CHAPTER ELEVEN

ECONOMIC AND LEGAL ISSUES

We have postponed until now full discussion of some of the more important and difficult issues to be resolved if the model of scholarly communication described here is to be viable. Some issues, principally of a technical nature, were identified earlier, in the discussion of the electronic publication of full texts. The issues discussed here might be described instead as more social or cultural in character, and their resolution will involve the participation of others besides specialists in computing, telecommunications, and other technical aspects of the electronic dissemination of information.

TENURE AND NONTRADITIONAL PUBLICATIONS

Among these factors is the nature of the reward system.[1] Scholarly achievement, a crucial basis for appointment and promotion at institutions of the type studied here, is measured by established methods and assumes publication by means of traditional vehicles: the scholarly monograph and journal article. What incentives are there for disseminating one's work over the Internet when the rewards are for publication by the traditional means? How might committees on appointment and promotion assess contributions made, perhaps collaboratively, to a database? One scholar has described the problem this way: "[T]enure is the incentive, and building up resources, such as a database, as opposed to, say, summarizing, is less valued in tenure considerations."[2] Moreover, electronic journals are few in number, and fewer still make use of the peer-review process. Some argue that scholars' inevitable tendency is to publish their manuscripts in the traditional way, because the status of the established vehicles is understood.[3]

COSTS OF UPGRADING LOCAL TELECOMMUNICATIONS

An issue of exceptional importance is the cost of upgrading local campus computing and telecommunications infrastructures.[4] The amount authorized by the U.S. Congress for the National Research and Education Network (NREN), the data superhighway that will succeed the Internet, will be only a fraction of the total needed to link university libraries and research laboratories electronically. An aide to Sen. Albert Gore, Jr., of Tennessee, one of the chief proponents of the NREN, has acknowledged that for every dollar appropriated by the federal government, state and local governments and private institutions may have to appropriate five to ten dollars. The federal government's role in making its appropriation was principally to encourage development of the entire infrastructure and bring the NREN only so far as "the gates of academe."[5] One observer suggested that it may require between 10 and 100 billion dollars to develop or improve campus networks.[6] A more fully integrated resource allocation process may be called for, and under such circumstances some renegotiation of the boundary between the library and computing-telecommunications budgets is almost inevitable.

STANDARDIZING ACCESS AND RETRIEVAL PROTOCOLS

There is, further, the substantial problem of the heterogeneity of access and retrieval protocols and means of representing information digitally.[7] In the culture of print, retrieval of material from an open-stack library usually consists of little more than determining the call number, which is as much a key to a physical location within the building as an element in a classification scheme, and locating the material. In an electronic environment, access and retrieval are clearly altogether different in kind. To facilitate access the National Information Standards Organization (NISO), the organization that created the International Standard Book Number (ISBN), has sought, first, to recast and expand bibliographic records so that their various elements are appropriate to the distinctive features of electronic products.[8] In addition to information about the author, title, publisher, and place and date of publication familiar from records for monographs, for example, the record for a computer program might include information about the medium,[9] the date of update or revision (this element reflects the dynamic quality of many electronic products),[10] and the system requirements (such information might include the specific model of computer on which the program is designed to run, the amount of memory required, the name of the operating system and its version, the software requirements, and so on).[11] Bibliographic records for databases, similarly, might specify "database" and the medium,[12] an indication as to the beginning date of the database with a reference to the effect that it is still being updated, and system requirements. The record for such a product might therefore have the following form:

World cultures [database on disk] (La Jolla, California: World Cultures, 1987- ). Quarterly. Computer disks: 5 1/4 in., double sided, double density, 320+ KB; program and data diskettes MAPTAB and programs and data utility diskettes MAP and SORT. Accompanied by: codebook; companion publication World Cultures Quorum. ISSN 1045-0564. System requirements: DOS 3.3; 256K RAM; disk drive.[13]

Clearly, such information is of extraordinary utility to readers, since retrieval depends upon the adequacy of the information concerning the medium and system requirements.

Of even greater importance are NISO's efforts to standardize access protocols themselves, to permit readers to negotiate their way among systems with various kinds of conventions and characteristics. Many of the online library catalogs available on the Internet, for example, employ distinctive search conventions. The NISO's Linked Systems Protocol (Z39.50) is a standard for system-to-system communication for purposes of retrieval of bibliographic information;[14] it permits a reader to use the conventions of his or her own institution's online catalog to access and retrieve information from other catalogs without having to master the idiosyncrasies of the search conventions of each.[15] At the Pennsylvania State University the Protocol also enables readers to search various subject databases like AGRICOLA, ERIC, and MEDLINE using familiar commands.[16]

