On the Necessity of Technologically Fluent Criticism

John M. Cox
Department of English
University of Michigan

December 2002

(NB: This article is part one in a series which attempts to bring technological methods to bear on problems which preoccupy literary scholars. In this part I attept to outline the difficulties inherent in such a process, and in part two I attempt to show what we can gain by working through these difficulties.)

Introduction
Mene: For Literary Scholarship, Computing
Mene: For Computing, Literary Scholarship
Tekel, Upharsin: Making Accounts

1. Introduction

Let me begin biographically. Over the past year I have been rather uncomfortably divested of a longstanding naivete regarding my areas of professional focus. I currently have two careers, working both within literary studies and within the computing discipline. This dual interest has been a longstanding feature of both my professional and private life for many years, but until recently I had assumed that they could both occupy my mental space with only the smallest amount of friction. This familiarity obscured the simple fact that none of my colleagues within literary studies understood what we do as computing professionals, and none of my information technology colleagues understood what we do as literary scholars. Not only that, but attempts by skilled practitioners of one discipline to bridge the gap between these two fields were either conspicuously absent, or seemed to me uninformed. I was struck with a disquieting thought: are these two disciplines so orthogonal to each other that they cannot relate but in the most inexact of ways?

This examination does not come purely from personal crisis. The topical intersections between computing and literary studies are growing more pronounced. From the literary vantage, I am thinking of Donna Haraway's "Manifesto for Cyborgs," Fredric Jameson's discussion of hyperspace in "Postmodernism and Consumer Society," and the work on hypertext done by both Stuart Moulthrop and Jerome McGann. From a computer vantage we have both writing, like Neal Stephenson's Cryptonomicon¹ and Larry Wall's recent paper on postmodern Perl², but more tellingly, we also have computer systems such as Project Gutenberg³ and Everything⁴, where computer technology and textuality come into direct contact. The bulk of computings relationship to literary studies is through systems like these, not through writing. Theory, by contrast, has a written preoccupation with technology, which is sensible given the increasingly determinative ramifications of both the computer and, more extensively, the network, on literary studies, and, indeed, on the printed word itself.

I must admit that I am anxious of proceeding. Beyond the fact that I will largely be speaking in generalizations, the task, it seems to me, is to speak as one situated between two disciplines, and as such it does not seem possible. There is a line, with literary scholars on one side and computer experts on the other, and in order for any one person to speak, they must first decide which side of this line they fall on. They choose their side—or their audience will choose for them, and each camp will, to some extent, dismiss an emissary from the other as an outsider, unschooled in discursive and philosophical nuances of the discipline. The stricture of this line is obviously suspect, but its determinate effect is not: there is immense resistance, in any discipline, towards acceptance of outsider judgments. By positioning this article in such a way that I select my fellow literary scholars as my audience—in such a way that I am speaking to them about computing—do I relegate myself to the other side of the line? To some extent the act of my speech must determine my camp: by speaking to literary scholars, as I am doing here, any balance I may have between the disciplines shifts, and as a speaker within the structure outlined above, I appear to gravitate toward computing. For the literary scholars, then, I become an outsider. This has a determining semiological effect: my words mean in the context of the point from which they are spoken, which, in turn, has an undesirable effect on my credibility. How, then, should I speak? Perhaps oscillation between the poles is the only method. Perhaps not. But it is an origin.

We will return to the central problem of this essay, the question of how it means, after a long detour. Before we can proceed to our interrogations, we must first clarify a few points regarding the computing discipline, and offer one caution.

