Reflections on the open access movement

One of the fundamental premises of higher education and research is that knowledge can grow and improve because, as a matter of institutional design and the motives of scholars, it is disseminated and shared widely through open channels. One of the key channels is the scholarly journal literature which researchers support by contributing their articles and reviewing services without the expectation of payments. Academic researchers know that by contributing to this intellectual commons, and by following its ethos of ‘share and share alike’, they will be rewarded when their work is credited by others. This puts them in a moral and institutional position to champion the system of ‘open science’ that stimulates critical scrutiny, spreads the benefits of disinterested learning and makes possible innovations and new ways of thinking that may improve society.

However, in recent years there has been a growing concern about the impact of commercialization and corporatization on higher education. There is increasing recognition that the ideals and social benefits of open science can be threatened when too many research projects and disciplines are expected to stimulate patentable innovations or become magnets for corporate sponsorship (Bok, 2003; Agres, 2003). But many now argue that open science is currently being threatened most of all by the domination of scholarly communication by commercial publishing giants. Huge price increases for many scholarly journals have led to a rash of cancellations of subscriptions by libraries and have strained the capacity of academic communities to access the articles, reports and reviews prepared by colleagues in their fields.

One of the responses to the current “serials crisis” has been the emergence of the “open access” movement in scholarly publishing. This is a movement of research librarians, scholars, research funding bodies and other stakeholders of the scholarly research process. Open access advocates have proposed that scholarly communities must take bold action to address the growing strains within the world of scholarly publishing and communication that are associated with increasing journal prices and the increased dominance of commercial publishers. The goal of the movement is to ensure the free, equitable, organized and wide distribution of the output of scholarship and research on the public Internet (Suber, 2003). Because authors and researchers want to give away their published work as a contribution to an intellectual and cultural commons, it is therefore both inefficient and morally unjustified to continue to accept the barriers of a ‘toll-based’ system for gaining access to that work unless those tolls are necessary for wide and effective circulation:

The adoption of the new system would of course be made possible by the pervasive acceptance of digitization and the accompanying transition from print to electronic distribution. It no longer makes sense, if it ever did, for researchers to transfer the copyright for their writing to journal publishers in exchange for its publication. The Internet offers a choice among the political economies of knowledge, and scholars have now to reckon on what they owe the public and themselves, in deciding the future of academic knowledge (Willinsky, 2002).

The realization of a new model would therefore require more than a change in scholarly publishing contracts: academics must embrace cheap and customized electronic access, distributed archiving of their work and the financing of journals by up-front publication subsidies instead of subscription fees.

In what follows I want to bring into focus both the implications and moral claims of the open access movement. On the one hand, its proponents have sharpened our awareness of the need for the reform of scholarly publishing and the changing social, technological and market conditions that have given rise to it. Because of their campaigns and public statements, even governments are reviewing how the growth strategies and pricing behaviour of commercial publishers have inhibited the ability of research libraries to offer full print and electronic journal collections. While acknowledging the apparent attractiveness of the open access movement, I want to inquire about what the proposed models might mean for the larger institutions of scholarship. Several questions emerge: What are the consequences of ‘disintermediation’ for the realm of scholarly communication and networks? Can we be sure that the new networks and filters that take hold in the new publishing model will really serve the goals of scholars and their communities, as well as being more equitable and better attuned to the public good? And finally, is commercialism in academic publishing as harmful to scholarly communities as it is to the research process itself?

The Shape of the Crisis

The depth of the “serials crisis” has been dramatized by astonishing reports of biomedical serials with annual subscription prices close to US$20,000 (Owens, 2003). In the past two decades the number of scholarly journals has increased significantly and the average price increase per year for individual titles was 9%, or nearly four times the rate of inflation (ARL, 2003). On the heels of some well-publicized mergers, as well as the expansion of niche publications and the growth of specialist sub-fields, the rate of price increases for commercial journals has far outpaced that of non-profits and university presses (Wellcome Trust, 2003). As a result, journal subscriptions consume a far greater share of library budgets than in the past, yet libraries have had to subscribe to a smaller share of available journals. Part of the problem is certainly due to significant declines in university library budgets relative to overall university budgets and the growth of the journal literature (Henderson, 2002). Commercial publishers now find themselves with a 75% share of the scholarly journal market. These companies realize that the biggest prizes are to be found in the scientific, technical and medical (STM) sector where prices average approximately four times those of the social sciences and humanities sector. Giant science publisher, Elsevier-Reed, several years after its controversial merger with Harcourt, now boasts an estimated 32% gross profit margin, a number that exceeds even the legendary 30% margin of Microsoft, the monopolist software producer (Wellcome Trust, 2003).

