I had originally intended this essay to cover three thinkers, René Girard, Jacques Derrida, and Claude Lévi-Strauss. As I began working on it the thinking grew knotty and the prose just grew and grew. So I’ve decided to break the essay into two parts. In this part I begin by negotiating a transition from the previous article in this series, which was about how the contemporary discipline of literary criticism emerged after World War Two. I then say a little about my years at Johns Hopkins by way of introducing our three thinkers. After that comes some empirical evidence about the mid-century transition in literary criticism. I then conclude with a look at René Girard. I’ll discuss Jacques Derrida, and Claude Lévi-Strauss in the next part and make some remarks about the aftermath of this mid-century Sturm und Drang.
Do we really have a discipline?
As the previous article ended, Michael Bérubé was informing us that that the disciplinary regime set in place by Northrup Frye would begin unraveling a decade later. That’s what this article is about. But I want to begin by reviewing where we’ve been.
Let us start with Cowen and his interest in the greatest economists. Cowen took the disciplinary existence of economics as a given: In the beginning there was Adam Smith, and the rest followed after. He had to do a bit of tap dancing to fit John Stuart Mill into his scheme, for we generally think of Mill as a philosopher, not an economist, but that is easily done. One might, I suppose, do the same for literary criticism. The deparments exist in colleges and universities; just go back as far as you can.
However, administrative continuity is one thing; intellectual continuity is another. The conceptual focus of academic literary criticism changed in the middle of the previous century. That’s what the previous article is about. Sure, the primary texts are still there, but what academics do with them has changed. There are continuities, to be sure, there always are. Think of astronomy – Lord! I hate what I’m about to do – and its Copernican revolution. The earth, moon, sun, and the other planets are the same things they were before and after the revolution, but our understanding of their relationships has changed. Something like that is what has happened to literary criticism, and the discipline almost knows it and is still thrashing about with the consequences.
Brooks & Warren, taken as a duo, are important because they focused the discipline’s attention on the texts themselves in the most concrete way possible, by gathering a bunch of them together for a reader intended for undergraduates. That in turn brought the teachers of those undergraduates to think about those texts in a new way, to search for the meaning held within each text. In his Anatomy of Criticism Northrup Frye both conducted an inductive survey of the literary field and, in his polemical introduction, explained how an interpretive focus on that field constituted a proper academic discipline. Then, a century and a half before them, Coleridge introduced concepts that brought the literary mind into conceptual focus. That’s a discipline.
It's the focus on meaning that became problematic. For one thing it turns out that critics kept coming up with different meanings for the same texts. How can we call ourselves an academic discipline if we can’t agree on the central objects of our discipline? The problem occasioned a lot of thinking about theory and method. What’s even more problematic, in time it became conceptually difficult to separate the roles of critics and writers in the literary system, if you will. It wasn’t at all obvious that that would happen, and it took a while for awareness of problem to come into view.
It’s in that context that we should consider the well-known symposium that took place at Johns Hopkins in the Fall of 1966: The Languages of Criticism and the Sciences of Man. Notice that phrase, “the Sciences of Man,” from the French “les Sciences de l’Homme.” The concept of the human sciences is European, not American, and covers a range of disciplines that would be divided between the humanities and society sciences in America.
In retrospect that event is recognized as a ‘tipping point’ in the course of American literary criticism. Such things do not tip in the course of four days (October 18-21). They are the culmination, in this case, of a decade of uncertainty about the conceptual nature of literary criticism. Before taking a look at that 1966 symposium, however, I want to step back and insert myself at the edges of the narrative.
A change in perspective
I wrote the previous posts in this series from the “view from nowhere” point-of-view that is the default stance for much intellectual writing. That stance really isn’t available to me for this post, which are about developments that happened early in my career, during my undergraduate years at Johns Hopkins and my graduate study in the English Department at the State University of New York at Buffalo. While I certainly did not play an active role in the events I will be describing, I was an interested, concerned, and involved bystander (see the appendix, "Skin in the game," in the next installment).
Though I never studied with René Girard, I heard him lecture in classes I took with Richard Macksey when I was an undergraduate at Johns Hopkins and I later met with him in connection with book collection that never came to fruition (it was to be a collection of structuralist essays on Shakespeare). The ideas of Jacques Derrida influenced me during my undergraduate years, especially “Structure, Sign, and Play in the Discourse of the Human Sciences,” the paper he delivered at the famous 1966 structuralism symposium. However, the influence of Claude Lévi-Strauss’s work on myth was to prove more intriguing, influential and enduring.
