Reconstructing Sociology: The Critical Realist Approach. By Douglas V. Porpora. New York: Cambridge University Press, 2015. Pp. 249. $35.99 (paper).The Relational Subject. By Pierpaolo Donati and Margaret S. Archer. New York: Cambridge University Press, 2015. Pp. 356. $37.95 (paper).

Abstract

Previous articleNext article FreeBook ReviewReconstructing Sociology: The Critical Realist Approach. By Douglas V. Porpora. New York: Cambridge University Press, 2015. Pp. 249. $35.99 (paper). The Relational Subject. By Pierpaolo Donati and Margaret S. Archer. New York: Cambridge University Press, 2015. Pp. 356. $37.95 (paper).Neil GrossNeil GrossColby College Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreA theoretical movement is afoot in American sociology. Critical realism, long a presence on the British and European scene, is gaining followers here as well, thanks to the efforts of a small number of scholars who find its ideas persuasive and who have been encouraging others to take a look. Douglas Porpora’s Reconstructing Sociology: The Critical Realist Approach is an attempt to win further converts on these shores, while Pierpaolo Donati and Margaret Archer’s book, The Relational Subject, offers a realist take on the study of social relations. There are scattered insights to be found in both books but also a lot of tilting at windmills.Porpora’s argument for critical realism is that it can counter “seven myths of American sociology” (p. 11) that he sees as pernicious. The first is that “ethnography and historical narrative are only exploratory or descriptive. They are not explanatory” (p. 11). This is a weird claim. Most American sociologists see ethnographic and historical work as crucial for the elucidation of causal mechanisms, which is central to explanation.“Myth 2: the appropriate posture in scientific social research is value neutrality” (p. 14). Presumably he has in mind here committed Weberians? But every sociologist who has read “‘Objectivity’ in Social Science”—which is to say, every sociologist properly trained—is already well aware that “an absolute fact-value distinction is a myth” (p. 15).“Myth 3: there is no truth. Everything is socially constructed” (p. 16). Yes, this is indeed the guiding belief behind editorial decision making at our leading journals.Porpora’s fourth myth is that “the most important scientific questions are empirical” and that there’s no space in the discipline any longer for theory (p. 20). It’s true the theory enterprise has changed over the last few decades. It’s lost some of its status and allure, and there’s skepticism of theory for theory’s sake. But I’m hard pressed to think of a serious empirical researcher who believes she or he can get by without well-formed concepts, which only theory can provide.“Myth 5: sociology can and should dispense with the Cartesian cogito” (p. 23). Here, Porpora argues against such hugely influential figures in contemporary American sociology as Foucault and Derrida. (OK, to be fair, also Bourdieusians and certain network scholars.) But while Cartesian dualism is hardly the rage, it’s not as though the typical sociologist doubts that people think and reason.What about “myth 6: there is no difference between structure and culture or structure and action. All is practice” (p. 24)? Leaving aside the fact that this presumes more theoretical savvy on the part of sociologists than myth 4 recognizes, I don’t get it. To say that structures are constituted in part out of practices as Bourdieu, Sewell, and others do is hardly to maintain there’s “no difference” between the two.“Myth 7: we can dismiss statistical analysis as the distinct methodology of positivism” (p. 27). Apparently, there are some in the critical realist camp who hold to this position. I don’t know enough critical realists to assess the factual accuracy of the claim, but if Porpora is right, it doesn’t speak well of his realist colleagues.Since most of these myths don’t amount to anything, I wasn’t sure why I should keep reading. In the end, though, I was glad I did, because Porpora offers a concise and engaging introduction to critical realism. As he describes it, critical realism is a “metatheory” intended to provide a critique of, and alternative to, covering law approaches to explanation, that is, those that understand explanation to mean accounting for facts by subsuming them under general causal laws of either a deterministic or probabilistic nature.Analytic philosophers have been debating the merits of the covering law theory of explanation for more than half a century now. According to Porpora, what critical realism adds to these debates is the distinction between “closed” and “open” systems. Closed systems are those “in which a single causal mechanism operates in isolation” (p. 43). Maybe you could find a law governing such a system, Porpora writes, but in the social sciences, and in many of the natural sciences too, we are faced with open systems where “an unlimited number of ever-changing causal processes operate simultaneously” (p. 43). There’s no chance of finding a general law in such a context—too much contingency and flux—so a different understanding of explanation is required. Critical realism recommends trying to identify the multiple and complexly interacting causal mechanisms and background conditions that generated some phenomenon of interest, where causal mechanisms are understood (idiosyncratically) as not necessarily observable powers of entities or persons that may be triggered under certain circumstances. Explanation becomes a matter of making sense of contingent conjunction.There’s a lot to commend a mechanisms-centered view of explanation, and sociologists learning the trade might benefit from thinking about the issues involved through Porpora’s lens. But to come back to questions of motivation, sociologists already pay close attention to mechanisms—and not because of the influence of critical realism. Who in sociology these days thinks they’re identifying laws? While many sociologists are after more or less general causal patterns in causal relations—as in, for example, the once again popular argument that the political right tends to mobilize in response to status anxiety—most clearly recognize the contextual limits to such patterns. (To continue with the example, the argument only makes sense in societies where democratic politics is structured according to a left-right dimension and in situations where the amassing of status rather than other scarce resources serves as the chief incentive to political action. And in any particular empirical case of right mobilization a multitude of factors will be at play.) Claims about probabilistic, within context patterns of this sort are not philosophically problematic, especially when accompanied by descriptions of the causal mechanisms generating the observed relationships. They’re not the same as the social laws of positivists of yore, which purposefully ignored time and place and with them the situatedness of social objects and their