THE modern university has been criticized almost since its inception for its indifference to students. Periodically university presidents and critics of the university call for faculty members to take teaching more seriously. The presidents of Stanford and Harvard recently took their faculties to task for their neglect of teaching. Over the years none of these exhortations has had a noticeable effect. They are accompanied by reiterations that the primary function of a great university is research. What's needed, critics say, is a ``better balance between teaching and research.'' Few of these exhorters have seriously questioned the ``publish or perish'' canon that so distorts the educational mission of the ``great universities.'' The issue is not that of achieving a ``better balance between research and teaching,'' but of exposing the fallacies behind the insistence on publishing as a condition for promoting faculty. The most fundamental fallacy is the notion that any conscientiously done piece of research is by definition a significant contribution to knowledge. This is false. Even if it were true that each little increment of research was a potentially useful addition to our knowledge in a field, the fact is our repositories of information are inundated by a flood of research. We are suffering in every disciplinary field from information overload. Research is driven, in the main, not by a passion for learning, but by the need-to-publish-in-order-to-be-promoted syndrome. In the face of what amounts to a public scandal, the pressures on academics to publish what is, in large part, mediocre or useless research increase rather than diminish. How best to dramatize this situation so that long overdue reforms can begin? One way may be to demonstrate the high cost of the present system. There is general recognition in the US of the accelerating cost of a college education. Perhaps, if it can be shown that a great part of this huge expense is not only unnecessary but an impediment to education, the public may be sufficiently aroused to demand reform. William Schaefer, a professor of literature at UCLA, recently reflected on his experience as editor of PMLA, the flagship journal of the Modern Language Association. Schaefer estimates that in a two-year period over 1,000 scholarly articles were submitted for publication - 500 a year. Only 50 were judged worthy of publication - about 5 percent. After reading this horror story, I thought to ``cost out'' the faculty time spent on research (what I term Faculty Research Time, FRT) for the 950 rejected articles). If we calculate that each article took two-thirds of a faculty member's time for six months, and said faculty member's salary was $50,000 per year, we get a figure of $16,000 in Faculty Research Time per article. Multiplying $16,000 by 950 articles equals $15,400,000 worth of faculty time, time that is, in essence, taken away from students. There are some 1,000 scholarly journals in the field of language and literature alone. Using the FRT formula for these 1,000, we get a figure of roughly $800 million plus estimated editorial costs of $250 million,for a total of more than a billion dollars. Remember, these are costs associated with scholarly journals in one field of the humanities - albeit the largest field. If we apply Smith's Faculty Research Time formula to the thousands of journals in other areas of the humanities and social sciences, the billions in cost are compounded. Appropriately included in the cost of scholarly journals is the expense to research libraries of ``acquisitioning'' and cataloging them. Research libraries are forced to pay extravagant charges to subscribe to scholarly journals. The average cost of subscriptions in the humanities, for example, rose from $90 in 1977 to $255 in 1989. In the social sciences it increased from $145 to $418 - about 280 percent. The high cost to libraries of subscriptions is, in practical fact, a form of coerced subsidization. There are some 300 research libraries in the US. If they each spend $700,000 to $3.5 million a year on scholarly journals and serials, we get an average of $1 million times 300 libraries - or $300 million more. In the sciences one finds even grosser inflation. Here the information overload is staggering. Walter Stewart of the National Institutes of Health estimates that over a million articles a year are published in the sciences. Most go unread. As long ago as 1960 a UNESCO Bulletin for Libraries noted ``the multiplicity of journals [in the sciences] results in a scattering of papers which makes it impossible for the scientist to keep informed of new developments, impossible for a library to cover the field completely, and impossible for abstracting services to include all the relevant articles.'' A ``highly specialized periodical is generally of interest to only 10 percent'' of those working ``in the subject area,'' one Bulletin author noted. The pressure to publish in the sciences may contribute to the growing number of cases of cheating or ``cooking the evidence,'' i.e., making false or exaggerated claims that cannot be validated. While scholarly journals consume half or more of the budgets of research libraries, the heart and soul of academic publishing, especially in the non-science fields, is the university press. In the humanities and the social sciences, a published book is usually essential for promotion to tenure. At one university the formula is said to be: one book plus six articles equals tenure; two books plus four more articles are required for a full professorship. The American Association of University Presses recently conducted a survey of monographs published in the humanities and social-sciences fields by 45 of its member presses between 1978 and 1988. The total came to 6,922. Several facts emerged. First, the number of titles published yearly had increased over the period while the average sales per volume had declined 33 percent, causing, in the words of the report, ``serious financial problems.'' The report, incidently, does not even touch on so-called ``fractional publishing,'' where only a few dozen copies of a particular monograph are printed, with more printed ``on demand.'' The cost of producing the 6,922 scholarly works published during the period 1978-1988 may be calculated using my FRT formula. The cost of research time, again with the median salary of $50,000, for the five years it takes a faculty member to complete a book-length manuscript is roughly $165,000. If the manuscript is selected for publication, publishing costs will run to some $10,000 per monograph. We thus have a cost of $175,000 to $200,000 per monograph regardless of its quality. Out of the 6,922 monographs published during that decade, I suspect it is generous to concede that a third (and these in the most neglected fields - China, Middle East, etc.) are useful. That leaves us with an excess of 4,600 works at a cost of $175,000 each, or $850 million. A recent survey of faculty publishing indicated that one-third of the manuscripts published by university presses in 1973 had been submitted to four or five presses. A number of authors had given up after submitting to a median number of 11 publishers. Manuscripts ``temporarily shelved'' - whose authors abandoned the effort to get them published - was 33 percent. If we assume that the number of faculty hours spent producing a manuscript that never finds a publisher is equivalent to the number of hours spent producing a marginally publishable monograph, that's $250 million for those poor abandoned monographs that will never see the light of day. Then there is the related matter of faculty salaries. In the period from 1970 to 1989, the median salary of a full professor at Harvard rose from $42,000 to $93,000, an increase of 112 percent over a period when the rate of inflation was about 50 percent. Other ``great universities'' have followed suit. A professor from an East Coast university, for example, was recently lured to the University of California at Irvine for a salary of $108,000 per year, plus $20,000 a year for research expense. High salaries are almost invariably accompanied by lighter teaching loads. Finally, we should not overlook the enormous sums distributed by various federal agencies to universities for research of interest to the government, most of it coming from the Department of Defense and NASA, plus substantial sums from the National Institutes of Health. Billions of dollars flow from these sources. Many critics consider these grants a corruption. Certainly they can be factored in as a cost to the general public of the vastly inflated research enterprise. Colleges charge a preposterously high overhead for administering such grants. Stanford University is presently being investigated by the federal government for gross misuse of such funds. To summarize: American universities are caught in an expensive and irrational process, one in which there is no justification between costs and returns. It is a ritual activity with little relation to the classic notion of research and scholarship, or to any identifiable intellectual mission. This tide of mediocrity threatens to overwhelm research libraries and information systems with the sheer weight of indigestible and largely inaccessible facts. It also diverts money and attention from areas needing more scholarly attention. There are numerous victims of this cancerous growth - professors, libraries, taxpayers, parents who pay preposterous tuitions, and, most important of all, undergraduate students who don't get the attention they deserve. The question is: What is to be done? Will we go on as we have until the great bloated mass collapses of its own weight, or do we have the resolution to effect the deepest, most difficult of reforms - the way we think about the aims of higher education? -PATHNAME- /usr/local/etc/httpd/plweb/DBGROUPS/paper/database/tape/91/mar/week10/esmith.