Modern computing can largely be understood as the product of two technologies: the individual computer itself, and networks of computers. In a practical sense, computers themselves came out of the Second World War and passed to industry shortly thereafter. Personal computers, which are associated with the advent of large-scale computer networks, largely through demand for information, began their heyday in the early 1980s. The Internet, much like the computer, has its roots in the military. The Advanced Research Projects Agency, formed during the United States' reaction to the launch of Sputnik in 1957, proposed a large-scale computer network called ARPANET in 1968, and, like the computer, this network passed into the public sphere. ARPANET is considered the grandfather of todays Internet. Though there would obviously be no Internet without computers, it is the Internet more than the computer itself which produced the current computing community. When we speak of this community, we are not necessarily or even chiefly discussing subjects who have formal training in computer science or programming, but rather those who have expertise in the area. Unlike many disciplines, in computing formal schooling is not necessary for advanced careers. This is not to say that those with degrees in computer engineering or science are not a boon or that degree holders are barred from the community, but rather that the barriers to entrance do not encompass degrees so much as demonstration of skill. This skill is often demonstrated by sharing computer source code, the collection of instructions given to a computer to tell it how to carry out a task, and this sharing occurs largely over the Internet. Certainly this particular network is not the only medium for the dissemination of computer code, but it is by far the most dominant. Commercial software manufacturers excluded, the computer discipline is dominated by this gift economy and open exchange of information. This exchange is so ingrained that most within the discipline are adherents of a movement called "open source" or "free software"⁵, which stipulates that public availability of source code is essential to both programmer freedom and good software development. Participation in this gift economy is often enough to demonstrate ones skill and gain access to the discipline, to be considered an expert. This division between institutionalized computing technologists and "free agent" computing technologists is a vital feature of computing in general. It exists simultaneously as an institutionalized and as an amateur practice. We will take this point up later, but for now it is enough to say that the culture of computing, with its ideas of agency, self-image, and means of knowledge, is dominant only within the amateur practice. Within computing institutions, both academic and industrial, the cultural practices of the institution are dominant. When we discuss the culture of computing within its institutions, we are speaking more to the institution and less to computing.

The caution we wish to offer before we commence the main part of our discussion is that it is not our intention to judge one discipline's value, or to offer prescriptive doctrinal suggestions, from the vantage of the other. Such discussions always process from a place of arrogance and go no further than accusation and dire cautions. I am not interested in rehearsing the old form, mene, mene, tekel, upharsin—at least not without variation. It is neither even-tempered nor, in the end, useful. Nonetheless, there are problems with each of these disciplines relation to the other, and we hope that enumerating them will not be construed as partisanship, but rather read as it was intended: as careful consideration and critique, with an ever-present goal of progress.

Our first question is, why this choice of disciplines? The simple answer, which I made with a nod toward Spivak, is that I have an accident-of-skill-facility with them. I may speak to both, at least individually, as a member; I may understand both from within the context of their respective historical and philosophical determinants. My instinct, which I am sure is anticipated, is to call my skill set a hybrid, a chiastic collision, a multiplicity in residence. Beyond that simple answer, this selection of disciplines is germane because the collision between the Internet and the printed word produces a crisis/opportunity unlike any we have seen in recent years. If one accepts the compelling thesis that the new publishing technologies provided by the Internet have the potential to dramatically restructure the written word, then the conclusion that it will also have far reaching effects upon semiology, écriture, and a host of other areas of focus and contestation for literary scholars is not far off. Literary theorists and scholars cannot afford to duck this change, and technologists within the computing discipline cannot afford to leave it unexamined.

It is useful at this point to closely examine the antagonism, practical and philosophical, between these two disciplines, so that we may return to our central questions regarding how meaning is affected by disciplinarity. By the time these analyses are finished, they will hopefully seem reductive, since they are intended to be an analysis of one discipline from the vantage of the other. The assumption that this clean separation of vantage is possible, that we can situate ourselves upon any objective ground with respect to disciplinarity, is a destructive misconception. Nonetheless, the approximation of this separation, the same approximation that we see enacted in the disciplines themselves, is a useful frame for our discussion. Everything in the following sections is bracketed within a vantage. I wish to underscore this, since the assertions will be heavily contentious, even to the point of being brazenly wrong, from an opposing point of view that makes different epistemological and valuative assumptions.