Those are the three thinkers at the center of this part of our story: René Girard, Jacques Derrida, and Claude Lévi-Strauss. All of them were French, through Girard spent most of his life in America. Jacques Derrida was frequently in residence at Yale and other schools and ended his career at the University of California at Irvine. The Nazi occupation of America caused Lévi-Strauss to leave France for in America between 1941, where he stayed until 1947 and then returned to France in 1948. None of them were primarily literary critics.
To be sure, Girard sojourned in literary studies from the late 1950s on into the 1970s. But he had trained as a historian and, by the time of that structuralism conference in the mid-1060s, he was moving through anthropology to become a grand social theorist. Derrida was a philosopher, though he often commented on literary texts. And Lévi-Strauss was an anthropologist, though he famously collaborated with Roman Jakobson, the great linguist, on an analysis of Baudelaire’s “Les Chats.”
What, then, are three non-literary critics doing in a series of posts ostensibly about the GOAT literary critics? They are influencing the course of literary criticism, that’s what. And that’s ultimately what this series of posts is about (for me). Why is literary study like it is? That the answer to that question depends critically on major thinkers outside of literary studies, that tells us something about the peculiar nature of a discipline focused on teasing out the meaning of literary texts.
Turning toward our three thinkers, Derrida arguably had more influence on literary criticism as a whole after 1970 than anyone trained as and writing primarily as a literary critic. It’s not clear to me that that is true of Lévi-Strauss, though he certainly had an influence and Derrida arguably made his bones with a famous essay about Lévi-Strauss. As for Girard, Tyler Cowen thinks he’ll go down in intellectual history as one of the major French thinkers of the last half century. Perhaps so. But that would be more in his person as a social theorist than as a literary critic. There his influence has been real, but limited. Thus he plays a somewhat different role in this story.
Note: In Appendix 1 I present some empirical evidence about the relative importance of these three thinkers.
Transition, the 1970s in literary criticism
Before discussing these three thinkers, however, I want to talk about what happened to academic literary criticism during this period in a general and empirical way. This is possible because of an important article that Andrew Goldstone and Ted Underwood published in 2014 in New Literary History: “The Quiet Transformations of Literary Studies: What Thirteen Thousand Scholars Could Tell Us”. Using topic modeling, a machine learning technique, Goldstone and Underwood analyzed complete runs of seven mainstream journals going back to the late nineteenth century.
This is not the place to explain how topic analysis is done – you’ll find an explanation in the article, or you may consult Appendix 2, where you’ll find what ChatGPT said about topic modeling. In this context “topic” is a term of art and refers to a group of words that tend to occur together in a collection of documents. It is up to the analyst to interpret the overall meaning of the topic.
This chart shows how Topic 16 evolved over time:
Here are the words most prevalent for that topic; criticism work critical theory art critics critic nature method view. The topic seems to be oriented toward method and theory and is most prevalent between roughly 1935 and 1985. The first half of that period likely reflects the rise of the so-called New Criticism while the second half reflects the developments I’ll be examining in this post, which came to a head at the time of that 1966 structuralism conference at Johns Hopkins. I should note as well that the study of poetry was prominent during the era of the New Criticism while disciplinary interested shifted toward the novel in the post-structuralist era. This is not covered by Goldstone and Underwood. I take it as a casual observation from personal correspondence with Franco Moretti.
Now look at Topic 20. Judging from its most prominent words, it seems weighted toward current critical usage: reading text reader read readers texts textual woolf essay Virginia.
That usage, reading as interpretation, becomes more obvious when we compare it with Topic 117, which seems to reflect a more prosaic usage, where reading is mostly just reading, not hermeneutics: text ms line reading mss other two lines first scribe. Note in particular the terms – ms line mss lines – which clearly reference a physical text.
This topic is most prominent prior to 1960 while Topic 20 rises to prominence after 1970. For what it’s worth – not much, but it’s what I can offer – that accords with my personal sense of things going back to my undergraduate years at Johns Hopkins in the later 1960s. I have vague memories of remarking (to myself) on how odd it seemed to refer to interpretative criticism as mere reading when, really, it wasn’t that at all.
Reading isn’t the only word whose meaning shifted during that period. Theory changed as well. Up into the 1960s and even the early 1970s “literary theory” was thinking about the nature of literature, as exemplified by the venerable Theory of Literature (1949), by René Wellek and Austin Warren. By the mid-1970s or so literary theory had come to refer to the use of some kind of theory about mind and/or society in the interpretation of literary texts. That is to say, it wasn’t theoretical discourse about literature. Rather it was a method for creating interpretations of texts, which are readings in the newer sense of the term. That’s where Derrida and Lévi-Strauss come in, as sources of interpretive tools, along with Marx, Freud, Adorno, Barthes, Deleuze, Foucault, and a cast of thousands. Well, I exaggerate the number, but you get the idea.
