Saturday, October 2, 2010

In the Tracks of Historical Materialism, by Perry Anderson

What exactly is Jacques Lacan [bottom left] doing in the photo on this book's cover? My copy from the library doesn't have the dust-cover, so all I can see are the product pictures from Amazon and the like, but wtf, Verso? It looks rather like, to use a lovely euphemistic phrase from Elizabeth Gaskell's Cranford, Monsieur Lacan is partaking in "a ceremony frequently gone through by little babies," but I imagine Freud, if not so much Lacan, might have a great deal to say about my interest in making that association.

At any rate, this is a great little book: one of the better transformations of a series of lectures into a book that I have read (although I hope to be blogging about another fine example in a few days). Delivered in 1982 as the second installment of the yearly Wellek Library Lectures, Anderson(not this Perry Anderson)'s subject is a sort of continuation (though "not exactly a sequel") of his 1976 Considerations on Western Marxism: "As I had already attempted a sketch in the mid seventies of the evolution of Marxism in Western Europe since the First World War, offering some predictions as to its likely future directions, it seemed opportune to review intellectual developments since then and to look at how my earlier conjectures had fared" (7). However, he notes, particularly within the period in question, "a survey of recent developments within Marxism was not practicable without some consideration of concurrent philosophical developments outside it, as they affected, or appeared to affect, its fortunes" (ibid.) and for that reason he devotes the second lecture to structuralism and poststructuralism, focusing on Lacan, Derrida, Saussure, Foucault, and Lévi-Strauss, and considers Habermas in the third lecture.

The second lecture is, I think the most interesting. Anderson begins by noting a "Latin [i.e., French, Spanish, and Italian] recession within the international map of contemporary Marxism" (32): where France and Italy were "the two leading homelands of a living historical materialism in the fifties and sixties" (30), they had become, by 1982, sites of "a precipitous descent" and a "massacre of the ancestors" (ibid.), with the new generation not only rejecting but anathematizing their elders and a consequent "demoralization and retreat" of any still-living Marxists. Anderson lays out a surprising hypothesis for this turn of events: "after French Marxism had enjoyed a lengthy period of largely uncontested cultural dominance, basking in the remote, reflected prestige of the Liberation, it finally encountered an intellectual adversary that was capable of doing battle with it, and prevailing. Its victorious opponent was the broad theoretical front of structuralism, and then its post-structuralist successors" (33).

Anderson next goes on to note that, rather unusually, "the passage from Marxist to structuralist and then post-structuralist dominants [sic—probably a transcription error for "dominance"] in post-war French culture has not involved a complete discontinuity of issues or questions. On the contrary, it is clear that there has been one master-problem around which all contenders have revolved; and it would look as if it was precisely the superiority of—in the first instance—structuralism on the very terrain of Marxism itself that assured it of decisive victory over the latter. What was this problem? Essentially, the nature of the relationships between structure and subject in human history and society" (ibid.).

Following some very lucid intellectual history regarding the debates between Sartre and Lévi-Strauss and the entrance of Althusser into the fray as well as a consideration of the impact of the events of May 68, Anderson lays out a "demarcation of a basic space in which structuralist and post-structuralist theories can be unified, as a series of possible moves or logical operations within a common field" (40). Anderson allows that this demarcation does not emphasize the internal differences of the thinkers commonly classed together by these terms, and that none of them makes all the moves he will describe. "Yet all their major themes and claims fall within the boundaries of this shared purlieu" (ibid.).

First is what Anderson calls "the exorbitation of language." It's a shame that term hasn't caught on, because what it names is brilliant: identifying the decisive origin of structuralism in Saussure's linguistics and more specifically in the langue-parole distinction (not in itself an original move on Anderson's part), Anderson then analyzes how Lévi-Strauss's "intrepid generalization of [this distinction] to his own anthropological domain" and specifically to kinship systems: "Once this equation was made, it was a short step to extend it to all the major structures of society, as Lévi-Strauss saw them: the economy itself was now added, under the rubric of an exchange of goods forming a symbolic system comparable to the exchange of women in kin networks and the exchange of words in language" (41). Then, Lacan joined the party, announcing that the unconscious was also "structured like a language."