The pressure to expand the absolute size of their journals collections (including site-licenses for electronic journals) has had an impact on the wider practices and priorities of scholarship itself. For many disciplines in the humanities and social sciences book publication is the mark of successful scholarly achievement. Yet the library budget squeeze has led to a decline in the market for specialized monographs of approximately 3-4% annually for the past ten years (ALA, 2003). In response, university presses and other publishers have adopted policies of requiring proposed titles to be more profitable and appealing to broader readerships. Many authors of books find that they cannot be published unless they guarantee sales from their courses as a condition of publication (MLA, 2003). The result has been a mismatch between the publishing opportunities for many scholars and the norms and aspirations of their academic traditions and careers. This secondary crisis reflects the more general strains on libraries and universities as they try to keep up with the demands of increasingly complex and extensive knowledge systems. Indeed, this has led to a kind of identity crisis for research libraries themselves; they no longer spend nearly as much of their budgets on resources that will accumulate over time, since they must expand services like interlibrary loans and also purchase access to electronic content that does not become part of a permanent collection (Gorman, 2002). Others focus on the way academics themselves have contributed to the discontents of their intellectual commons, particularly by encouraging excessive levels of publishing merely in order to score high on research audits and performance indices. At its worst, this sort of “Brownie point scholarship” (Adams, 2001) may actually favour productivity and quantity over creativity and quality.

Commercial Publishers and the Intellectual Commons: From Symbiosis to the ‘Big Deal’

There is little disagreement in the open access debate that, in order for an academic journal to succeed, it must be seen to be a worthy representative or trustee for the internal goods of the community of scholars who are its readers, authors and content filterers. Until recently academic communities and stakeholders have complained little that commercial publishers have failed to hold up their end of the bargain (Berry 2001). In their public debates over open access and pricing, conventional publishers have therefore been energetic and persuasive in describing how their specific capabilities as commercial entities have allowed them to create value within this mutually beneficial relationship:

An effective journal publisher will achieve the highest sale value by achieving the maximum use value for authors and readers. High use value is reached through standards of editorial quality; by investment in recruitment and implementation of rigorous peer-review processes; by creating and developing journal titles that give so-called brand context to the work; by ensuring reliability of published products and services through uniform standards of redaction, style, and presentation; by dependable and timely production, manufacturing, and distribution of the finished product; by metatagging of digital content to enable its delivery via online systems and its accessibility via abstracting and indexing services and its mining by search and-retrieval software; by computer support for round-the-clock online hosting and by marketing and selling the information to achieve the widest possible international exposure. Publishing is a business of relationships, cemented by a continuing exchange of value between the publisher (as vendor) and its customers (libraries and individual readers), as well as by commerce with its suppliers and partners, authors included (Crawford, 2003).

Crawford’s comments have an undeniable salience; this sense of mutual interest between commercial publishers and academic communities runs so deep that previously independent academic society journals are increasingly relying on joint ventures with large publishers. Moreover, even if most print-based journals are eventually replaced by electronic-only versions, the sustainability of a large number of worthy publishing activities and processes may indeed require access to specialized business and entrepreneurial skills.

There are even broader claims made on behalf of the advantages of commercial publishers. Arie Jongejan (2003), CEO of Elsevier Science, has argued that commercial journals are uniquely suited to serving and identifying market needs and wider scholarly ambitions that cannot easily be met by journals managed by scholarly associations. Changes in scholarship and science have paralleled the growth of the ‘knowledge society’ and the identities of many disciplines and academic sub-fields have been shaped by the closer relationship between academic research and rapid social change. Because of their role as market innovators these publishers have the resources to support and reshape scholarly communities in a way that offers the market skills and patience needed to support new academic trends and movements. Jongejan also believes commercial publishers’ political freedom and flexibility gives them an advantage over the more “conservative” academic societies in responding to the needs of interdisciplinary communities and international linkages.

In the last few years several studies of the journals market have begun to challenge the perceived symbiosis between the commercial publishers and academic communities. The growth of new areas of knowledge has coincided with the rise of neo-liberal policies, which has placed a strain on the usual sources of funding for a wide range of academic activities. As a result, library budgets have simply failed to keep pace with the size and scope of scholarly output as well as the increasing demands of managing it (Kyrillidou & Young, 2003b; Henderson, 2002). As the number of journal titles increases, and as libraries fall further behind in their ability to subscribe to the same proportion of the literature, the for-profit publishing firms have responded by developing new strategies. Threatened by the simultaneous developments of journal proliferation and tighter competition these publishers have resorted to many of the standard business strategies typically used to respond to maturing markets. Increasingly publishers have found that the prospects for new or ‘marginal’ journal titles are viable only to the extent that they can replace other journals offered by either competing commercial publishers or by the non-profits (Wellcome Trust, 2003). The predatory approach they have taken is to offer ‘bundle’ pricing for portfolios of journals, a strategy that has become infamously known as the ‘Big Deal’ (Frazier, 2001). This has been augmented by mergers among some of the leading publishers as well as the introduction of integrated electronic delivery and indexing systems that are linked advantageously to their own portfolios of journals. Taken together, these approaches allow publishers to restrictively tie the marketing of subscriptions for high prestige ‘must-have’ titles to their more marginal titles. It also allows them to achieve economies of scale and scope by integrating electronic and print-based distribution, ultimately creating barriers to entry for others. Finally, it allows them to ‘lock down’ libraries into long-term arrangements that are hard to alter but which remain attractive in the short-term due to discounted volume rates and continuing access to highly valued titles.

