality by defining it and comparing statistics to the definition, and compare degrees of inequality according to the definition, but this is not the same as saying that “equality” is possible and, therefore, that one’s explanations of inequality describe real mechanisms.
The radical claims of perspectives rejecting conventional methods are, in a sense, a response to radical claims for the conventional practices; claims that they alone produce objective knowledge, or that they are uniquely scientific. These are similar claims to exclusive validity, and there are others. One claim of this kind that matches up with the claimed superiority of perspectives associated with particular groups is the appeal to personal experience. One of the purposes of appealing to “personal experience” is to underline the fact of ownership: you cannot have my personal experiences, and I cannot have yours. We can only testify about them. But the basis for judging them is itself conditioned by our personal epistemic viewpoint which is in turn valid for us, based on our personal experience. We cannot escape this circle. We have to confess and reflect on our “positionality,” but doing so does not allow us to escape it. It only serves to qualify our statements by acknowledging that everything we say is from this limited point of view.
The more sophisticated response has been to try to reconcile the two-considerations of fact and considerations of what counts as fact―in some higher vision. This is the strategy of the most famous statement on objectivity in social science, Weber’s ([1904] 2012), but also of Mannheim’s sociology of knowledge and its idea of the free-floating intellectual ([1929] 1936); [1924]1986). These alternatives have not resolved the issues. Why not? Accounts of objectivity, including the reconciliations provided by Weber and Mannheim, can be dismissed as themselves being ideological, as indeed the concept of objectivity is now being itself dismissed as a manifestation of “whiteness” and patriarchy, on the basis of alternative consciously constructed ideologies.
The normal response to this problematic has been either to define objectivity in a general way that purports to be non-ideological (though this may take a complex form that acknowledges the valuative or ideological constitution of objects), or, alternatively, to affirm the truth of a particular ideology as a condition of genuine sociological knowledge - to say that the problem of capitalism, or the oppression of women, is always fundamental. This is the path to post-truth: it reproduces the conflict or variability that arises on the level of fact at the level of theories about what a fact is. And in each case there is a strategy of selectivity. The conflicts about interpretations of the facts are conflicts about which facts to select as anchors for our interpretation; at the level of theories about facts, philosophies or methodologies (often called, misleadingly, in these discussions, epistemologies) the conflicts have to do with which facts are regarded as the “real” or probative ones, and which have some lesser status.
These conflicts are irresolvable, because they involve considerations about how to resolve them that are variable, and on which there is no agreement about how to resolve them. Once we have reached this stage, there is no higher “methodological” appeal - no method that can generate common facts to settle the differences. There is perhaps an ethical appeal: to say that some “position” is morally deficient, and therefore unworthy of consideration, or the older Marxist claim that it represents the final class in history, whose victory would vindicate both the validity of the theory and its alternative to bourgeois science. We can call this the post-modernist denouement: the end of the possibility of agreement on the basis of the facts, and the possibility only of conversion.
With these steps we arrive at post-truth. But we need never arrive at this point if we acknowledge the nature of the variability in interpretation, and our limitations in overcoming them. To do this is not to embrace post-truth, but to understand the limitations of the kinds of knowledge available to us as sociologists, and the limited probative value of the “facts” we can appeal to.
Closer to Practice
Despite the assertion that there are alternative “epistemologies” to Whiteness and that these have already been developed and are being developed, there are no such things. “Story-telling” has sometimes been presented as such an alternative in anti-racist training. But story-telling has always been used in sociology: indeed, Weber’s Protestant Ethic ([1930] 1958) is a prime example. It is as vivid as it is because many of its episodes are taken from his own family history. Similarly for “experiential knowledge”: Weber experienced the ethic he writes about: that is what makes him so good at describing it. Sociologists have historically employed a form of research and research presentation that has used some mixture of statistics and quotations from respondents which provide some interpretational context for the statistics. This is the form of all the classic works of sociology, the empirical studies of American sociology in the interwar years, and the classic studies of later years. In the later part of the twentieth century there was a certain amount of causal modelling and more recently statistical modelling of other kinds that depended on, but did not exhibit the basis of, their assumptions about the plausibility of causal connections. These two elements, numbers and testimony, are forms of evidence. They are the quantitative and qualitative sides of the discipline. And participant observation adds the element of experiential knowledge.