2. Mene: For Literary Scholarship, Computing

We will start with the least surprising and contentious of our analyses: that of computing from the point of literary scholarship. On the most basic level, there is a cultivated disquiet between the academy and industry, as well as between what is institutionalized as the sciences and what is institutionalized as the humanities. This disquiet is largely born out of methodological, valuative, and epistemological disagreements. These overarching problems apply to the relationship between literary scholarship and computing, though they are not particular to it. While we do not wish to ignore these factors, we also do not wish to turn this discussion into partisan bickering between the mythic categories of "pure academy" and "real world," which are reductive and fundamentally misconceived. To that judgment, we wish to offer only a pithy question: where does the academy exist but in the real world? Our project here is to collapse the idea that any such clean distinction is desirable, even if it were possible. In the specific case of literary scholarship and computing, even that often appears to be an intractable problem.

It is rare for a literary scholar to also be a computer expert, with a clear understanding of how both individual machines and complex networks operate. Literary scholars are nonetheless willing to extend the glance of their critical eyes to the effects of computing, and will both critique it as a discipline and as an institution which produces social effects (see, for examples of this, the work of Moulthrop and McGann). There is a general consensus that, though computers offer unique opportunities for the dissemination and presentation of information formerly contained in more restrictive media, the ramifications of this technology upon textuality are problematic at best. There is also evidence for the sociological assumption regarding age and computing expertise, namely that the later an individual comes into contact with computers, the less likely they are to develop fluency and comfort with them. For disciplines whose admission requirements require many years of schooling, and consequently, tend to be constituted by an older population, the overall familiarity with computing tends downward.

This is not to say that either discourse about computing or discourse that makes use of computing metaphors is completely alien to literary scholarship. Critics like Haraway and Jameson, are most willing to incorporate computing technology metaphors into their discussion. I am thinking specifically of Jamesons discussion of "hyperspace" in his discussion of the Bonaventure Hotel, and of Haraways use of the cyborg as a creature of hybridity in her "Manifesto for Cyborgs."

Notwithstanding these discussions, computing largely remains a hinterland for literary scholars. This is due, in a large part, to the newness of it all; by and large the only people comfortable with computing technology are too young to be ensconced within the literary discipline. Taken alone, however, age is too simple a demarcator, given that it cannot be applied as a rule for judging computing proficiency with any rigorous degree of accuracy. There are other, farther-reaching and more thoroughly internalized features of the literary discipline that cause the distance between it and computing. Foremost among these is the suspicion, commonplace in literary theory since the advent of post-structuralism, of stable, unequivocal truths, and of definitions that are not in the service of ideology rendered natural. For a literary scholar, the tenet of software libre⁶, that "information should be free," is highly suspect. Information is never free, but is instead a field of contestation between warring ideologies. Individual freedom, which those in the computing discipline value so highly, is itself suspect, since the individual is constructed to some important degree by social institutions, ideologies, etc. The technologists belief in individual liberty and agency is a feature of mistakenly uncomplicated humanist and essentialist viewpoints. These viewpoints lead the technologist to a different judgment on "truth" than that commonly held by the literary scholar, and the technologist's judgments often seems to breathe from the mouth of arrogance. Those in the computing discipline are very willing to dismiss judgments on technical issues from those outside the discipline as misinformed or untrue, where they may merely be proceeding from different epistemological assumptions. For example, in the discussion of free software sketched above, a common defense of the statement "information should be free" might be that any in contestation simply do not understand current mechanisms for information dissemination, and if they did, they would not raise their objections. This is often held as sufficient, even if no attempt to explain why these mechanisms necessarily have a liberating effect. Furthermore, in the harsh light of social critique, technologists often appear to be more interested in what can be done and less interested in the structures technology creates, which may have deleterious effects upon valued, preexisting structures. Upon, for example, the codex⁷.

