Email this article 
Cyberhope or Cyberhype?: Computers and Scholarly Research
Abstract: With the expansion of electronic products and services for libraries, two questions continue to demand our attention. What is the impact of new tools on scholarly research? And how is scholarship affected by libraries' decisions to remake or reorganize themselves so as to accommodate new technologies? This paper looks at the social and political implications of various electronic tools, especially in the humanities, and particularly in literary fields.
Résumé: La prolifération de produits et services électroniques disponibles dans les bibliothèques soulève deux questions qui ne cessent de requérir notre attention. Quel est l'impact de nouvelles technologies sur la recherche savante? Et quels sont les effects sur la recherche lorsque les bibliothèques décident de se refaçonner ou de se réorganiser pour mieux s'accommoder de nouveaux outils technologies? Cet article se propose de discuter des questions sociales et politiques soulevées par un certain nombre d'outils électroniques dans les humanités et surtout dans les domaines littéraires.
As we transform our libraries into postmodern institutions that offer an increasing number of electronic products and services, it is crucial that we understand the way these transformations will affect the disciplines we support. Computers do not simply increase access to scholarly material. They reshape the objects of study, alter research possibilities, and even begin to redefine what constitutes research. This paper is part of an ongoing exploration of what this means, especially for humanities fields. It is also an attempt to explore some of the social and political implications of electronic tools for scholarly research.
Not long ago, in pursuit of these goals, I singled out a special issue of Computers in the Humanities (Fortier, 1993) as illustrative of a misguided determination to harness the computer to literary research (Manoff, 1996). I cite it again as it offers an interesting example of the complexity of the problem. Titled A New Direction for Literary Studies? it is a collection of articles demonstrating and discussing possible uses of literary computation. There is a plaintive subtext: Why are there not more literary scholars doing this kind of work and why does so little of it appear in prestigious literary journals?
One participant proposes that humanists could benefit from the use of high-speed computers if only they would learn to address different kinds of issues. Instead of focusing on particular texts or particular authors, they should formulate new questions involving large corpora because this is where computation can be most productive (Olsen, 1993). My response was that it made no sense to redesign the field of literary studies to accommodate the abilities of high-speed computers. This is precisely how not to make use of a new technology. Four years after the publication of this proposal to jump-start literary computation, it "remains largely dominated by a narrow range of concerns and strategies and has... failed to be a serious presence in even the narrowest fields of literary studies, let alone a factor in the broader intellectual conversation" (Renear, 1995, p. 389). Our success in implementing new technologies will depend on our ability to think critically and learn to recognize the ways in which some computer applications may be fruitful and some may not.
Does the failure of literary computation mean that electronic technology has nothing to contribute to literary studies? Not at all. It has been a godsend to the creators of scholarly editions and has sparked a renewal of interest in textual editing. As Jerome McGann (1995) points out, "Scholarly editions comprise the most fundamental tools in literary studies." Nevertheless the print medium has not been hospitable to the creation or publication of such editions. A print edition of, for example, an early novel, has required editors to either choose one among many available versions or to compile a "best" version from numerous sources. As there are frequently multiple versions of important works of literature, selected pieces of alternate versions might be provided in footnotes, endnotes, or an appendix. But to include in full all extant variants of most literary works between two covers has simply not been an option. Even where there is sufficient space to include a few variants, the textual apparatus tends to be cumbersome and rarely allows for easy comparison of different versions of the same passage.
But electronic editions entail no such constraints. They can typically accommodate all versions of a particular text and can be made to offer split screens where one might be able to view and compare four variants of the same passage in four different windows. Such editions may also incorporate facsimiles of original editions, allowing scholars to study something as close to the original artifact as one might get without traveling to a special collection to view it. And though there can be no substitute for handling primary source materials, computer facsimiles allow for the magnification and manipulation of screen images for very close scrutiny of particular documents.
The ability to represent multiple variants of primary texts has come at a moment when theoretical concerns have focused on textual instability and multiplicity (Tanselle, 1996, p. 52). Just when literary theory has subverted the notion that it is either desirable or possible to create an "ideal" version of a text, the technology is allowing scholars to create editions where they are not forced to choose between variants or assemble best possible editions. According to Ronald Tetreault (1996), even the idea of a "base text for collation" is "a relic of print culture." This is an instance where the advantages of the electronic medium are incontrovertible.