The rest of this essay, especially the second part coming up in a later post, is about those shifts.
Finally, and in view of recent events at Harvard and elsewhere, I must admit that I’ve ‘plagiarized’ from myself in what I’ve said about Topics 20 and 177. Those words come from a working paper that has a more complete discussion of this transitional period based in part on the work of Goldstone and Underwood. Here’s the paper:
Transition! The 1970s in Literary Criticism, Version 2, January 2017, https://www.academia.edu/31012802/Transition_The_1970s_in_Literary_Criticism
Rene Girard and Mimetic Desire
René Girard was the first professor I heard lecture when I entered Johns Hopkins in the fall of 1965. At one point during freshman orientation we were given a choice of attending one of five lectures. I attended Girard’s lecture on cultural relativism – I forget what the other four were. I remember only two things from that lecture: Girard spoke with an accent and it was the craziest damn thing I ever heard. Yes, that’s how I thought of it at the time, but this guy was a Hopkins professor, so there must be something to it.
In the spring of 1966 I took Richard Macksey’s course on the autobiographical novel. He invited Girard to lecture on mimetic desire, the central idea of his first book, published as Mensonge romantique et verité in 1961 and translated into English as Deceit, Desire, and the Novel in 1966. Here’s a brief account of the argument by John Pistelli:
He posits, therefore, a fundamental geometric relation governing the plots of great novels: the protagonist seems to desire something (wealth, status, a lover, etc.), but in fact really desires the object through a mediator whom the protagonist wishes to emulate or usurp. Desire, then, is mimetic—in the sense of mimicry—and triangular—because it doesn’t go from subject to object but from the subject through the mediator to the object. Girard’s first and clearest example is Don Quixote, who learns to want what a chivalric knight wants through reading about the hero Amadis of Gaul in medieval romances; his quest is less to possess the knight’s rewards than to become Amadis through this possession. Given Don Quixote’s reputation as the inaugural European novel, it’s no surprise to find that the pattern continues in later fiction: Girard’s main examples are the works of Stendhal, Flaubert, Dostoevsky, and Proust.
That book put Girard on the intellectual map, making him a force to be reckoned with.
And he used that force to organize the 1966 symposium at Johns Hopkins: The Languages of Criticism and the Sciences of Man. Girard provided the visibility and the international connections while Richard Macksey and Eugenio Donato did much of the actual organizing. Here’s a few observations Cynthia Haven offered about that symposium in the chapter, “The French Invasion,” in her biography of Girard, Evolution of Desire: A Life of René Girard (2018):
At that historical moment, “structuralism” was the height of intellectual chic in France, and widely considered to be existentialism’s successor. Structuralism had been born in New York City nearly three decades earlier, when French anthropologist Claude Lévi-Strauss, one of many European scholars fleeing Nazi persecution to the United States, met another refugee scholar, the linguist Roman Jakobson, at the New School for Social Research. The interplay of the two disciplines, anthropology and linguistics, sparked a new intellectual movement. Linguistics became fashionable, and many of the symposium papers were cloaked in its vocabulary.
Girard never saw himself as a structuralist. “He saw himself as his own person, not one of the under-lieutenants of structuralism,” said Macksey. Yet structuralism would have had a natural pull for Girard, who was already moving away from literary concerns and toward more anthropological ones by the time of the symposium. Indeed, in this as in other matters, he was indebted to the structuralists. His own metanarratives strove toward universal truths, akin to the movement that endeavored to discover the basic structural patterns in all human phenomena, from myths to monuments, from economics to fashion.
It's that drive toward “more anthropological” issues that interests me.
For it is through his readings in anthropology that Girard was able to enlarge mimetic theory to embrace society-wide dynamics of conflict and the resolution of conflict through the mechanism of sacrifice. Just how he was able to do that is not my concern here, any more than I was concerned about the validity of Northrup Frye’s archetypal criticism in the previous essay in this series. That led Girard to write his second book, Violence and the Sacred, which came out in English translation in 1977. And from there, apparently – for my own attention was elsewhere – he came up with what is somewhat derisively called a “theory of everything,” albeit one that was convincing enough that he was elected to the Académie Française in 2005. I prefer to think of him has a grand social theorist in the 19th century manner of, say, Karl Marx, Herbert Spencer, or somewhat later, Oswald Spengler. Whether history will remember him as such in 50 years, that’s another matter.
“But,” one might ask, “why did he do it?”
“Well,” might come the response, “there is curiosity.”
“But that’s not enough given the strictures of the modern academy.”
