Wednesday, March 01, 2006
What We Talk About When We Talk About Genre
In a course I am co-teaching at the City University of New York’s Graduate Center, Louise Geddes, one of the students, posted some questions about genre theory:
I’m reading WHAT IS PASTORAL by Paul Alper, and he has an interesting discussion about the difference between mode and genre. Alpers cites Alistair Fowler, in defining genre (although Fowler prefers the term “kind") as being “marked by a complex of substantive and formal features that always include distinctive (though not usually unique) external structure” (Alper, 45), before going on to explain that the generic repertoire includes “meter, size, and style to values and attitude.” (45). Alper then talks about mode, saying that "’mode’ is thus the term that suggests the connection of “inner” and “outer” form; it conveys the familiar view that form and content entail each other and cannot, finally, be separated,” and also that a mode can include a number of different genres. He also heavily cites Angus Fletcher, who says that “mode is the literary manifestation, in a given work, not of its attitudes in a loose sense but of its assumptions about man’s nature and situation” (50).
I have several questions: Firstly, how are we defining genre in this class? At the risk of putting things into an “inner” or an “outer” category and creating oversimplication, it seems as though we’re going to be looking at a combination of the two - is this a reasonable assumption? For example, the Arcadia could be in the epic genre, but the romance (and pastoral) mode, with a healthy helping of eclogues dropped in. If “mode” balances form and content, then is genre redundant? Is this simply the argument that put genre study out of vogue for so long? If so, how is this concern being redressed? I think that this is a fascinating discussion, but I’m a little perplexed by our terminology.
My answer follows.
You ask some wonderful, challenging questions. In response, I have several overlapping and partially contradictory answers, a cowardly cop-out (the sincerest way, I have found, to honor good questions), and a few further questions of my own. I have numbered these items to give them a spurious aura of structure and rigor.
1. You are right to be puzzled about terms like genre, sub-genre, and mode. My hope is that this course will make these terms “a little harder to see,” to quote Wallace Stevens.
2. Early modern literary theory did not distinguish clearly between genre and mode, although such a distinction seems to be implicit in remarks like Polonius’s about “tragedy, comedy, history, pastoral, pastoral-comical, historical-pastoral, tragical-historical, tragical- comical-historical-pastoral, scene individable, or poem unlimited.” Of course this list is a parodic attack on the systematizing tendencies of treatises on literary kinds like Scaliger’s.
It is worth adding a few notes about how early modern critics thought about genre. Most importantly, they defined many kinds at least partially by their forms, but their dominant tendency was to define kinds by their social functions. George Puttenham says that pastoral was a kind whose purpose “to counterfeit or represent the rustical manner of loves and communication, but under the veil of homely persons and in rude speeches to insinuate and glance at greater matters, and such as perchance had not been safe to have been disclosed in any other sort, which may be perceived by the Eclogues of Virgil, in which are treated by figure matters of greater importance than the loves of Tityrus and Corydon.” To take an unforgettably bizarre and offensive example, Puttenham says that the purpose of the epithalamion was to drown out the “outrageous shrieking” of the bride.
3. I agree that no distinction between the “outer form” and “inner form” of literature is sustainable. But it is sometimes useful to make over-simplifications like this. The question is what the pay-off might be for knowingly making a particular over-simplification.
4. Part of the problem with discussions of genre has been their ahistorical essentialism. When considered through different lenses, a work may belong to several different genres. These lenses themselves change over time. It is only retrospectively that one can label Robinson Crusoe as a novel and group it with Pamela and Joseph Andrews.
5. One must become comfortable with the fact that many works cannot be confidently assigned to a single genre. But this does not mean that we cannot examine the ways those works use generic codes. Often these classification-resistant works are the most productive objects of inquiry for genre criticism.