Tetreault, who is editing a new electronic edition of Wordsworth to be published by Cambridge University Press on CD-ROM, argues that the print medium is inadequate to present the work of someone like Wordsworth who was an inveterate reviser. His poems were essentially works in progress. Print editions are merely a snapshot of a particular moment and cannot map the transformation of Wordsworth "from Romantic rebel to Victorian sage." Wordsworth scholarship has acknowledged that the 70-year-old poet still revising his early works bears little resemblance to the 20-year-old who first composed them. Tetreault claims that an electronic product "may be the most effective way yet to represent Wordsworth in development." Not only can it include numerous variants so as to represent change over time, but it can also provide linkages between versions for easy comparison. Whereas the print medium is more conducive to presenting a stable self, the electronic medium lends itself to representing an evolving and fragmented self, much more in sync with contemporary theories of identity.
But even as technology is providing faster and easier ways to navigate between texts and parts of texts, this activity demands a new kind of literacy. Users must develop a familiarity and level of comfort with reading and manipulating documents on a computer screen. They may find it easier to move between passages in an electronic edition, but they may also feel that the context of any particular passage may be much less immediately obvious than in a printed volume where one sees where any page is in relation to the whole. Many people do not have the ability to look at a computer screen and intuit all the possible options available at a given moment. Many find it difficult to follow a path through a series of computer documents without getting lost and without losing the ability to retrace their steps. Even those who have grown up with computers need to develop more advanced forms of computer literacy in order to make the best use of hypertext archives.
Electronic editions and software for the manipulation of humanities texts are still in the early stages of development. Although Tetreault has some good things to say about the DynaText software available from Electronic Book Technologies that is the basis of his forthcoming edition of Wordsworth, he nevertheless declares it "a blunt instrument when it comes to the presentation of poetry," unable to provide the kind of line-by-line navigation he would like. Current editors of these new editions are being forced to improvise in ways that will perhaps be less necessary as more software specifically tailored to literary applications becomes available. Work of this sort is being done at the Princeton / Rutgers Center for Electronic Text in the Humanities and at a number of similar institutions.
In order to distinguish between more or less productive applications of electronic technology, I propose that we attempt to discriminate what I will call technology-driven applications from content-driven applications. Of course, the distinction is one of emphasis or degree. Nevertheless, I think it might be useful to reflect on the difference between technology that meets a pre-existing need (for example, the ability to represent multiple versions of a text simultaneously) and technology in search of a need (for example, literary computation requiring projects seeking to analyze huge bodies of text). Douglas Greenberg, of the American Council of Learned Societies, has drawn a similar distinction. He claims that humanists must "continue striving to adapt technology to fit their values as scholars and teachers, rather than permitting technology to reshape their values" (1993, p. 6). Undoubtedly, scholars' needs will evolve with the growth of technology, but surely it makes more sense to focus our energies on providing tools for which a need already exists, rather than creating tools in search of both needs and users.
Nor can scholars make productive use of new tools just because we make them available. This fact is highlighted by a recent project conducted by the Getty Information Institute (formerly the Getty Art History Project). In order to study "how advanced humanities scholars operate as end users of on-line databases" (Siegfried, Bates, & Wilde, 1993, p. 273), a small group of such researchers was offered a few hours of Dialog training. (Established in 1972 and recently taken over by Knight-Ridder, Dialog is a leading provider of on-line searching and data retrieval, supplying access to several hundred databases.) They were then provided fully subsidized access to all humanities-related Dialog databases. Complete logs were kept of all searches and scholars were interviewed "in depth." Though not the ostensible purpose of the study, to me the most striking finding was that Dialog searching proved to be of limited interest or use to the humanities scholars involved. As Marcia Bates, one of the project's co-ordinators, has observed: "Though the Getty scholars had the better part of an academic year to search and could do all their searching for free, only eight of the 27 scholars searched more than two hours total during all that time (1994, p. 335). And, she further notes, "Five scholars did no searching on their own after training" (1996, p. 516).
The study raises several important questions, some perhaps unintended by its sponsors. What struck me as I read the six articles analyzing a massive accumulation of data was: why on earth was the decision made to concentrate so much time, money, and energy on examining Dialog searching by humanists? As the articles acknowledge, "It is difficult to master and retain Dialog searching well enough in a fairly short time to be really successful with it" (Bates & Wilde, 1995, p. 17). And, more importantly, Dialog databases are not strong in humanities resources:
Available databases often proved not to be what the scholars wanted. They wanted access to more European literature, earlier literature, and primary research materials. Most databases cover only literature going back to the early 1970s, which is often just a small part of the range that a humanities scholar is interested in. (Bates, 1996, p. 518)
None of this should have been news to the people who selected Dialog as the basis of their project.