“But it wasn’t always thus, was it? Earlier thinkers had more latitude, no?”
“Yes, but...”
Seeing the world through texts
And so the conversation might go. I want to add something further into this speculative mix, something we’ll return to in the second part of this essay. Let’s turn to an unpublished essay by Cowen from January of 2005: “Is a Novel a Model?” Here is his abstract:
I defend the relevance of fiction for social science investigation. Novels can be useful for making some economic approaches – such as behavioral economics or signaling theory – more plausible. Novels are more like models than is commonly believed. Some novels present verbal models of reality. I interpret other novels as a kind of simulation, akin to how simulations are used in economics. Economics can, and has, profited from the insights contained in novels. Nonetheless, while novels and models lie along a common spectrum, they differ in many particulars. I attempt a partial account of why we sometimes look to models for understanding, and other times look to novels.
Setting social science aside, isn’t that one of the reasons people read novels, to learn something about how the world works?
Consider a passage from a great mid-century critic, Kenneth Burke. This is from a midcentury essay, “Literature as Equipment for Living” (1938) that has been reprinted in The Philosophy of Literary Form (1973, p.298):
…surely, the most highly alembicated and sophisticated work of art, arising in complex civilizations, could be considered as designed to organize and command the army of one's thoughts and images, and to so organize them that one “imposes upon the enemy the time and place and conditions for fighting preferred by oneself.” One seeks to “direct the larger movements and operations” in one's campaign of living. One “maneuvers,” and the maneuvering is an “art.”
Whatever our interest in the imaginary creatures in these fictions, whether the imagined world is realistic in the various ways writers have advanced or fantastic in many other ways, we are even more interested in understanding and living in the world around us. Literary texts display and highlight patterns of behavior which we then see in the people and events around us.
And just how do we learn of the world? There are, of course, the events which we experience directly. But there’s a much larger range of events that we learn about through various media, including texts of various kinds, articles in newspapers and magazines, and nonfiction books of various kinds, including histories and biographies. To a first approximation are not these various texts, fictional and nonfictional, more like one another than any of them is like the actual events? A pattern is a pattern whether one finds it in fictions or nonfictions.
To be sure, Burke was addressing himself to the ordinary habits of ordinary readers while Cowen was speaking to social scientists. Those are different audiences, no? Still, social scientists are also ordinary readers and there is a point where ordinary readership and social scientific cogitation blur into one another. Girard may be writing as a social scientist, albeit in an older formation of the role, but when he is writing on religious topics, as he often is, can one separate the scholar from the devout Catholic, for he is that?
Recall that Girard was trained as a historian, as someone who uses texts as evidence about events in the world. Is it so difficult to imagine that, having discerned mimetic patterns in a variety of important fictional texts, Girard would then go on to discover those same patterns in nonfiction texts? Once he’s done that, well, there are lots of nonfiction texts one can read about the world.
There are limitations to what we can learn about the world simply by reading texts, whether they are fictional or nonfictional texts. If that were not the case, there’d be no need for empirical investigations. Cowen knows this, that’s why he’s a social scientist. Girard certainly knew it as well. Yet some have found his grand theorizing a bit fanciful. Cowen is aware of the problem and cautions us against some aspects of Girard’s thinking in his post on Girard’s importance:
Where is Girard weakest: His theory of language, his overemphasis on the destructive nature of mimesis, excess claims to have discovered universal mechanisms, just making lots of stuff up, and not knowing enough economics or empirical anthropology.
That seems quite a lot. But I’ve not come here to bury Girard. If you’re interested in that, one of his colleagues at Stanford, Josh Landy, has done so in his essay, Deceit, Desire, and the Literature Professor: Why Girardians Exist. Nor am I interested in praising him. I’m interested only in situating him in an account of what happened in literary criticism during the 1960s and 1970s. I’ll say more about that in my next essay, where I discuss Derrida, Lévi-Strauss, and others.
Appendix 1: By the numbers
Google Ngram counts are a crude and somewhat problematic tool. But it’s a tool that’s readily available and easy to use. As long as we realize that we’re skating on thin empirical ice we should be OK.
I want to use Ngram counts as a rough measure of the relative popularity and influence of Girard, Derrida, and Lévi-Strauss. Just how that translates into importance and profundity, if it does so at all, that’s another issue about which I will say nothing.
