Seen through a sociologist’s eyes, the literary system can look very strange indeed.
I thought I understood this principle, having quaffed vigorously of the heady Bourdieuean mead and embraced Bourdieu’s ways of making questions of literary value, literary reading, and literary belief historically and socially relative. But I have been spending months coming to terms with a new version of the lesson, having been reading and thinking over the British sociologist John B. Thompson’s recent book Merchants of Culture: The Publishing Business in the Twenty-First Century. I’m going to discuss it in this post and a subsequent one, and my lit-sociological comrade Lee Konstantinou will address Thompson too, continuing the dialogue from our earlier discussion of Wendy Griswold’s work on the reading class. In this post, I will give a quick summary of this remarkable and important book and then state what I see as its principal challenge to literary study.
Thompson’s book is an analysis of contemporary trade publishing, both fiction- and non-fiction, in the US and the UK, on the basis of interviews with the people of the world of publishing (writers, editors, publishers, corporate executives, retailers) and statistics about the publishing industry. I approached the book with the idea that Thompson would provide a guide to the social context of recent writing: his project is to explain the “logic of the field” of trade publishing, and I assumed that this would involve explaining the way institutions support different versions of literary ideas, different kinds of literary value, and so on. I imagined that Thompson would enrich my sense of how production, mediation, circulation shape what kinds of books get made and produced in publishing.
And in one sense this is indeed what Thompson does. Here is a rough summary of his findings. Contemporary trade publishing is shaped by three structural transformations that have taken place since roughly the 1970s: the growth of retail chains (bookstores—also supermarkets) as the dominant way books are sold; the power of literary agents to command large advances for big authors and serve as indispensible mediators between authors and publishing houses; and the consolidation of the medium- and large-sized publishing houses under a few corporate umbrellas. These transformations have produced what Thompson calls “the polarization of the field” (146): there are now a few very big, market-dominating trade publishing corporations, and there are many, very many, small publishing operations, but there are almost no medium-sized trade publishers.
Both large and small publishers are, in the polarized field, ever more preoccupied with the need to come up with the “big book,” the bestseller, the title that exceeds expectations; and in order to find the big book, publishing becomes ever more devoted to the signs of potential bigness: “track” (the track record of an author’s sales); “platform” (the built-in potential audience of web-followers, affiliated organizations, etc.); and—in one of Thompson’s neatest phrases—a “web of collective belief” (74) that transforms an agent’s “hype” into the self-reinforcing “buzz” of publishers’ commitment to the possibility of a book’s big sales. The last part of Thompson’s book explores the consequences of this logic of publishing, in terms whose judicious neutrality does not obscure the signs of the industry’s crisis: an ever-shrinking “window” of visibility for new books, in which they either take off as bestsellers or fall into neglect; an ever-stronger emphasis on frontlist over backlist; and diminishing “diversity in the marketplace” of books sold to readers in spite of ever-proliferating “diversity of output” (and quantity of output) from the field of publishers (389).
Thompson cuts through some prominent clichés of contemporary book-talk. For example, he gives a powerful argument that the Web era’s impact on trade publishing is not the instant revolution it is sometimes thought to be. Amazon, though important and growing, is still small in market share compared to both the chain sellers and the mass-merchandisers; the 800-pound monster of book retail is not Amazon but Costco and Walmart (or, in the UK, Sainsbury’s and Tesco). E-books are growing fast but are still minor on the balance sheets of trade publishers. The true digital revolution in publishing, Thompson argues, has already happened: is it digitization of the process of publishing, not the product (321).
This process/product distinction leads me to what is most startling about Thompson’s book for a literary scholar. The truth is that we are much more used to making arguments about the product than the process. And Thompson’s attention to publishing turns out not to require much specificity about the products—that is, the texts of books, or even the way readers read books. Merchants of Culture is about the producers and distributors of the sorts of books literary scholars study—or the latter-day descendants of the sorts of books those of us who work in earlier periods study—but it says practically nothing whatever about either writing or reading. Instead, as Thompson himself says in his conclusion:
For the vast majority of writers or aspiring writers, this system seems like an alien beast that behaves in unpredictable and erratic ways, sometimes reaching out to them with a warm smile and a handful of cash, invting them to join the party and holding out the prospect of riches and fame, and then suddenly, without much warning or explanation, pulling back, refusing to respond or perhaps cutting off communication completely. (374)
As for reading, it proves to be the great black box of publishing. Thompson repeatedly emphasizes that one of the distinguishing traits of publishing as a business is the unpredictability of the success of any given book. Selling books is not like selling widgets, because readers’ taste remains, in the view even of the sales executives and marketing managers of big publishing corporations, very hard to predict. Only top author “brands” and certain very well-defined genres have any predictability of reader response, but these domains of lesser uncertainty are not enough to sustain the big publishers’ business.