Certainly, some explanation of their choice of Dialog would be in order. How much ought one to generalize from the results of a study of a set of databases difficult to search and missing important material? This seems to be a case where the assumption was made that an electronic product would inevitably prove useful to scholars if only they could be taught to use it. But the response of many of the participants was, in effect, a polite "no thanks." Librarians who assume that Dialog or any other product is necessarily meeting the needs of their constituents might well look for concrete evidence.
In a recent article addressing the use of information technology by humanists, Stephen Wiberly & William Jones speculate that the limited use such scholars make of on-line databases may have to do with the high volume of citations found in their professional reading (1994, pp. 505-506). Humanities scholarship tends to include so many bibliographic references that they may feel no need to initiate systematic database searches. Humanities scholars may assume that in their specialized fields such as Renaissance drama or U.S. colonial history, they would inevitably come across a reference to any important book or article of interest. Wiberly & Jones also convincingly argue that "humanistic evidence is not easily categorized and entered into a relational database and not readily subjected to quantitative measure or statistical analysis." Furthermore, they claim that there are no generally accepted software packages for analyzing the kinds of evidence humanists collect (1994, p. 505). They conclude that we must not assume that information technology provides the same solutions for everyone.
In any case, the narrow coverage afforded humanities materials in Dialog databases should remind us how important it is to consider precisely what we are making available through our new electronic tools and, perhaps more importantly, what we are not making available. In 1995, I found it a cause for some concern that electronic databases in academic libraries were getting disproportionately heavy use at the expense of print indexes, despite their often limited coverage. We seemed to be devoting a considerable amount of our resources to providing electronic access to some fields rather than others. Moreover, the products we did offer contained little in the way of alternative perspectives. Our CD-ROMs were providing relatively little oppositional material, either of the Left or Right. The technology had the effect of reducing access to alternative or non-traditional periodical literature as students increasingly determined that consulting print indexes was too time-consuming (Manoff, 1996, pp. 220-221).
In some ways the situation has improved although the larger issues have yet to be resolved. Most major print indexes have been converted into CD-ROM products and new indexes have been created as well. Vendors are now searching out less mainstream indexes to create or digitize and hawk to the library market. Libraries therefore have many more options than they did in the recent past. The Alternative Press Index is now available on CD-ROM; so is the Black Studies Database (The Kaiser Index to Black Periodicals since 1948); and Ethnic Newswatch provides full-text access to scores of minority newspaper and magazines. Two years ago there were no electronic indexes specifically devoted to women's studies; now there are four (Contemporary Women's Issues, CD-ROM, Beachwood, OH: RDS, Inc.; Women's Resources International, CD-ROM, Baltimore, MD: NISC; Women's Studies on Disc, CD-ROM, New York: G. K. Hall & Co.; and Women "R," CD-ROM, Stanford, CT: Softline Information, Inc.).
The boom in women's studies indexes has created its own problems. The CD-ROM market will not sustain four such indexes for long, even if they do cover different materials. Some of these may not survive long enough for libraries to sort out which of them are the best products and the most appropriate for their institutions. Will shoestring operations like the Alternative Press Index survive in the electronic age? We cannot be certain. What is clear is that many producers of less mainstream material exist pretty close to the edge. If they do not get library support they will probably not last.
But even if the more diverse resources survive in CD-ROM format, access may continue to be problematic unless their producers can keep up with the technology curve. Right now, academic libraries are most interested in offering electronic databases through their own campus networks so students and faculty can consult them on their own computers regardless of location. No doubt, this provides greatly expanded access. Most libraries prefer these databases to be Internet accessible because this means they do not have to mount tapes locally or provide much technical support.
Although some small producers of electronic databases have found it relatively easy to establish an Internet presence, others are struggling and have so far been unable to create Web versions of their CD-ROM products. As many libraries are not much interested in mounting additional CD-ROMs on crowded local area networks, producers of new CD-ROMs unable to quickly meet the demand for Internet access may not survive.