observation. (For example, “Increasing size generates structural differentiation in organizations along various dimensions at decelerating rates” [Peter M. Blau, “A Formal Theory of Differentiation in Organizations,” American Sociological Review 35: 201–18]; original emphasis). So why the need for critical realism?At one point, Porpora insists that sociologists are on the quest for general laws—even if they deny it—whenever they concern themselves with counterfactuals. His argument is that if you conceive of a causal effect as the average difference in outcome for some set of entities between those exposed to a treatment and control condition, you’re positing general laws and assuming a closed system. It is true that advocates of counterfactual causal reasoning in sociology take randomized experiments as their model. But within context, probabilistic generalizations also entail assumptions about what would have likely happened in the absence of the interventions they describe, given the contingent features of the cases at hand. Bringing these assumptions into focus does not move us into bad general law territory.Like the Porpora book, Donati and Archer’s The Relational Subject is undermotivated. The authors’ foil is what they call “North American relational sociology” (p. 12). They have in mind the work of thinkers like Mustafa Emirbayer and Harrison White, as filtered through the writings of Canadian sociologists Christopher Powell and François Dépelteau—energetic champions of an Emirbayer/Dewey-inspired relationalism. There is much about North American relational sociology that Donati and Archer find objectionable—so objectionable that they characterize one of Powell and Dépelteau’s volumes as a “theoretical jihad” filled with “frenzied rhetoric” (p. 23). What seems to most perturb Donati and Archer is Dépelteau’s call for a “single-level ontology in which relations are simply the transactions between interdependent individuals” (p. 23). Donati and Archer see two problems here. First, they think such an ontology prohibits meaningful analysis of the macro. Ignoring the question of whether Dépelteau gets Emirbayer right, the notion that transactionalism leaves no room for the macro would come as a surprise to the coauthor of The Racial Order (Mustafa Emirbayer and Matthew Desmond, The Racial Order [University of Chicago Press, 2015]) where the macro takes center stage.The second feature of relational sociology that disturbs—and on this point, Donati and Archer agree with Porpora—is its antihumanism: its description of human beings as emergent from social relations rather than the other way around, along with what Donati and Archer see as the correlative tendency to downplay the human capacity for reflexive moral action. Again, I don’t think they have much of a point. Yes, network-oriented sociologists sometimes get carried away in their theoretical proclamations, characterizing humans as merely nodes in a network. But the best relational sociology does no such thing. It attends to agency, morality, affect, and the phenomenology of human experience. To anyone who has read the haunting ethnographic work of Matthew Desmond, Emirbayer’s student, the charge that relational sociology is antihumanistic will ring hollow.As an alternative to North American relational sociology, Donati and Archer champion their own version of relationalism. Elements of this are worth a look, though it’s not apparent what these add up to.One intriguing chapter, “The Plural Subject versus the Relational Subject,” offers a thoughtful critique of work by analytic philosophers of social science. Michael Bratman, John Searle, Margaret Gilbert, and Raimo Tuomela each give well-developed accounts of what it means when humans think and act like a “we.” But as Donati and Archer show, these accounts end up reproducing individualist and holist positions taken by social theorists of decades past, often with no attention to relevant sociological work done in the interim. Rather than focus on “wes” who are plural subjects—whether because of some kind of shared intentionality or joint commitment—Donati and Archer urge attention to relational subjects. Against the (alleged) antihumanism of transactionalism, they begin with “a subject [who] is, first and foremost, an agent or actor apprehended in his or her singularity as a human person” (p. 53). Then, a relational subject is a subject who at least some of the time is oriented toward what Donati and Archer see as the “distinct stratum” of reality comprising the social—a stratum that isn’t about aggregation or “overshadowing” structures (p. 57) but relations between people. Individuals partake of relational subjectivity when they turn themselves toward these relations, thereby becoming invested with “emergent properties and powers” (p. 58). As Donati and Archer put it, “To speak sociologically about ‘collective subjects’ is [therefore] to refer to groups of individuals who act ‘collectively’ in the sense of constituting a collectivity that evaluates objectives (discernment), deliberates about realizing its common concerns (deliberation), and commits itself to achieving them (dedication)” (pp. 61–62). This approach would probably not satisfy philosophers like Searle or Gilbert, but sociologists will appreciate that here real social processes, like deliberation, rather than toy model concepts from philosophy, become the foundation for theorizing collectivities.But what does this mean for explaining stuff in society—you know, the thing sociologists are supposed to do? Beats me. The book goes on and on with endless tables and charts and typologies, covering everything from “relational phases of the self” to connections between the “cultural system” and the “sociocultural system,” with about as much discussion of “morphogenesis” and “morphostasis” as you’d expect from Archer. The occasional attempts at empirical application fall flat. When I got to Donati’s chapter on the 2008 financial crisis—a chapter where he refuses to engage the impressive scholarship produced by economic sociologists, economists, anthropologists of finance, and others, preferring to give a theoretical account that loosely weaves together ideas of relational subjectivity with the systems theory of Niklas Luhmann—I gave up.The world is in flames. We need good, clear, accurate, and powerful explanations for what’s happening so that we can figure out how to smartly move forward. Maybe a sociologist will read some critical realism and get inspired to produce a brilliant explanation she or he wouldn’t have otherwise. I hope so. But neither of these two books makes a convincing case that critical realism is the royal road to sociological truth. Previous articleNext article DetailsFiguresReferencesCited by American Journal of Sociology Volume 123, Number 1July 2017 Article DOIhttps://doi.org/10.1086/692468 For permission to reuse a book review printed in the American Journal of Sociology, please contact [email protected]PDF download Crossref reports no articles citing this article.