More can be said about the rapidly changing nature of the scholarly journal industry. The point here, however, is that the so-called ‘Big Deal’ is attractive to the commercial publishers most of whom need to find survival strategies for the large numbers of journals they carry which are easy targets because of their low impact ratings and specialized focus (Bergstrom, 2001; McCabe, 2002). In the course of trying to avoid the threat of increased journal cancellations, while simultaneously keeping prices high, these publishers have undermined their reputation as benign partners to buying institutions. In the meantime, commercial publishers are now acquiring competing titles at a faster pace than they are creating new ones, thereby weakening the case that their unique function as innovators in the world of scholarly communication justifies higher prices (Willinsky, 2003; OFT, 2002).

The Search for a New Model

According to open access advocates the current pricing crisis will awaken scholars to the fact that many opportunities for reform lay in the hands of academic communities themselves (Suber, 2003). Optimism about the inevitable path toward open, and even unpriced, access to the peer-reviewed literature rests on many foundations and practices that are already evolving (Willinsky, 2003b). In the first place, the manner in which scholars use and retrieve journal literature is increasingly being satisfied by cheaper, more flexible and rapid access on the Internet. Publishing agreements often specifically allow researchers to freely post pre-print versions of their articles on their institutional web sites, usually with the proviso that links will be made to the official version. In addition, there is a growing receptivity to the moral argument for open access - the belief that the great bulk of the work of scholarly communication is intended for the public good – and that, accordingly, publishers now have a moral responsibility to recognize these ambitions (Gannon, 2004). Some publishers have responded by freeing electronic access to articles after a six or twelve-month delay and the major biomedical publishers are now making electronic databases of their collections freely available to poor countries (Owens, 2003; Morris 2003).

The most systematic developments have been the direct organizing initiatives taken by the research communities themselves. In 1998 the Scholarly Publishing and Academic Resources Coalition (SPARC) was launched by the American Association of Research Libraries (ARL) in order to pursue advocacy and other projects in support of more affordable access to scholarly literature. SPARC’s main goals are to strengthen and expand not-for-profit publishing and to organize consortia of member libraries to provide a counterweight to the increasing market power of for-profit journals. In addition, SPARC’s member libraries make purchasing commitments to SPARC-organized alternative journals with the long-term hope that pricing pressure on libraries will be relieved and that “journal proliferation” won’t have an offsetting effect (Case, 2000). SPARC has also helped libraries develop “scholar’s portals” and other ways of aggregating access to published content so that commercial electronic indexes and cross referencing products will not be able to be used to privilege the for-profit journals. Along the same lines, Canada and the UK have launched national site licensing projects with the help of government start-up funds. There is some evidence that pooled buying power may be moderating average prices in the last two years (Kyrillidou & Young, 2003).

Universities such as Stanford and Cornell made headlines in late 2003 by cancelling their ‘Big Deal’ subscriptions and creating growing interest among academics in open access alternatives. Some high profile editors have defected to such journals. The Directory of Open Access Journals lists 836 tiles that allow free electronic access. Many are recent start-ups and have a wide array of editorial mandates and reviewing procedures, although a large number are not integrated with the mainstream indexing services to which libraries subscribe. The most well known open access journals are those offered by BioMedCentral as well as the recently launched journal Public Library of Science (PLoS) Biology. These journals are perhaps best known for their attempts to compete with leading biomedical journals using an ‘author-pays’ model of financing, which is meant to sustain high quality peer-reviewing and editing systems. PLoS Biology charges authors of accepted papers a publication fee of US$1500, although there is concern that the low level of this fee is unsustainable, since it is subsidized by a large foundation grant that might not be available to other initiatives (Guterman, 2004).

