Such a highly non-linear problem would clearly benefitfrom the computational power of many layers. Unfortu-nately, back-propagation learning generally slows downby an order of magnitude every time a layer is added toa network.

Other conventions mentioned by experienced users include keeping e-mail correspondence short, and acknowledging receipt of mail. "I hate it when I get messages with three or four questions, some of which I can handle immediately, and others which require more time," writes Jane N. Ryland, president of CAUSE, an association of university computer administrators. "If a message is more than one screen long [about 20 lines], I find it difficult to respond without first printing a copy of the message, which defeats much of the benefit of electronic messaging."

while the technology has boomed, the conventions and etiquette of computer interaction are still nascent.

No Bias, No Merit: The Case against Blind Submission

But perhaps the greatest change is the one that renders the key opposition of the essay-between the timeless realm of literature and the pressures and exigencies of politics-inaccurate as a description of the assumptions prevailing in the profession. There are of course those who still believe that literature is defined by its independence of social and political contexts (a "concrete universal" in Wimsatt's terms), but today the most influential and up-to-date voices are those that proclaim exactly the reverse and ar- gue that the thesis of literary autonomy is itself a political one, part and parcel of an effort by the con- servative forces in society to protect traditional values from oppositional discourse. Rather than reflecting, as Ransom would have it, an "order of existence" purer than that which one finds in "actual life," lit- erature in this new (historicist) vision directly and vigorously "participates in historical processes and in the political management of reality" (Howard 25). Moreover, as Louis Montrose observes, if litera- ture is reconceived as a social rather than a merely aesthetic practice, literary criticism, in order to be true to its object, must be rearticulated as a social practice too and no longer be regarded as a merely academic or professional exercise (11-12

Reviewers who receive a paper from which the identifying marks have been removed will immediately put in place an (imagined) set of circumstances of exactly the kind they are supposedly ignoring. I

Nevertheless, there were a few who questioned that definition of fairness and challenged the assumption that it was wrong for reviewers to take institutional affiliation and history into consider- ation. "We consider a result from a scientist who has never before been wrong much more seriously than a similar report from a scientist who has never before been right. . . . It is neither unnatural nor wrong that the work of scientists who have achieved eminence through a long record of important and suc- cessful research is accepted with fewer reservations than the work of less eminent scientists" (196). "A reviewer may be justified in assuming at the outset that [well-known] people know what they are do- ing" (211). "Those of us who publish establish some kind of track record. If our papers stand the test of time . . . it can be expected that we have acquired expertise in scientific methodology" (244). (This last respondent is a woman and a Nobel laureate.)

A similar point is made by some of the participants in a discussion of peer review published in the Behavioral and Brain Sciences: An International Journal of Current Research and Theory with Open Peer Commentary (5 [1982]: 187-255). The occasion was the report of research conducted by D. P. Peters and S. J. Ceci. Peters and Ceci had taken twelve articles published in twelve different journals, altered the titles, substituted for the names of the authors fictitious names identified as researchers at institu- tions no one had ever heard of (because they were, made up), and resubmitted the articles to the jour- nals that had originally accepted them. Three of the articles were recognized as resubmissions, and of the remaining nine eight were rejected. The response to these results ranged from horror ("It puts at risk the whole conceptual framework within which we are accustomed to make observations and con- struct theories" [245]) to "so what else is new."

c. To the question "What is criti- cism?" Ransom answers with what it is n

Before that meeting was held, however, a second letter arrived, written this time by "my" president, informing me that the constitution was in fact already available in the form of the constitution, ready made as it were, of the Milton Society, which also, the letter went on to say, was to provide the model for a banquet, a reception, an after-dinner speaker, an honored scholar, and the publication of a membership booklet, the chief function of which was to be the listing of the publications, recent and forthcoming, of the members. The manner in which work on Spenser is to be recognized and honored will have its source not in a direct confrontation with the poet or his poem but in the apparatus of an organization devoted to another poet. Spenser studies will be imita- tive of Milton studies; the anxiety of influence, it would seem, can work backwards. Moreover, it con- tinues to work. The recent mail has brought me, and some of you, an announcement of a new publication, the Sidney Newsletter, to be organized, we are told, "along the lines of the well-established and highly successful Spenser Newsletter," which was organized along the lines of the well-established and highly successful Milton Newsletter

Here is the real message of the letter and the real rationale of the Spenser Society of America: to multiply the institutional contexts in which writing on Spenser will at once be demanded and published. It so happens that the letter was written before the society's first meeting, but as this sentence shows, the society need never have met at all, since its most impor- tant goal-the creation of a Spenser industry with all its attendant machinery-had already been achieved