What is the responsibility of libraries in such cases? It is certainly in the interest of libraries to contribute to the sustainability of diverse resources. But libraries are also driven by technology-related requirements and increasingly choose not to buy products for reasons having less to do with their content than with their electronic format. Traditionally, libraries have been able to offer more marginal and diverse materials by acquiring small and alternative press titles. Many such presses depend on library dollars. But it has proved more difficult for libraries to collect alternative materials in electronic formats. They will therefore have to devise new strategies if they are to continue to maintain broad collections and access and not cede their role as a major outlet and support for independent publishers.
Libraries will also need to continue to apply pressure to the producers of mainstream electronic databases to broaden their coverage. All-purpose indexes available on campus networks are often the first and sometimes the only databases consulted by students. It is therefore critical that producers of databases like OCLC's FirstSearch, Expanded Academic Index, and UnCover (formerly Carl UnCover) be made aware of our concerns. We need to pay attention to what particular areas of study are better or less well served by these multidisciplinary indexes. Librarians can be effective in pressing for the inclusion of additional material. OCLC, for example, agreed in 1996 to add a number of women's studies journals to its ArticleFirst and ContentsFirst databases at the urging of the Association of College and Research Libraries (ACRL) Women's Studies Collection Development Committee. And several task forces of the American Library Association's (ALA) Social Responsibilities Round Table have effectively lobbied index publishers to broaden their coverage or have been successful in convincing vendors to include more independent presses in their approval plans.
The proliferation of electronic resources is slowly eroding the traditional library model where collection decisions are made by individual subject specialists. When large expensive electronic databases become available, especially multidisciplinary databases, collection decisions require input from a number of subject specialists and also staff with technical expertise as well as staff knowledgeable about licensing and copyright. Technical staff may simply declare certain products so difficult to mount as to eliminate them from the running. Many libraries are turning over the decision-making for large electronic databases to committees representing various kinds of expertise. Broader input may lead to better decisions; but it also leads to the bureaucratization of the collections process, a dilution of the input of subject specialists and all the expeditiousness of committees divvying up scarce resources.
A further bureaucratic complication is that more libraries are participating in consortia as a way of achieving price breaks for expensive electronic products. As libraries enter into discussions with one another over which products to purchase as a group and as they work out the details of site-licenses, passwords, numbers of simultaneous users, etc., individual libraries are further drained of authority over decisions. In the brave new world of electronic resources, it is becoming less clear precisely where responsibility for choosing important new materials lies.
However, consortia may also provide a way for libraries to use their joint purchasing power to support progressive electronic products, to apply pressure to vendors to improve other products, and to write licensing agreements that represent their interests. If libraries can learn to make the most of their very substantial combined purchasing power, they may be able to create an environment more conducive to the building of effective collections and access.
We also need to be aware that the Internet is radically altering the kinds of material available to students and researchers. Regardless of the size of their libraries and whatever their intentions, institutions with network access are making available a great deal of progressive as well as reactionary and alternative material. In this way, the Web, in particular, has begun to compensate for some of the limits of the print environment by becoming a place to find material previously excluded from intellectual debate or material simply deemed unmarketable by commercial presses. Although many people are aware that there is a considerable amount of junk on the Internet, few are aware of the amazing breadth of material available. There are countless Web sites devoted to ethnic, minority, and religious groups; to every imaginable political perspective and ideology; to interest groups and disciplinary subspecialties of all sorts. Many of these sites do contain valuable and interesting information, although plenty do not. As our local and national collections shrink as a result of serial inflation and as we collect more narrowly because we cannot afford to do otherwise, we should recognize and make the most of the compensations offered by the Net.
Individual librarian subject specialists can cull the best Internet resources from their disciplines and assemble impressive collections of material. Library organizations like ALA's ACRL Western European Specialists Section (URL: http: //www.lib.virginia.edu /wess) and its Women's Studies Section (URL: http: //www.library.yale.edu /wss) have been putting together Web pages with important discipline-related material. Also the WWW Virtual Library (URL: http: //vlib.stanford.edu /Overview.html) provides very large collections of discipline-specific Internet resources.