The last essential point here is that literary scholarship is the realm of the professional. This is a common enough point within disciplinarity studies⁸ and one that would benefit little from my belaboring it here, except to underscore the specifics in this case: the strictures of professionalism within literary scholars seems to many computer experts to be a deliberate territorial move born out of the desire to stake off a part of the intellectual landscape as off limits to those without a permit.

3. Mene: For Computing, Literary Scholarship

Here we will use the word "technologist" as a convenient, though inexact, marker for those within the computing discipline. Technologists have a rather standard set of objections to the methods, modes, epistemology, and assumptions of literary scholarship. Their central objections have been raised by many groups before, not the least of which is that of literary scholars themselves⁹: our type of scholarship amounts to little more than insular bickering over ephemeralities, which have little to no bearing on the outside world; our methods disqualify as undeveloped, uninformed, and under-nuanced the thoughts of those who have not spent years gaining advanced degrees; in short, that literary studies is comprised of a circular, exclusive group of ensconced thinkers whose work has little practical effect. The more particular objections of technologists arise chiefly when their own discipline is the subject of study, as in Moulthrop, or when their field of expertise is used as a metaphor, and put to work within literary scholarship. We see this in Haraway. I single these two scholars out because it was their work which crystallized the nagging objections I felt as a technologist whenever I saw literary scholars make computer technology either the subject of focus or a mode of expression and examination. I hasten to add that those expecting a wholesale refutation of Moulthrop's hypertext or Haraway's cyborg based on these objections will be disappointed. As a literary scholar, I found that, though I objected to their treatments of technology in point of fact, this was not sufficient grounds to dismiss or even thoroughly complicate their arguments.

Let us take the case of fact first. Haraway writes:

Pre-cybernetic machines could be haunted; there was always the specter of the ghost in the machine. This dualism structured the dialogue between materialism and idealism that was settled by a dialectical progeny, called spirit or history, according to taste. But basically machines were not self-moving, self-designing, autonomous. They could not achieve man's dream, only mock it. They were not man, an author to himself, but only a caricature of that masculinist reproductive dream. To think they were otherwise is paranoid. Now we are not so sure.

In fact, technologists are very sure. They will argue that the current existence of an intelligent, self-designing machine is fever-dream. It was not possible when Haraway was writing; it is not possible today; it is not yet an imminent facet of the future. While it is possible in theory, and is a dream likely of realization, right now it is merely exciting fancy, relegated to science-fiction and always cast as a dark portent, a terrible consequence of man's hubris. To them, Haraway's passage exposes a fundamental misconception about computers: that they are in any way intelligent. All computers do is follow instructions, blindly and without question. If they exhibit anything that looks like intelligence, it is only because their instructions, given to them by a human, are very sophisticated. The much-touted Deep Blue, which beat a human grandmaster at chess, did so because it was capable of running through every possible move, every possible outcome, and following human-designed rules for selecting which possibility yielded the most advantageous results, not because it was intelligent¹⁰. This is not an easily-discarded problem; it posits, for example, the discomfiting idea that emulation of intelligence may be indistinguishable from intelligence provided it is sufficiently complex. There would then be the philosophical problem of telling what difference, if any, can be said to exist between intelligence and a perfect emulation. For the purposes of reading Haraway, however, the technologist is likely to say that Haraway either has a fundamental misconception about the inner workings of computers, or she is twisting the truth. The technologist does not need to spot Hegel in her sentences to make this judgment, and would not consider it relevant even if it was apparent. The important thing about the passage is its mistaken judgment on computing, and that this judgment was presented as fact.