But for me another, absolutely basic fact was even more revelatory: Publishers’ customers are not readers. The primary customers for trade publishing houses are, rather, the buyers for retail booksellers. Because bookstores can return unsold stock to publishers, by an arrangement that is eighty years old in the US and, I think, a about a century old in the UK, the individual reader is in a sense the final customer. But the audience the publishers must most urgently convince to buy consists of the retailers. And, as Thompson shows, because contemporary bookselling has become a highly consolidated business, dominated by mass-merchandisers and big chain bookstores, the “audience” constituted by those large retailers heavily influences how the publishers produce: they need to produce the kinds of books those sellers can reliably and profitably sell.
Thus the field of trade publishing sees everything through mediators’ eyes. This is as true of writing as of reading: writers lie outside the logic of the field, but agents lie within it. I can make my point with a diagram, which is a drastic and tendentious oversimplification of Thompson’s intricate charts of the “supply chain” and “value chain” of publishing (15-16):
Must literary studies confine itself to the margins of the publishing field?
I thought that, following the example of book historians, I would know how to make sense of an analysis like this literarily: I would be able to see how these contours of publishing processes shaped what D.F. McKenzie called “the sociology of texts.” But to encounter mediators at both ends of the process of literary circulation is to feel, as I have tried to show in my diagram, that literary scholarship as I know it hovers only on the very edges of the circle.
Thompson shows, compellingly, that trade publishing’s self-reinforcing “logic of the field” (Thompson, too, draws on Bourdieu) joins agents, publishers, and retailers in relations of cooperation and competition—but those relations include writers and readers only in a remote way. Thus it is no criticism of Thompson’s sociology to say that his account of trade publishing barely features the primary producers—writers—and the ultimate receivers—readers. Indeed, Thompson himself, in his methodological appendix, admits to interviewing more writers than his study really needed, because they were fun to interview; but still:
They [writers] offered a very different and crucially important perspective on a field to which writers both belong and don’t belong, like some distant cousin who is tolerated but not really welcome at the family gathering. A proper study of the worlds of writers would be a wonderful project in its own right, but this was not the project on which I was embarked here. (411)
Thompson is not likely to have assumed that readers are passive consumers, either; he is the author of The Media and Modernity, a powerful theoretical statement about the importance of the way audiences use and “appropriate” media forms. But within the logic of the field they are marginal, the source of varying sales figures and vacillations of taste, the objects of the folk wisdom of agents and editors and booksellers—but that is all.
And so I find myself, in thinking over Thompson’s arguments, oddly but emphatically at sea about how to appropriate his work for literary scholarship. I have some scattered ideas for a next post, but I hope this post (and Lee’s, to come, and, we both hope, the discussion in comments) will spark ideas for more possibilities. The questions are: What can we say about the books that are produced, and the ways those books are appropriated, when we know what Thompson and other analysts have to tell us about the logic of publishing? Or, better yet: how would literary scholarship have to change in order to analyze intelligibly the relation between the field of publishing and the literary field? What are the questions to which the answers would include both arguments and data from Thompson’s repertoire and arguments and data from the literary-historical repertoire?
That’s enough for Round 1. But I have much more to say about Thompson, both about specific findings that I think do have immediate significance for literary study, and about the limitations of what he has to say. So, some teasers for the sequel…
- The narrowness of the heroic age (and the heroic ideal) of publishing
- The Wild West is east of Boston
- The mystery of “Quality”
- accredited visibility: what Oprah has that you, “literary critic,” don’t.
Thanks for the rundown, Andrew. This book sounds fascinating and not a little terrifying. I'm reminded of Hershel Parker's Kraken Edition of Pierre (RIP, Maurice Sendak!), which attempts to excise what Parker sees as the corrupting influence of Melville's interactions with his publishers. Maybe that was the part that should have been expanded.
