This post started off as a comment on Elena Pierazzo’s note drawing attention to a very useful blog entry by Lisa Spiro on collaborative projects in the humanities. But it’s grown a little large for that, and my purpose here isn’t really to respond to that entry — which, as I say, is tremendously useful and thorough — but more to comment on collaboration in the humanities in general. Or rather, how it tends to get talked about — on the rhetoric of collaboration in the humanities.
Because I’m struck by how readily it’s assumed that collaboration in the humanities is an unreservedly Good Thing. And by how seldom it is that, despite this, such collaboration actually occurs.
The numbers are pretty clear. Going with the figures Spiro provides, 2% of the papers published in the journal American Literary History are co-authored, and 4% of those in Mind; and while one might argue that two (albeit very respected journals) hardly constitute an adequate sample, I’d imagine most humanities scholars would say that these figures chime with their own sense of the state of co-authorship in their own fields.
It just doesn’t happen — with the exception, that is, of the digital humanities. According to Spiro, fully 64% of the posters and papers at DH 2008 were co-authored. Not only that, but 35% of DH 2008 posters and papers even contained the word “collaboration” itself.
How do we explain this difference?
One mode of explanation — and one understandably very popular with digital humanists— would be the teleological, and it runs a little like this:
Large scale multiple-author collaboration on the model of the lab-based sciences is the Way of the Future. In the dark, benighted past humanities scholars toiled alone in their tiny cells/the ivory tower, unable to communicate with each other and thus endlessly falling prey to their own subjective delusions and/or pointlessly duplicating the labours of others. Now, however, information technology allows them the to exchange ideas and information easily, freeing them from the shackles of their own limited perspectives and datasets. Soon — very soon — such scholars will at last realise how trapped they are by their own working practices, throw off their Dark Age informational shackles, and come together to aggregate, integrate, and crowdsource their way to shared knowledge structures on which all can agree and in which all participate. Digital humanists, as the chief architects and user-base of the technological infrastructures that allow such collaboration to occur, are riding the leading edge of this technological shockwave; the rest of the discipline’s practitioners will inevitably follow.
As the above summary might indicate, I am no great fan of this mode of explanation, which does little justice to the way in which humanities scholars do in fact tend to work together — a process described by John Unsworth (cited also by Spiro) as “cooperation” rather than “collaboration”. In Unsworth’s words
In the cooperative model, the individual produces scholarship that refers to and draws on the work of other individuals. In the collaborative model, one works in conjunction with others, jointly producing scholarship that cannot be attributed to a single author.
You can read the rest of his arguments here (and Spiro provides a pretty good overview and commentary), but essentially to Unsworth and the other teleologists, “cooperation” is just a treacle-slow, embryonic, and ultimately inadequate form of “collaboration”: speed up the whole sluggardly process of “referring to” and “drawing upon” the work of others and you’ll get “collaboration” proper, while realising a bunch of other benefits besides.
This is, to my mind, a gross oversimplification of the matter. It’s true that humanities scholars at times — lexicographers, I’m looking at you, here — create problems of information exchange for themselves that could easily be fixed by a technological/collaborative shot in the arm. But it’s more often the case that the divide between “cooperation” and “collaboration” reflects fundamentally differing conceptions of and approaches to information — and very often because the information under review is itself different in character.
Let’s think, for a moment, about what needs to be true about an information set for collaboration to be an inherently beneficial approach in understanding it. A number of responses are possible here, but I’d say at a minimum the information set has to be modular, in at least three senses.
- it is made up of smaller constituent parts, whether very high in number (a dictionary is made up of lots of individual entries for words) or relatively low (a single codex may be made up of several smaller texts).
- these parts must be separable from each other. The separation may be clean (“persons” form a clear-cut unit of resolution for prosopographies) or otherwise (translators often work in teams of two, with one person assuming chief responsibility for the original and the other for the target language). To a large extent the cleanness of the separation will determine the character of the collaboration.
- these individual data must to some extent be self-assembling: the way that they cohere together to form the information set must be clear to all.
This last point starts verging into philosophical territory: on a strong positivist reading, the features that characterise the data and the ways that they inter-relate must be objectively definable; more weakly, such definitions must be a matter of common consensus. However, this is not a bone I wish to pick here, and I don’t think it actually has much bearing on the matter at hand. It is possible to use ontologies without believing in them, and there generally remains enough consensus within disciplines for a large number of humanities project to fit the mould above: prosopographies, dictionaries, large-scale translation projects, corpus linguistics work — the data for all of these fit the criteria outlined above reasonably neatly, and there is a penumbra of other enterprises that are close enough to benefit from some aspects of electronic collaboration thus conceived.
What’s missing from all of the above, however, is a factor absolutely crucial to a large number of humanities endeavours — and that’s the unifying sensibility. For an informational enterprise to meet the criteria for a good collaborative project, it simply cannot depend critically upon a single coherent intelligence that unifies its data set and organises it along non-obvious and non-trivial lines.
And this is a problem for the humanities. Because I would say that the vast bulk of humanities work — including, for the most part, the stuff that is of greatest interest to a non-academic audience, that has, in the bureaucratic word-of-the-day, the greatest “impact” — depends to a large degree on this kind of unifying sensibility.