To cite a specific example of the value of Internet resources, at MIT I maintain a Web site with links to material we could not dream of offering in the print environment. For years the MIT libraries have been approached by students wanting us to subscribe to newspapers and magazines in their native languages from their cities and countries of origin. But, for the most part, all we could afford was material in the few languages taught at MIT, primarily in support of course work. In 1996 I took over from a departing faculty member in the department of Foreign Languages & Literature a Web page with over 600 links to newspapers and magazines in eight languages (Manoff, 1997). I cannot say I was enthusiastic about the prospect of maintaining it, but it was a perfect complement to our rather weak collection of foreign language magazines. Our statistics suggest that it is the most popular page on the MIT libraries Web site, and I get a good deal of e-mail from people telling me how much they appreciate the service. All it costs is my time and storage space on the libraries' Web server. (The conversion of the newspaper Web site from the format in which the faculty member had set it up into a clearer and more attractive site took about 10 hours. Upkeep requires about an hour per week. Creating a Web site like this from scratch with over 600 links could easily take about 60 hours. Many more hours than that have been devoted to the site thus far.)
We are just beginning to see the effects of the availability of a broader range of resources in particular disciplines. Looking again to literature, perhaps the most profound way in which the Internet is influencing this field is in the area of canon formation. Now that many more authors and works are being made available electronically than were available in print, critical attention is being focused on the ways in which this may be helping to redefine how a particular period or genre is studied.
At the December 1996 Modern Language Association (MLA) conference, a session was devoted to exploring these kinds of issues. Titled Reconfiguring Romanticism in the Information Age, the papers presented at the session addressed the convergence between the availability of on-line resources and canon revision. In the proposal for the session, Alan Liu (1996) describes how there is now a critical mass of electronic projects in Romanticism being provided by places like the University of Virginia, Oxford University, the University of Alberta, and the University of Pennsylvania. In the paper environment, faculty depend on textbook anthologies to define a period or genre. Anthologies constitute a statement about who are the central figures and which works are deserving of attention. In Romanticism, which has been undergoing canon revision for several years, there has been a move to include more women and other previously underrepresented authors in new anthologies. But the constraints of a print volume mean that there is just so much space; devoting more of it to previously excluded authors means devoting less of it to "major" figures. The results are often an unhappy compromise. But in electronic archives or "anthologies," no such compromise is necessary. Including the work of additional authors does not require eliminating the work of others. This has raised a number of intriguing questions that were addressed by the MLA session:
... what will happen to the canon in a medium that at least in principal does away with space limitations, rethinks the logic of the "page"... diminishes the role of capitalized middlemen (editors and publishers), has no permanence, and resists hierarchical structure. For example, who will be canonically "marginal" on a Web that technically has no "center"? (Liu, 1996)
Having experienced the dilemma of the out-of-print status of material they would like to teach, many faculty members are not just sitting around waiting for publishers or librarians to solve their problems; they are taking matters into their own hands. So, for example, Elizabeth Fay at the University of Massachusetts, is putting together what she calls the Bluestocking Archive (URL: http: // fay.english.umb.edu /archive / toc.html). She is mounting a collection of texts on the Web by or relating to eighteenth-century British women authors who were contemporaries of the better-known male Romantics. Fay claims she began her archive because the material "was no longer in print and not otherwise available for classroom purposes." She considers it to be "both a scholarly and a teaching tool" (Fay, 1996).
Like many others who are developing electronic resources, Fay sees their potential to transform or reconfigure the fields in which they are created. She describes her archive as providing a way to "rethink the bounds of the Romantic period" by making clearer the connections between the Bluestocking Circle and the work of High Romanticism. It allows for a consideration of how these women authors created a female version of Romanticism as well as the tools for reconsidering "basic beliefs about the autonomy, rebellion, and masculinity of Romantic texts." Electronic access to material not previously available is thus helping to rewrite the history of various literary periods.
Many other faculty and graduate students are assembling alternative collections on the Web. Paradoxically, it is this new technology that is allowing access to old and forgotten works of literature that can now be taught for the first time. However, the simple availability of more material is no panacea. Electronic archives may be broader than print anthologies, but a semester still has the same number of weeks. Faculty must still decide what to include on a syllabus. Nevertheless, they do have much more choice about which texts to assign and students have more choice about what to read.
One of the most well-known and systematic attempts to recover more marginal material is the Women Writers Project at Brown University (URL: http: //www.wwp.brown.edu /wwp_home.html). Devoted to the work of authors from 1300 to 1850, this is a major undertaking. So far, they have digitized about 200 texts, most previously out of print. They are also pioneering the establishment of standards for textual encoding as well as contributing to research on literary databases and text management systems.