Eventually, efforts to standardize (or to facilitate access) will be extended to the primary information itself, currently formatted in a bewildering variety of ways, specifically as contrasted with print, which requires minimal standardization.[17] Under the worst circumstances a reader may have to master as many different user interfaces, software packages, and system conventions as there are electronic products to be accessed. Such efforts would fulfill the objective of organizing "[a]ccess to diverse technical environments and information resources ... to appear to be coming from a single system."[18] Natural language is increasingly being incorporated into interfaces to facilitate retrieval; expert systems will operate within heterogeneous environments and make the requisite translations among computer languages so that material retrieved from various databases has a uniform appearance to the end-user.[19]

CHALLENGES TO COPYRIGHT LEGISLATION

There are, finally, the critical issues resulting from the challenges to copyright legislation the new technologies implicitly pose. These issues are of such importance that we will consider them here in detail.[20]

American copyright practices have their origin ultimately in the U.S. Constitution, where in Section 8 of Article 1 it is stated that "Congress shall have the power ... To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries."[21] The original intent, therefore, was to encourage intellectual productivity, specifically by granting "to Authors ... the ... Right to their ... Writings." By the late 20th century, of course, it had become the practice for many scholars to assign copyright to their publishers; the incentives for scholars, therefore, are now "diffusely economic," as Henry Riecken points out, "through the bearing that scholarly publication has upon professional recognition and occupational success":

[T]he producers of scholarly work rarely receive more than a token pecuniary reward.... Instead, they are usually supported by an institutional salary paid in part for the performance of duties other than producing publishable scholarly work, but a salary whose continuance and ... magnitude is dependent in part upon the quantity and quality of published scholarly work---i.e., tenure and promotion are granted partly if not mainly on the basis of scholarly achievement evidenced by published work. Such achievement can also be necessary and sufficient to persuade private foundations, government agencies, or the scholar's employing institution to provide money for research, study, travel, and publication.... Instead of rewarding the producer, the economic arrangements nourish the distributor of scholarly work. Publishers ordinarily retain copyright to the work, sell permission for its use in other publication (e.g., anthologies), and attempt to collect fees for photoduplication.[22]

Publishers, one might say, are assigned copyright in return for the substantial contribution they make to scholarly communication.[23]

What rights does the 1976 copyright law specifically grant, and what are their implications for academic libraries and the process of scholarly communication? The exclusive rights of the copyright holder include the right to (1) make copies for sale or distribution; (2) prepare derivative works (scripts based on novels, for example, or translations into foreign languages); (3) distribute copies by sale, transfer of ownership, lease, or lending; and (4), in the case of a literary, musical, dramatic, choreographic, or audiovisual work, perform or display it publicly.[24]

The holder's rights are then limited by a series of provisions balancing those rights with the rights of information users. Section 109 of the law, for example, limits somewhat the holder's right to control the distribution of a work: the lawful owner of a particular copy of the work is permitted to sell or otherwise dispose of it or display it publicly. This doctrine, known as the first sale doctrine, provides the legal foundation for interlibrary lending. Similarly, Section 107 was intended, in the language of the House and Senate Committee Reports, "to restate the present judicial doctrine of fair use," which permits the use of excerpts from a protected work for scholarly, critical, or journalistic purposes. The statute states that

the fair use of a copyrighted work ... for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement.... In determining whether the use made...is a fair use the factors to be considered shall include

the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
the nature of the copyrighted work;
the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
the effect of the use upon the potential market for or value of the copyrighted work.[25]

For our purposes the most important provisions are stated in Section 108, which specifies the conditions under which copying of library materials is permitted. All such copying, first, is limited to single copies, which must be undertaken without any purpose of "commercial advantage." Moreover, before a library can "copy ... a published work ... for the purpose of replacement of a copy ... that is damaged, deteriorating, lost, or stolen," it must first determine "that an unused replacement cannot be obtained at a fair price."[26] Finally---and this provision is of crucial importance---Section 108 states:

The rights of reproduction and distribution under this section extend to the isolated and unrelated reproduction or distribution of a single copy ... of the same material on separate occasions, but do not extend to cases where the library...

is aware or has substantial reason to believe that it is engaging in the related or concerted reproduction or distribution of multiple copies of the same material, whether made on one occasion or over a period of time, and whether intended for aggregate use by one or more individuals or for separate use by the individual members of a group; or
engages in the systematic reproduction or distribution of single or multiple copies ... of material described in subsection (d):[27] Provided, That nothing in this clause prevents a library
... from participating in interlibrary arrangements that do not have, as their purpose or effect, that the library ... receiving such copies ... for distribution does so in such aggregate quantities as to substitute for a subscription to or purchase of such work.[28]

We have quoted these provisions at such length because they have important consequences for libraries' efforts to form consortia and share resources with other institutions. The phrase "systematic reproduction" is left undefined in the statute but is explained by example in the report of the Senate Judiciary Committee, to which the bill was referred for consideration:

While it is not possible to formulate specific definitions of "systematic copying," the following examples serve to illustrate some of the copying prohibited....