Ah yes, textual change and communal production as "corruption" of the purity of an authorial intent. It reflects a sort of Will to Urtext in the credentialed literary reader, especially in the true devotees of canonical authors. But our whole book culture tends to submerge the collective labor and institutional mechanisms that produce a book, leaving us with author and text (and context, maybe). I guess I want alternatives to a scandalized discourse of corporate corruption or commercialization (as though publishing had ever been anything other than an enterprise whose interests included money) or interference with authorial autonomy (as though the work of circulating books had ever been anything other than communal). Thompson provides a strong antidote to those discourses but leaves it mysterious how to connect up his analysis with one that would systematically open the covers of books, or ask how readers open the covers of books. More in Round 2.
Thompson repeatedly emphasizes that one of the distinguishing traits of publishing as a business is the unpredictability of the success of any given book.
Arthur De Vany (Hollywood Economics) has made the same argument, in some mathematical detail, about the movie business. The number of titles produced in a year is obviously much smaller than in book publishing and the production costs of a single title are considerably higher, but unpredictability rules. There's always that odd little film by unknowns the finds an audience and earns many times its production and marketing costs and even the biggest film, with the biggest names, and the biggest PR campaign, can die in a weekend.
Though I have no source to cite, I'm pretty sure that's true of music as well, and the plastic arts. All of these cultural goods are highly unpredictable. The individual works vary enormously from one to the other and the details matter.
De Vany mentions in passing that the pharmaceutical industry is like that as well. The number of substances that could be of medicinal value is huge and, short of clinical testing, it's very difficult to find out which ones will actually work. Clinical testing, alas, takes time and is very expensive. Big Pharma was hoping (starting two decades ago?) that computer technology would allow them to discover drugs more quickly and cheaply. And, while high-tech has become deeply enmeshed in the business, the results have not been forthcoming.
I'd say that the unpredictability of publishing is the key factor. The only reliable way to find out whether or not a title has an audience is to put it before an audience. By the time you've done that, your time and money are spent. If it fails, you can't get it back. And there's no quick and cheap way to put another title out there, and another, and still another.
You might think that there's a way to research the readership, find out what it wants, and tailor titles to suit. Well, it turns out that the best way to conduct this research is simply to put books out in the market place. There's really no cheaper way to get "information" out of the readership about what it wants.
This is the sort of thing that looks like evolutionary dynamics. The authors over there to the left in your diagram, Andrew, are cranking out random variations. The readers at the right are the environment in which those variations must survive. The purpose of the publishing industry in the middle is to pick some of the proposed variations (based on guesses about the readerly environment) and develop or "amplify" them to the point where they can be set before a readership. The industry is using its resources to gamble on what readers want.
Some years ago an undergraduate teacher of mine, Arthur Stinchombe, published a book, Information and Organizations (UCalPress 1990). Uses examples from different industries, he argues that organinzations take the form best suited to resolving the major uncertainties facing their business. Given that perspective, one could approach publishing with the question: Given the unpredictability of given titles, how do you organize a business to publish books? Well, as Thompson has provided an analysis of the business, the question becomes what that analysis reveals about how the business copes with the deep uncertainty of the product.
Q. Andrew, does Thompson give any numbers on the percentage of books that break even or turn a profit vs. those that lose money? I'd guess that less than a quarter break even or better, but I don't know.
Thank you Andrew; I'm putting this on my summer reading list. I just finished teaching a Bourdieu-inspired capstone class in which I made my students research the field of recent fiction and then create a book list for the rest of the semester. I sent groups of students to interview local retailers to get their sense of which books were selling. One of the interesting results, which seems like it relates to the polarization of the field, was that while independent stores were more likely to stock some personal favorites that the owners and staff would then try to sell to regulars, they also all relied heavily, of course, on the books that sold themselves (Stieg Larsson, paranormal romance, etc.). I find the retail and publishing side of things easier to fold into scholarship (I'm in the process of working out some ideas about how pulp editors created contemporary genre fiction by attending both to their own evolving visions of the genre and the raw numbers of circulation and fanmail). The thing that feels truly obscure and crucial is the realm of the agent. I know anecdotal stories about how power agents can make and break careers but have no idea to get at this realm other than counting names of agents that get thanked in blockbuster books.
In reference to the question of corruption, I can't help but think of Gordon Lish's role in creating Raymond Carver's style (and through him the styles of innumerable MFAs past, present, and future). I haven't yet heard anyone call that one desecration or corruption...
Sounds like a fantastic course, Leif. I love the project of sending the students out on fieldwork in bookstores. One of the differences this highlights is what Thompson describes as the intense commodification of the most visible store space in the retail chains and mass-merchandisers: they actually get a big "co-op" budget for advertising from publishers that independents can't. It totally is tied to field polarization: Thompson shows how over time the bestsellers sell ever more copies but "the number of titles that sell in modest but acceptable quantities is declining" (391).