In late 2001 a conference in Budapest convened by George Soros’ Open Society Institute drew up a series of principles that almost certainly comprise the most ambitious vision of open access scholarly publishing. The Budapest Open Access Initiative (BOAI) proposes that the growth of electronic and distributed access to scholarly research reports and articles has rendered traditional models of publishing obsolete. The document (BOAI, 2001) proposes that the only legitimate and useful role for copyright in contemporary academic journal publishing is one that protects the rights of authors. The costs of basic academic publishing (peer review plus near-zero cost electronic distribution) could be covered much more cheaply if the institutional budget amounts currently devoted to subscription were to be reallocated directly to publishers. The BOAI anticipates a system where some portion of access provision to peer-reviewed literature is achieved not only through the traditional journal form, but also through the growth of self-archived repositories of papers provided by universities and linked, indexed and cross-referenced by search tools that could seamlessly integrate scholarly literature on the Internet.

More recently a group of research sponsors and activists issued the “Bethesda Statement” in summer of 2003 encouraging scholars and their institutions to ensure that the transition to open access publishing is accompanied by policies that preserve the needs of scholars and the integrity of academic publishing. In addition to persuading scholars to contribute to open access journals, the Bethesda Statement recommends that all stakeholders be educated about the need for reform and adjust their practices and norms accordingly. For instance, research sponsors should be prepared to allocate resources to support author payments for publication, and collegial bodies should accept the academic legitimacy of articles published in free online peer-reviewed journals. The Bethesda meeting demonstrated a strong commitment to open access by leading biomedical research sponsors and institutes such as the The Wellcome Trust, Max Planck Society and the Howard Hughes Institute. However, it appears that non-profit publishers from scholarly associations have not yet signed up as full allies of the open access movement. Subscription revenues for society-based journals are used to support membership services and activities. Hence representatives from the not-for-profit sector have proposed that scholarly associations should only experiment with new models where possible and when they don’t conflict with the publishing practices preferred by their members as well as the maintenance of high standards of review and editorial excellence (Morris, 2004).

Perhaps most interesting of all, direct attempts at legislative and regulatory reform have also emerged. In July of 2003 the ‘Public Access to Science’ bill was proposed by U.S. Congressman Martin Sabo with the support of the PLoS (Zandonella, 2003b). This Act would exempt the publication of government-funded research from copyright protection, thereby requiring free access as a matter of law in many cases. Rep. Sabo’s public statement echoed the views of other advocates by suggesting that it was wrong for the public to pay twice for research it has funded out of its taxes and which is intended for the public good. The possible implications of the approach taken by the Sabo proposal are many, and have stirred some controversy (Zandonella, 2003a). What rights would authors of scholarly papers retain? Would the copyright exemption cover derivative works, review articles and so on? How does one draw the line between publicly supported scholarly endeavours and non-public or individual creations? Since the Bill protects public access for U.S. citizens, does it adequately address the fact that much science and scholarship is the creature of international collaboration (or can be improved by international access)? What part of a scholar’s work and output falls under the category of hired work? If publishers contribute value-added features to ‘works’ and to the publishing process, which of those features – if any – should be sufficient to yield them a grant of exclusive rights of distribution? Since governments frequently try to create ‘spillovers’ between publicly funded research and commercial applications why should free and open access be reserved for scholarly publications alone? In the light of the unwieldy nature of the copyright issues involved it is doubtful that the Sabo proposal will become law. Nevertheless, legislative horizons may be opening in other countries. In early 2004 the UK’s Science & Technology parliamentary committee was conducting an inquiry on scientific publications, which presumably might consider some form of government support for open access.

Will open access create better performing scholarly networks?

It would be both presumptuous and premature to predict the course of the open access movement, or to fully anticipate what consequences its success might have in refashioning scholarly publishing processes and activities. However, it is fair to say that whatever experiments or new models are chosen, they will have to justify themselves as worthy communication tools for scholarly communities and as publication systems that adequately further the wider ambitions of open science.

One approach, associated with Stevan Harnad (2003) is what can be called ‘distributed archiving.’ He argues that in the Internet era scholars can self-organize virtually all of the functions of publishing currently “outsourced” to journals, with the exception of peer review. In his vision scholarly exchange will reach its full potential when scholars post their work in distributed open archives, which could be indexed across repositories and collections using metadata that allow rational and organized “harvesting” of contributions across open networks. One effect will be that journals that currently use their market power and ‘brand’ recognition to charge high subscription fees will have to compete with this integrated public system of archives if they want to prove their value to the world of scholarship. In the meantime the web of self-archived literature will continue to grow because the increased ‘impact value’ and efficiency of open archiving will be irresistible to researchers trying to make their work well known (Harnad, 2003c).

Harnad (2003b) insists that these developments will not result in the kind of information anarchy that many fear from expanded practices of self-publication on the public Internet. Instead, software mechanisms will be available to rank the importance of scholarly papers, partly usurping and partly demystifying the authority of the traditional journal over the non-peer review activities of publishing such as marketing, branding and collection-building. For instance, meta-tagging of content can ensure that peer-reviewed articles accepted by journals – or peer review bodies - can be distinguished from ‘pre-print’ postings. In the end, Harnad forecasts, open access publishing will be supported by those looking to maximize their “impact” and by research patrons who will want to measure that impact through the more advantageous operation of “scientometric performance indicators.”

