The Anxiety of Disintermediation

Reading today's post to Andrew Keen's Great Seduction blog, "The Children's Crusade," reminds me of how much I hate the word "disintermediation" and the cult formed around it by Internet evangelists. Regular readers of this blog should know by now that I see a lot of value in intermediation, and I feel most strongly about the need for an intermediate party that stands between a writer and a reader. For those who find that language too elusive, the intermediate party I have in mind is a talented editor; and, as I previously asserted, this refers specifically to the job of the editor to identify the author's voice and make sure it is speaking clearly to the reader base.

Where communication is concerned, the Age of Disintermediation is the world of the direct channel from the writer to the reader; and the fallacy behind drinking too much Web 2.0 Kool-Aid is a simple one. If one writer has even a moderately large number of readers (whether or not they are out on some "long tail"), then, for all intents and purposes, that "direct channel" is one-way. There is no doubting that communicating through face-to-face conversation can be very informative; but, as Erving Goffman pointed out in Forms of Talk, the "unit of dialog" is not a "semantic unit of text" but a "move," whose content is both linguistic and paralinguistic. The more people are participating in that "dialog" (whose etymology implies only two participants), the harder it becomes to communicate through such moves and the more likely it is that the whole encounter will end in confusion, rather than understanding.

So it is that the writer benefits from the skilled editor. That editor may not understand the writer's area of expertise, but the editor does understand that the text will be distributed to a population of readers and must therefore communicate through means other than the moves of dialog. The editor draws upon knowledge of that population of readers to work with the writer to yield a text, which, while not a simulacrum of face-to-face conversation, will communicate to the reader with the effectiveness of personal dialog. This is no easy matter, but the best editors are the ones who know something about the nature of the reader base and can advise the writer productively on the basis of that knowledge. This is intermediation at its most effective; and, hopefully, it illustrates why I get so aggravated by those who would do away with it.

Now, as my father used to say, "one is not a statistic." Is editing an exceptional case of intermediation, or is it a representative one? Keen's post has offered, perhaps unwittingly, another example on a slightly more elevated plane. He began with the following quote from David Edgerton, the founding director of Imperial College's Centre for the History of Science, Technology and Medicine:

Thinking about the use of things, rather than of technology, connects us directly with the world we know rather than the strange world in which "technology" lives. We speak of "our" technology meaning the technology of an age or a whole society. By contrast, "things" fit into no such totality, and do not evoke what is often taken as an independent historical force. We discuss the world of things as grown-ups, but technology as children.

Keen takes this as a point of departure to serve up his own take on disintermediation; but, while his approach was somewhat in the vein of my argument about the one-way nature of the direct channel, I found an another example of the need for intermediation in Edgerton's text.

As I read this passage, thinking about the use of technology is a matter of thinking about the use of abstract concepts, rather than concrete things. Computation is such an abstract concept, as is communication; and, for most domains of thought and discourse, most of us are not particularly comfortable reasoning with those abstract concepts but have no problem using many of the "things" derived from those concepts. Consider an example from mathematics: "Triangle" is an abstract concept; but any triangle I draw on a sheet of paper is a "thing." I can do things with the "thing," such as measure the lengths of the sides and the sizes of the angles, whereas the "triangle" concept can only tell me about certain "rules" that those sides and angles must "obey." However, if I want to talk to you about the "triangle" concept and you are not a mathematician, chances are that what I say to you will be easiest for you to understand if you can relate to a picture that I draw. That "thing-picture" intermediates between my talking about the "triangle" concept and your understanding what I am trying to tell you; and, as we shall see shortly, it intermediates by giving you something to perceive.

By analogy, then, we tend to discuss the world of technology by using products of that technology (perceivable things) as intermediaries. Saying that we discuss technology "as children" may be a bit overly pejorative; but it is one way to capture the idea that we really do not deeply understand technological concepts. I would prefer to say that we understand it as naïfs, which is to say that our understanding is simplistic but, like naïve physics, can still be functional.

As I see it, the implication of this example may throw light on a connection between disintermediation and the argument that Nicholas Carr was trying to make in his "Is Google Making Us Stupid?" article. As Gerald Edelman has demonstrated in his Neural Darwinism research, our very capacity for understanding is grounded in our ability to form perceptual categories, which means that our understanding of concepts is grounded in those things (which we happen to perceive) that are instances of the concepts we come to understand. Thus, in a context far broader than mathematics, the very workings of the "wet brain" depend on the concrete to intermediate between the mind and any abstraction than needs to be understood. When Carr is complaining that he can no longer read War and Peace, part of that complaint has to do with the extent to which reading provides a surrogate for experiences through which perceptual categories are formed, while the more "efficient" artifacts delivered by Google search results do not serve in such a surrogate capacity (nor were they intended to do so). Thus, by gradually insinuating its role as our "window on the world," Google is (probably unintentionally) subverting our brain's hardware for forming perceptual categories through either the "directly perceived world" or the "world perceived indirectly through text." If Google is making us stupid, it is doing so by allowing the brain functions that "make us smart" to atrophy by not giving them opportunities to "do their thing." (This is at least partially analogous to some of the arguments about why watching too much television makes us poorer readers.) Furthermore, when it comes down to exercising the capacity for perceptual categorization, it may be not only Google that is making us stupid but also a whole set of behavior patterns grounded in disintermediation! No wonder the word aggravates me so much!