Thursday, January 01, 2004

(Reading roundup: Devitt, Bazerman, Yates & Orlikowski on Genre Assemblages)

Originally posted: Thu, 01 Jan 2004 11:35:45

So I've been reading a variety of genre-related articles and chapters lately, gearing up for the big genre discussion in the book I'm working on. If you know my previous work, you know that I'm really very interested in how genres are brought to bear on activities and how they jointly mediate them; I've been using the term "genre ecologies" to discuss these complexes or clouds of genres. Well, others have used terms to describe assemblages of genres as well. Until now I've sort of vaguely nodded to this other literature, but I think that genre ecologies do work that they don't. So finally I'm going to draw some distinctions.

To do that, I've been reading about genre sets, genre systems, genre repertoires, and (of course) genre ecologies. On first glance, they all seem to be interchangeable. But poking around in some of the canonical texts of each, I see some fairly important differences.

Amy Devitt's 1991 essay "Intertextuality in tax accounting: Generic, referential, and functional" is the canonical text on genre sets (at least as far as I know -- it sure is referenced often). In her examination of how accountants get things done, she posits that texts form networks of interaction for the accountants. Each text connects to the previous text in a chain of actions. "In examining the genre set of the community, we are examining the community's situations, its recurring activities and relationships." And she adds, "This genre set not only reflects the profession's situations; it may also help to define and stabilize those situations" (p.340). Her focus, in fact, tends to be on that sequential and stabilizing work (p.341), and that leads her to examine the official (disciplinarily developed, stabilized, and regulated) texts that do the most to perform this work. Although she counsels us to "examine the role of all texts and their interactions in a community" (p.354, my emphasis), she only examines the "products" of the work -- memos, correspondence, tax provision reviews -- not unofficial genres such as transitory annotations, notes, aides memoire, etc. These official genres bound and enable professions, in Devitt's account, but they appear to serve serial and strictly communicative functions, not self-mediational ones. Remember when you were doing math in high school and the teacher told you not just to write your answer, but to "show your work"? Genre sets don't show the work; they don't expose the unofficial genres that play such a large part in distributing cognition. They're scriptures, not scribbles. Objects, not tools. They are asymmetrical in Latour's sense.

They are also, as Charles Bazerman points out, focused on individual perspectives. In "Systems of genre and the enactment of social intentions," he seeks to extend the notion of genre sets by talking about genre systems. "These are interrelated genres that interact with each other in specific settings. Only a limited range of genres may appropriately follow upon one another in particular settings, because the success conditions of the actions of each require various states of affairs to exist" (pp.97-98). Like genre sets, then, genre systems are made up of sequences of genres that hand the baton of communication onward. Each genre is required in order for the next one to be produced and used. That, of course, once again implies official genres and focuses away from the informal, unofficial assemblages of genres that we often bring to bear on our work. Unlike genre sets, genre systems involve "the full set of genres that instantiate the participation of all the parties" (p.99) -- but that "full set" appears not to be so full when we consider that unofficial genres are squeezed out.

JoAnne Yates and Wanda Orlikowski develop this notion of genre systems in "Genre systems: Structuring interaction through communicative norms" (2002). They still share Bazerman's view of genre systems as sequences of interrelated communicative actions that structure collaborative work (structure is particularly important to them), and they see these genres as being "linked or networked together" to constitute "a more coordinated communicative process" (p.14). Indeed, genre systems do not just support a social activity, they comprise it.

But this sequential understanding is not enough, for clearly people bring assemblages of genres to bear simultaneously on a problem. In their 1994 article "Genre repertoire: Structuring the communicative practices in organizations" (yes, there's that word "structure" again), Orlikowski and Yates mention that genres do not just sequence, they overlap. Through these two sorts of coordination, genres work together to produce a more communicative practice. Members of a community "tend to use multiple, different, and interacting genres over time. Thus to understand a community's communicative practices, we must examine the set of genres that are routinely enacted by members of the community" (p.542), and this set of genres is what Orlikowski and Yates term a genre repertoire. The authors recognize that this repertoire changes over time as new genres are improvised or otherwise introduced, and they suggest that explicating these changes over time can help us to understand changes in the community's communicative practices and organizing processes.

Orlikowski and Yates edge away from the rigidly sequential understandings of genre sets and genre systems here (although you notice that they return to that conception in 2002). The notion of genre repertoire is developmental and accounts for overlapping as well as sequential communicative actions. But at the same time, genre repertoires emphasize individual and group communicative performances: you perform a genre, but it doesn't perform you. That is, they still reflect an asymmetrical understanding of genre that exclusively deals with communication rather than mediation or (more broadly) distributed cognition. And because of the firm emphasis in communication, particularly communication in repeated enactments across a group, genre repertoires still emphasize the official rather than unofficial genres.

Looking at these frameworks has helped me to clarify some of the things I've been trying to do with genre ecologies. I find Orlikowski and Yates' distinction between overlapping and sequencing to be useful, though oversimplified. My genre ecology work has tended to emphasize the overlapping aspect. It has also emphasized mediation rather than communication -- and in doing so, has explored unofficial and idiosyncratic genre use more thoroughly than has been typically done in these other frameworks. At this point I won't go into the articles on genre ecologies that Mark Zachry and I have written, nor the related concept under the same name that Freedman and Smart wrote about in 1997. But in their emphasis on compound mediation, genre ecologies deal with a quite different phenomenon than do genre sets, systems, and repertoires: they deal with activity, whether individual or group. Any person or group can make connections and build assemblages (or ecologies) of heterogeneous genres to mediate their own or others' activities. Notice how communication drops out here, at least as a separate category, because mediation includes influence over one's community as well as oneself. This is an interesting but terribly underdeveloped point, and one that I'll have to think about for a while.