Again, none of this is particularly original, and neither exactly is the following, although the clarity of its extension all the way to Derrida is tremendously helpful:
After such fundamental expansions of the jurisdiction of language, there inevitably followed a host of lesser adventures and annexations: clothes, cars, cooking, and other items of fashion or consumption were subjected to diligent semiological scrutiny, derived from structural linguistics. The final step along this path was to be taken by Derrida, who—marking the post-structuralist break—rejected the notion of language as a stable system of objectification, but radicalized its pretensions as a universal suzerain of the modern world, with the truly imperial decree, 'there is nothing outside of the text', 'nothing before the text, no pretext that is not already a text' (42).
"Fundamental expansions of the jurisdiction of language" is, I think, tremendously apt, an excellent definition or re-articulation of the "exorbitation of language" which isolates precisely how and where to put our finger on what, exactly, was the import of structuralism and post-structuralism and how we might, if we wish, shrug off its grip—simply rein in its "exorbitation," acknowledge that there are domains of human existence which are not best analogized to or analyzed by structural linguistics.

Anderson also gives us an excellent reason why we might be justified in doing so:
It was Saussure himself, ironically, who warned against exactly the abusive analogies and extrapolations from his own domain that have been so unstoppable in past decades. Language, he wrote, is 'a human institution of such a kind that all the other human institutions, with the exception of writing, can only deceive us as to its real essence if we trust in their analogy'. [Wow!] Indeed, he singled out kinship and economy—precisely the two systems with whose assimilation to language Lévi-Strauss inaugurated structuralism as a general theory—as incommensurable with it… Saussure's whole effort, ignored by his borrowers, was to emphasize the singularity of language, everything that separated it from other social practices or forms… In fact, the analogies that were to be promptly discovered by Lévi-Strauss or Lacan, in their extension of linguistic categories to anthropology or psychoanalysis, give way on the smallest critical inspection. Kinship cannot be compared to language as a system of symbolic communication in which women and words are respectively 'exchanged', as Lévi-Strauss would have it, since no speaker alienates vocabulary to any interlocutor, but can freely reutilize every word 'given' as many times as is wished thereafter, whereas marriages—unlike consversations—are usually binding: wives are not recuperable by their fathers after their weddings. Still less does the terminology of 'exchange' warrant an elision [sic?] to the economy… No economy… can be primarily defined in terms of exchange at all: production and property are always prior… Far from the unconscious being structured like a language, or coinciding with it, Freud's construction of it as the object of psychoanalytic enquiry precisely defines it as incapable of the generative grammar which, for a post-Saussurian linguistics, comprises the deep structures of language: that is, the competence to form sentences and carry out correctly the rules of their transformations. The Freudian unconscious, innocent even of negation, is a stranger to all syntax (43).

Anderson drives the point home further: the langue-parole relation is a "peculiarly aberrant compass for plotting the diverse positions of structure and subject in the world outside language" (44) for three reasons:
  1. because the rates of change are so different as to be fatally incommensurate—language alters itself far more slowly than do the economic, political, or religious structures which are supposedly so assimilable to linguistic models;
  2. language as a structure is fairly rigid relative to the "inventivity" of the subject—that is, "utterance has no material constraint whatever: words are free, in the double sense of the term. They cost nothing to produce, and can be multiplied and manipulated at will, within the laws of meaning. All other major social practices are subject to the laws of natural scarcity: persons, goods or powers cannot be generated ad libitum and ad infinitum. Yet the very freedom of the speaking subject is curiously inconsequential: that is, its effects on the structure in return are in normal circumstances virtually nil. Even the greatest writers, whose genius has influenced whole cultures, have typically altered the language relatively little" (44).
  3. Speech is "axiomatically" understood to be produced, when intelligible, by a single subject: some individual speaks for a collective (or individuals speak sequentially on behalf of the same collective); when collectives actually speak collectively, it is generally considered a din or a cacophony or something similar which indicates, as Anderson, that "plural speech is non-speech" (ibid.). This is in great contrast to "economic, cultural, political or military structures which are first and foremost collective: nations, classes, castes, groups, generations. Precisely because this is so, the agency of these subjects is capable of effecting profound transformations of these structures" (44-45).
Anderson concludes, "This fundamental distinction is an insurmountable barrier to any transposition of linguistic models to historical processes of a wider sort. The opening move of structuralism, in other words, is a speculative aggrandizement of language that lacks any comparative credentials" (45).

