Stirrings in the attic

Introduction

Reading Kevin Cox’s (2013) paper reminds me of returning to a dusty chest in an attic. The chest has not been opened for quite some time, but longing to explore its contents and reconcile some philosophical memories, it is dusted off, and its ontological and epistemological heirlooms reassessed. The putative antiques in question are those of critical realist and historical materialist geographies, and while turning out to be extremely valuable for social research in the 21st century, the significance of their differences seem to be somewhat forgotten in human geography, at least explicitly it seems. It is therefore with great pleasure that I accompany Professor Cox (2013) into the attic as he sets out to recover an issue that sparked considerable debate in the 1980s, namely the relationship between historical geographical materialism (HGM) and critical realist geographies. He is to be congratulated for recovering this debate and reminding us in systematic fashion of some crucial ontological and epistemological distinctions.

Cox (2013) divides his paper into three parts; the first (and the bulk of the paper) is devoted to demonstrating how HGM and critical realism (CR) differ on the grounds of five issues. In the following response, I discuss the various components of this adjudication, and provide a very brief commentary on each of them. In light of these distinctions, the second part of Cox’s (2013) paper seeks to discuss the work of both Doreen Massey and Andrew Sayer and then contrast their approach with that of David Harvey’s. The third part calls for “a limited—a very limited –rapprochement” between CR and HGM. Ultimately, I support this reconciliation, but I conclude by relating Cox’s (2013) discussion of CR and HGM to some more recent literature in geography, especially economic geography and political economy (the areas of geography with which I am most familiar). While I suggest that Cox’s (2013) discussion would be enhanced through engagement with more literature over the last decade (especially the concerns of post-structuralism in geography) rather than conducting what appears to be an isolated debate, I maintain that his analysis reminds us of significant distinctions between CR and HGM. In doing so, it forces us to reflect on the ways in which we approach our own work. In short, there is considerable value in reopening the attic door.

Cox (2013) begins his paper with what he views as the unresolved debate between HGM and CR, lamenting that geographers have “moved on” and “exhort others to do likewise.” Yet he remarks further “just how one can ‘move on’ without their critical resolution challenges credulity.” He then turns to a number of criticisms of CR, including to the notions of necessity and contingency, on positing an opposition between general processes and “the locally specific” when undertaking case study research; the failure of CR to say much about social struggle, and as both Harvey (1987) and Smith (1987) pointed out in the 1980s, the dangers of a shift in emphasis on the particular rather than the universal. Nonetheless, Cox (2013) argues that there are ways in which certain assumptions in CR can facilitate empirical research, and it raises the question how empirical research should be undertaken in the vein of HGM. He claims that Marxist geographers who undertake empirical research show little attention to issues of abstraction or explanation, and he concludes that if CR is to be mobilized in any way, it should be “entirely subordinated to HGM as a method.” However, Cox (2013) does not want to completely sell CR’s family silver, for “Critical realism should not be entirely rejected.”

Considering the debate

With the above in mind, let me turn to the five elements that differentiate CR from HGM. The first is abstraction. Cox (2013) reminds us of a crucial distinction between the method of abstraction in CR and HGM. In the former, critical realists are said to move from the concrete (the “many-sided”) to the more “one-sided.” However, Sayer’s (1984[1992], 138) insistence on “rational abstractions” (abstractions that isolate “a significant element of the world which has some unity and autonomous force such as a structure”) rather than the “chaotic conceptions” of positivism and its formal relations, turn out, for Cox (2013), to involve as much “abstractions from the real” as positivism. Though it appears that Marx pursues the same approach, Cox (2013) maintains that Marx’s analysis involves similar abstractions only insofar as they could offer a deep critique of capitalism as neither ahistorical nor natural. In contrast then, the kind of abstraction Marx uses are historical or systematic. As Roberts (2001) has argued, one of the key differences between Marx’s abstractions and CR’s is that the latter involves a certain transcendentalism divorced from specific social formations. One might add to Cox’s (2013) analysis that Marx’s abstractions include what Ollman (1993) refers to as extension, levels of generality, and vantage point. For example, in Marx’s works, “extension,” especially temporal extension, figured centrally in his critique of classical political economists’ narrow (temporal) conception of value.

