ACRL News Issue (B) of College & Research Libraries 206 / C&RL News ■ April 2004 SCHOLARLY COMMUNICATION Open access in the real world Confronting economic and legal reality by Rick A n d erso n Open access to scholarly information is a hot­button issue that quickly triggers heated dis­ cussion— especially if the topic arises in a mixed group of librarians and publishers. Sometimes the discussion ends up generating useful ideas and prac­ tical solutions to real-world problems; too often, it leads to nothing more than facile phrasemaking or spluttering accusation. Open access has becom e an increasingly im­ portant and potentially divisive issue in recent years as journal inflation rates have increased. For many librarians and scholars, journal price infla­ tion is itself the central problem and open access is the solution. According to this view, the fact that libraries have to pay for access to some schol­ arly information is acceptable, but prices are too high and are increasing at an insupportable rate; the establishment of competitive open-access jour­ nals will force commercial publishers to moderate their profit-seeking behavior.1 Some believe that scholarly information is a public good and ought to be available to the pub­ lic at little or no charge.2 Others believe that a ll information is inherently free and no one ought to have to pay for access to it.3 For still others, the primary problem lies in the fact that academics are producing most o f the scholarly articles in the journal marketplace, and that those articles are then being sold back to the very institutions that produced them; 4 according to this view, the prob­ lem is not that journals cost money, but that the institutions that provide the content are having to pay excessively for access to the very content that they created. Then there is the question of whether access to information that has been created with the support of public funds should be restricted at all. These and other issues surrounding open ac­ cess are important, and they deserve serious con­ sideration. Serious consideration however, requites the recognition of certain legal and economic re­ alities. While choices made by authors, publishers, and librarians do have an effect on the informa­ tion marketplace, their choices and actions have little or no effect on the deeper economic reality in which that marketplace exists. That reality is determined in fundamental ways by two simple facts over which the human players in the infor­ mation economy have little control, and a produc­ tive and intelligent conversation about open access must proceed from a recognition o f these facts. In addressing them here, it will seem to some readers that I am belaboring the obvious, and to them I apologize— but I think a careful treatment of these points is necessary, because while many in the library profession recognize them as self- evidently true, there are some who regard them as blasphemous. Th e m y th o f fre e in fo r m a tio n First, there is no such thing asf ree inform ation. Most people (including most proponents of open ac­ cess) understand this implicitly, but it might be worthwhile to discuss why this is so. Information is not the same thing as ideas or concepts. Ideas may be free, but they do not become information A b o u t th e a u th o r Rick Anderson is director o f resource acquisition at the University o f Nevada-Reno, e-mail: rickand@unv.edu © 2004 Rick Anderson mailto:rickand@unv.edu C&RL News ■ A p ril 2004 / 207 until a person creates a symbolic representation of them. Doing so is a process o f labor, and the cre­ ator w ho undertakes that labor is incurring a cost. T h e creator m ay th en ch o o se to distribute the information at no charge to others, but that does not m ean that the information is fr e e —it means that its creator has chosen to absorb the costs of creation and distribution rather than try to re ­ cover them. W hen information can b e generated and distributed very cheaply, or when unrestricted distribution o f it will benefit the creator in som e way, h e or she might ch o o se to m ake it freely available to others, and, in fact, people do so ev­ ery day (on the open Web, in casual conversation, via e-mail, etc.). But the costs o f producing and distributing that information have not magically disappeared in these cases; they have b e e n ab­ sorbed by the author. Authors are less likely to give away informa­ tion that requires a great deal o f time and effort to create. O ne way or another they will usually try to recover their costs or even realize a profit, as will the m iddlem en w ho turn the raw informa­ tion into a publishable product and then distrib­ ute that product. Freelance authors usually re­ coup their costs by trading copyright for money. Scholars (most o f w hom are paid a salary to cre­ ate information) usually try to trade their copy­ rights for enhanced reputation or professional ad­ vancement by submitting their writings to presti­ gious publishing houses or journals. Either way, those who produce labor-intensive, high-quality information usually try to get something in return for their labor. T he cost o f creating information, the cost o f preparing it for publication, and the cost o f dis­ tributing it are all quite distinct from each other, and each type o f cost may ri se or fall independent o f the other. W hen information was published primarily in print format, distribution costs were very high; in the electronic realm, they are rela­ tively low. B ut even in a publishing system in­ creasingly characterized by online distribution, the cost of p rep arin g information for distribution remains significant.5 In the case o f many scholarly journals, it is true that editorial tasks are performed by contributors at no cost to the publisher. How­ ever, even when articles and editorial services are provided at no charge, the remaining costs of pre­ paring information for publication are consider­ able. These include a publisher’s staffing and over­ head costs, w hich are often substantial and may actually increase when a journal moves from print to online. Consider, for example, the added costs o f robust and long-term archiving (which is inte­ gral to the open access concept and not an issue with which publishers typically concern themselves in the print realm), the cost o f hiring a Web mas­ ter and other content maintenance staff, and the costs o f acquiring and maintaining servers. A jour­ nal publisher that employs its ow n editorial staff has even greater costs to recover. One more complicating factor in the scholarly information econom y is the fact that many jour­ nals, particularly those published by nonprofit so­ cieties, bear the weight of recovering other costs in addition to those incidental to their ow n cre­ ation and distribution. O ften subscription rev­ enues support legitimate organizational activities that have nothing to do with