So far, my focus has mostly remained well within the confines of a strictly historical materialist argument, emphasizing the lasting effects of the adoption of a certain mode of production, the hypertext markup language or HTML, on its textual artifacts, in this case [webcomics] that might otherwise seem to represent the arbitrary choices of their creators and for which decision McCloud chose in Reinventing Comics to take them to task. We see in the relationship between the infrastructure imposed by HTML on actual webpages (and therefore on elements within said pages) what of Althusser’s three types of causality would be called “structural effectivity” (along with “mechanistic” and “expressive” effectivities), that is, the effect of the whole on its parts only realized (meaning understood as well as made manifest) in those particulars. However, in Althusser’s work (as well as Jameson, who follows him closely in Political Unconscious) both effects and structure are regarded, at least for the purposes of analysis, as merely there and not necessarily motivated by any agent or conspiracy of agents. This is not to say that either Althusser or Jameson ignore questions of agency, though how historical forces are motivated is more specifically addressed in Althusser’s work, but rather that, for Jameson in particular, these questions, which are regularly raised by those who valorize the unitary author and his/her authorship, can be set aside so as to consider those seemingly external factors in their work which they may not ever acknowledge or in some cases explicitly resist. The logic underlying Jameson’s mode of analysis is that literature always says far more than it intends, and often in order to see this “far more” one must set aside what authors claim to have done in their work.
According to this way of thinking, both the base, the material conditions or mode of production (read: the economic order), and the superstructure, things like culture or ideology, simply are. This particular limit of Marxist analysis, vulgar as it may be, is one reason why I have preferred infra– (emphasizing, as it might, the “hell” of that particular Latin prefix as opposed to the blander sub-) over the traditional Marxist super-structure, so as to address a rather thorny question of how [textual] structures are motivated and how they likewise motivate their effects, rather than regard them by way of an aloof empiricism that refuses to address such questions. Up until this chapter, this problem has largely been addressed through the lens of Heidegger, but the situation of [webcomics] and HTML could benefit equally from a Freudian perspective. Beginning from our dear closet Nazi Martin, we might say that what the Marxist approach fails to recognize is how the being of the base, be it material or coded (as with languages), only is insofar as it is understood and is in fact conditioned by those supposedly super-structural particulars (or particularities) that Marxism for the most part only ever regards as aftereffects or afterthoughts.
In the present example, the periodical design basis manifest only awkwardly and by degrees in HTML would likely be regarded in historical materialist terms as merely there, and for the kinds of analysis Marxists typically endeavor upon, I should repeat so as not to seem too dismissive, this assumption serves well enough. Indeed, so far, my own argument has not proceeded past this point, in that the identification of a certain “periodicality”—even scare quotes make such a neologism hard to swallow—in the development of hypertext documents in the early history of the Web posits just such a structural causation or homology with such effectivity, since in Althusser the “structure” is not really plural, namely that it makes perfect sense that the kind of [comics] one sees regularly on the Web would be fundamentally similar to what one saw at the time (i.e. the 1990s) in newspapers, because in each case the underlying basis, the design framework, was analogous to its old/new media counterpart. The mere fact of this analogy is easy enough to observe in the design rationales the W3 consortium members produced contemporaneously to the implementation of certain features in HTML.
What is not readily apparent in any given webpage, its code, or even in the proclamations of principles and best practices is how the historical fact of what happened in the early history of HTML reflected a clear choice, if not always a conscientious one, of one design ideology over another, an ideology of form that was not merely an expression of a base or lower order, in the present example the “code itself” and in McCloud’s the presumed “nature” of computer hardware, but rather conditioned it to be a certain way, meaning the code conformed to the ideology, not the other way around.
Here I turn not to the later, more familiar Freudian tripartite model of the psyche, the so-called Id-Ego-Superego, but rather to a much earlier understanding of the relationship between the unconscious and conscious mind contemporaneous to The Interpretation of Dreams and characterized by what he refers to as the primary and secondary processes. The primary process is the means by which the unconscious produces that which can be known at all in conscious awareness. So, the unconscious, which we might presume to desire something, say, sexual gratification, manifests initially as the expressible, if socially unacceptable, preconscious desire to rape a dog. The conscious/preconscious mind, upon which human culture has had some effect, recognizes this desire as abhorrent and attempts to alter it so that it comes out as, if not acceptable, then at least something less likely to cause one extreme emotional distress, like say playfully wrestling with a family dog. This repression by way of transformation—and in Freud’s own excursus many such transformations layer on top of one another so as to put the initial desire at a far remove—is key to what he calls the secondary process, crudely understood here as the latter transformation of what was originally manifest, a transformation that, I hasten to note, conforms to expectations seemingly “from without.”