In the case of Moulthrop, the technologist is likely to raise terminological objections. In particular, Moulthrops use and understanding of the word "hypertext" in his "Hypertext and the Laws of Media." Technologists are not likely to use the word "hypertext" in their discussions, favoring instead the term "link." The reason for this difference is rooted in what each group wants out of the term. Moulthrop is interested in how hypertext "create[s] and enact[s] linkages between stored bits of information," in how hypertext adds an extra, uncertain and nonlinear dimension to our existing conception of text. In this light, the term seems very well-chosen, as it highlights the extension of text beyond its conventional forms in much the same way that a hypercube extends beyond the easily-conceived spatial nature of its three-dimensional cousin. Literary scholars use "hypertext" to refer to any readable gestalt that operates similarly. Technologists, on the other hand, prefer "link" because they do not see the technology as explicitly textual: not sharing the literary scholars broad definition of "text" as inclusive of things outside the written word, the term "hypertext" seems misapplied when this linkage can occur between, for example, an image of a cello and a snippet of a cello suite. "Link" describes only the connection between the conjoined items and the markers for that connection. It does not apply to the entire text. Beyond this definitional squabble, there is also a dispute as to how hypertext/the link operates. For Moulthrop, hypertext has a profound deheirarchizing, decentering, "infinitely convoluted" nature and effect. After all, the author can no longer direct the reader in a contained, linear sense. Moulthrop is quick to note that to a large extent, this is the same as it ever was: after all, conventional textuality hardly forecloses on intertextual reference, hardly prevents the reader from skipping around within a text or setting it aside in favor of another. In this he anticipates a large part of the objection technologists raise against this understanding of hypertext. However, technologists, particularly those who specialize in the creation of material on the Internet, would also object to the assertion that the nature of the link itself, and not the method with which it is employed, necessarily creates a decentering effect. Web design, the subfield of computing most concerned with hypertext, is largely predicated upon the use of the link as an organizing technology, and one that can and should be employed to direct the reader quickly and clearly to intended and desired information or text. For them, the link is put to organizational use. The link is neutral when it comes to centering or decentering in the sense that it may be applied either to centering or decentering with equal ease and effect. Moreover, many web sites are designed with the intention to provide content in mind, and as such their design, much like that of a good, compelling book, encourages the user to remain with the text rather than setting it aside and visiting another. Nor is anything "hypertextual" necessarily dehierarchizing: the Everything project, which has been put to use most famously as a platform for the development of a user-generated encyclopedia, and which we referenced earlier as a good example of a circumstance where technology and textuality are in direct contact, is extremely hierarchical, with users vying for the appreciation of their peers, which can result in appointment as an editor of the encyclopedia, as a tastemaker and standard-setter. The technologist is likely to view a conception of the link as decentering or dehierarchizing as yet another misunderstanding, or, at the very least, another generalization, presented as fact.

What is likely to irk the technologist most is not what ends these misunderstandings are put to, but that they exists in the first place. This looks backwards to a literary scholar, who is likely more interested in tracing out the effects, and focusing on both its reason and result. For the technologist, however, the mistake appears as a mark of laziness, of an unwillingness to learn. This is a curious judgment to make of a scholar, and its reasoning is deeply rooted within one of the most fundamental assumptions technologists have regarding their discipline: that it is eminently learnable by very nearly everyone. Technologists do not believe, of course, that everyone can do what they do: someone who never has contact with a computer due to poverty is not likely to become a computer expert. In the main, however, technologists believe that most people, given the availability of computers and the time in which to learn, are completely capable of teaching themselves into expertise. They have a great deal of evidence for this assertion: many of computings canonical figures either have no formal computing education (Eric S. Raymond, for example) or were dropouts who decided more could be done outside the academy then within in (the most famous of these, who many technologists are reluctant to claim as one of their own, is Bill Gates). Not only that, many technologists are self-taught. In the interest of full disclosure, your author admits that he falls into this camp. This belief, whether or not it is grounded in fact, has a powerful determinative effect on technologists: when they see ignorance of computing, they assume it is willful.