A library with a collection of journals in biology informs other libraries with similar collections that it will maintain and build its own collection and will make copies of articles from these journals available to them and their patrons on request. Accordingly, the other libraries discontinue or refrain from purchasing subscriptions to these journals and fulfill their patrons' requests for articles by obtaining photocopies from the source library.[29]

As Robert Oakley noted, the language of the Senate Committee Report "is similar to the analysis of fair use that suggests that the most important of the four factors is the one dealing with the effect of the use on the potential market for the work."[30]

Clearly, the legislation was intended to address precisely the kind of arrangement typified by the Colorado Alliance of Research Libraries, for example, and for that reason CARL is properly careful to compensate publishers so as not to violate copyright. It is difficult to contrast the relative costs to individual institutions of either maintaining journal subscriptions locally or distributing and sharing such resources, absent full studies comparing the underlying cost structures of the two models. It is difficult, therefore, to gauge the precise economic effects of compliance with copyright as one of the costs underlying the distributive model.[31] (We should also note that a local decision to cancel a subscription and share resources with other institutions might lead to an increase in unit price, given the marginal cost structure underlying the printed scholarly journal; the price is therefore a moving target, which makes it more difficult still to predict and contrast costs. Producing journals electronically may have a different economic effect in this respect.)

Electronic information technologies, for the first time in history, permit acceptable alternatives to the purchase of entire printed books or journals and greatly complicate the enforcement of copyright legislation. As Paul M. Gherman and Paul Metz suggested, "[T]he critical point about the print medium is that it converts intellectual property into a physical commodity whose use can be limited and monitored and whose replication and redistribution is inconvenient and unsatisfactory. These limitations in reproducibility and transportability, while serious drawbacks to libraries and their users, are highly valuable strategic advantages to publishers."[32] Electronic technologies afford possibilities one simply could not have envisioned in the culture of print. Indeed, one might say that the most fundamental assumptions underlying copyright legislation are as much an expression of that culture as are literary writing and the other genres sharing its values and characteristics. In Williams and Wilkins v. U.S., for example, the Court of Claims held that "it is almost unanimously accepted that a scholar can make a handwritten copy of an entire copyrighted article for his own use;"[33] the "effect ... upon the potential market for ... the copyrighted work" specified as a factor to be considered in attempting to determine fair use would clearly be minimal. The making of multiple copies by means of a high-speed photocopier, on the other hand, would have considerable economic effects.[34]

The instances of copyright violation or careful compliance suggested by the practices of Kinko's Graphics Corporation on the one hand and McGraw-Hill, Inc. on the other give some sense of what is possible. In the first instance, a U.S. District Court judge rejected Kinko's claim that photocopying excerpts from books and selling them in anthologies constituted fair use.[35] In the second instance, as we have seen, texts stored electronically in McGraw-Hill's database can be assembled in an almost endless number of ways; McGraw-Hill, however, has been granted permission by copyright holders to reproduce the texts.[36] In both instances a new technology was used to create personalized information products that were not possible in an earlier era.

The 1976 law recast and extended (or, to use the more highly charged language of one observer, "stretched"[37]) the provisions of the existing legislation in an attempt to cover such new situations. The legislation also implicitly recognizes, however, that under some circumstances enforcement is difficult if not impossible: Subsection (f), Clause (1), of Section 108 states that "[n]othing in this section ... shall be construed to impose liability for ... infringement upon a library ... or its employees for the unsupervised use of reproducing equipment located on its premises: Provided, That such equipment displays a notice that the making of a copy may be subject to the copyright law."[38] Photocopying services offices dutifully post such notices,[39] but publishers suspect---correctly, to be sure---that infringements regularly occur. In this way "the widespread violation of copyrights and the shortened life of patent rights have become the unenforceable Prohibition of our time," observes Harland Cleveland.

Sony v. Universal Studios suggests some of the other possible complicating effects of new technologies. In that case it was ruled that "the practice of recording a program to view it once at a later time, and thereafter erasing it" constituted fair use: "Time-shifting enables viewers to see programs they would otherwise miss because they are not at home, are occupied with other tasks, or are viewing a program on another station at the time of a broadcast they desire to watch."[40] Oakley suggested the argument made in the Sony case might easily be applied to other situations, specifically when the electronic transmission of a protected work is at issue; in that instance the justification might be "location shifting" rather than "time shifting":

[S]uppose the only copy of a 1963 medical journal needed by a physician in Oregon for research purposes was held by the National Library of Medicine in Bethesda.... [I]f that material were transmitted electronically and read with no permanent copies being made, there is little difference in circumstances from the Sony case, and a court might find fair use. If, however, researchers used such arrangements to create their own paper or disk-based libraries then the copying would clearly be beyond what is permitted under Sony.