Oxford University Press has published 10 printed volumes from the Brown project and other volumes are planned for this year and next. All 200 titles are available in either electronic or paper format directly from the Brown project. Unfortunately, the textbase will not be available over the Internet for another three years. Nevertheless, the accessibility of this material for teaching and other scholarly purposes has had a significant impact on the discipline. According to Robert Hamm & Rebecca Wood (1996a), texts "are now being added to the Renaissance canon" that surfaced only as a result of the Women Writers Project.
Increasingly the Web is becoming a place to explore questions of marginality and canonicity, rather than just being a place to find alternative resources themselves. And again individual scholars or small groups of scholars are building Web sites devoted to addressing these issues. A few examples include Electrifying the Renaissance (Hamm & Wood, 1996b), The Victorian Canon (Jones & Raley, 1996), The Romantic Chronology (Mandell & Liu, 1997), and Women of the Romantic Period (1997). All of these sites are being used in support of teaching.
The Victorian Canon Web site, for example is devoted to exploring questions of taste and aesthetic value and is "concerned with the shifting categories of the `high' and the `low' or the `canonical' and the `noncanonical' " (Jones & Raley, 1996). It raises some interesting questions about the current popularity of Victoriana; it poses questions about the relation of the Victorian canon to Victorian kitsch and proposes syllabi to address these questions. In doing so, it demonstrates the connection between the presence on the Web of a great deal of alternative material and the presence of many sites addressing the implications of this proliferation. Over time, these new approaches to assembling texts and secondary material will inevitably reshape the way scholars conduct their research. Julia Flanders (1996), the text-base editor of the Brown project, makes a similar argument about the transformative effects of electronic technology: "The computer can no longer be regarded simply as a tool which assists in doing what we already do, but must be understood as a medium in the true sense: an integral part of our systems of communication, with the potential for profound influence on our habits of thought and work."
Recognizing this potential, a number of small groups of scholars are trying to harness electronic technology to create new models for both publishing and research. One of the newer and more ambitious of such projects is a Web site called Romantic Circles (URL: http: //www.inform.umd.edu:8080/ RC). Steven Jones (1996), one of the general editors, declares it to be nothing less than an attempt to "help transform the way Romanticists conduct their business, both in the classroom and in their work as critics and scholars." The Web site is devoted to the study of Byron, Mary Shelley, Percy Shelley, Keats, and their contemporaries. Its goal is to provide not only new scholarly electronic editions, but also a variety of other scholarly resources and space for critical exchange.
One striking aspect of the Romantic Circles Web site is the collaborative nature of the enterprise. The editors are determined to set a precedent in literary studies for the kind of collaborative and "radically distributed" work much more common in the sciences. In order to make this possible, they are attempting to create a way to offer traditionally accepted "means of vetting scholarly work -- namely peer review and editorial control" (S. Jones, 1996) in the electronic environment. To this end they have assembled a distinguished group of editors and advisors:
Especially given the slipshod quality of much do-it-yourself "publishing" on the Web, it seems absolutely essential that users know that all texts and research tools at our site have been produced under careful editorial control. Local Area Editors are responsible for guiding and vetting the production of texts and resources in their sections, and the General Editors have similar responsibility for the site as a whole. Area Editors, along with the General Editors and the long list of prominent Advisors, form the kind of Editorial Board that is recognized by the discipline. (S. Jones, 1996)
Not surprisingly, this project is raising questions not unlike those being addressed by many of us concerned with the future of electronic scholarship. Jones identifies four central issues confronting the project: (1) How are editorial control and common standards maintained while creating a large scholarly archive with many contributors? (2) What role can /should such a site play in the electronic marketplace? What kind of relationship might it have to university presses? (3) Will it be possible to remain a scholarly Web site in what is rapidly becoming a mass medium? and (4) How will a site like this affect the current canon and system of hierarchies in its field?
One particular advantage of the model being offered by the Romantic Circles project is a commitment to keeping scholarly production within the control of its producers. Though the editors anticipate that there may be some resources on their site for which there would be a fee, they do not envision anyone making excessive profits off their labours. This could also be said for all the humanities projects mentioned in this paper. Libraries should welcome and encourage this new recognition on the part of faculty that they bear some responsibility for the way their scholarly output is marketed and distributed.