In the case of metaphor, a large part of our work has been done for us. Mary Poovey writes of the bipartite relation between metaphor and discipline, stating that metaphors both organize the discipline in which they are used, and have their meaning fundamentally changed by that discipline. Given this, we must suspect, or at least read closely, to what extent the metaphor-in-discipline is determined and is determining. When Haraway imports the cyborg, for example, we can say both that the term can no longer be understood to strictly mean what it does within the context of cybernetics or within computing. The computer expert reading Haraway finds a familiar term but does not find a familiar meaning, and the first reaction is that Haraway has distorted the term to further her own ends. The metaphor, in other words, is a bad metaphor. It is very easy to feel justified in this judgment: after all, an outsider has appropriated a term and put it to work. It comes as no surprise that their unfamiliarity with the term results in an inexact use. Within this framework, it not obvious that it can be the difference in discipline, and not in expertise, that has the distorting effect upon meaning. It is also not obvious that the metaphor, bad or not, is having an effect back upon the user, organizing understanding within the users discipline, structuring that discipline's understanding of what the metaphor describes. Instead, the objection simply is that the user cannot be bothered to understand the nuances of a term before applying it, and that user loses credibility from those within the discipline from which the metaphor originated.

4. Tekel, Upharsin: Making Accounts

Here is where we reach the crisis point. So far, this discussion has been structured in such a way that its rhetoric protected me from appearing to cast authentic, personal judgments upon one discipline from the point of view of the other. The stance of examination from a carefully demarcated and easily differentiable vantage (computing from the view of a literary scholar or its reverse) is a comfortable shield, but it will not suffice for the remainder of this analysis, where both disciplines simultaneously come under analysis and the danger of appearing as an outsider, simultaneously, to both factions becomes the threat. I am most likely to be dismissed by technologists, perhaps unavoidably, since I have not chosen them as my audience. This seems paradoxical, but in electing to speak to literary scholars, I must, as a consequence of earning qualification to speak, also elect to analyze in the literary fashion. It is impossible for me to adopt a rhetorical stance of impartiality when the qualifications of expression make this requirement.

The mutual dismissal between computing and literary scholarship, which structures their refusal of outsider speech, method, and epistemology, can be traced to the way in which entry to and speech within the field is mediated. Mary Poovey has recently discussed the movement within literary scholarship from a discourse in which amateurs were welcomed to one that requires professional qualifications and forecloses upon amateur discussion. As recently as the 18th century, literary scholarship made no demands upon degrees, and the writing of amateur scholars appeared in print alongside that of their professionally-trained colleagues¹¹. This holds true for science as well: the age of the amateur laboratory has largely passed. This is not so for computing. Two of the technologies that have, in the past three years, been most influential in computer users' day-to-day lives started as dorm-room projects. I am speaking here of Napster, the file-sharing service that came to its end under litigation from the music industry, but which inspired a legion of similar services before its assets came up on the auction block, and of Google, an Internet search engine so good that it, more than anything else, gives sense, order, and researchability to the new publishing medium. A search engine may seem like an odd candidate for a highly-influential technology, but without it the usability of the Internet would suffer an extremely sharp decline. At the time of this writing, Google currently functions as an index for 3,083,324,652 web pages, and this number constantly increases. These technologies eclipse all other recent advances in terms of social effect, including those bankrolled by large research universities and industry Research and Development departments. It is clear that foreclosure upon the amateur, that pernicious consequence of institutionalization, is not at all complete within the computing discipline, which, as we have discussed, is characterized and actively conceived as a field without a dominant institutionalized and institutionalizing principle.