6. There are some wonderful things in Paul Alpers’s book on pastoral, but I had trouble with his chapter on the concept of literary mode. I found it difficult to follow, and I found the payoff somewhat disappointing. Alpers says that most theories of pastoral center on place, on Arcadia and the Golden Age; he confidently and somewhat arbitrarily decides that the crucial defining feature of pastoral is the presence of shepherds. Perhaps some enterprising eco-critic will make a case for focusing on the sheep. Annabel Patterson takes an antithetical approach: she finds it so difficult to define pastoral that, despite the title of her book, Pastoral and Ideology, she decides to evade the task entirely and simply say that the works she discusses are in one way or another imitations of Virgil’s Eclogues. I think the lesson here is that rather than trying to identify the essential feature of pastoral, one should talk about the full range of conventions, which of those conventions are present in a particular work, and how those conventions are creatively modified. Even a partial list would include the song contest, the pipes, the names, the poetic genealogy, the otium, the locus amoenus, the gift of a sheep or a cup, the lamenting of unrequited love, the echoing woods, the fall of night at the end of the eclogue, the debate between old and young shepherd, and shepherds and sheep more generally. As one example of how to approach the analysis of the way a particular feature gets used in different works, I would point to Alastair Fowler’s discussion of names in pastoral. Fowler has observed that the set of character-names in pastoral remained small and stable: if you encounter a figure named Lycidas or Corydon or Damon, you are almost certainly reading a pastoral, or a pastoral inset in a larger work. A pastoral name is a strong, clear generic signal. Therefore, even a small innovation in the naming of pastoral figures signals a generic shift, a reconfiguration of pastoral. To name a pastoral shepherd “Piers,” as Spenser does, announces the introduction of a Christian allegory into an eclogue, and it also establishes a vernacular genealogy for pastoral where none had existed: it makes Piers Plowman into a proto-pastoral. A grand generalization about a mode cannot account for subtle modulations of this kind. One needs an understanding of the full generic repertoire and its history.
Proposed definition of pastoral: “We are such stuff as dreams are made of, and our little life is rounded with a sheep.”
I have found Richard Slotkin’s (1984) “Prologue to the Study of Myth and Genre in American Movies” (Prospects 9:407-432) a very useful discussion of genre, albiet in a slightly different context than the one above (so perhaps with necessarily a slightly different meaning?). One of the nice things about Slotkin’s approach is that it does try to be historical--to say something about how and why genres emerge and develop.
What’s a stake? That is, why does the definition of genre matter? What role does genre play in the literary system?
I ask because I tend to think of genre as being loosely comparable to the notion of a species in biology. The notion of a species plays a certain role in the causal “logic” of evolution. Do genres play a certain role in the causal “logic” of the literary system, or are they simply a classificatory convenience?
Classification is tricky. We seem to have a strong preference for tree-like classification systems, where each item can have one and only one parent. If one believes that genres are to be arranged in a tree-like classification system, then a given text can be of one and only one kind. If you feel it necessary to treat various texts as belonging to more than one kind, then either you aren’t looking at them properly, or a tree-like classification system isn’t appropraite for literary works.
And this takes us back to “causal logic.” The causal mechanisms in the biological world naturally lead to tree-like phylogeny for a good many life forms—namely, multicelled animals and most multi-celled plants; single-celled organisms are rather different. In this case the “causal logic” is that of reproduction and adaptation. A tree-like classification scheme is a natural fit to tree-form phylogeny.
It is not at all obvious that a tree-form phylogeny obtains in the literary system, not at the level of the text. Though I’ve thought a bit about genres, I haven’t really gone to town on it, as it were. I read Fowler’s Kinds of Literature some years ago and was not happy, but my unhappiness didn’t prompt me to make an attempt to sort things out.
I’ve given somewhat more though to the problem of “kinds” in music, with jazz as my focus, but there it’s not clear to me what “kinds” of the “genre” type would be. Stylistic categories are by far the most important ones. Different types of performing ensemble are recognized—solo piano, piano trio, quartet-quintet-sextet, big band (say 12+ members). And different forms are recognized, 12-bar blues, 32-bar “standard,” the rest. And different kinds of arragements—head arragements (consisting of head, improvised solos, head), vs. more elaborate forms. And different tempi corresponding to different emotional tones, slow ballads, up-tempo swingers and such.
But the joint product of all that hasn’t yielded any strong sense of “kinds” that are comparable to genre. I don’t know what a jazz parallel to the pastoral would be.
Were I to have at it on genre in literature, I’d certainly wander down the well-trod path of attempting to distinguish between form and content. Yes, I know the distinction is a tricky one, but . . . Given some such distinction, it is clear that the pastoral has both formal and content elements in it. If that’s what it is, then that’s what it is. The notion of a sonnet does not seem to imply specific content. Does it make sense to think of the sonnet as a genre, or family of genres? It certainly is a family of kinds. What I"m interested in is constellations of recurring elements. What I’m not interested in is insisting that literary kinds can be adequately classified in trees or that kinds can be defined purely in formal terms, or content terms.
"I don’t know what a jazz parallel to the pastoral would be.”