What is really at stake is not “epistemologies,” but conclusions. The alternative points of view disagree with the conclusions reached by “colonial” thinkers about such things as the reasons for non-development, and disagree with the policy implications of these conclusions. The case of Black sociology as opposed to the “epistemology” of White sociology, is exemplary. Black “epistemology” is defined, according to its practitioners, by (1) the assumption that race is a central organizing force in society; (2) the need for historical and global contextualization of the social phenomena being studied; and (3) the assumption that sociology is a tool of emancipatory knowledge that challenges racism, injustice, and oppression, serving to empower racial and social justice projects. (cf. Brunsma and Wyse, 2018; Ladner, 1973; MacLean and Williams, 2008; Staples, 1976; Steinberg, 2007)
The first “assumption” is a conclusion, and a social theoretical claim, but is called an assumption solely for the reason that it is not, and probably cannot be, established by another means. The second is a nod to another theory, but an unstated one, that there is a global historical process of racial oppression tied to capitalism. This too is a conclusion without a justification, so to it put into the box of epistemology is to put it where it does not belong: it is not part of a theory of the nature of knowledge but an interpretation of the factual world. Emancipatory is a term with many possible meanings, but here it seems to mean “validates particular beliefs that enable people to make claims on others for different treatment,” or “supports policies that are hoped to achieve results, such as equality, that are taken to be desirable.” The idea that sociology is, or should be, emancipatory, is thus a value choice about the kinds of conclusions that sociology should reach.
This mode of thought is familiar from Marxism, which has the same general structure, but which is more obviously circular. Here it is from Žižek:
To anyone minimally acquainted with Western Marxism, the answer is clear: Hungarian philosopher Georg Lukacs demonstrated how Marxism is “universally true” not in spite of its partiality butbecauseit is “partial,” accessible only from a particular subjective position. We may agree or disagree with this view, but the fact is that what Foucault was looking for in far-away Iran - the agonistic (“war”) form of truth-telling - was already forcefully present in the Marxist view that being caught in the class struggle is not an obstacle to “objective” knowledge of history but its condition.
The usual positivist notion of knowledge as an “objective” (non-partial) approach to reality which is not distorted by a particular subjective engagement - what Foucault characterized as “the pacifying, neutralizing, and normalizing forms of modern Western power” - is ideology at its purest - the ideology of the “end of ideology.” (Žižek, 2021)
This is the thinking that converts a conclusion - in this case about the end of history in the success of the proletarian revolution - into a premise, and therefore an “epistemology,” in the sense that it by definition converts anything opposed to it, in this case the idea of objectivity, into a counter- ideology - “an ideology at its purest.” But it pretends to avoid the conclusion that everything is ideology by claiming truth for itself, in the form of the idea that class struggle is a condition of truth. But this is also a conclusion, and one that may be true but is never defended, but simply asserted or assumed.
There is some truth to the argument that objectivity is an ideology, however, as well as its claim to non-partiality. In some sense, any model or account can be converted into an ideology. But it requires adding something to do so. If we look at some epistemic examples, we can see how this works. Return for a moment to Kuklick’s (2014) observation about the differential evaluation of archives and personal testimony and acquaintance between historians and anthropologists. Each discipline is practicing, and to a large extent defined by, a particular form of epistemic discipline: they do not regard, or accept as genuinely probative evidence, what the other discipline does. If this kind of discipline is taken as an absolute, it becomes what we may call “ideological.” And in this form it fits nicely into the progression to post-truth. But in practice and in the normal way in which people think about these topics when they are not subjecting themselves to professional discipline-and this common sense mode is always there as a check on the results of epistemically disciplined reasoning-they acknowledge that the other kind of evidence is evidence of something, and that this evidence is relevant to the conclusions one is drawing. There is a good example of this in the history of sociology itself. Lloyd Warner, in his Yankee City studies, hewed to the anthropological practice of relying solely on the testimony of his informants. As Stephan Thernstrom later showed, the historical testimony of his informants was simply false-a kind of local mythology (1965). The common sense conclusion to this would be that the discipline Warner practiced was mistake: not an epistemology, but a prejudice about what facts mattered that led to a failure to comport with facts he ignored.