the journals them­ selves, such as annual conferences or member ser­ vices. For-profit publishers have legitimate costs that must b e m et as well, though in som e cases they could lower their prices significantly and still recover those costs with revenues to spare. (What kinds o f activities nonprofit societies ought to be funding from subscription revenues and to what degree commercial publishers should b e allowed to seek profit in the scholarly-information mar­ ketplace are contentious topics best suited to an­ other essay.) T h e m y th o f in fo r m a t io n a s a p u b lic g o o d There is a second basic reality o f the information econom y that bears on the open-access question, and this one is more controversial than the first: Inform ation is not a p ublicg ood. Too often, commen­ tators on the econom ics o f scholarly information seem to confuse the con cept o f “a public good” (meaning something that is owned by the public) with “the public good” (meaning the general wel­ fare). The fact that something is good for people, or the fact that its broad distribution w ould b e beneficial to the general public, does not make it a public good. What makes something a public good is legal public ownership. A municipal park is one example o f a public good; it is paid for and main­ tained with public funds, and belongs equally to all citizens. Most o f the information in scholarly journals does not fit this definition. Is it p ossible for inform ation to b e a public good? Yes, if it is created and owned by the pub­ lic. The information contained in a government document is a public good, which is why govern­ ment documents are exem pt from copyright. In­ formation produced b y private individuals and organizations, how ever, is not a public good. 208 / C&RL News ■ A p ril 2004 W hatever its potential benefits to the world at large, however much the public might benefit from its free distribution, m ost information is never­ theless the intellectual property o f those who cre­ ate it. We can object to that fact if w e wish, but it is still a legal fact. This begs a question, o f course, and it is one that bears heavily on the open access issue: What about information that is produced with govern­ m ent funding? If the public is underwriting the creative costs o f that information, should it be freely available to the public, as som e have ar­ gued?6 Like all “should” questions, this is one that has more than one possible answer. One reason­ able answer is, “It depends on how much govern­ ment funding was involved.” Information created at just about any university or college (public or private) can b e said to have b e e n supported to some degree by public funds— does it follow that every physics professors research article and every English instructor’s short story collection should pass immediately into the public domain? Perhaps a certain threshold o f government funding should b e set, such that inform ation created w ith the substantial support o f public funds becomes pub­ lic property by definition (as Congressman Martin O . Sabo has proposed in H.R. 2613, the Public Access to Science Act, which is still in committee as o f this writing). Until such a proposal becomes law, however, there is a legal reality within which w e must w ork: m ost information created with the support o f governm ent funding is, in fact, copyrightable, and the copyright in most publicly funded information is held by the author. Im p lic a t io n s f o r o p e n a cc e s s B oth o f th ese facts— that inform ation is inher­ ently costly, and that information is not usually a public good— pose challenges for the idea of open access. For information to b e made freely and per­ manently available to the public, the costs o f cre­ ation, publication, and distribution must be absorbed by som eone other than those w ho wish to use it. The Internet eliminates most distribution costs, but not all o f them, and does not affect creative costs or publication costs to any appreciable degree. D o these inconvenient facts mean that open access is not desirable or that it cannot work? No. But they do define some limits to our options, and a recognition o f those limits should lead us to have patience with publishers that are moving more slowly towards that model than w e might like. Any proposal that is built on the premise that information is inherently free, or that online pub­ lication can be undertaken without significant cost, will not work in the real world becau se both o f those premises are demonstrably incorrect. If w e do not b ear in m ind the intrinsic costliness o f information, w e will have a very hard time dis­ cussing intelligently the intractable econom ic re­ alities that govern its creation and distribution, let alone formulating pricing and distribution strate­ gies that will provide maximum public benefit. Nevertheless, the problem s that have led to the current controversy over journal pricing and open access are real and need to be solved in some way. Clearly the status quo is insupportable; if the price o f scholarly journals continues to rise at the present rate, research libraries will quickly lose the ability to meet the needs o f their patrons. At the sam e time, it is not reasonable to e xp ect all journal publishers (even nonprofit ones) to move immediately into an open-access m odel simply because the Internet has lowered the cost o f dis­ tribution. We w h o are trying to act in the b est interests o f the public and m axim ize access to quality information must balance zeal with rea­ son. Most journal publishers are operating in good faith and need our patience and support, as well as our exhortations, as they work through the diffi­ cult process o f moving from the old information world into a new one. N otes 1. “Why is SPARC Needed?,” in the Frequently Asked Questions section o f the SPARC Web site at www.arl.org/sparc/core/index.asp?page=eO (accessed March 4, 2004). 2. Richard Edwards and David Shulenberger, “T he High Cost o f Scholarly Journals (and What to Do About It),” C h an g e 35, no. 6 (November/ D ecem ber 2003): 10. 3. John Perry Barlow, “The Economy o f Ideas,” W ired 2, no. 3 (March 1994): 84. 4. Academic Senate Committee on Libraries, Columbia University, “Crisis in Scholarly Commu­ nication: Motion re Stanford’s Reaction to the Seri­ als Crisis.” Version referred to is dated February 19, 2004, library.cpmc.columbia.edu/hsl/scholcom/ clibresolution.html (accessed March 4 , 2004). 5. Donald W. King and Carol Tenopir, “E co­ nom ic Cost Models o f Scientific Scholarly Jo u r­ nals,” paper presented at the ICSU Press Work­ shop, Kebl College, Oxford, UK, April 1998, ww w .bodley.ox.ac.uk/icsu/kingppr.htm (a c ­ cessed March 4, 2004). 6. Edwards and Shulenberger, “The High Cost o f Scholarly Journals.” ■ http://www.arl.org/sparc/core/index.asp?page=eO http://www.bodley.ox.ac.uk/icsu/kingppr.htm C&RL News ■ April 2004 / 209