Within this concept we also have an interaction with the idea of liberal self-image and agency. This results in a mutually reinforcing chiasmus which enshrines amateurism within the computing discipline as the genuine mode of operation within the field, as the proper way, not only to be, but to know and to practice. It is certainly true that there is institutionalization within computing: there are senior research scientists, tenured professors, division chiefs, corporate technology officers—an entire hierarchy, with all the modes of discernment that it implies. These institutions are responsible for a great number of fundamental computing technologies. We have already discussed what is perhaps the best example: the origins of computers and networks within the United States military. But it is likewise true that a great deal of work in computing happens outside of these conventional institutions and their hierarchies, or at least happens only indirectly and incompletely under their aegis. The Linux kernel¹² and the GNU Project¹³ are useful examples of this type, as are Napster and Google. As these projects have matured, they have adopted their own hierarchies and institutions while still maintaining their adherence to computing culture and methods of conception and operation: Napster and Google incorporated, while the Linux and GNU Projects have a much more ad-hoc and unconventional but nonetheless binding organizational structure, with leaders in positions to say which contributions to a given subproject are included and which are not. As one might expect, the later projects, which have not adopted the institutional markers of a corporation, maintain a higher degree of confluence with computing culture at large (they are heavily predicated, for example, upon the gift economy outlined above). It is especially here that amateurism is considered both viable and desirable. Many who contribute their work to these projects self-identify, not as a "programmer" or "computer expert," but as a "hacker"¹⁴.

The friction between this amateurism and the professionalized institution of literary scholarship is very pronounced, with each discipline decrying the authorial values of the other. The academics are insular, removed, and snobby, while the technologists are unexamined, arrogant, ad-hoc, and sold on the illusion of their independent operation as subjects. Both feel excluded by the methods of the other, by the jargon, strange assumptions, and incomprehensible tools. What makes this difference particular is that they disagree, not only on what is true and on what is valuable, but on how things become true and how things are valued. Assuming that we find such a thing desirable, the question becomes, what is the desired relationship between the disciplines, and what is the best method for establishing that relationship?

The bearing of technology upon text is the fulcrum. In the midst of a medium shift that is already beginning to have far-reaching effects upon meaning, power structures, the role of information, and so on, it is foolhardy to leave the medium unexplored, or only shoddily explored. The need for examination of these new technologies is very present, both in their relation to literature and literary studies, and in their relations outside of our discipline. Neither literary studies nor the computing discipline, as they currently are conceived and operate, have all that is necessary to do this. The computing discipline is structured in such a way that it is unlikely to formulate what literary scholarship is likely to see as the necessary questions: it is not the case, for example, that the concept of "text" within computing is likely to give rise to the questions of textuality that literary scholars find so essential, to those questions of authority, semiotics, hierarchy, power, and so on. Literary scholarship is not structured in such a way that it is likely to be fluent in the intricacies of extant computing systems, and this makes it unlikely that literary scholars will formulate questions or answers that take these intricacies into account. The consequence of our current structures, then, is confounded inquiry. If we accept that this inquiry is vital for our continued work as either technologists or literary scholars, we must dismantle the parts of these structures which confound us. Currently the largest hurdle is the stricture of the line we discussed at the onset of this article, which separates technologists and literary scholars into disciplined camps that dismiss the authority of what they see as external discourse. The discourse we are producing within each of these groups is not sufficient to the task we set for it, to the examination of technology within these contexts.

To entirely do away with this confound, we need to refigure our concepts of authority and allow speech from those outside our disciplines to permeate our discourse and inform our examinations. I would like to end with practical considerations that may help toward this goal. It is doubtful that either discipline would drastically work their epistemology or methods; it is likewise doubtful that this is necessary or entirely desirable. There will not be perfect accord, nor anything that closely resembles it, but such a thing is hardly necessary. The tension itself is illuminating, after all. Given that, we need not throw up our hands. As technologists, we can work to be less stubbornly abstruse in our software and in our explanation of that software. There is a large push within programming toward increased usability, toward manufacturing computers that work more as appliances than intricate, baroque machines requiring high degree of expertise, and part of the fallout of this will be that the opacity of the most common forms of computing technology will decrease. If we truly believe that our discipline is eminently learnable, and we have no interest in maintaining or creating barriers to this learnability, then this elimination of opacity is already a part of our agenda. As literary scholars, we can increase the attention we pay to the technical specifics of any fact or metaphor we wish either to discuss or employ. This is already part of our agenda, as we value nuanced and detailed scholarship. We do not have to accept the valuative systems or interpretations of technology of technologists, but we would do well to accept their authority on matters that are wholly technical, such as the intelligence of computers or the neutrality of links within hypertext. This eliminates disputes on matters of fact, and defuses easy objections. As experts in either field, we can be suspicious of the notion that the external is entirely uninformed, and that our easy objections are sufficient for wholesale dismissal. This increases our abilities of interrogation, since we are less likely to discard arguments prematurely and leave them, on the whole, unexamined. If we can begin our discussions without the need for an apologia for that discussion, without the necessity of skirting rejection because of our outsider status, our analyses will only benefit.