But there is much more to the idea of epistemic discipline. Here is another: the slogan that the novel never lies. There is a large kernel of truth here: novels have to be plausible to their audiences. So in some important sense they must be “true” to the topics they discuss. Even better, they are historical artifacts, so we can get some sense of the intended audience, which serves as a kind of validator of the relevant truth. Jane Austen may not have been a professional ethnographer, but her novels are evidence that is in some ways better: other people recognized the truth of the conventions and attitudes she describes. “Theory,” oddly, is another form of epistemic discipline. When one generalizes, for example about ethic and racial conflicts, one tests the explanations that claim to capture the essence of the phenomenon, such as “white supremacy.” By seeing the case as an example of something larger, involving Black and Brown people or caste and religion as well, is to provide a means of selecting from the possible “essences” that is not a matter of personal preference or local passion.
Marcel Mauss struggled to articulate this point in his own ethnographic work, in trying to distinguish the kind of knowledge that led to what he took to be the object of sociological knowledge, the total social fact. And in the course of doing so he addresses the issue of partiality. He came up with an awkward formulation, but one which captured the problem. As Claude Lévi-Strauss paraphrased him,
An appropriate understanding of a social fact requires that it be grasped totally, that is, from outside, like a thing, but like a thing which comprises within itself the subjective understanding (conscious or unconscious) that we would have of it, if, being inexorably human, we were living the life of the indigenous people instead of observing it as ethnographers. (Lévi-Strauss 1967 [1950], pp. 30, 31)
What he is getting at is that we need to understand our “scientific” external facts from the inside of the experience of those who are enacting it-we must understand them. And he even points to Jung as someone who at least points to the common unconscious inheritance we might be drawing on in this understanding of people living alien forms of social life.
Understanding, in short, is itself a kind of evidence. And including understanding is a form of epistemic discipline: if we don’t understand them as if “we were living the life” we can’t understand the total picture. We can leave the term “total” aside, and see this form of discipline as characteristic of the most mundane sorts of quantitative analysis in sociology. If we generate numbers about something, and show statistical relations between two or more sets of numbers, we can’t do things like make causal inferences successfully without some sort of understanding of what the people in the numbers are doing, and why. This usually takes the minimalist form of making assertions about the plausibility of a causal connection between variables, and leaving the bulk of the probative work to the correlations themselves. But the causal claims depend on our understanding the people involved.
The fact that we need this minimal understanding of others to separate causal correlations from “mere” correlations shows that these forms of evidence cannot be abstracted from, and elevated above, the other forms of evidence we have. We can do this for “methodological” reasons, that is to say we can impose a particular form of discipline on ourselves and see what results. The ethnomethodogical tradition of Conversational Analysis tried to do this by restricting its evidence to content that was part of the recorded conversation itself. But what it discovered was that this didn’t help with such issues as the role of gender, and didn’t yield much in the way of theoretical results. “Positive” economics did something similar: it said that we should ignore the reality of human decision making at the experiential level and limit ourselves to abstractions, radical simplifications, that have predictive power. If we treat these strategies as “methodological” we are acknowledging that we are not using them to make claims about ultimate reality, but seeing how far we can go and what results we can get within a particular form of epistemic discipline.
Alternatively, we can invoke the same purism about forms of evidence, on the grounds of a particular theory of “science.” This does ideologize these disciplinary practices, in precisely the way that leads to the post-truth problematic. Put differently, the kind of epistemic discipline that demands numbers exclusively as evidence, or alternatively demands perfect understanding to the exclusion of other forms of evidence, is a kind of impoverishment. Discipline has a price- as we can see in the case of reliance on archives versus personal testimony. We would have a worse history if we submitted to this discipline and ignored personal testimony: what the material in the archives meant to the people “living the life” is lost. But merely relying on the testimony of informants loses something too: their understanding of their world, and of themselves, is limited, and limited in ways that the material found in archives can and often does reveal.