Anything that consists chiefly of interpretation, whether of a literary work, a historical document, or a chain of events; anything that attempts to give shape to an information set by imparting to it a narrative structure; and, most crucially, anything that aims not just to elucidate our current understanding of an information set, but to create a new perspective upon it, or to claim that the boundaries of the set have been badly defined: all of these absolutely require some unifying sensibility as a precondition for their existence.
This sensibility may, of course, be shared by more than one individual; it may come about as the result of collaboration within a team; and digital technology may work to facilitate the formation of such a shared sensibility. But there is no compelling reason for any of these conditions to hold. And, most of all, there is no reason to suppose that what is required for this unified sensibility to arise is speed.
Indeed, I would say that it is the very difficulty and delicacy of creating such a shared sensibility that is the chief reason that humanities scholars (outside the digital humanities) have tended to cling to such an allegedly outmoded cooperative workflow as “referring to” and “drawing upon” the work of others. For to fully understand and appreciate the perspective of another scholar upon some information set, it is not enough (provided that this perspective is indeed valuable, i.e., non-obvious, non-trivial, and obtaining at some kind of scale) simply to present the information set to the reader and await its self-organisation in his or her head.
Rather, the reader must engage with the scholar — by which I mean, follow the scholar’s arguments; learn to perceive where they are going, and why; follow their footnotes through, to discern what information they have included, and what left out; familiarise him- or herself with the information set, its characteristics as described by other writers, and by the scholar; come to sense the presuppositions left unstated in the scholar’s work, and how these interact with those that are explicit; and so on and so forth, iteratively, until the reader has come to appreciate all he or she feels it is possible to gain from the work being studied. And there is no reason to believe, in the case of truly significant works and scholars, to believe that this process will be a quick one. Or even that the assessment of when one is “done” with a particular work or author will remain stable.
What the reader gains from this, however, is significant. First, there is of course a knowledge of whatever factual detail is contained in the actual matter of the work being studied. Second, there is the question of sensibility: of coming to understand something at a depth, or from a perspective, not understood before, but now acquired.
And third, and perhaps most pragmatically, there is an epistemological or, one might say, a research, benefit in coming to understand another scholar’s perspective. And that is, to the extent that one understands the perspective of another scholar, that scholar’s intellectual acts become legible and to some extent predictable. This post is already long enough, so I’m not going to go into detail here; but suffice it to say that a tendency to ignore this to my mind constitutes a critical failure of insight among the “teleologists”, who very often portray navigation amongst a large set of facts and choices as somehow preferable and easier than engaging with a single intelligence that mediates them.
In one sense, in fact, it might be said that traditional modes of scholarly “cooperation” resemble the process by which friends and acquaintances are made, albeit solely on an intellectual plane. One “gets to know” the experts in one’s particular field; one comes to understand their perspective, their “take” on controversial areas; knowing this, one also appreciates their quirks and foibles, their strengths and shortcomings; and one furthermore gains some limited ability to adopt this perspective — to assess and appreciate new facts and events through their gaze, and to evaluate how useful such a perspective is within this new context.
Or, to continue the analogy — the difference between “cooperation” as practiced by traditional humanities scholars, and “collaboration” as recommended by Unsworth (and, for that matter, various funding bodies) is not entirely unlike the difference between old-fashioned face-to-face interaction, and e-mediated social transactions like Facebook Friendship and Twitter-following. There’s nothing inherently good or bad about either of them. But they have different strengths, work at different speeds, and serve different requirements. In the final analysis, they do very different things.
In a way, I feel this should be obvious. But it’s something that we, as humanists, seem to have forgotten — still basking, apparently, in the afterglow of the invention of the web, two decades ago now.
Malcolm Gladwell, not normally a Luddite, recently penned a piece on the difference between physical and electronic activism — and observed that, while social networking is good for broad and shallow activist work, sustained or radical change requires commitment of a depth that social networking and Facebook friendship don’t normally support.
The analogy with humanities work is broad, but not entirely a bad one. Electronic tools in the humanities can make what ought to be easy, easy. Fact-gathering, sharing documents, identifying and collocating resources — all of this you want to be as straightforward and simple as possible. But that doesn’t mean your job is done — any more than “Liking” a political party is the same as joining that party, or even voting for it. The only way all that easy stuff is going to have any meaning is if you then go on to do the hard, difficult, challenging stuff afterwards: the work that’s needed to make sense of it all and that motivated you to do the easy stuff in the first place.
In this context, the worst possible abdication of responsibility is to claim — as teleological arguments implicitly or explicitly do — that the hard stuff doesn’t matter anymore. That all the little pieces of easy work will auto-organise themselves to do the hard work for you. Or worse, that the hard stuff just doesn’t exist.
Mainstream pundits, a decade after the techno-utopianism of the first dot-com bubble has faded, have started to notice this — and to decry it, in journals as leisured and stuffy as The New Yorker. In tech circles, the Web 2.0 brand started tarnishing four years ago.
It’s time the teleological stream within DH thought — still coursing boldly into the future, as defined in 1998 — caught up.