Notes

1: Neal Stephenson is a science fiction writer popular among computer enthusiasts for his insightful and accurate portrayals of real-world or near-real-world computer systems. Stephenson himself is well-versed in computing. Cryptonomicon, his largest book, is chiefly concerned with early computer systems and cryptography, though it contains an extended passage which examines the interrelations between academics and computing professionals.

2: Perl, created by Larry Wall, is a quick-and-dirty programming language often used on UNIX and UNIX-like operating systems to automate repetitive tasks, manipulate and extract information from large amounts of text, and many other diverse tasks. "Perl, the First Postmodern Computer Language," is available online.

3: Project Gutenberg is a freely-accessible computer archive of classic texts which can be reproduced indefinitely. Its creators liken the advent of the Internet to that of the printing press.

4: The Everything Engine is an "information management system" which tracks the movements of people through its database and generates links which associate written pieces of information with each other. Over time the links in the database begin to map associated concepts. An example of this application may be found in the wild at http://everything2.com, where the system has been put to use as infrastructure for the development of an online, user-generated encyclopedia of everything.

5: There is a massive schism within the computing community over whether open source or free software is the superior methodology. The distinction between these two schools of thought is neither germane to or within the scope of this discussion, but interested readers are encouraged to read a discussion of these ideas, albeit one that is biased in favor of free software over open source. The basic distinction is that the free software movement is more concerned with individual freedom, and the open source movement is more concerned with a successful software development methodology.

6: A not-quite-canonical term used to make clear the distinction between software that is free monetarily but that does not have freely available source code, and software that has free source code, but is not necessarily free monetarily. One popular way of phrasing the distinction is "free as in freedom, not as in beer."

7: McGann, "The Rationale of Hypertext".

8: For a full discussion of the emergence of professionalism and the decline of amateurism within literary scholarship, see Mary Poovey, "The Model System of Contemporary Literary Criticism," Critical Inquiry, Spring 2001.

9: I am thinking specifically of Stuart Halls question in "Cultural Studies and its Theoretical Legacies": "Against the urgency of people dying in the streets, what in Gods name is the point of cultural studies?" We could easily bring this question to bear on literary studies as a whole.

10: Deep Blue, a chess-playing supercomputer designed by IBM, beat Garry Kasparov, considered the best chess player in the world, in 1997.

11: Mary Poovey, "The Model System of Contemporary Literary Criticism," Critical Inquiry, Spring 2001.

12: The Linux kernel, which is the basis for the worlds most popular free operating system, was originally written by Linus Torvalds in 1991. Torvalds, a self-taught computer expert, created the first version of the kernel "just as a hobby." An operating system is the fundamental software a computer runs that tells it how to do its most basic tasks and manage its resources. Without operating systems, computers cannot run any other programs. Popular operating systems in use today include UNIX, Linux (or "GNU/Linux." There is an ongoing dispute as to which name is correct), Microsoft Windows, Macintosh OS X, BSD, and Solaris.

13: Launched in 1984, the goal of the GNU ("GNU's Not UNIX," a recursive acronym which can be unpacked ad infinitum: "GNU's Not UNIX," "GNU's Not Unix Not UNIX," and so on) Project is to create an entirely free UNIX-like operating system. UNIX is an extremely complicated, sophisticated, and baroque operating system created by Bell Labs that today has many variants, such as HP-UX, Solaris, and BSD.