There is, however, an issue left over from Mauss - the idea of a total social fact, or a total understanding to which we can compare a partial one. Is this possible? Or is it a misleading ideal? Do we give up on the idea of knowledge, and fall back into post-truth relativism if we reject the idea of total facts and total understanding or knowledge? Or are “total” and “partial” simply misleading ways of talking about the same problem? This is a point to which we will need to return.
Morality
The discussion so far assumes, tacitly, that there is a more or less clear common understanding of what sociology and the sociological perspective are about, or to put it differently, what the object of sociology is. This understanding is challenged by the idea that sociology should be “emancipatory.” Presumably “emancipatory” and “total” are not the same thing. If so, a total understanding of a total social fact would be emancipatory by definition. Presumably there are non-emancipatory sociologies, at least there were in the past, or alternatively that to be an emancipatory sociology one must do something different than plain sociology, or certain other kinds of sociology. Demanding that sociology be emancipatory, rather than treating it as intrinsically emancipatory, raises the question of what the difference between emancipatory and non-emancipatory sociology is, and what their relationship is to the “whole,” whatever that might be.
One way of thinking of this idea might be this: sociology is subject not only to various kinds of epistemic discipline, but also to a kind of moral discipline. And one might also try to connect this to truth: to say that a failure to include certain moral considerations in one’s sociological accounts, such as the omnipresent social phenomenon of oppression, is, like the partiality that results from restrictive epistemic discipline, itself a source of partiality that undermines the project of sociology and its claims to truth, and therefore its epistemic validity.
One thinks here of Weber’s discussion of ideal-types: that they are inherently one-sided, and therefore partial. This is a “methodological” point rather than an epistemic one. But Weber goes to the epistemic and then to the ontological level when he comments that the world, meaning the total, is a meaningless chaos which a culture selects from and makes order out of. The chaos, for Weber, extends to the social: we don’t have a non-chaotic total social object. We only select, just in the way that historians also select their facts from the mass of historical material - and can only select: it is impossible to have the whole (Roth 2021). This itself is a kind of relativism, or even post-truth. But it salvages truth as an internal fact. Once we make this selection, we can say things about cause, facts, and so on, and also compare our one-sided ideal-types to the rest of what is selected by culture, a selection that is tacit and largely given to us by our ordinary social experience-out of living the life we live, in Maussian terms. And we can compare our selection to the selections of others.
But here we arrive at a conflict, or a conundrum. Is there a conflict between sociology based on the standard of living the life and a sociology based on the recognition of emancipation as a value? There is an important clue. Living the life is largely tacit: the sociologist or ethnographer tries to articulate this as she encounters an alien social world, but is at the same time dependent on their own culture or tacit background, in ways that I will explain shortly. From this point of view, the project of understanding and explaining in sociology is necessarily partial, in the broader sense that Weber tries to capture with the notion of culture as a selection from the meaningless chaos of the world. We understand, and impute plausibility to causal claims, on the basis of our tacit cultural background. We explain to each other, an audience of similarly situated people: WEIRD or Western, Educated, Industrialized, Rich, Democratic people of the twenty-first century. But it is also partial in the “methodological” sense that constructing ideal-types or concepts is even more limited than this-one-sided, as Weber says of his own abstractions.
Is “emancipation” an additional kind of discipline, similar to the kinds of epistemic discipline based on particular conventions of fact production? If so, it makes the perspective more “partial” by adding a kind of restrictive discipline that makes be selective in favor of the facts that produce emancipation, so it makes us say less, not more. Or does it add to the content of the discipline, by including something neglected by the wrong kind of sociology, namely the massive fact of human oppression in all times and places. The second answer would expand sociology; the first would constrict it.
