Our Tautological America
I just posted at the Valve what I’ve pasted below, a post on the tautological rhetorical construction of Walter Benn Michaels’ Our America, and a proposed defense of this as a project. It’s actually a revision of a post I wrote here and have been thinking about for some time, but I was partially spurred to finish because Andrew Seal posted an interesting reading of the more recent, less sympathetic Walter Benn Michaels (and his interest in The Wire) which y’all might want to check out. I have a real respect for Michaels’ work (that I’ve read), and the fact that I disagree with as much of it as I do only illustrates that: he does raise the problem of how “culture” often obscures issues of class, and although my perspective is almost diametrically opposed to his (being interested in how often “class” becomes an excuse to ignore social formations that do not reduce to economics, like race or culture) it seems to me like such a opposition is only possible because of a shared investment in social justice as an ideal for literary criticism and a shared set of critical terms that are being used to that end. I think most people who viscerally disagree with him miss that about his work. But in re-reading that post, what really struck me the most is how I didn’t mention what I find to be the primary virtue of that book: that the readings are really, really good. So often, in lit crit circles, the real proof of a theoretical pudding is how it’s readings taste in practice. And as yucky a mixed metaphor as that might have been, that’s what keeps Michaels’ book worth reading: his readings taste great!
What I like most about Walter Benn Michaels’ Our America: Nativism, Modernism, and Pluralism is the underlying claim that “nativism…is simultaneously a modern and a modernist phenomenon.” This is one of those arguments that has to get made again and again, so powerful is the “March of Progress” narrative: even when we acknowledge that racism (or whatever -ism we’ve chosen to focus on) isn’t merely a fragment of a broken past that we can rely on the passage of time to wash away, the very metaphors embedded in our very language tempt us to oppose “retrograde” and “reactionary” prejudices to ideas which are more “progressive” or “advanced.” And the quietism and complacence this allows, too, is tempting; after all, if racism were somehow limited to parts of the country which haven’t yet been blessed with the fruits of modernity, ignorant hicks who haven’t yet been brought into the modern world, then we can simply wait for it to be washed away by the natural progress of time, resting easy while the March of Progress takes care of what would otherwise be an uncomfortable and difficult project. Formulations like “race is still with us” or “post-racial” encourage us to think this way, to think of “Race” as a thing perpetually on the decline.
Michaels’ claim that American nativism and modernity are mutually constitutive is incompatible with this narrative, and he coins the shotgun-marriage term “nativist modernism” to express his sense of the structural intimacy between. But there’s a begged question at the root of this nags at me: how to locate this “nativist modernism” in actual, you know, history.
For example, Michaels writes:
“The Sound and the Fury repeatedly insists that what people and things do or mean is a function of what they are; it insists, that is, on identity as the determining ground of action or significance. In this, as the following pages will make clear, it is typical of the major American texts of the 1920’s and in particular of those texts…that belong to the discourse of what I will call nativist modernism.”
Note the shell game this sentence plays: by starting with a text that everyone agrees is “major,” he can then go on to call what it does “nativist modernism” as a first step towards saying that what it does is typical of other texts which are, therefore, also “nativist modernism.” If you sniff a tautology here, your critical nose has not led you astray, for The Sound and the Fury is typical of a group of texts which is defined by being similar to the typical example, which turns out to be, ta-da, The Sound and the Fury! It is, in short, similar to the texts which are similar to it.
I wonder if this is a problem. After all, since the texts with which he concerns himself are a strategically limited subset of the larger cultural conversation (but the strategy for selecting that subset is not explicitly addressed), Michaels’ claim for the twenties as the pivotal moment of change in how racial and cultural categories were imagined becomes very difficult to locate. And there is an unacknowledged dependence on the status of “major” texts as major: by talking only about already canonized works, he effectively closes off his argument from having any reference to other texts around it (except insofar as other texts contribute to these major texts). This allows him to cherry pick examples that fit his thesis, letting the underlying claim that these examples represent anything to be made only by implication.
