Draft Only -- Not for Citation or Distribution -- Comments Welcome!



J L Lemke
City University of New York


What happens to the academic journal publication system and its symbiotic relationship with our publish-or-perish paradigm when new information technologies make print publication obsolete? What new balance among author, user, and community interests should be struck under these new conditions?

New information and communication technologies, supported by local and global computer networks, will very shortly make it possible for scholars to publish their own work and gain access to the work of colleagues without the intermediary of print publication, publishers, and academic journals. At present academic authors receive little or no royalties from journal publishers for their work, and may even pay per-page subsidy charges to some journals. Scholars gain access to colleagues' work through very expensive library subscriptions to academic journals, which depletes funds to subsidize research itself. Nevertheless, the fact is that most published articles are probably never read, cited, or used after print publication. In the past there was no other effective way to distribute academic research globally; today there is.

There are two related functions that have attached themselves to the academic journal system. One is the peer-review process, which filters manuscripts, both encouraging more careful work and discouraging radical innovation. Its primary benefit to authors are reviewers' suggestions for improvements, and to users it is the degree of confidence we can have in the reliability of published work, as well as the concentration of all relevant work in a small number of journals. The other function is that of evaluating scholarly production for purposes of hiring, tenure, promotion, and confidence in the reputation of grant applicants. In effect, much of the evaluation which university committees particularly are supposed to do is delegated to the editorial processes of academic journals, because specialization has made it impossible for even departmental colleagues to evaluate one another's technical work.

Both of these functions could be equally well served outside the system of academic print journal publication. One model of course is the simple transfer of the status quo to the online medium, but this brings at present no benefits to anyone except the publishers, whose costs of production would be greatly reduced. Subscriptions to online journals however continue to be and promise to remain as high as the traffic will bear, and the political reality is that these prices are artificially supported by the coercive constraints imposed by the evaluation function (i.e. publish-or-perish).

An alternative model would be for groups of well-respected scholars in each specialty to operate an independent peer-review system, perhaps on submitted electronic manuscripts, but also perhaps by automated searching for new material posted to the internet in their specialty fields, and posting comments on the quality and importance of these works that would become available to anyone interested. It should be technically feasible for anyone interested in finding good quality work on a topic to do a search that provides not only links to relevant documents but also links to available evaluations of these documents. With some degree of standardization by simple ratings categories, and even with pseudonyms for evaluators, users could introduce a (modifiable) filter that would apply their own ratings of the reliability of evaluators (based on past experience) to select and weight the evaluations of each document.

The payoff to evaluators under this model would be influence in the field, but it would be an influence whose extent was determined by the user community as a whole, and in a way that did not impose majority views on anyone. The benefit to authors would be in the critiques provided, and to end users, obviously, in the filtering process itself.

Any group seeking an evaluation, for example for promotion purposes, of a candidate's work, would need only to create its own profile, selecting and weighting the reliability of members of the available internet evaluator pool, and it would then have evaluations, both in the form of ratings and in the form of specific comments on each document submitted by a candidate. This would have the great benefit over the present system that the actual grounds of the evaluation would remain public and subject to dispute, rather than being forever interred upon acceptance or rejection of a manuscript by journal editors.

Power in general would be more widely distributed and less concentratable in a few hands under this model.

What happens to the money formerly accruing mainly to journal publishers? One can make two arguments here, I think. First, that scholarly publication is not intended as a profit-making activity; universities and granting agencies already subsidize academic work and the results belong in the public domain in order to be most useful to society at large. University libraries would save vast amounts of money under the model I am proposing; that money should become available to further sponsor research itself. The other argument is that scholars deserve to derive some monetary benefit from their work, especially if it is uniquely valuable in the view of users and the market.

Agreements need to be reached between scholars and their primary sponsors regarding rights to the work they produce. Where the primary sponsor is a governmental agency, funded by the whole community, I believe all work should enter the public domain. Where this is not the case, scholars, as generally the weaker partners in sponsorship arrangements, need to be guaranteed in law that their share of proceeds from their work will never be less than 50%. For there to be any proceeds at all, authors and sponsors must agree to withhold information from users unless a fee is paid, and the market will then determine the value of the information.

Unfortunately, however, for traditional models of copyright and patent, information is not a material commodity. It is never truly "transferred" in the sense that when it moves to one place or ownership it is thereby necessarily lost to another; it can only be copied. And it is in practice nearly impossible to control the unauthorized copying of information without continuous monitoring of all digital information in the world-system, which would be a clear threat to political freedom in itself, but is fortunately beyond foreseeable technology not only in the matter of scale, but also as regards the ability of monitoring systems to recognize all reasonable transformations of the original form of the information which preserve its essentially usable content. Once the information is out, its price will therefore rapidly decline, until that price is low enough that paying it for convenience of access, the moral pleasure of supporting the original creator's work, and perhaps most of all for a guarantee of the reliability of the information, beats the risks of using a (possibly corrupted) unauthorized copy.