The second answer, which is to say that a full sociology must include, or even make central, the fact of oppression, seems at first to be an expansion. But to absolutize the notion of oppression de-sociologizes the concept itself. In Mauss’s terms, oppression is both objective and subjective, part of the externally recognized world of social fact and of “living the life.” But we know, and it should be a basic lesson of sociology, that what people experience as just, oppressive, and so forth is highly variable. Justice, as Hans Kelsen famously observed, is exceptionally variable. So if oppression is unjust treatment, it will vary wildly as well. So the sense in which oppression is universal is no more than the sense that in every system of justice there are also injustices, disagreements about justice, and so forth. To properly sociologize these requires not just looking at them from a WEIRD perspective, but to live the life and also identify, from the external point of view, the various features of the world of which they are a part.
What would sociologizing actual oppression, or in Maussian terms, “living the life” terms, look like? Sociology begins with “moral” concerns, but in the “living the life” sense these are our moral expectations, often tacit, that are violated in a new social context-where people behave and act in puzzling ways. The puzzles have a typical genesis: they begin with some external fact, such as poverty or political differences, or demographic or disease patterns, where there is a difference. As I have already noted, sociology studies these differences in more or less conventionalized ways. So some apparent differences can be explained more or less in terms of these conventions. One can explain a demographic age distribution on the basis of a prior age distribution, for example, and show that the underlying reproductive and life cycle patterns are the same. These ways allow for the construction of “facts.” But more importantly, the “living the life” aspect is the next stage in the process, when the external explanations are not enough: when we arrive at a difference, for example in reproductive patterns that can’t be explained by such purely demographic considerations as a prior distribution, we turn to the practices, in a broad sense of the term, that produce these differences.
This is the case with oppression. We can identify external facts-life spans, income differentials, and any of many more markers of our concept of oppression-and compare “their” oppression to ours. But the “externality” of these facts is, in a sense, deceptive: statistics seem like solid facts, but they depend on contextual understanding to be constructed. Methods are normally “mixed,” whether they are labelled that way or not. And in this case and indeed generally, there is an internal or subjective element to our construction of the facts: they are WEIRD definitions; often contested WEIRD definitions. They are part of our life, as we live it. It is a factual, sociological question whether they are part of the lives of our subjects, and also parts of the larger fabric of these lives. This is, I take it, Mauss’s point about total social facts.2 So to truly sociologize oppression we can start with our WEIRD definitions, but once we get to the point where we are sociologizing about the subjective “living the life” differences, we are faced with the need to “relativize.” We recognize that ours is not the only way to think of justice or oppression, and that it has no special grounding outside of its sources in our own subjective experience, our living our own life. This is not post-truth, because it is not a universal relativization: it is a very specific one, tied to a particular topic and the particular fact of the diversity of practices that are part of the context of this particular topic in these particular societies.
Differences in interpretation, or explanation, in the face of sociological puzzles, often have “moral” implications, in the sense that they fit or fail to fit with someone or some group’s moral preferences and judgments. They may or may not be flattering, or fit with larger ideas about “justice” or the social good. So the temptation to absolutize these preferences and judgements is always present, and sometimes overwhelming. Indeed, it can be confused with the lesson of sociology itself, which as Comte put it, was the fact of our social dependence in virtually every domain of life. And we can turn this into an ideology, even a religion, with little difficulty, as he himself did.
The confusion can be more subtle, and more pervasive. The traditional issue in American sociology has been with individualism, which sociologists have generally not liked. But other topics, such as racial attitudes and evaluations, have a prominent “moral” aspect, and larger ideas about justice and the social good that have motivated some of the study of these topics are also ideological ideals. In one sense, these ideals are simply ways of defining a problem, though WEIRD ways. And they are problematic, in that the idea of a completely egalitarian society is an ideal, not an actual society with a complex of practices and forms of life. There are no social orders that are “just” or fully collective and none that are purely individualistic. So one can compare actual social orders that are more or less individualistic, in the external sense, that is, so to speak, our external point of view based on our concepts, not an external point of view as such.