. It follows then that the machinery of the institution does not grow up to accommodate needs that are independently perceived but that, rather, the institutional machinery comes first and the needs then follow, as do the ways of meeting them. In short, the work to be done is not what the institution responds to but what it create

t "[ilf Northrop Frye should write an essay attacking archetypal criticism, the article would by definition be of much greater significance than an article by another scholar attack- ing the same approach" (Schaefer 5). The reason, of course, is that the approach is not something in- dependent of what Northrop Frye has previously said about it; indeed, in large part archetypal criticism is what Northrop Frye has said about it, and therefore anything he now says about it is not so much to be measured against an independent truth as it is to be regarded, at least potentially, as a new pronouncement of what the truth will hereafter be said to be

what the Waddington-Lewis example shows (among other things) is that merit, rather than being a quality that can be identified independently of professional or institutional conditions, is a product of those conditions; and, moreover, since those conditions are not stable but change con- tinually, the shape of what will be recognized as meritorious is always in the process of changing too. So that while it is true that as critics we write with the goal of living up to a standard (of worth, illumi- nation, etc.) it is a standard that had been made not in eternity by God or by Aristotle but in the profes- sion by the men and women who have preceded us; and in the act of trying to live up to it, we are also, and necessarily, refashioning

Everyone is aware of that risk, although it is usually not acknowledged with the explicitness that one finds in the opening sentence of Raymond Waddington's essay on books 11 and 12 of Paradise Lost. "Few of us today," Waddington writes, "could risk echoing C. S. Lewis's condemnation of the concluding books of Paradise Lost as an 'untransmuted lump of futurity"' (9). The nature of the risk that Wad- dington is about not to take is made clear in the very next sentence, where we learn that a generation of critics has been busily demonstrating the subtlety and complexity of these books and establishing the fact that they are the product of a controlled poetic design. What this means is that the kind of thing that one can now say about them is constrained in advance, for, given the present state of the art, the critic who is concerned with maintaining his or her professional credentials is obliged to say something that makes them better. Indeed, the safest thing the critic can say (and Waddington proceeds in this es- say to say it) is that, while there is now a general recognition of the excellence of these books, it is still the case that they are faulted for some deficiency that is in fact, if properly understood, a virtue. Of course, this rule (actually a rule of thumb) does not hold across the board. When Waddington observes that "few of us today could risk," he is acknowledging, ever so obliquely, that there are some of us who could. Who are they, and how did they achieve their special status? Well, obviously C. S. Lewis was once one (although it may not have been a risk for him, and if it wasn't why wasn't it?), and if he had not already died in 1972, when Waddington was writing, presumably he could have been one again. That is, Lewis's status as an authority on Renaissance literature was such that he could offer readings with- out courting the risk facing others who might go against the professional grain, the risk of not being listened to, of remaining unpublished, of being unattended to, the risk of producing something that was by definition-a definition derived from prevailing institutional conditions-without me

Instead they would be dismissed as being a waste of a colleague's time, or as beside the point, or as uninformed, or simply as unprofessional. This last judgment would not be a casual one; to be unprofessional is not simply to have violated some external rule or piece of decorum. It is to have ig- nored (and by ignoring flouted) the process by which the institution determines the conditions under which its rewards will be given or withheld. These conditions are nowhere written down, but they are understood by everyone who works in the field and, indeed, any understanding one might have of the field is inseparable from (because it will have been produced by) an awareness, often tacit, of these con- ditions

convention is a way of acknowledging that we are engaged in a com- munity activity in which the value of one's work is directly related to the work that has been done by others;

one bothers to define it, except negatively as everything apart from the distractions of rank, affilia- tion, professional status, past achievements, ideological identification, sex, "or anything that might be known about the author"

Everyone agrees that intrinsic merit should be protected; it is just a ques- tion of whether or not the price of protection-the possible erosion of the humanistic community-is too high. In what follows I would like not so much to enter the debate as to challenge its terms by argu- ing that merit is not in fact identifiable apart from the "extraneous considerations" that blind submis- sion would supposedly eliminate. I want to argue, in short, that there is no such thing as intrinsic merit, and indeed, if I may paraphrase James i, "no bias, no merit

. Predict- ably, Schaefer's statement provoked a lively exchange in which the lines of battle were firmly, and, as I will argue, narrowly, drawn. On the one hand those who agreed with Schaefer feared that a policy of anonymous review would involve a surrender "to the spurious notions about objectivity and absolute value that . . . scientists and social scientists banter about"; on the other hand those whose primary concern was with the fairness of the procedure believed that "[jiustice should be blind" ("Correspon- dence" 4). Each side concedes the force of the opposing argument-the proponents of anonymous re- view admit that impersonality brings its dangers, and the defenders of the status quo acknowledge that it is important to prevent "extraneous considerations" from interfering with the identification of true merit (5)

verningthis function is the belief that there must be - at a particular levelof an author's thought, of his conscious or unconscious desire — apoint where contradictions are resolved, where the incompatibleelements can be shown to relate to one another or to cohere arounda fundamental and originating contradiction. Fin

Assuming that we are dealing with an author, is everything hewrote and said, everything he left behind, to be included in hiswork

Bazerman (1984, 1988) has chronicled the evolutionof the acknowledgment during the 19th and 20th centuriesin the journal literature of experimental physics, showinghow it became, to paraphrase Grafton (1997, p. 233), anintegral part of the rhetoric of narration and annotation