Anderson traces this initial illegitimacy to two consequences: what he calls "the attenuation of truth" (ibid.) and "the randomization of history" (48). These deserve their own expositions (which will have to be shorter and held for a subsequent post), but the "exorbitation of language" is, I think, the crucial leg of Anderson's answer to the hypothesis he poses. The form of that answer, if I may anticipate my next post, is a little questionable, though. The hypothesis that Anderson presents—that structuralism/post-structuralism drove historical materialism out of business because it had more satisfactory resolutions to the structure-vs.-subject problems—is disproved by a demonstration of its insufficiency (and even superficiality) in dealing with that problem at all, much less resolving it. Yet Anderson takes this as a dead end for what he calls "intrinsic" answers "from within the logic of the ideas of the time" (56) to the question of why historical materialism declined in France so rapidly and so thoroughly and so, he argues, we must turn to "extrinsic,"i.e., geopolitical answers—roughly, "the fate of the international communist movement" (68). One of the extrinsic factors that Anderson barely considers, however, is the surprising eagerness of American academics to take up French structuralism/post-structuralism, a factor which arguably has less to do with the fate of the international communist movement and more to do with issues and conflicts internal to U.S. academia; it is notable (and come to think of it, surprising) that one of the primary vectors of "French theory" was Fredric Jameson's The Prisonhouse of Language. The impact of American popularity on these thinkers' prestige within France, however, is rather difficult to assess—I started reading François Cusset's French Theory—which purports to explain this—awhile ago and just got bogged down in its small-bore trivialities—viewed as gossip, I'm sure it's fascinating to anyone who was involved or on the outskirts, but to someone whose introduction to Theory actually was Derrida's death, it seems pretty turgid intellectual history.

Returning to Anderson for one last point, however: I am not sure that Anderson's dismissal of "intrinsic" reasons for the triumph of structuralism/post-structuralism is completely warranted. It is not that I disagree with his conclusion that structuralism/post-structuralism failed to engage with (and a fortiori to resolve) the structure-subject problem in a superior or even equivalent manner compared to Marxism's own efforts. It is not that I am not convinced that successful engagement or successful resolution is the only "intrinsic" factor that counts when accounting for the triumph of one idea or one system over another. Could we not see structuralism/post-structuralism's victory over historical materialism as the result of an exhaustion of a writing style as much as or more than of a set of answers, that writing in the structuralist (or even more, in the post-structuralist) fashion became "intrinsically" more pleasurable, not to mention more capable of commanding attention? This is also not a particularly original insight (post-structuralism is all about style, who knew?) but it is an angle that Anderson neglects, and I think his oversight is fairly serious.


LML said...

'there is nothing outside of the text', 'nothing before the text, no pretext that is not already a text'

Derrida didn't, I think, believe that market economies are produced like or function in the same way as texts; rather, our perception of their workings, since mediated by language, is limited by the ad-hoc nature of the signifying process. We think linguistically; language is only capable of provisionally situating external reality; hence, if we make truth claims that do not account for linguistic slippage, we are philosophizing badly. Marx's analyses of markets are essentialist and, therefore--useful and accurate as they may be in all sorts of ways--must be reworked by the conscientious Marxist (enter Althusser). This seems obvious and uncontroversial. Or is Anderson saying something different and I'm the one being dense?

Andrew Seal said...

That's a great point. I think Anderson is saying something slightly different; I imagine he (and I) wouldn't disagree with the need for re-working that you are pointing to.

Perhaps it is merely that he is taking Derrida's "il n'y a pas de hors-texte" more literally than we've begun to take it; where (as I think you're doing, though correct me if I'm wrong), I believe we have more recently tended to think of linguistic slippage as a sort of mediation effect, it seems to me that many (both critics and proponents) around the time Anderson was writing took Derrida's pronouncement as a truth-claim about the actual nature of the market, or colonization, or gender, etc. It wasn't (for them) that we can't think or talk about these things without using language, but that these things have no effects on us or on the world except through language, that the "textuality" of these things isn't a product of our handling them linguistically, but rather a condition of their actuality: gender is a kind of language, the market is a kind of language, etc. I think Foucault's "discursivity" often got wrapped up in this process (part of the conflation of "French Theory") such that it looked like all "discourse" was at least purely linguistic, if not in fact "textual."

I believe that is what Anderson is referring to as the "exorbitation of language"--it's not something like "we must always take into account the instability of our own linguistic practices," but the belief that the instabilities of the market, of gender, of colonialism only occur as the instabilities of language.

Obviously, a Marxist insists that there are objective contradictions in capitalism (among other things) that are not merely the product of linguistic slippage--it's not just the endless deferral of meaning that reproduces the class structure.

Adam Kelly said...

I haven’t read Anderson’s book, but I would say from what you present that there’s a fair bit wrong with his argument, which might help to explain some of the contradictions you alight on. Believe it or not, I was teaching Saussure and Barthes only today, so I feel I should offer my two cents.