Blogged with Flock

Sunday, December 28, 2003

Reading:: A Thousand Plateaus: Capitalism and Schizophrenia

Originally posted: Sun, 28 Dec 2003 05:16:43

A Thousand Plateaus: Capitalism and Schizophrenia

by Gilles Deleuze, Felix Guattari

(Note: A later review of this book is available.)

A Thousand Plateaus is a deeply influential book which, though first published in French in 1980, is being increasingly cited in rhetoric, technical communication, science studies, educational psychology, and the like. In in, Deleuze and Guattari introduce themes such as rhizomes, deterritorialization, bodies without organs (BwO), and the like. Through their highly metaphorical language, imaginatively written explanations, and their attempt to write the book as a rhizome -- even inviting readers to begin reading at any point rather than at the beginning -- Deleuze and Guattari perform as well as describe an approach to language that relies on heterogeneous, multiply linked assemblages.

I found the book to be excruciating and abandoned it on p.178.

Is it worthwhile to review a book of which I have read less than half? Perhaps so, especially because I strongly suspect that most people who cite this thing haven't read it all the way through either. When I see references to Deleuze & Guattari, they almost always focus on the notion of rhizome discussed in the first (and most readable) chapter. In any case, I'll take the authors at their word when they say in Chapter 1 that the book is written as a rhizome, an organic structure (or rather anti-structure) that is so interlinked that any part is as the whole.

This notion of rhizome, as I hint above, is the best traveled and most used concept from the book, at least in the literature I've read. "Any point of a rhizome can be connected to anything other, and must be," the authors tell us (p.5). Rhizomes may rupture in spots, temporarily, but everything is connected so these disruptions do not destroy the rhizome. The rhizome is an anti-genealogy: it does not rely on historical development, it links through and across states of being and ontologies. One can see how writers such as Bruno Latour and John Law, who have been applying the notion of heterogeneous networks to science and technology studies, would be intrigued by the concept. Frankly, I think they attempt to apply the concept much more concretely than Deleuze and Guattari do -- to actual empirical cases, not just to endlessly circulating semiotic systems -- and in doing so they expose some of the problems with the concept.

What problems? Here's Latour, from his chapter "Social theory and the study of computerized work sites":

Rhyzomes and heterogeneous networks are thus powerful ways of avoiding essences, arbitrary dichotomies, and to fight structures. But ... their limit is to define entities only through association. ... they become empty when asked to provide policy, pass judgement or explain stable features. ... Their dissolving power is so great that after having dissolved the illusions of critical postures, there is not much that is left and they even may turn into a somewhat perverse enjoyment of the diversity, perversity, heterogeneity and multiplicity of the unexpected associations they deploy so well. (1995, p.304).

The genius of rhizomes -- the point that we can put Cartesian distinctions behind us and examine the multiple and multiplying interconnections among people, practices, artifacts, ontologies, etc. -- is also its Achilles' heel. What do you do with them after you cry for the umpteenth time, "aha, there's no Cartesian divide!" You descend into the joyless, wearily self-indulgent spiral of deconstruction that Latour so rightly condemns as "perverse" here and in stronger terms in Pandora's Hope. Stability is lost -- or, rather, completely rejected -- as is any interest in or explanation of sociocultural development. The latter point is particularly corrosive to empirical examinations of work, and I think has fought against some of the more interesting work in science and technology studies.

Of course, Deleuze and Guattari didn't come up with the notion of rhizome with those studies in mind, and it's actually not very clear to me exactly what they want to apply rhizomes to. The examples in subsequent chapters, though, tend to be semiotic systems, endless regimes of signs forming infinite, infinitely circular networks (see especially Chapter 5). It seems to me that what the authors have given with one hand -- the notion of a rhizome composed of heterogeneous nodes across categories -- they take away with the other by immediately collapsing these heterogeneous nodes unto dreary, labyrinthine regimes of signs. We're back to the prison-house of language from which Latour, Law, Callon, and others worked so hard to release us. Behind our backs, Deleuze and Guattari restored the asymmetrical relationship between humans and nonhumans by insisting that the whole is interpreted. So we spiral back into an endless post-Freudian discussion of how pores resemble vaginas, a colorless set of instructions that a masochist might leave his dominatrix, and so forth.

That seemingly endless discussion involves a lot of neologisms and a lot of highly metaphoric language. Even some of the words seem endless, especially when you say them out loud. "Deterritorialization." "Bodies-without-organs." Frustratingly, it's very difficult to figure out these metaphors in spots, making the book read like a mystical text rather than an imaginatively constructed one. When Deleuze and Guattari say that language is not a map but a tracing, do they mean the same by trace that I do -- a sociocultural line of development? Who knows? As with any mystical, metaphoric text, A Thousand Plateaus requires the acolyte to read without immediate comprehension, to soak in the wisdom through multiple readings, to work out an endless regime of signs in his or her own life. Tons of words, teaspoons of insight.

I'm just not up for that, so I just stopped reading. Yes, I'll likely cite the first chapter as apparently everyone else does. But I can't imagine reading the rest of this excruciating book of my own free will.

Blogged with Flock