We seem, in short, to be left with an argument about texts whose importance has been inherited, not argued. For example, when Michaels claims that Dixonian racism is a change from Red Rock and from plantation fiction, and then claims that Hemingway is a change from Dixon, I’m at a bit of a loss. Those are certainly useful ways of situating Dixon vis-à-vis other important writers, but it doesn’t tell me anything about how they map onto larger historical movements in American society. What are these changes representative of? What are they in response to? Faulting the book for not answering questions it didn’t set out to ask is, of course, a bad-faith move, but I want to note that to the extent the book justifies its study of these texts by their “major” status, its argument is unable to extend outward from those texts or tell me something about the larger cultural conversation in which they were only a part.
So let me ask again, rhetorically, is this a problem? In a Modernism/modernity forum on Our America, Marjorie Perloff noted the “inevitable reductionism of Michaels project,” and observed that it was “structuralist” rather than “historical.” I’m uncomfortable with this distinction; while I’m obviously scratching at the same itch she was, instead of faulting Michaels for not being “historical,” it seems more useful to think about what kind of histories are being produced by this methodology, and I’m not sure “structuralist” is the right word for them. After all, the gesture of canon-formation that’s going on here is precisely a production of history: by dismissing older principles of selection, Michaels produces a new (albeit anachronistic, but aren’t they all?) historical framework, “nativist modernism.”
As Charles Altieri notes in the same forum, the logic by which a literary historical period is created (the 1920’s as the period of nativist modernism) is one of the weakest links in Michaels’ argument: a period of place and time can only be conflated with a particular form of discourse by selectively ignoring all the many and important ways that they don’t link up. But what is this category he’s trying to establish? When Michaels writes that “I believe that any account of nativist modernism would end up making American literary history central,” he isn’t talking about the authors themselves so much as he’s talking about the contexts into which we place them that give them meaning. Yet aren’t we just talking about how the contexts that give them meaning are a function of the way those authors have already been conceptualized in the first place?
On the one hand, you cannot categorize a writer without reducing them. You can call Eliot a “modernist” and you’d be right, but that’s not all he is. And this gesture can be repeated: is Henry James an American? Is Chinua Achebe an African? A Nigerian? Is the pope Catholic? To each question, yes and no. Yes, of course yes, but also, let’s not forget that saying “yes” blinds us to the ways that it is also “no,” since whatever principle of selection you choose to define that category will exclude the things it excludes. James was an American, but also a European. Achebe is an African, but he is also a Nigerian who lives in the US and participated in the Ibo attempt to secede as Biafra. Et cetera. I think Altieri is therefore right to say that “where conflict and uncertainty should be, Michaels doggedly and brilliantly imposes a single dominant set of beliefs,” thereby producing a reductive vision of the literary scene in that period. But, on the other hand, that seems a bit beside the point; even if the thing Michaels is doing is reductive, isn’t it exactly the kind of reductiveness that category making requires, the soil in which canon formation grows, and without which the gesture of contextualizing and historicizing cannot occur? My point, I guess, is that it’s a tautology, but a useful one.
I was thinking about this the other day because I was trying to think about whether or not I’m a historicist, or rather, I was trying to figure out how to think critically about the critical paradigms I find I’ve absorbed practically by osmosis while in grad school. And I wonder if part of the virtue/problem with a term like “historicism” isn’t that it muddles an important distinction between a historian‘s methodology and a literary critic’s — if you’ll pardon me the conceit that such platonic ideals might exist — that one might locate, were one so inclined, in the scholar’s relationship and responsibility to the data.
For example, the day before last Halloween, Eric Rauchway at the Edge of the American West posted “three guidelines for social science analysis,” which were:
1. Anecdote is not the singular of data.
2. If you found something surprising, you’re probably wrong (but conversely:)
3. Don’t believe everything you think.
From the perspective of my perspective, I found each of these statements both deeply sensible and methodologically impractical. After all, how often does literary criticism have recourse to the empirically unsound — yet suggestive — anecdote? All the time, to put it bluntly. And while the skeptical humility of the third guideline is probably always a good thing, things “you think” are, themselves, grist for the literary critic’s mill, a task in which the point is not to distinguish representations from the real, but to figure out how it happens that we come to imagine the things we imagine and what it means that we do. Finally, when I’m trying to teach my students to read critically, I tell them that if something is surprising or counterintuitive, they need to learn to embrace that surprise, that strangeness, learning not just to think about why that strangeness seems strange, but also to inhabit that experience of vertigo, the friction between reality as we imagine it and as it is made and re-made through our fictional imaginations.