Cox (2013) then turns toward the issue of internal and external relations. While Marx’s analysis entails a conception of internal and external relations, they are quite different from CR’s. For Marx, internal relations involve a relation between a subject and object, or the subject’s “objective conditions of existence.” As Cox (2013) notes, this can be thought of as a subject–object relationship centered on the property relation and bound up with nature and social relations. Marx’s approach is historical—what is an internal relation between people and their objective conditions of existence comes to appear as external only under particular historical circumstances, for example within capitalism. In contrast, in CR, internal–external relations can apply independently of the subject–object relationship. The empiricist abstractions of CR are ahistoric and separated from the property relationship. Cox (2013) is quite right to point this out and it is supported by other analyses of Marxism and CR (see e.g. Brown et al., 2001).

The third issue is the question of totalization or pluralization. For Marx, the world is a unity in which the nature of the parts is bound up with the whole or totality. There is a constant tension between part and whole, in which the whole changes in relation to the parts. While CR has a similar relational view of the world, it makes an important distinction between internal and external relations or the highly vilified “necessary” and “contingent” relations. In this respect, CR and HGM are sharply opposed, and as Cox (2013) insists, Marx’s conception of totality or totalization does not imply a deterministic view of the world, since inherent to totality is change and contradiction, especially contradictory production relations.

Again, I think few would disagree with Cox’s (2013) adjudication between HGM and CR. At the same time, Roberts (2001) has stressed how Bhaskar’s The Pulse of Freedom is an attempt to “dialectize” CR and argue that critical realists have always relied on a conception of totality and one that has contradiction as its essence. The more crucial point I think is that, just as there are many Marxisms, there is no singular CR, as theorists revise its initial statements associated with Bhaskar’s (1975) original formulation. Second, and perhaps quite evidently, “totality” is presumably rejected by most self-described post-structuralists, let alone “totalizing” discourses.

The fourth distinction is contradiction and change. Here, Cox (2013) points out that contradiction (essential to Marx’s philosophy of internal relations and dialectical method) represents a fundamental distinction between HGM and CR. Regardless of some ambiguity about whether CR incorporates a notion of totality or contradiction, and despite Cox’s (2013) clear preference (and this author’s) for situating contradiction at the heart of social analysis, post-structuralists, for example, are unlikely to adopt such a view, since presumably, for most post-structuralists, each of these opposites (in HGM) or dualisms (in CR) are subject to deconstruction, and therefore have no “essence” in themselves (see the discussion for example in Elden, 2008).

For Cox (2013), causation and determination represent the fifth difference between CR and HGM. Both CR and HGM subscribe to similar elements of causality (e.g. anti-atomism and a sort of structuration); yet if examined more closely, significant differences can be ascertained. In CR, the critical realist notion of causal structure involves a separation between an underlying reality and the empirical world. While HGM seems to rest on a similar dichotomy, Marxism relies in fact on a distinction between essence and appearance, and appearances are, as we know from Marx’s works, actually ideological forms rooted in the social relations of capitalism. Again, I would only add that in Marxism, references to determination may raise eyebrows given his dialectical method, but Marx uses determination to underline asymmetric relations within capitalism (Ollman, 1993).

The final section of Cox’s (2013) paper is devoted to exploring the relationship between, on one hand, the work of Doreen Massey and Andrew Sayer and, on the other hand, Harvey. Cox (2013) favors the work of Harvey and HGM over CR for three reasons: (1) the geographies envisioned are more dynamic and open-ended, which involve a constant “becoming;” (2) “geography” or “space” are “necessary” rather than “contingent” to the “social process,” and (3) that the geographies are necessarily capitalist ones. Having sketched these preferences, Cox (2013) concedes that a very limited rapprochement between CR and HGM is possible, provided that human geographers embrace CR only insofar it espouses a conception of “structures” as central to its approach. Yet, these structures must be understood within an immanent critique of capitalism; that is neither ahistorical nor natural. I share with Cox (2013) the need to situate structures within the logic of capitalism, but this is unlikely to convince those who, again, have difficulty accepting the existence of structures, if not the totality of capitalism.