The distributed archive approach extends what has been going on for more than a decade in a few specialized cases, such as the ArXiv project in theoretical physics. Projects like ArXiv have been criticized by some sceptics as suitable only for specific fields and disciplines. Others see them as inadequate substitutes for the permanent and durable archiving function of libraries and the various professional management functions of publishers (Blume, 2000). Questions have also been raised about whether distributed scientific literature can live up to the open access promise of “levelling the playing field.” Meyer and Kling (2002) have found that, in the case of ArXiv, meaningful participation of researchers in developing countries has not been noticeably improved by open access to scientific literature, primarily because better access does very little to boost institutional resources and support for these researchers who must instead migrate toward elite research centres.

But I would argue that there are other troublesome dimensions to this approach to open access. In the first place, it is narrowly based on efficiency considerations; it assumes that the role of mediation in the space of scholarly communication is losing its function for modern scholarly communities that now have the tools and motivation to self-organize. From this perspective, traditional journals are slow and outworn ways of packaging together functions like peer review with other activities like distribution, marketing, exchange of commentary, and so on. Hence, although journals will still be necessary, they may have to “scale-down” to become mere peer review “services” (Harnad, 2003b). Yet Harnad cannot explain why journals would still survive in any meaningful way at all, since, in his system, they would only be able to sell “add on” services like printing which he says no one will need. Neither can Harnad explain why academics in some (perhaps most) disciplines are still attached to journals as authoritative organizers of the literature, and not simply as review services. To put it simply, many academics still like to browse journals (on-line or in the library stacks) rather than simply search for articles through indexes and Google. Although scholarly journals have their imperfections and will undoubtedly evolve to reflect the changing networks in which they are embedded, they can still play a significant and widely appreciated role as a platform for scholars to locate themselves in relation to intellectual trends and tendencies (Morris, 2004). In an open archiving universe peer-reviewers and editors may be reluctant to be recruited by journal publishers if they feel they are contributing to a downgraded process of filtering that is only loosely attached to a final aggregated product with value of its own.

Many open access supporters would probably say that open archiving is only transitional, and that journals can and will survive with financing by publication charges rather than subscriptions. They argue that the shift to unpriced and open access does not eliminate journals, but merely changes the type of transaction by which journals collect revenue. Under the ‘author-pays’ system, there is nothing preventing the emergence of more journals like PloS Biology which collect revenues from publication fees paid by authors and/or their research patrons, instead of subscriptions. Indeed, many journals already ask authors to pay for page charges and colour charges.

Despite its apparent elegance, the author pays model has many critics (Guterman, 2004). For instance, some observers argue that the current multi-billion dollar total price tag of worldwide scholarly publication would not be reduced by open access, but rather simply shifted from library subscription budgets to research budgets. If universities paid for author charges, their budget for supporting journal publishing might actually increase. By paying the up-front publication charges they would be covering the journal revenues previously derived from subscription sales to other users such as corporations and other non-academic subscribers. Furthermore, the toll-free availability of online versions could also make the economics of print-based journals unsustainable.

Of course, it cannot be denied that the author-pays model has the potential to produce many of the intended transformations; commercial publishers would have to cede some of their market power, and, more importantly, unhindered access to valuable research on the public Internet would be achieved. However, this still leaves the problem of the inherent conflict of interest of funding publication by the author. At least some journals would have an incentive to relax reviewing standards to stay afloat. Representatives from PLoS Biology have said that their journal will maintain the highest standards of integrity, will offer fee waivers for needy authors and ensure that the reviewing and submission process would be blind to such requests (Doyle, et. al, 2004). However, the problems appear to lie at a much deeper level than that of editorial policy. Assuming fees would be charged for published papers only, the cost of managing and coordinating peer-review would be eased by accepting more papers. Authors who feel they would not qualify for a waiver, or who could not receive support from institutions or funding agencies, might be deterred from submitting altogether or might instead look to sponsors at the margins of the academic culture, such as think tanks and corporations. The resulting situation would be full of ironies; the author pays model of open access could be far more susceptible than commercial publishing to worries about selective bias and conflict of interest, as well as the influence of political and commercial agendas (Crawford 2003; Okerson, 2003).^¹