The time periods and social contexts are off by too much to compare those two. But wouldn’t the traditional response be that jazz is a musical genre with subgenres such as Dixieland, free jazz, bebop, hard bop, fusion, big band, jug band, and so on? Reviewing my own loose bundle of genre definitions (marketing term, beleagured-and-contested territory, debate-and-mutual-interest society for producers, assumed expertise for consumers), they seem to match.
"It is worth adding a few notes about how early modern critics thought about genre. Most importantly, they defined many kinds at least partially by their forms, but their dominant tendency was to define kinds by their social functions.”
That still sounds right. Genre is what we call it when function is perceived to delimit or prompt form. The difficulty is that functions vary over time while texts stay set.
Since literary scholars by definition have greater access to texts than to writers, editors, publishers, or audiences, some confusion is inevitable. I’ve heard academic responses to genre range from puzzled to puzzling to apalling. (English department head to Samuel R. Delany: “I don’t read that science fiction shit.")
Even in historic genres, it’s possible to make a conscious effort to switch imaginative contexts. I’m guessing that’s what Matt wants to do with his class. Along those lines, I liked Fowler’s book more than I expected. Genre markers are (sometimes unintended) signals to a community. Spotting the markers isn’t enough to tell us what they’re doing. Fowler seemed to get that.
A more recent good example is Richard A. McCabe’s “Annotating Anonymity” in Ma(r)king the Text. Rather than just describing the archaisms and explanatory apparatus of The Shepheardes Calendar as pastoral conventions, McCabe shows how they took advantage of (and sometimes had to work against) the specific model of Servius’s Virgil commentaries.
I’m now finishing Memory in Oral Traditions by David C. Rubin.* (Thank you, forgotten person who recommended it to me.) Rubin explains certain generic conventions of epic, ballad, and counting-out rhymes as directly molded by the mechanisms of multi-generational verbal transmission, and backs it up with evidence.
On a more negative note, Franco Moretti’s “Graphs” and “Trees” are strictly parasitic on texts which he keeps strictly at arm’s length. Not what I’d call a healthy relationship. More successful** interdisciplinary materialist approaches go beyond the confines of the English department. And although the nora project limits itself to textual analysis, it’s with the reasonable aim of noticing textual aspects missed by received opinion.
Books don’t compete with each other in a closed bibliosphere. Popularity and genre are social, not strictly textual. They can only be understood by looking outside a text itself. And when you do so, I think you find something more like chaos theory than like biological evolution.
* Although Rubin acknowledges that his book’s body can get a bit dull and repetitive, he has a winning way with an endnote. Here’s my favorite from Chapter 9: “The one inconsistency in order indicates that the pattern was not always followed. It is not an error in the epic. The claim of a fixed-order script is mine, not the poets’.”
** “More successful” by my lights, that is; obviously not more successful in terms of public attention.
The thing is, Ray, I don’t think musical styles are comparable to literary genres. Things like “hard bop” and “fusion” seem parallel to things like early modern or late romantic. “Jug band music” is like, I don’t know, say the African-American toast. Big band music is like symphony music.
When I talk about jazz and pastoral, I’m not looking for a pastoral sensibility within the jazz world—you could find it, though, in Dave Brubeck’s Japan album. I mean there isn’t anything in the jazz world the functions like the pastoral has done in the literary world. If jazz had a form understood to be, say, “in praise of limber-limbed ladies’ and that you had examples in a variety of styles—say traditional, swing, and early bop—for performance by various types of ensemble—solo piano, piano trio, vocalist accompanied by piano and piano trio—that would be functionally comparable to the pastoral. But there isn’t anything like that.
There might be something like that in Western Classical though. I’m thinking of things like coronation marchs, or Kyrie’s in masses, and such things. But even there I’m not sure of the parallel.
But “the pastoral” is a special case, consisting of a chain of references between temporal-geographical clumps over thousands of years. _No_ musical genre can have that kind of history.
If we instead restrict ourselves to similar timespans and cultural spans—for example, between twentieth-century popular music and twentieth-century popular literature—I think analogies are easier.
Welll, maybe. But I don’t even see musical parallels to things like comedy, tragedy, and romance. Are there some parallels? Yes. But it’s not the same.
And we’re only thinking in Western terms. What about the music and literature, for example, of the Indian subcontinent? I know very little about literature and only a little more about the music. And the classical music is a mix-and-match affair of rhythmic patterns, called talas, and musical patterns, called ragas. There isn’t an exact parallel for either of those in Western music after, say, the early modern period. The thing about ragas is that they are often specific as to season of the year and time of day for performance. And the performance is improvised.