The irony of these two processes is that in analysis, we only ever see this process in reverse, symptomatically, and so the deconstruction of the secondary process only ever approaches the primary by way of inference and speculation, a fact which later greatly contributed to denigration of Freud by a psychological profession that sought to be ever more rigidly empirical in an attempt to justify itself as a valid science. Whether Freud’s is an acceptable or useful model of the mind is not at issue here. What is of concern is how the secondary process shows us the way in which supposed aftereffects, what Freud would call “the subtle work of culture,” work themselves back into what seems at first glance to be the point of departure, the base, the code.
When style sheets or, more specifically, cascading style sheets (CSS) were implemented in 1996, it was done, presumably, for practical concerns: accessibility, so that a page would render appropriately on different platforms such as a monitor or Braille-based device; flexibility, so that designers might have greater control over the appearance of websites; and functionality. In an environment where bandwidth limitations were a real concern—I can hardly imagine a modern webpage with its swaths of scripts and embedded objects loading in a reasonable amount of time over a 28.8 kbps connection—separating out elements from the numerous pages that make up a particular site which are common to all of those drastically increased performance. In the earliest iterations of HTML, the code for stylistic and typographic changes was entirely written into the hypertext document itself, meaning, as you navigated a site’s many pages, the formatting, even if it was the same for each page, would have to be completely reloaded each and every time. Style sheets made it so that formatting could be loaded just once, initially, thereby speeding up the loading of subsequent pages that followed the same style sheet.
As above with tables, CSS was introduced as drawing from and speaking directly to the concerns of print publishing. “CSS1 is a simple style sheet mechanism that allows authors and readers to attach style (e.g. fonts, colors and spacing) to HTML documents. The CSS1 language is human readable and writable, and expresses style in common desktop publishing terminology.” (REC-CSS1) However, CSS was also an expression of a larger issue in computer science and the development of programming languages, namely “separation of concerns,” which by the mid 1990s was consensus among programmers, developers, and their fellow travelers, but when Edsger Dijkstra explicated the concept in his widely circulated typescript “EWD 447: On the role of scientific thought” (1974), such an atomization of concerns was still very much a matter of debate, at least in practice.
Dijkstra, quoting from one of his own letters, describes there what, according to his “taste” (an odd word choice, given the subject) is “characteristic for all intelligent thinking,” namely that “one is willing to study in depth an aspect of one’s subject matter in isolation for the sake of its own consistency, all the time knowing that one is occupying oneself only with one of the aspects.” The kind of analysis Jameson proposes, then, would be completely anathema. As Dijkstra says, “nothing is gained—on the contrary!—by tackling these various aspects simultaneously.” This is not to say that Dijkstra disregards the importance of thinking about how aspects relate to one another, but as he presents the issue, of being “one- and multiple-track minded simultaneously,” such relations are rendered yet another separate concern. Left there, this might be well enough as manifestation of a certain prevalent ideology in the sciences, but Dijkstra presses his point in a moment where he presumes himself to be obviously in the right but according to the standards of humanistic inquiry demonstrates his point to be woefully naïve.
If you so desire, you may observe here scientific thought in action: I do, for instance, not deny political aspects, I would be a fool if I did so! The anti-intellectualistic backlash against “the technocrats” that is so en vogue today, is inspired by a —largely unjustified— fear for the power of him who really knows how to think and by a —more justified— fear for the actions of him who erroneously believes to know how to think. These political considerations, however, have nothing to contribute to the technical problem of ordering one’s thoughts effectively, and that is the problem that I want to discuss “in isolation, for the sake of its own consistency”.
That “political considerations” would have no role whatsoever in “ordering one’s thoughts effectively” is a rather alarming statement on its face, but it also points to the way in which totalizing “sciences” (Marxism too!), which—Dijkstra is in agreement here—are a matter more of “how” than “what” one knows, fall prey to a certain obliviousness that cannot see how in many ways the very issues separated out have sweeping ramifications for how the atomized aspect under consideration might be understood “in itself.”
Now, we have wandered back somewhat to Heidegger, where what any given thing is must be understood as a totality of interrelations, but so too can we see an analogue to Freud’s primary and secondary processes. As first implemented, HTML represented precisely the opposite of Dijkstra’s separation of concerns, for it manifested, along the lines of the primary process, a desire to create a kind of document that made use of network architecture. The later implementation of CSS projected a prevailing and pre-existing computer science design ideology, separation of concerns, back onto what historically first emerged with little regard to this “ideal.” In fact, the effect of such a repression by way of transformation, if that is what we wish to call it, or the “deprecation,” to use the more localized jargon, of an older way of doing things is in no way complete. The return (or uncanny eternity) of the repressed in this case means you can still markup a hypertext document in precisely the manner CSS was and is meant to supplant.