There is, however, a problem at the outset. Will Google Ngrams conflate the name of our French thinking with the American manufacturer of blue jeans? Noting that the thinker has an accent in his name, while the blue jeans manufacturer does not, I did a plot of the two names from 1960 through 2019 (the last date for the English-language book collection). Here’s the result:
We can see that the Ngram plotter recognizes a difference. The plot for the French thinking, “Lévi-Strauss,” makes sense to me. Given his general importance, it’s not surprising to see the number of references increase after his death (2009). And that curve is consistent with the ones we’ll see shortly. I have no idea why the curve for “Levi-Strauss” takes the form it does, though I have no reason to believe that it only involves the jeans manufacture. This does not make me happy, but there’s nothing I can do about it.
So, let’s plot the curves for Derrida, Girard, and Lévi-Strauss.
The curves for Derrida and Lévi-Strauss seem plausible to me. That is, independently of this exercise, it has been my impression that Derrida has been more influential than Lévi-Strauss.
But the curve for Girard does not make sense. Why should it be roughly level since 1960 when the book that put him on the intellectual map, Mensonge romantique et verité, was only published in 1961. I conclude that René Girard is not the only Girard in the corpus.
We’ll have to resort to searching on first-and-last names together:
The curves for Derrida and Lévi-Strauss are roughly the same as before, which is encouraging. Note, though, that in the previous chart, Derrida peaks at just below 0.000550% while in his one he peaks at just below 0.0000700%. I would expect that a search on first-and-last names would produce fewer hits than one on last names alone, but I have no intuitions about whether or not that difference should in fact be an order of magnitude.
On the other hand, the curve for Girard now makes sense. It starts at or close to zero in 1960 and is close to Lévi-Strauss in 2019, as it was in the previous chart, but Girard is now below Lévi-Strauss rather than above him. That change, however, is minor in comparison to fact that are at roughly the same distance below Derrida.
I would expect the relative positions of those three thinkers will change in coming decades. But I can see no point in making an argument on the matter at this time about how I think that should go.
Note: You might want to compare Derrida’s visibility over Lévi-Strauss and Girard with his visibility over Northrop Frye, Stanley Fish, George Steiner, Hillis Miller, Harold Bloom, Noam Chomsky, Robert Penn Warren, Cleanth Brooks, and John Crow Ransom in charts in an earlier post. However, in those charts I only used Derrida’s last name while I used first (middle) and last for the others. That may have skewed things in Derrida’s favor. When I have time I’ll redo those charts.
Appendix 2: Topic Analysis
I asked ChatGPT to tell me about topic analysis using a succession of four prompts. I then asked it to gather that information into a single statement. Here’s that statement.
Basics of Topic Analysis
Overview:
Topic Analysis is a natural language processing (NLP) technique used to identify underlying themes or subjects in a collection of text documents. It involves the application of algorithms to discover clusters of words that frequently co-occur, representing topics present in the corpus.
Key Steps in Topic Analysis:
- Data Collection:
- Gather the relevant text documents for analysis, ensuring the data is in a suitable format.
- Data Preprocessing:
- Clean the data by removing irrelevant information, tokenizing text, removing stop words, and performing stemming or lemmatization.
- Document-Term Matrix (DTM) Creation:
- Construct a matrix where each row represents a document, and each column represents a unique term, capturing word frequencies.
- Topic Modeling Algorithm (e.g., LDA):
- Apply a topic modeling algorithm, such as Latent Dirichlet Allocation (LDA), to infer topics based on word distributions.
- Number of Topics:
- Choose the number of topics for the algorithm to identify, guided by prior knowledge or cross-validation.
- Topic Inference:
- Run the algorithm to infer the distribution of topics for each document and terms for each topic.
- Interpretation and Analysis:
- Analyze the results to understand the main topics present in the corpus. Interpret topics based on associated words.
- Visualization:
- Create visualizations, such as topic proportion plots or word clouds, to represent topic distributions.
- Refinement and Iteration:
- Refine the analysis based on insights gained, and iterate the process if necessary.
"Bag of Words" Model:
- The "bag of words" model simplifies text representation by focusing on word frequencies and disregarding word order.
- It involves tokenization, counting word frequencies, and creating a Document-Term Matrix (DTM).
- Despite its simplicity, the model is effective for various NLP tasks and serves as the input for topic modeling algorithms.
Understanding Topics:
- In the context of topic modeling, a "topic" is a cluster of words that co-occur frequently in documents.
- Analysts interpret topics based on associated words, providing semantic understanding and connecting topics to real-world concepts.
- Domain-specific knowledge is often crucial for meaningful interpretation.
Importance of Interpretation:
- Interpretation by analysts is necessary as numerical representations provided by algorithms lack inherent meaning.
- Analysts bring context, domain expertise, and semantic understanding to assign meaningful interpretations to topics.
- Interpretation ensures quality control, relevance, and insight generation from the analysis.
No comments:
Post a Comment