If we do restrict ourselves as you suggest, I think problems still arise. Consider 20th century pop music and literature. What we have now when we go into stores are marketing categories, which are real categories in the cultural system. They set the terms in which these goods are marketed and sold. For fiction we’ve got categories like romance, fantasy, mystery, science fiction, and subtypes of that. That’s not quite like rock, country, hip-hop, disco, folk, world, world-beat, pop, blues, etc. and their subcategories.
Now maybe blues functions something like the pastoral, or perhaps the sonnet. On the one hand, we have a whole pile of blues styles and traditions. But we also have the blues as a partidcular set of conventions within other styles. Every variety of jazz up through the 1960s has blues forms that are recognized as such by the musicians, but these are only one set of forms within those styles. And the same with classic rock and roll. “Jailhouse Rock,” for example, is a standard blues.
Categories like world and world-beat music are each internally diverse. World music is most likely subdivided by continent and country. In each category you get whatever music is made within that location, all treated as, say, Brazilian. So that music is simply marketed as Brazilian music here, but it has its own world types just like American vernacular music does.
I’m sorry, Bill; it must seem as though I’m being deliberately obtuse, but my obtuseness is genuine. When I reflect on the uses and problems of the word “genre” as I’ve heard it applied to music, I can’t help but see similarities. Genre as marketing category, often resented by the marketed artists, check: sections in the shops, as you note; links in Amazon; the New York Times has a rock critic, a mystery critic, a jazz critic, and a science fiction critic.... As web of influences and shared habits which can rigidify into a recipe for hackwork, check. As contested territory, check: the arguments I hear surrounding Wynton Marsalis remind me a lot of sf and poetry wars. As betrayed territory, check: Bob Dylan and Miles Davis, Kurt Vonnegut and Jonathan Lethem, all moved to “the mainstream”. There’s even the same oddity of artifacts being shoved into new “genres” retroactively after a context shift, although Anglo-American readers and publishers pay so little attention to translations that our literary shifts tend to be temporal (gothic novels repackaged as horror; high-art responses to folklore repackaged as fantasy; popular theater repackaged as high art) more than geographic. (But consider domestic bafflement at how our “commercial trash” is reconceptualized in Europe.)
Obviously there are vast differences between a performing medium and a printing medium, but I can see how you could carry over some concept of “genre” if you wanted to. I guess the question is “Why do you want to?” Or, as you asked, “What’s at stake?”
But it’s also a difference between a medium that is about words and the experiences you can create through those words and a medium where there are no words, or where words are secondary or, at the very most, in partnership with something that has as much effect on the result as the words do. All art forms have discussions about hackwork vs. substance, formula vs. originality, technical virtuosity vs. authenticity. And they will have these discussion in whatever categorical terms are available.
Take swing vs. bebop. That’s different from romance vs. mystery. People who have a passion for romance novels may be indifferent or even repelled by mystery novels. They can read and understand them perfectly well; they just don’t like them.
But when bebop emerged many swing musicians and fans objected because they simply and honestly did not hear it as music. It didn’t make sense to them. It was noise. That’s like someone who likes romance being unable to follow the plot of a mystery novel because they don’t know what the actions and intentions of the characters are. That’s quite different from being uninterested in such plots.
Correct me if I am wrong (my knowledge of the initial reception of jazz artists is sometimes shaky), but doesn’t the move from swing to bebop constitute an avant-garde transformation of a more or less popular art form into something DESIGNED to appeal to aficionados and to repel the ignorant masses? There are certainly many analogous transformations in literature, like, well, the modernisms of Joyce, Proust, Rilke, and Eliot.
Incidentally, the distinguished medievalist Helen Cooper (who holds a professorship originally created for C. S. Lewis) just pointed out to a group of academics how strange it is that certain romances like Bevis of Hamtoun (or Hampton) remained popular in a virtually identical form for five centuries. This poses a bit of a problem for literary historians who want to historicize literary works.
But Bill, I _have_ heard analogies to the “noise” thing in literary genres. Poetry maintains a venerable tradition of attacks on self-indulgent or random or fraudulant noise. Delany and others who have tried teaching science fiction to traditional English department types have reported that readers have to gain comfort and expertise with the idea and practice of world-building; it’s not something we’re born with. And hasn’t anyone else heard English teachers complain about how hard it is to get kids nowadays to understand what’s going on in a Victorian novel?
That’s one of the justifications for teaching literature or writing criticism, right? That it’s possible to help someone learn a new reading protocol?
Whether listening to / making music or reading / writing, though, exposure and practice are more important than studying lists of characteristics. When we talk about the formal properties of a genre, we’re generally talking about unconscious pattern learning from a set of prototypes.