This is a circle we cannot, as sociologists, escape. Nor should we. Sociology does not provide us with what Daniel Dennett calls a “skyhook.” Dennett had in mind the alternatives of evolution and intelligent design. A skyhook was a self-grounding external doctrine which “solves” problems in the inside of a discourse from outside of it (Dennett, 1995). In the context of sociology, a skyhook would be a way of escaping “living the life” in its complexity and interrelations and in its diversity and variation, and explaining selectively those things which fit with the premise we are wedded to, such as the idea of class struggle, or gender oppression, or whiteness. And this takes us back to the original problem of post-truth. Post-truth depends on skyhooks-on the idea that on can resolve conflicts between ideas by absolutizing premises from outside the discourse.
Where does this leave sociology as a science? And where does it leave the question of its similarities and differences with the rest of “science” as it is currently practiced? I began with causal models. What I have tried to show is that these models depend, in the end, on freezing social concepts, and assumptions about practice, which are, in the chaos of which Weber spoke, not freezable, or freezable only in a practical sense, a sense of temporary utility. And this is what holds for contemporary “science on demand” as well. Its models, for example, of a disease, are good enough for our immediate purposes, such as inventing a therapy. But this style of science implicitly abandons the quest for the whole, the non-utilitarian business of fitting the pieces together. It stops with its practical successes.
Sociology faces the same issues. It has its practical uses, its limited frozen models. And it has a larger “whole” which is very different from these models. Mauss was haunted by the whole, both in the sense of the total social fact, which is a problematic construction (though one we might salvage by treating it as a methodological device for defining an intermediate target for explanation), but also by the larger whole indicated by his appeal to Jung. The larger whole is either the meaningless chaos of Weber, or a coherent total science of which sociology is a tiny subdivision, but one related to and dependent on the rest. But this is a distinction without a practical difference: the total science is not one which we will ever achieve.
Michael Oakeshott had a nice phrase for the necessary limitations and epistemic disciplines, and the assumptions, or as he called them “platforms,” which we must construct to frame any problem, in science or in thought generally. The term was “arrests.” It is a reminder that it is only a temporary stop, a provisional one, not an absolute one. That is the lesson that both post-truth thinkers and absolutists of various kinds, including the moral absolutists, have failed to learn: that we can achieve results, but only by stopping inquiry at a point which may be surpassed.
References
- BELLAH, Robert. Habits of the heart: Individualism and commitment in American life. Berkeley: University of California Press, 1985.
- BOON, James A. Other tribes, other scribes: Symbolic anthropology in the comparative study of cultures, histories, religions, and texts. Cambridge: Cambridge University Press, 1982.
- BROWNLEE, Fred. Preface, in: JOHNSON, Charles S.: To stem this tide: A survey of racial tension areas in the United States. Boston and Chicago: The Pilgrim Press, 1943.
- BRUNSMA, D.L. and J.P. WYSE. “The possessive investment in White sociology”. Sociology of Race and Ethnicity, v. 5, n. 1, pp. 1-10, 2018.
- DENNETT, Daniel C. Darwin's dangerous idea. New York: Simon & Schuster, 1995.
-
DUBOIS, W.E.B. The Philadelphia Negro: A social study. New York: Schocken Books, [1899]1967. https://archive.org/details/philadelphianegr001901mbp
» https://archive.org/details/philadelphianegr001901mbp - FREEMAN, Derek. Margaret Mead and Samoa: The making and unmaking of an anthropological myth. Cambridge, Mass.: Harvard University Press, 1983.
-
GEERTZ, Clifford. “Thick description: Toward an interpretative theory of culture”, in: The interpretation of cultures. New York: Basic Books, 1973. https://philpapers.org/archive/geettd.pdf (accessed 15 June 2021)
» https://philpapers.org/archive/geettd.pdf -
HARDING, Sandra S. “How standpoint methodology informs philosophy of social science”, in: S.P. TURNER and P. ROTH (eds.). The Blackwell guide to the philosophy of the social sciences. Hoboken, NJ: Wiley Blackwell, 2003. https://doi.org/10.1002/9780470756485.ch12
» https://doi.org/10.1002/9780470756485.ch12 - HARDY, H. In search of Isaiah Berlin: A literary adventure. London: Taurus Parke, Bloomsbury Pub., 2018.