Not that I say it like that to college freshmen, exactly. But because it’s something I try to practice and model for them, that second guideline is both sensible and deeply problematic. Does it mean that one is wrong to be surprised (that you were right the first time and the surprising fact is wrong) or that the surprise indicates an original wrongness (your original assumption was wrong and the surprise comes from observing its failure in practice)? Probably both; presuming your own wrongness seems to me to be a way of imposing a necessary humility on the social science analyst, since an inclination to deduce the truth from the anecdotal needs to be replaced by something more like an inductive process, more attentive to the grain of historical experience and record than to one’s personal expectations and apprehension. It is to prevent one from falling too deeply into the trap of the brash James Watson, who — as Ahistoricality noted in that comment thread — was to have said that “No good model ever accounted for all the facts, since some data was bound to be misleading, if not plain wrong.” As Francis Crick glossed him, this was because a theory which fit all the data has most likely been built to do so, and would thus be open to suspicion: reality is simply more complicated than are the models we build to apprehend it. Yet the danger is that judging ourselves as competent to select which data is “wrong” will privilege the see-er over the seen, and (potentially) to cherry pick the data in service of a homogeneous theory one seeks to impose on the heterogeneity of reality. This is bad social science methodology. But might it be a good methodology for literary analysis?
The gesture of canon-formation, for example, seems to me to be exactly this kind of process; whether we wish to explore culture in terms of kitsch (as Moretti disapproves of Adorno for disapproving of Veblen for arguing) or we wish to understand low culture as a thing identified by its failure to live up to the standard of high culture, the inherent implication of a canon is to draw an evaluative distinction, and in doing so, to base itself on a categorical logic less derived from the texture of the texts (the historian’s empirical data) than of what we make of those texts, or what is made of them for us. Michaels, for example, does not argue inductively for the “modernity” of his selected texts, and this non argument seems quite typical to me in a landscape of literary criticism structured by anachronistic terms like “romanticism,” “medieval,” and “modernism,” ahistorical constructs that the textural record does not require or necessitate (or, in most cases, even inductively suggest).
Yet don’t we impose terms like these for reasons which seem to us to be good ones? Edward Said’s sense of “the job facing the cultural intellectual,” for example, was “not to accept the politics of identity as given, but to show how all representations are constructed, for what purpose, by whom, and with what components.” And even so un-Saidian a critic as Michaels is using this kind of playbook in a particular sense, after all; major texts are worth reading not because their “majorness” requires subservience, but because they are the tradition with which we must contend. To the extent that we have inherited them as important, we have also inherited the task of understanding why they’ve been deemed so, and re-evaluating that gesture. Which makes me wonder whether it’s possible to follow a program like Said’s without doing other than what Michaels has done, without tautologically presuming the canon one produces. Orientalism, too, didn’t so much reject the canon as presume it so that it could be reassessed and critiqued (though he’s been repeatedly and willfully misread on this point, I think).
I wonder, then, if the fact that Michaels’ book is an argument about texts whose importance has been inherited rather than defended, maybe this is actually its virtue. It’s an intervention into a tradition which, precisely because it presumes The Sound and the Fury‘s importance, raises the stakes for what we conceive of that novel to be, and makes the question something of importance. It can, of course, only be a legible or relevant intervention within that tradition, and can only be a meaningful categorical standard insofar as we have already agreed to live within the social reality constructed by the idea of modernism; without an investment in the term “modernism,” and whatever it entails, Michaels’ argument goes nowhere. Yet if we do have an investment in the particular construction of the social which it represents, then maybe it’s exactly this non-empirical logic that makes it a useful gesture. Tautological, but useful.