So while administrators, publishers, and librarians debate the future of scholarly publishing, others are writing their own blueprints. Some institutional sites experimenting with new ways to provide authoritative resources include the Institute for Advanced Technology in the Humanities (URL: http: //jefferson.village.virginia.edu / ) at the University of Virginia, Labyrinth: Resources for Medieval Studies (URL: http: //www.georgetown.edu /labyrinth / labyrinth-home.html) at Georgetown University, Project Bartleby (URL: http: //www.columbia.edu /acis / bartleby / ) at Columbia, the American Memory project (URL: http: //cweb2.loc.gov / ) at the Library of Congress, the Victorian Women Writers Project (URL: hhtp: //www.indiana.edu /~letrs /vwwp / index.html) at Indiana University, the American Verse Project (URL: http: //www.hti.umich.edu /english /amverse) of the University of Michigan Press, and Project Perseus (URL: http:www.perseus.tufts.edu / ) at Tufts University.
Two particularly salient features of many of the new humanities initiatives are the desire to experiment with new forms of collaboration and to pioneer new models of graduate training. The Orlando Project (URL: http: //www.ualberta.ca / ORLANDO), at the University of Alberta, for example, is seeking to create new models for research. As described on their Web site, the final product will be a five-volume scholarly history of women's writing in the British Isles in both printed form and on CD-ROM. The project is an experiment in team research, with its co-investigators, research collaborators, and advisory panel drawn from universities in Canada, the United States, and the United Kingdom. Graduate students are participating in authoring, researching, and data entry, and they are gaining new kinds of experience in team research and advanced humanities computing. Similarly, the Women Writers Project (URL: http: //www.wwp.brown.edu /wwp_home.html) at Brown, described above, is "exploring the educational advantages of integrating undergraduate and graduate students into a technology-intensive interdisciplinary research project" (Overview of the Brown Women Writers Project, 1996). Both of these are devising new apprenticing models for humanities scholarship as well as new forms of collaboration.
Although initiatives like these are clearly cause for interest and excitement, it is important that we maintain our perspective on what we can accomplish with technology. At the moment, major impediments to the use of electronic text are the limitations of the computer screen. For the foreseeable future, print will continue to be much easier to read. So despite the fact that electronic text provides many more options for interaction, print will remain a preferred means of reading for material of any length. For this reason, even projects like the forthcoming Pennsylvania Electronic Edition of Frankenstein on CD-ROM (due in 1998) will be accompanied by a "thoroughly traditional paperback of the 1818 text bundled with the computer disk." This electronic Frankenstein will perhaps be the most heavily annotated edition ever produced, running to the equivalent of about 20,000 pages in print without its multimedia material (Lynch, 1996). And yet its producers believe that users will require a paper volume for portability, extended reading, and the opportunity to write in it as they see fit.
New technology is posing new challenges for all of us. Librarians, in particular, will need to educate themselves so as to be less dependent on technical experts who would define the library of the future in the narrower terms available to them. Although many libraries have offered their staff training in particular applications of technology, they have not provided the kind of comprehensive information technology training that Michael Harris & Stan Hannah (1993) claim is more often available in the private sector. "The almost universal lack of an information technology program in libraries continues to be a major obstacle to the effective utilization of information technology" (1993, p. 31). I agree with Harris & Hannah that we need to develop better and broader information technology training programs for librarians, but I think that it is equally important to cultivate staff with an understanding of the information needs of all the disciplines we support. As libraries have come to place a premium on managerial and technical skills as a way to cope with the new electronic environment, it is essential that we also recognize the value of subject expertise and faculty input.
Librarians are constantly being exhorted to embrace change, to stop resisting technology, and to look to the future -- specifically technology -- as a solution to the complex problems facing libraries today. But this is simply not enough. More than technical skills and an appreciation of technology will be required to make informed decisions. Our success in implementing new technologies will depend on our abilities to discriminate between the potentially useful and the merely new. This will require a deep-enough understanding of the nature of our scholarly communities to gauge the potential value of all the electronic tools and products that vendors keep telling us we cannot do without. A healthy scepticism about particular products and services will serve us well. Our greatest libraries were built by scholars and librarians with a vision of what both present and future generations would require to conduct their research. This kind of vision will also be required to build even the most virtual of future libraries.
References
Bates, Marcia J. (1994). The design of databases and other information resources for humanities scholars: The Getty Online Searching Project Report No. 4. Online & CDROM Review, 18(6), 331-340.