Anderson: “It was Saussure himself, ironically, who warned against exactly the abusive analogies and extrapolations from his own domain that have been so unstoppable in past decades. Language, he wrote, is 'a human institution of such a kind that all the other human institutions, with the exception of writing, can only deceive us as to its real essence if we trust in their analogy'.”

This would be fine, except that Saussure is explicit in the Course on General Linguistics that linguistics is only one branch of semiology, the science of signs in society more generally. It was this move that was taken up by Barthes and Levi-Strauss in the extension of the basic structures of linguistics outside of that domain, and it is a move wholly warranted, and indeed called for, by Saussure’s text.

Saussure did not, as Anderson claims, spend a great amount of effort emphasizing the singularity of language – his whole point was that we should see it as a system like other systems, and that singular examples of it - parole – were simply a distraction from the study of the underlying system.

Turning to Anderson’s three arguments that “drive the point home further,” which I won’t quote but will just respond to:

1. Rates of change are irrelevant to the structuralist because the analysis of synchronic and diachronic elements should be distinct; this is, in my view, one of the major differences between the methodological aims of structuralism and of a Marxist dialectics.

2. I don’t understand how this is an argument against structuralism. The subject is not “inventive” in either structuralism or Marxism. Again, the fact that words are “free” has nothing to do with language as a system, which still delimits the subject as economic circumstances do.

3. Again, the focus on speech is a red herring. Language (not parole) is indeed collective, something Anderson fails to acknowledge. And agency – changing the language – can by definition only happen collectively, in a similar way to the structures Marx is interested in.

More after the jump...

Adam Kelly said...

The nub of the issue is, I think, that if you want to relate structuralism, post-structuralism and Marxism to each other historically, then lining up the first two (as nearly indistinguishable) against Marxism is the wrong way to do it. The key figure here is Barthes. If you read Mythologies, you see very clearly that Barthes saw in structuralism the chance to offer a more scientific basis for Marxist ideas regarding base, superstructure, etc. After the introduction, most of Mythologies is a straightforwardly Marxist analysis of culture – in other words, penetrating the cultural surface to the underlying economic logic.

What Derrida did, and why he was able to convert Barthes to post-structuralism, was to undermine this surface/depth distinction, questioning the whole notion that we could attribute surface factors to a universal and isolatable structure operating underneath. This assumption that the simple always underlay the complex was, for Derrida, inaccurate phenomenologically and metaphysically, and always depended on the re-introduction of a transcendental signified (in the case of Barthes, and Jameson after him, the notion of History) that could never be located in the text being dealt with.

So the point is – and Anderson is right on this at least – that structuralism and Marxism shared the same ground, because both were attempting to understand culture scientifically, i.e. trying to find laws that could explain cultural phenomena. Post-structuralism is a break with both, because it deconstructs the relation between singularity and law, between particularity and generality, assumed by those systems of thought.

But before I get onto the spectre of to Specters of Marx, I’d better quit! Comments welcome.

Andrew Seal said...

This is extremely helpful; I am not so familiar with Saussure's work as I am with the uses to which he has been put, so it's really great to get a better picture of what he said that Lévi-Strauss and Lacan were working from. I checked the Saussure quotes which Anderson uses to make his case; they are not from the Course, but rather from his (published) notebooks--again, I'm not familiar enough with Saussure to know the significance of this origin, but it might be a useful thing to throw in at this point.

I'm not quite willing to give up Anderson's argument, though. While it may in fact have little to do with what Saussure would have thought about Lévi-Strauss's and Lacan's extensions of his theories to anthropology and psychoanalysis, I feel that Anderson is right in identifying an "exorbitation of language" as something shared by structuralism and post-structuralism, even if he does not sufficiently account for the divergences between the two. Furthermore, I think that this exorbitation is the crux of an argument that can be constructed between the structuralists and post-structuralists on one side, and the historical materialists on another. There is a great quote I'll be using for the second post on this book, but the gist of it is that, for Anderson, structuralists and poststructuralists alike participated in a shift from thinking about language as a medium to language as a foundation--or, in the case of the poststructuralists, as that which made the idea of foundations impossible, which for a Marxist doesn't really all that different from calling it a foundation.

This is probably little more than a re-statement of what it is you're refuting, and I probably share too many of Anderson's assumptions to critique him properly, but I hope I'm at least making his ideas sound more like something which can be worked with.

And please hit me with your thoughts on Specters of Marx!

Adam Kelly said...

Well, I'm teaching "Differance" next week, so I think I'll hold back on old Jacques until I've re-absorbed myself thoroughly!

More anon.

Bernard Morgan said...

Can anyone provide a detailed review about this book? Any kind of review is very much appreciated.