A related problem is that even if an “author pays” model were to maintain integrity and improve accessibility it might reinforce or increase inequalities among disciplines and sectors of the academic world. Publishers would no doubt sell ‘real estate’ in their journal’s pages at a rate governed by the prestige of the journal or the marketability of the field. The greatest disadvantages would be felt in the humanities and some social sciences, areas where authors typically receive less research grant funding than do their counterparts in the sciences. Some leading advocates of open access acknowledge that the poorer disciplines are less likely to benefit in the beginning from the author-pays model (Suber, 2004). By contrast, the current subscription system has limited the ghettoization of the Humanities and social sciences since journal revenues are principally derived from library subscription budgets, which, in turn, reflect usage rates and intellectual interests within the collegial community rather than the capacity of different fields of research to win grants and support. In fact, the management of library subscription budgets provides one of the most tangible devices by which publishing outlets for less marketable disciplines are cross-subsidized by their more marketable cousins. Similar kinds of inequalities would be reinforced among countries and regions. Researchers from poor countries or poor institutions could be forced to publish with lower prestige journals charging lower author fees. Although open access will solve the unequal buying power among universities and readers, it might have the offsetting effect of worsening the weak bargaining power of institutions hoping to attract leading researchers. Large research universities possess the greatest ability to support their authors, which means that they will likely be able to attract a greater concentration of promising and productive researchers from less endowed sectors and jurisdictions (Guterman, 2004).

Open Access as false comfort

I do not want to minimize the achievements of the open access reform movement. It has pointed to the many shortcomings of a publication system that is failing to address the needs and aspirations of scholarly authors, readers and cash-strapped university systems. Yet many of the representatives of the movement sometimes assume an evangelical tone that exaggerates the dysfunctions of the current publishing model. The 2003 press release announcing PLoS Biology includes an alarming warning about high-priced medical journals preventing breast cancer patients and rural doctors from taking full advantage of the fruits of modern science (PLoS, 2003). This leaves the misleading impression that unpriced ‘desktop’ access to journal articles - normally intended for small scholarly audiences - is itself synonymous with realizing the ideals of a healthy intellectual commons and its ability to serve the public good.

It is therefore important to place the open access movement into perspective. No one can be against the ideal of open and free environments for basic research and learning. My reservations – as tentative as they are – have to do with the fact that the open access movement is overstating its radicalism and understating the real challenges of open science. Take for example the following quote by Peter Suber, who represented open access advocates at the UN-sponsored World Summit of the Information Society in December 2003:

For the first time since the rise of commercial publishing of scientific journals, scientific communication can be in the hands of scientists, who answer to one another, rather than corporations, who answer to shareholders (Suber, 2003).

Here Suber presents the case for open access as if it was on the same plane as addressing the conflict between commercial interests and the disinterested norms of exploratory inquiry. In fact, the open access movement has thus far done very little to address the way science and scholarship has been reshaped by the commercialization of research itself, which is almost certainly a greater long-run threat to openness than today’s publication system. Today, countless researchers working in the ‘open knowledge’ system of universities are expected to obtain patents for their discoveries and innovations so that they can be sold or jointly developed with industry. Some have insisted that commercialization initiatives can make industry and society more innovative, providing better avenues for translating the pre-market, ‘blue sky’ activities of university researchers into socially useful downstream applications. But the ‘patent race’ also changes the institutional structures of science (Kelty, 2001), since in many cases it creates systemic disincentives and contractual barriers to sharing or disclosing research information. Perhaps worse, it can violate the spirit of disinterested inquiry by displacing research topics that have weak market support, such as community health or the challenges of sustainable development (Bollier, 2003).

Open access is meant to return science to scientists themselves, but many leading researchers in the scientific community may either benefit from ‘technology transfer’ arrangements or see no conflict at all between commercialized research and the goals of open science. The open access movement itself offers neither an incentive nor a moral pressure for these researchers to tackle the more deeply rooted institutional obstacles to open and free scientific inquiry.

Conclusions and Reflections

Many of the reforms proposed for scholarly publishing are likely to create interesting reconfigurations of scholarly communication. In some respects these reforms would simply follow from the technological transformations and market pressures that have already gripped academic communities and their institutions. In some cases the new publishing system will retain the present form of the traditional journal and the peer-reviewed literature, and, in other instances, reforms will attempt to harness the power of distributed networking in order to create a self-organized model carried on by institutions and groups of scholars themselves. For their part, the defenders of the commercial publishers will continue to make strong appeals to their political independence, their historical role in supporting new and untested areas of scholarship, as well as their ability to invest in many aspects of the publishing process valued by the communities they serve.

Nevertheless, it is certain that the large publishers – especially in the STM fields - cannot continue to be complacent about their traditional symbiosis with academic communities as new alternatives become available. Commercial publishers therefore can be expected to move more quickly to show their willingness to embrace many of the goals of the reformers. For example, the largest biomedical publishers already provide much of their journal content free (in electronic form) to developing countries through the UN project HINARI. Most publishers are also in the process of differentiating their pricing schemes – either on their own, or in conjunction with library consortia - in order to make access more sensitive to ability to pay (Morris, 2003). These reforms are positive, but they do not represent great sacrifices since they involve giving away content for free to users who can’t or won’t pay for subscriptions.