Bates, Marcia J. (1996). The Getty end-user online searching project in the humanities: Report No. 6 -- Overview and conclusions. College & Research Libraries, 57(6), 514-523.
Bates, Marcia J., & Wilde, Deborah N. (1995). Research practices of humanities scholars in an online environment: The Getty Online Searching Project Report No. 3. LISR, 17(5), 5-40.
Fay, Elizabeth. (1996). MLA 96 session on the Web and the canon: The bluestocking archive. URL: http: // fay.english.umb.edu /archive / fayblue / talk.
Flanders, Julia. (1996). Editorial method and the electronic text. URL: http: //www.wwp.brown.edu / NASSR / Argument.html
Fortier, Paul (Ed.). (1993). A new direction for literary studies? [Special issue]. Computers and the Humanities, 27(5-6).
Greenberg, Douglas. (1993). Remarks. In Technology, scholarship and the humanities: The implications of electronic information (pp. 35-36). Summary of proceedings of conference, Technology, Scholarship, and the Humanities: The Implications of Electronic Information, September 30-October 2, 1992. Santa Monica: American Council of Learned Societies and the J. Paul Getty Trust.
Hamm, Robert, & Wood, Rebecca (Eds.) (1996a). Canon revision and the Renaissance. URL: http: // humanitas.ucsb.edu / depts /english /coursework /rar/crandr.html
Hamm, Robert, & Wood, Rebecca (Eds.) (1996b). Electrifying the Renaissance: Hypertext, literature, and the World Wide Web. URL: http: // humanitas.ucsb.edu / depts /english /coursework /rar/
Harris, Michael, & Hannah, Stan A. (1993). Into the future: The foundations of library and information services in the post-industrial era. Norwood, NJ: Ablex Publishing.
Jones, Jennifer, & Raley, Rita. (1996). The Victorian canon. URL: http: // humanitas.ucsb.edu / depts /english /coursework /raley /canon /victoriana.html
Jones, Steven. (1996). The Romantic Circles project and emergent forms of scholarly production on the Web. URL: http: //orion.it.luc.edu /~sjones1/nassr 96.htm
Liu, Alan. (1996). Reconfiguring Romanticism in the information age. URL: http: //humanitas.ucsb.edu / liu /mla 96 / descrip.html
Lynch, Jack. (1996). Workshop of filthy creation, cyberspace division. URL: http: //www.english.upenn.edu /~jlynch / Frank / frank.html
Mandell, Laura, & Liu, Alan (Eds.). (1997). Romantic Chronology. University of California at Santa Barbara. URL: http: // humanitas.uscb.edu /projects /pack /rom-chrono /chrono.htm
Manoff, Marlene. (1996). Revolutionary or regressive? The politics of electronic collection development. In R. Peek & G. Newby (Eds.). Scholarly publishing: The electronic frontier (pp. 215-229). Cambridge & London: MIT Press.
Manoff, Marlene. (1997). Foreign language newspapers and magazines. MIT. URL: http: //libraries.mit.edu / humanities / flnews
McGann, Jerome. (1995). The rationale of hypertext. University of Virginia. URL: http: // jefferson.village.virginia.edu / public / jjm2f / rationale.html
Olsen, Mark. (1993). Signs, symbols and discourses: A new direction for computer-aided literature studies. Computers and the Humanities, 27(5-6), 309-314.
Overview of the Brown Women Writers Project. (1996). Brown University. URL: http: //www.wwp.brown.edu /overview.html /
Renear, Allen. (1995). Understanding (hyper) media: Required readings [Review essay]. Computers and the Humanities, 29(5), 389-407.
Siegfried, Susan, Bates, Marcia J., & Wilde, Deborah N. (1993). A profile of end-user searching behavior by humanities scholars: The Getty Online Searching Project Report No. 2. Journal of the American Society for Information Science, 44(5), 273-291.
Tanselle, Thomas. (1996). Reflections on scholarly editing. Raritan, 16(2), 52-64.
Tetreault, Ronald. (1996). Electrifying Wordsworth. URL: http: //is.dal.ca /~tetro /nassr 96 /wordsworth.html
Wiberly, Stephen E., & Jones, William G. (1994). Humanists revisited: A longitudinal look at the adoption of information technology. College & Research Libraries, 55(6), 499-509.
Women of the Romantic Period. (1997). URL: http: //www.cwrl.utexas.edu /~worp