- JOHNSON, C. S. Patterns of Negro segregation. New York: Harper Row, 1943.
- KUCKLICK, Henrika. “History of Anthropology”, in: BACKHOUSE, R. E., and P. FONTAINE (eds.) A historiography of the modern social sciences. Cambridge, New York: Cambridge University Press, 2014.
- LADNER, J. (ed.). The death of white sociology. New York: Vintage, 1973.
- LÉVI-STRAUSS, Claude. Introduction to the work of Marcel Mauss, trans. F. BAKER. London: Routledge: Kegan Paul, 1967[1950].
- LYND, R. S., and H. M. LYND. Middletown: A study in American culture. New York: Harcourt Brace & Co., 1929.
- MACLEAN, V.M. and J.E. WILLIAMS MacLean, Vicky M., and Joyce E. Williams. “Shifting paradigms: sociological presentations of race”. American Behavioral Scientist, v. 51, n. 5, p. 599-624, 2008.
- MANNHEIM, K. “Conjunctive and communicative thinking”, in: D. KETTLER, V. MEJA, and N. STEHR (eds.), trans. J. J. SHAPIRO and S. WEBER NICHOLSON. Conservatism: A Contribution to the sociology of knowledge. London: Routledge and Kegan Paul, [1924]1986.
-
MANNHEIM, K. Ideology and utopia: An introduction to the sociology of knowledge, trans. L. WIRTH and E. SHILS. New York: Harcourt Brace & Company, [1929]1936. https://archive.org/details/ideologyutopiain00mann (accessed 16 November 2020)
» https://archive.org/details/ideologyutopiain00mann - PUTNAM, Robert. Bowling alone: The collapse and revival of American community. New York: Simon and Schuster, 2000.
- RIESMAN, David. Individualism reconsidered, and other essays. Glencoe, IL: Free Press, 1954.
- RIZVI, Sana. ”Racially just epistemologies and methodologies that disrupt whiteness” International Journal of Research and Method in Education, v. 45, p. 225-231, 2022a.
- RIZVI, Sana. “Racially just epistemologies and methodologies that disrupt whiteness II” International Journal of Research and Method in Education, v. 45, p. 323-329, 2022b.
- ROUSSEAU, J.J. The social contract and other political writings, trans. Q. HOARE, ed. C. BERTRAM. London: Penguin, [1762]2012.
- STAPLES, R. Introduction to Black sociology. New York: McGraw-Hill, 1976.
- THERNSTROM, Stephan. “‘Yankee City’ revisited: The perils of historical naïveté”. American Sociological Review, v. 30, n. 2 (Apr), p. 234-242, 1965.
- TURNER, Stephen. Sociological explanation as translation. Rose Monograph Series of the American Sociological Association, New York, Cambridge: Cambridge University Press, 1980.
- TURNER, Stephen. 1984. La Explicación Sociológica Como Traducción, trans. M. USABIAGA; Preface by J.P. MAYER. México: Fondo de Cultura Económica, 1984.
- WAQUANT, Loïc. “Scrutinizing the street: Poverty, morality, and the pitfalls of urban ethnography”. American Journal of Sociology, v. 107, n. 6, p. 1468-1532, 2002.
- WAX, Murray L. “Tenting with Malinowski”. American Sociological Review, v. 37, n. 1, p. 1-13, Feb. 1972.
- WEBER, Max. “‘Objectivity’ in social science and social policy”, in: H. H. BRUUN and S. WHIMSTER (eds.) Max Weber: Collected methodological writings, trans. H. H. BRUUN. London/New York: Routledge, [1904] 2012.
- WEBER, Max. The Protestant ethic and the spirit of capitalism, trans. T. PARSONS. New York: Charles Scribner’s Sons, [1930]1958.
-
ŽIŽEK, S. “The real reason why the Taliban has retaken Afghanistan so quickly, which Western liberal media avoids mentioning”. RT 17 August, 2021. https://www.rt.com/op-ed/532207-zizek-taliban-retake-afghanistan-reason/
» https://www.rt.com/op-ed/532207-zizek-taliban-retake-afghanistan-reason
