If I am reading David McCourt’s arguments correctly, he lauds constructivism’s early promise, laments its wrong conventional turn, and expects a turn toward practices, processes, and relations to revive its original potentials. But constructivism never promised as much as McCourt would like. It has not made a wrong turn, so much as developed in many different directions, including toward pracitce theory. I am also not so convinced that practice theory and relational approaches will lead to the promised land either.
McCourt writes that the "true value" of constructivism is that it keeps "IR scholarship sensitive to the social and cultural contexts in which international politics takes place." He indicates that the same value is offered by practice theory and relationalism. But constructivism, especially in the U.S., has "narrowed" over time, leaving us with three inadequacies: a structuralist bias, a focus on language over practice, and a "preference for causal over constitutive claims." Moreover, practice theory and relationalism can get constructivism beyond these sins.
First, McCourt offers no evidence that constructivism has "become less exciting with time," or that "scholars have moved away from constructivism over time," or that it needs "reinvigoration." In fact, as the international TRIPS survey has recently shown, constructivism has become and remains part of the IR research and teaching Holy Trinity, along with realisms and liberalisms. In fact, I would argue that instead of narrowing, constructivism has spawned a lively and growing interest in not just narrowly constructivist IR, but social theory and IR in general, even in the United States. Indeed both practice and relational theories have been advanced by, and found an audience among, U.S. scholars. Moreover, identity, norms, and culture strike me as pretty capacious categories for research, allowing for coverage of much territory, including McCourt's preferred approaches.
Indeed, as far back as 2002, Iver Neumann reminded constructivist IR scholars that Foucault's conceptualization of discursive formations both "sayings and doings," steering us away from an over-reliance on exclusively textual evidence. To be fair, it should be pointed out that Foucault paid a great deal of attention to words, language, and texts. Culture is not so constraining a concept if one remembers it includes the cultural practices to which the practice turn points.
Second, I am not pretty certain that we cannot ever resolve key questions around structure and agency, ideas and materiality, and causality and constitution. Constructivists rightly raised these as problems that demanded sustained and agonistic attention. We should prefer this permanent state of theoretical anxiety to what will always end up as strained and unconvincing resolutions. Field theory, network analysis and Actor-Network Theory do not do an obviously better job. We should not expect them to be antidotes to fundamental socioal-theoretic problems, but rather as additional tools with which to think through these issues.
Third, I was surprised by McCourt's observation that there is a preference for causal over constitutive stories in constructivist IR. As one of the most consistent supporters for the use of neo-positivist methods combined with an interpretivist epistemology among constructivist scholars, I can report that my position is a rather lonely one. In an H-Diplo roundtable review of my book, Reconstructing the Cold War, Patrick Thaddeus Jackson lamented that he could not assign the book to his students because of my methodological approach. I am afraid I am only going to further disappoint Jackson, as I am now embarked on creating a large-N intersubjective national identity database for all great powers from 1810 to the present, so that quantitative scholars can use national identity as a variable in their models.
Which scholars have identified subjects only as subjects after they are in a social relationship with another subject? To be sure, most of us argue that meaningful subjectivities do not emerge until such interaction, but it is not as if many of us wait for that subjectivity to appear before we specify units of analysis or actors or agents of interest, whether they be states, movements, or individuals. In making the case for relationalism, McCourt cites Norbert Elias to the effect that "individuals live first in interdependence with others." This seems to me as no different than basic constructivist insights "old school" constructivists dervied from Berger and Luckmann (1966)?
Fourth, I do not see how practice theory and relationalism are going to resolve this issue. McCourt tells us that "processes and relations [are] analytical primitives" for relationalism. But aren't relations always between two entities? It is one thing to claim that processes matter; it is another to dispense with entities altogether. Of course we should not exclusively concentrate on agents or structures, practical or discursive knowledge, reflex or reflection, representational or non-representational knowledge, knowing how or knowing that, etc.
I find it ironic that McCourt advances field theory as a way past constructivism's structuralist bias, since one of the fundamental theorists of field theory, Pierre Bourdieu, as cited by McCourt, has been roundly criticized for being a structuralist.1 Bruno Latour's Actor Network Theory (ANT) is singled out for restoring material agency, if we can call it that. Is that an unalloyed improvement? One of ANT's more notorious actants are the bridges over the Long Island Expressway from New York City to Long Island. They are so low as to prevent buses from passing under them. Clearly they have an effect upon who can easily travel to Long Island's beaches, favoring those with cars. But are we to attribute agency to these low spans of concrete for keeping poor people (of color) from easily traveling to these beaches, or should we adduce agency to Robert Moses, who designed them that way deliberately? Again, we should not exclude either consideration, but surely concentrating on material agency has more pernicious ethical consequences than concentrating on Moses.
The bottom line is that constructivism was never a panacea for many of social theory's eternal dilemmas, though it offered significant advantages over the liberal and realist alternatives. And it is still broad enough to easily accommodate practices, processes and relations, although these latter are only going to add to the conversation, not settle the argument.
1 For example, Paul DiMaggio, "On Pierre Bourdieu,” American Journal of Sociology 84:6, 1979, pp. 1460-74; Rogers Brubaker, "Rethinking Classical Theory," Theory and Society 14:6 1985, pp. 745-75; William H. Sewell, Jr., "A Theory of Structure: Duality, Agency, and Transformation," American Journal of Sociology 98:1 1992, pp. 20-1; Barbara Farnell, "Getting out of the Habitus," The Journal of the Royal Anthropological Institute 2000, 6:403, 425-6; Terry Eagleton, "Doxa and Common Life," New Left Review 1992, 191:111-21; Axel Honneth, "The Struggle for Symbolic Order," Theory, Culture, and Society 1986, 3:35-51; and Morag Schiach, "Cultural Studies and the Work of Pierre Bourdieu," French Cultural Studies 1993, 4:213-23.