In other areas progress may not be so easy to achieve, predict or even clearly define. For instance, publishers seem eager to retain their market power as ‘bundlers’ of electronic content and managers of access systems (Wellcome Trust, 2003). Libraries that purchase electronic journal content through site licenses – normally through a consortium - are not given access rights to a permanent archive beyond the life of their subscription or access license (Spendlove, 2004). The commercial publishers are unlikely to simply hand over electronic content that can then be freely distributed through the access systems they themselves have developed. In part, this explains the attraction of the ‘open archive’ alternative, which involves open systems and software for linking, searching, downloading and indexing across self-archived material on the public Internet. A number of universities are already setting up repositories of their researcher’s scholarship with standardized tags, and services like Google are currently testing a system for creating federated searches and access across these collections. But once this approach reaches critical mass it could simply lead to a parallel system that leaves traditional journals in limbo. Journals would still be indispensable both for peer review and for mapping and filtering the world of scholarly achievement, yet fewer subscribers would want or need to pay for them (Morris, 2003; Meyer, 2004).

We can conclude that, where scholars and institutions are motivated, and can gather the necessary support, open access journals and archives, or at least cheaper and more liberal access policies, will compete with the more traditional publishing model. The deciding factor in this competition may turn out to be whether the interests of the various constituencies can find a convergence point. This may be difficult since scholars’ own interests are divided; they hope for unfettered access as readers, but also seek credible publishing outlets as authors. For their part, institutions and libraries want cost-effective solutions, but they do not want to invest in solutions that compromise reliability and quality. Finally, there will be questions – like many of the ones I have raised above – about whether the “new political economy” of academic publishing will have a desirable impact on the full range of scholarly norms, ambitions and identities.

Adams, K. M. (2001). Impact Factors: Aiming at the Wrong Target. Cortex, 37(4), 600-603.

Agres, T. (2003). The Costs of Commercializing Academic Research. The Scientist, 17(16), 58. Available: http://www.the-scientist.com/yr2003/aug/prof1_030825.html.

ARL. (2003). Monographs and Serial Costs in ARL Libraries, 1986-2002. ARL Statistics. Available: http://www.arl.org/stats/arlstat/graphs/2002/2002t2.html.

Bergstrom, T. (2001). Free Labor for Costly Journals? Journal of Economic Perspectives, 15(4), 183-198.

Berry, S. R. (2001). The Rationale for "Full and Open Access" of Scientific Information. In S. R. Berry & A. S. Moffat (Eds.), The Transition from Paper: Where are we Going and How Will we Get There? Cambridge, MA: American Academy of Arts and Sciences. Available: http://www.amacad.org/publications/trans.htm.

Bok, D. (2003). Universities in the Marketplace. Princeton: Princeton University Press.

Bollier, D. (2003, June 13, 2003). Preserving the Academic Commons. Paper presented at the 89th Meeting of the American Association of University Professors, Washington, D.C. Available: http://www.aaup.org/events/archived/2003/03Bollier.htm.

Case, M. (2000, March 22-24). Capitalizing on Competition: The Economic Underpinnings of SPARC. Paper presented at the The Economics and Usage of Digital Library Collections, University of Michigan. Available: http://www.si.umich.edu/PEAK-2000/case.pdf

Crawford, B. D. (2003). Open Access Publishing: Where is the value? The Lancet, 362(9395), 1578-1580.

Doyle, H., Gass, A., & Kennison, R. (2004). Who Pays for Open Access? PLoS Biology, 2(4). Available: http://www.plosbiology.org/plosonline/?request=get-document&doi=10.1371%2Fjournal.pbio.0020105.

Frazier, K. (2001). The Librarians' Dilemma: Contemplating the Costs of the 'Big Deal'. D-Lib Magazine, 7(3). Available: http://www.dlib.org/dlib/march01/frazier/03frazier.html.

Gorman, M. (2002). The Economic Crisis in Libraries: Causes and Effects. In R. E. Abel & L. W. Newlin (Eds.), Scholarly Publishing: Books, Journals, Publishers and Libraries in the Twentieth Century (pp. 257-271). New York: Wiley.

Guterman, L. (2004). The Promise and Peril of 'Open Access'. The Chronicle of Higher Education, 50(21). p. A10.

Harnad, S. (2003a). Self-Archive Unto Others as Ye Would Have Them Self-Archive Unto You. University Affairs, December, p. 41.