Oakley's final sentence, of course, states the central point; depending upon what kinds of controls are in place under the circumstances he describes, individuals might well find it easy to engage in what are infringing practices, knowingly or not, just as now photocopying machines enable them to do so. In the culture of print, the only kinds of transactions possible are exchange transactions. Electronic technologies, in contrast, permit sharing transactions in the following sense: if a particular volume owned by one library is lent to another, the first still owns it; if, however, the text is available electronically, it is more easily shared; absent adequate technical controls limiting use to a read-only option, both can own it in some sense, yet only the first has compensated the publisher.

The new technologies have even more far-reaching effects. To return to the examples of the Colorado Alliance and the consortium formed by the three Virginia universities: one might say that it is electronic information technologies that make such consortia viable in the first instance, that in the absence of the technologies such arrangements would be very difficult to undertake and sustain. They depend on electronic bibliographic tools that permit institutions to share information about collecting practices with one another, build complementary collections, and facilitate rapid document delivery. Such arrangements are clearly systematic, and copying of materials for other institutions within the consortium would therefore be in violation of the 1976 law, as explained by the language of the Senate Judiciary Committee Report, if publishers were not duly compensated. But what if a single institution decided unilaterally to reduce expenditures for the acquisition of printed materials in order to improve its interlibrary loan services and acquire electronic bibliographic tools to locate materials held elsewhere? Would such an internal reallocation be considered systematic and in violation of copyright? It would certainly constitute an alternative to purchase of materials and would therefore affect publishers' revenue streams.

Copyright practices have been crafted in response to the traditional model of scholarly communication and implicitly assume many of its elements, including the technology of print. The legislation has been extended piecemeal in an effort to apply it to an emerging model fundamentally different in kind. The 1976 law, one might say, attempts through a series of controls to preserve the effects of the conditions created by print (little alternative to purchase other than hand copying, exchange rather than sharing transactions, and so on), but the characteristics of the new technologies are such that altogether different conditions are created.

The new technologies serve to raise more basic questions, still John Garrett notes that "the idea of a work changes fundamentally in an electronic world. Questions that have never been answered in the print world---such as whether there is a core unit of a work (word? sentence? paragraph? image?) in which copyright ownership remains imbedded forever, like Chomsky's grammar---will need to be debated and resolved."[41] Francis Fisher suggested, similarly, that under the terms of the current law,

works that were derivative of protected works were... protected, but derivative works were generally adaptations of the entirety or a large portion of it.... But now the new technology makes it much easier to use pieces of a work. It is easy to search an enormous database and locate the exact piece of information that could be usefully combined into a new product. The combining is made easier because the pieces of expressed ideas to be combined are already in the same electronic form as the new product. Indeed... the pieces may be used simply by being pointed at by the re-creator.[42]

Under such circumstances what responses are possible? What new models might be considered that would be appropriate to the distinctive characteristics of electronic products, as contrasted with those of print products? A number of observers have suggested a transactional approach in which information users would be charged on a per-use basis.[43] Computers, they rightly argue, are "good at counting things," and in principle a fee structure of this kind would therefore be easy to manage. There are disadvantages, however. Consensus on precisely what constitutes use would be extraordinarily difficult to achieve.[44] There would, moreover, be significant costs involved in tracking use.[45] Finally, there is the difficulty of controlling access to information in electronic form;[46] if photocopy machines already provide one with an alternative to purchase, what will be the effects of the ready availability of large amounts of information in electronic form? "Digits," declares Fisher, "ooze from their containers."[47]

Others have suggested accordingly that a more effective alternative would be licensing.[48] In situations where use is difficult to control, a license, by statutory means, grants blanket permission to use works in return for the payment of a fee to a central agency.[49] Many of the difficulties of the transactional approach are thereby avoided; on the other hand, one would lose the data one could otherwise use to determine prices.[50]

There are, finally, the proposals that universities (1) claim joint ownership of scholarly writings with members of their faculties, remunerating them and prohibiting them from assigning copyright to a third party; (2) request that faculty members first submit manuscripts to publishers whose pricing practices are, in effect, more consonant with larger educational objectives; and (3) grant unlimited copying to libraries and individual scholars and specify that such permission has been granted in the copyright statement.[51] These proposals, of course, are extensions of the broader proposal that universities reclaim responsibility for disseminating the results of faculty scholarship.[52]

"Law is born old," Nino Tamassia, the Italian legal historian, used to say,[53] and we envision a situation where the new technologies will continue to challenge the assumptions underlying current legislation and permit possibilities that existing copyright practices were not specifically designed to address and to which they are not well suited. Under such circumstances, new legislation will be crafted in response to situations predating it, as Tamassia's apt metaphor suggests.