Your awesome pulp project is a good example of the kind of publishing circuit / literary history analysis I can grasp. Note how it involves editors and periodical circulation. Somehow the scale is more tractable. In the next post I want to talk, too, about how the early 20th century emerges in Thompson as a kind of "heroic" age of publishing in which a certain kind of house with a personality closely tied to a single charismatic owner emerges (and then gets swallowed up in the consolidations of the 1960s). There too I find it much easier to think about contemplating a Knopf list or a Liveright list and having things to say about the kinds of books and literary strategies those publishers are part of than I do when I look at, say, the dozens of imprints owned by Penguin Putnam.
You're spot on about the mystery of agents. Thompson interviews Andrew Wylie--more on that hi-lar-ious section in post #2. They become central to the operation of the field after 1970 but their gatekeeping functions are hard to suss out.
More tomorrow here. Thanks for the comment!!
Looking forward to your post, Lee.
This course sounds amazing, Leif. What did students take away from it? (I don't suppose you could get any of them to weigh in here?)
Andrew, I'm with you on the contemplatability of a Liveright or a Knopf. In some ways this notion—or this literary moment—restores the primacy of an author, or let's say, meta-author (that is, the publisher), who controls the list. To privilege the early twentieth century for the ways in which it enables such a reading (whether or not the charismatic publisher model is quite appropriate) seems to be a bit of a retreat. For the mind-boggling publishing conglomerates we have now, we have to compass a notion of corporate meta-authorship (the way that Pixar authors an animated feature). At that point, I suppose, narratives fail.
Tocqueville writes that
One feels rather called out in advance by Tocqueville: historical conditions themselves shape what seems trackable, what appears to be a cause, who gets to be a meta-author and (therefore) who gets to be an author.
And meanwhile…Natalia, "corporate meta-authorship" is a great name for the problem. It's not authorship, it's the organization of many parallel tracks of authorship-publication-marketing. It won't have the qualities of intention or style that an individual author or a house supposedly dominated by a single charismatic publisher has. But it must have some qualities.
The Tocqueville passage sets up a binary pair that just cries out to be overcome. The whole point of a "field theory" like Thompson's or Bourdieu's is to try to work out how agents and structures interact and remake one another. Thompson is definitely producing "midlevel concepts" (maybe upper-midlevel) or little-t theory, which I embrace but which I find harder to use, literarily, than Grand Theory.
Meanwhile, Lee's new post on Thompson addresses what we're discussing with some terrific ideas, so let's continue the conversation in comments there!
Thanks for starting this series, Andrew. I hope to post my response to this soon, ending my long blogging drought.
"...diminishing “diversity in the marketplace” of books sold to readers in spite of ever-proliferating “diversity of output” (and quantity of output) from the field of publishers (389)."
Retailers aren't presenting that diversity to their customers?
I spent some time, Andrew, thinking about this post, trying to clarify something that seems implicit in your argument, but not quite explicit. I ended up posing a question:
That question (perhaps perversely) assumes that the logic of book publishing is more or less transparent to the forces governing the circulation of literary texts in a society. Whatever those forces are, book publishing does not resist them. Rather, it channels them and extracts a fee from them for that service, but it does not modify or interfere with those forces.
On THAT assumption, one need not know anything about publishing to understand the circulation of literary texts. Or, if you will, knowing about publishing is not going to tell you anything very interesting about the circulation of texts.
In a similar way, printing is transparent to book publishing. Paper doesn’t care what’s impressed on it, printing plates don’t care what’s engraved on them, and printers don’t care what they print as long as the bill is paid. That’s what I mean by transparency. Knowing something about inks, papers and printing presses is not going to tell us anything interesting about telephone books and experimental fiction.
Your questions, though, seem to proceed from the opposite assumption, that the publishing industry is NOT transparent to the forces of the literary field. It modifies and resists. Hence, knowing about publishing will tell us things about literary culture that are otherwise mysterious. And perhaps you are correct.
The question of whether or not publishing IS transparent to literary forces strikes me as a rather tricky one. I’ve blogged a bit about it, talking of censorship on the one hand, and the economics of print runs on the other.
It seems to me that the real mystery, though, has to do with the dynamics of demand for literary texts. How is it, for example, that the publication of a single text, in a small print run, Harry Potter and the Philosophers Stone, gave rise to a new demand in millions of readers? Publishers surely don’t know, but I’m not sure that literary critics do either. Just what ARE these literary forces?