Harnad, S. (2003b). Open Access to Peer-Reviewed Research Through Author/Institution Self-Archiving: Maximizing Research Impact by Maximizing Online Access. In D. Law & J. Andrews (Eds.), Digital Libraries: Policy, Planning and Practice: Ashgate Publishing. Available: http://www.ecs.soton.ac.uk/%7Eharnad/Temp/digital-libraries.htm.

Harnad, S. (2003c). Mandated online RAE CVs Linked to University Eprint Archives:Improving the UK Research Assessment Exercise whilst making it cheaper and easier. Ariadne(35). Available: http://www.ariadne.ac.uk/issue35/harnad/.

Henderson, A. (2002). Diversity and the Growth of Serious/Scholarly/Scientific Journals. In R. E. Abel & L. W. Newlin (Eds.), Scholarly Publishing: Books, Journals, Publishers and Libraries in the Twentieth Century (pp. 133-162). New York: Wiley.

Jongejan, A. (2003). Commerce and Research: Why Commercial Publishers are Good for Research. Research Information. Available: http://www.researchinformation.info/riaut03elsevier.html.

Kelty, C. M. (2001). Free Software, Free Science. First Monday, 6(12). Available: http://www.firstmonday.dk/issues/issue6_12/kelty/.

Kling, R., & Meyer, E. T. (2002). Leveling the Playing Field or Expanding the Bleachers?: Socio-technical Interaction Systems and arXiv, Center for Social Informatics Working Paper. Available: http://www.slis.indiana.edu/CSI/WP/WP02-10B.html.

Kyrillidou, M., & Young, M. (2003). ARL Statistics 2001-02: Research Library Trends: Association of Research Libraries. Available: http://www.arl.org/stats/arlstat/02pub/intro02.html

McCabe, M. (2002). Journal Pricing and Mergers: A Portfolio Approach. American Economic Review, 92(1), 259-269.

Meyer, M. (2004). Open Access Ignoring Lessons of Dot-Com Bubble. Nature Web Focus. Available: http://www.nature.com/nature/focus/accessdebate/15.html.

MLA. (2002). The Future of Scholarly Publishing. Profession 2002, 172-186. Available: http://www.mla.org/issues_scholarly_pub.

Morris, S. (2004). Open Access and Not-For-Profit Publishers. Nature Web Focus. Available: http://www.nature.com/nature/focus/accessdebate/2.html.

OFT. (2002). The Market for Scientific, Technical and Medical Journals: UK Office of Fair Trading. Available: http://www.oft.gov.uk/NR/rdonlyres/A56C7602-C0BD-428D-BED2-36784363243B/0/oft396.pdf

Okerson, A. (2003). Towards a Vision of Inexpensive Scholarly Journal Publication. Libri, vol. 53. Available: http://www.library.yale.edu/~okerson/Libri.html.

Owens, S. R. (2003). Revolution or Evolution?: A shift to an open access model of publishing would clearly benefit science, but who should pay? EMBO Reports, 4(8), 741-743.

PLoS. (2003, June 26, 2003). Public Library of Science Acts to Increase Public Access to Scientific Research; New Bill will Ensure Public Access to Federally Funded Research Results. Public Library of Science. Available: http://www.plos.org/news/announce_wings.html.

Spendlove, P. (2004). The Future of Research: A popular initiative is changing how scholars - and libraries - work. Quill & Quire, November, pp. 16-17.

Suber, P. (2003). Open Access to Science and Scholarship. Geneva: World Summit on the Information Society. Available: http://www.earlham.edu/~peters/writing/wsis.htm

Suber, P. (2004). Open Access in the Humanities. SPARC Open Access Newsletter. February 7. Available: http://www.earlham.edu/~peters/fos/newsletter/02-02-04.htm#humanities

Velterop, J. (2003). Should Scholarly Societies Embrace Open Access (or is it the kiss of death)? Learned Publishing, 16, 167-169.

Wellcome Trust. (2003). Economic Analysis of scientific research publishing. Cambridgeshire: Wellcome Trust. Available: http://www.wellcome.ac.uk/en/images/SciResPublishing2_7445.pdf.

Willinsky, J. (2002). Copyright Contradictions in Scholarly Publishing. First Monday, 7(11). Available: URL: http://firstmonday.org/issues/issue7_11/willinsky/index.html.

Willinsky, J. (2003a). Scholarly Associations and the Economic Viability of Open Access Publishing. Journal of Digital Information, 4(2).

Willinsky, J. (2003b). Nine Flavors of Open Access Publishing. Postgraduate Journal of Medicine, 49(3), 263-267.

Zandonella, C. (2003a). Open Access Law Introduced. The Scientist. June 27. Available: http://www.biomedcentral.com/news/20030627/04.

Zandonella, C. (2003b). Sabo Bill Assessed. The Scientist. July 16. Available: http://www.biomedcentral.com/news/20030716/04/.