David Rubin’s book summarizes some great research on a genre which happens to be both literary _and_ musical: the traditional ballad. There’s plenty that’s germane to this discussion, but I’ll just mention one piece of it. Subjects who were exposed to ten examples of traditional ballad became significantly better than the norm at identifying where the texts of previously-unseen ballads had been tampered with. Not too surprisingly, they also did a better job of listing characteristics of a ballad, and, when challenged to compose an original ballad on a set subject, came up with something that included more characteristics of the genre. But what’s interesting is that the characteristics they consciously listed were far fewer than the ones they unconsciously reproduced.
As is unfortunately typical, the musical side of things didn’t get much attention. But Rubin does talk about another experiment in which expert tradtional ballad singers were asked to compose a new ballad about a newspaper story, and says that (except in one case of obvious parody) both the tunes and lyrics seemed original. The newspaper story was adjusted quite a bit to fit ballad lyric conventions, though, and a year later, the singers had a hard time telling their own lyrics from other singers’ lyrics. I’d imagine that the new music was recognizably “ballad” as well—that is, that it was equally “generic”.
More or less, Matt, and we don’t think of those literary transformations as matters of genre. Nor should we think of the jazz transformations as matters of genre. You have blues tunes in swing, you have them in bebop; you have rhythm changes in swing, you have them in bebop; you have ballads in swing, you have them in bebop; and so forth.
Yes, Ray, we have the “noise” (or crap) thing in literature. But we don’t think of that as a matter of genre, it’s something else—like period or style. We thing of SF as a genre; but is the Victorian novel a genre? Aren’t “SF” and “Victorian novel” categories from two different classification schemes? One is based on what happens in the book while the other is about when it was written.
The word “genre” _does_ get used for many different sorts of classifications. (Look at how Moretti uses it.)
I’ve spent a fair amount of foul breath contesting the idea that “genre” is always “based on what happens in the book”. (Are “literary ballads” a different genre from “traditional ballads” or not?) But even on those terms, I think what confuses contemporary youth about “Victorian novels” aren’t their obscure references to forgotten events, but their seemingly roundabout way of telling a story. That’s happening in the book, right?
I’m still confused as to what you’re after in this distinction, so maybe it would be best if I dropped the subject. I don’t seem to be adding much at this point.
The word “genre” _does_ get used for many different sorts of classifications.
And that’s most of what I’ve been after, I think. And I think that’s much of why the term is such a troublesome one. It IS being used for (possibly irreducibly) different types of classification and we’re looking for the ONE RIGHT type of classification. Maybe there isn’t one right scheme. Maybe there are several good schemes. Maybe schemes need to be taylored to the body of texts.
Beyond that, we also have the notion of mode, which is also used in classifying literary works. I know Frye talked of mode, but don’t know whether he introduced it or inherited it. (I just glanced at the opening of the mode chapter and that suggests the Fry has introduced the term, though he grounds his introduction in Aristotle). I supposed the notion is roughly that genre and mode are orthogonal to one another so that one could set up a matrix like this:
_________ mode1 mode2 mode3 mode4 mode 5 . . . . . .
. . . . . .
We could then have different kinds of works in each cell in the matrix. Some cells might, in fact, be empty while others would have only a few examples and still others would have lots of examples.
But if the space of literary kinds is, say, a 10 dimensional space, and neither mode nor genre capture one of those dimensions very well, then it’s still all rather confused. What would we get if we used statistical techniques to classify a whole body of texts? This is the sort of thing you could explore in a nora-type environment.
There are at least two families of issues: 1) Historically, how have literary works been classified? In particular, how has the notion of genre been used? How has the classification system affected the production and circulation of literary texts. 2) What are the REAL kinds?
That second question has a whole mess of presuppositions built-in to it. And one or more of those presuppositions can be located in the fact that I separate it from the first set of issues. Are there any REAL kinds at all, or are all classifications more or less arbitrary constructs? Clearly literary works differ from one another in many ways and it is important for us to understand those differences.
At the moment I’m not deeply concerned about identifying the REAL dimensions of difference or even clarifying just what that question might me. For one thing, I don’t think we have good enough analystic descriptions of literary works. Beyond that it’s all I can do to think about things in these terms. I think someone looking at the problem historically should be alive to the idea that one notion of classification, genre, is being used in various ways, to recognize various kinds of textual difference. And it would be useful to have general discussions of the problems of classifying things.
Here’s an exceedingly specialized genre, haiku devoted to spam (the canned meat substance):