The debate in light of (economic) geography in the 21st century

I enjoyed Professor Cox’s (2013) journey into the attic. It reopened my own memories of vigorous debates that raged while in graduate school in the early 1990s, the very moment when CR seemed to relegate HGM to the shuttered closets of philosophical and theoretical discourse. Kevin Cox’s (2013) analysis has shed considerable light on adjudicating between my tattered copies of Marx’s Grundrisse, the three volumes of Capital and Sayer’s Method in Social Science: A Realist Approach. It is important to recognize, however, that he is far from the first to interrogate this, even long after the 1980s mêlée. For example, Brown et al. (2001) edited an entire volume on this matter, especially changes in the nature of CR, and Peter Nielsen and others devoted considerable efforts to the relationship between Marxism and CR in the Cambridge Journal of Economics in 2002. In any case, for me, the question is not whether his exposition of their differences is “new” or “correct,” nor whether his calls for a limited rapprochement between CR and HGM has merit (for what it is worth, I agree for the most part with his careful and judicious assessment), but rather the stripped down, skeletal version of his analysis and his insistence that we must (re)turn perhaps unproblematically to HGM, shorn of any reflection on totality, dialectics, objectivity–subjectivity, internal–external relations, or causation and determination (again Elden, 2008, offers a brief tour of some work on these questions by geographers during the 2000s).

And I am probably a more sympathetic interlocutor, wary of dismissing a capitalist totality, the unity of opposites, or pursuing abstract philosophy for itself, “merely” in the service of explanation. Yet at the risk of reiterating the readily apparent, it seems that most critical (economic) geographers have indeed “moved on,” certainly from the debate between CR and HGM. This for me is not best described as “challenging credulity” but reflects periodic academic fashions and cycles, long-standing ideological positions or discursive strategies, and just plain aversion to HGM or CR, for a range of reasons too numerous to mention here. Consider above all, Gibson-Graham’s (1996) adamant post-Marxist reading of totality. Sure, some economic geographers over the last few years have re-engaged explicitly with CR, attempting some sort of combinatory geography between it (especially, CR’s notion of retroduction), an emphasis on practices, actor–network theory, and even power (Jones and Murphy, 2011). Yet, “practice-based” economic geographies, as Jones and Murphy recognize, may border on empiricism, and I would add that they are potentially at odds with Marxist abstractions, and the method of abstraction. In other words, practice-oriented (economic) geographies often rely on ethnographies that eventually rely on abstractions through a process of induction and grounded theory. Such studies may therefore be non- or anti-Marxist, non or anti-Hegelian, and even non-Realist in their possibly uncritical acceptance of “appearances.” While Jones and Murphy attempt to enlighten “practices” with an explicit engagement with CR and the last 20 years of actor–network theory in economic anthropology, geography and sociology, then others have simply absorbed CR and let its features remain in the shadows as an implicit under-laborer for empirical analysis. I have in mind here how an explicitly Marxist analysis of “capitalism” as a totality seems to have morphed into more critical realist analyses of neoliberalism/neoliberalization (see e.g. Brenner et al., 2010). Despite the recognition of structures such as “rule regimes,” studies of neoliberalization appear to be inspired more by the Foucauldian idea of governmentality and an emphasis on state and non-state practices, than with attention to, for example, “the international law of value” and the capitalist/class bases of states. What happened, for example, to the Marxist critique of “development,” concepts of “surplus value extraction” (one of Marx’s key discoveries borne of a philosophy of internal relations, a dialectical movement from the abstract to the concrete back to the abstract) and by extension, the “rising organic composition of capital”? They seem to have largely disappeared into the dark recesses of at least economic geography’s attic. While this occlusion of capitalism by neoliberalization has unfolded, other economic geographers have embraced “relationality”—a preoccupation that can be more productive than clinging to Euclidean or strictly territorial geometries, but which nonetheless trudges down what I would call a vague, to not say vacuous path (for similar reactions, see Jones, 2009; Sunley, 2008) that inter alia, seems to suffer from amnesia, neglecting the whole history of dialectics.

On the subject of dialectics, the engagement with Marxism or HGM over the last decade is a different story. To begin with, Marxism and especially the question of “value” seem to be all the rage in political ecology (e.g. Mann, 2009), not to mention elsewhere (for some discussion in geography, see Barnes, 2010; Elden, 2008, and the special issues of Antipode (on Marx, biopolitics, reproduction and “race,” 43(5), and the special issue on dialectics in Environment and Planning A, Volume 40, 2008)). I wish Professor Cox (2013) had juxtaposed the earlier history of HGM—the valuables in the attic if you will—with some newer Marxist items on the floors below, but bravo nevertheless to Kevin Cox (2013) for his analytical effort to resurrect a debate over some of critical geography’s philosophical, ontological, and epistemological heirlooms.