No one will pay more than a small fee for most academic publications today, probably not much more than twice the present cost of hardcopying. Indeed probably no one will ever pay anything at all for most articles. A small percentage of articles will be in demand, but by and large, unless they are the product of already well-respected scholars, demand for them will grow only after they are distributed fairly widely and gain a reputation by being cited and evaluated favorably in the field. By that time many copies would already be in existence, tending to keep the price of authorized copies low. However, even at a low price, a large market would accrue non-negligible revenue to the authors. Marketing strategies can be foreseen in which portions of a work will be made public and fees charged for the remainder if the public portion creates demand for the rest. One can hope that our most distinguished scholars will make their work available initially at low prices, relying on volume demand to compensate them.

Beyond the normal run of academic publications, of course, there is the matter of information production which has further commercial potential, i.e. information which can form the basis for material benefits, ranging from routines for the synthesis of useful medicines to software code to perform useful tasks. In this domain many other considerations apply and our society will be a long while yet working out the proper balance between creator, investor, user, and community interests. But this area has only a very small overlap with the general domain of academic research and publication.

Somewhere in between the usual and the commercial cases we find the current and future analogues of textbooks, popular texts, standard handbooks, and other large-market academic products. I think for example of information hyperwebs that might be of great value to students or newcomers to a field. I believe the principle which is coming to be accepted here is that one may reasonably charge for "value-added". If the same or equivalent information is already available, piece by piece, in other forms which are in the public domain or are available at prices generally affordable by typical users, but the form in which this information is assembled by the creator has special value to some users, then it is not unreasonable that its price too should be determined by the market, unless the creators wish to donate its value to the community at large.

I certainly believe that it is in the interest of society as a whole to subsidize research by a large number of scholars selected by a variety of criteria in a variety of institutional frameworks, and to provide such subsidies for very long terms, perhaps including for life, in order that research be able to take the long view, and that there be freedom to pursue high risk ideas. Such subsidies are by and large rather cheap, and even if the probability of a return is small in each case, the total value of the return in aggregate is usually very great indeed. Every winner justifies many losing bets in this game. Publicly subsidized research, again, should enter the public domain, though in the case of substantial commercial potential, it is not unreasonable that the public treasury should benefit, provided that commercial developers are not profiting at the expense of the public by charging them a higher price than that at which the government could provide the same benefit (without special subsidy). So it would be unconscionable for the government to sell databases produced with public monies to vendors who would then re-sell them at a profit back to the community whose taxes have already paid for them. If the government can make these databases available at minimal cost, it should, and in any case its contracts must provide a limit on the profit that can be extorted from the public for returning to it what it already owns. On the other hand, it is perfectly reasonable that if substantial investment is needed to turn raw information into commercial products, that this investment be entitled to a reasonable rate of profit, some of which is returned to the treasury as licence fees for the base information.

Once we allow, however, that information may be held ransom for its price, care must be taken to protect the community as a whole and its weakest members. It has long been an informal (i.e. non-Constitutional) pillar of modern democracy that the essential information needed by citizens to make political decisions and to have an opportunity to contribute to the welfare of society should be freely available to all either through the public schools or through free public libraries.

It is possible that both these institutions will perish in the next century. Schools because they are both incredibly expensive and not very effective in accomplishing their stated objectives for large portions of the population, and public libraries because the internet will soon be the only library on the planet. There is no reason why it will not ultimately be the only school as well. It is the institutional frameworks here which are likely to be superseded, their essential functions will remain. There will still have to be library services such as access and reference, and educational services such as instruction and assessment, but they will be provided in very different ways. What concerns me here is the possible scenario in which the information have-nots, who are also financial have-nots, will have no guaranteed access to any information beyond what all citizens will no doubt still get by way of general educational provision. Basic education will provide information and skills with reasonably long lifetimes of use, but citizenship and productive membership in society require current and updated information of much more specialized kinds as well. That information is now available to a large extent through the free public libraries, but it is possible that all such access to information will be for-fee-only in the not too distant future, unless we make a clear policy commitment otherwise.

What categories of information should the community guarantee free or at-cost access to for all its members? this is a pressing and largely still unanswered question.

It is the generalization of my closing question regarding academic publication: Are there criteria by which we can determine that an academic document, otherwise available only at high cost to each user, is of sufficient importance to be taken by the community, with generous compensation, in order to make it available freely or at-cost to all citizens? Such criteria and procedures may become necessary when there are no public libraries which can make a material copy of the work freely available to all comers seriatim. In the regime of material commodities, the library may decide to purchase some number of physical copies, but then can make these available to an unlimited number of patrons for no further cost. In the future digital information regime, it may become possible to charge a high cost for each user and to prevent copying and distribution by intermediaries, including libraries. I hope this will not happen, and it is probably not commercially feasible today, but there are great economic incentives at present to develop such technologies. Under these conditions it will be only by a total taking of the rights to the information that the community can ensure that the information will not be denied to citizens who cannot afford its price. I believe we must establish a right of eminent domain with respect to information.

There is no conclusion to this exploratory essay because it is in many respect the hub of an information and analysis web that could grow to be as complex as the social networks which participate in these issues.