Then I saw this article about the Rijsmuseum in Amsterdam, which reminded me that even in Pandora’s box, hope remained in the bottom — some museums are bucking the trend and creating reuse-friendly policies for images of public domain works.

Whenever I am asked about the process of getting “copyright permission” from museums to use images of artworks in a new project, I start by explaining three basic principles:

1. Many of the artworks held by museums around the world are in the public domain, either because they were created before copyright came into existence in the late 17th and early 18th centuries or because any copyright they had has expired.  And even for works that are still protected by copyright, most often the museum that owns the art work does not also hold the copyright.
2. Even a photograph or digital scan of a public domain artwork that is created by the museum may lack copyright protection under a U.S. District Court decision that said that such “slavish” reproductions — reproductions that add nothing but attempt *just* to accurately reproduce the work — are not original enough to gain any copyright other than whatever protection the underlying work enjoys (which, if the work is public domain, is none).  This is only a District Court case, but it has never been appealed or, to my knowledge, otherwise challenged so it probably should guide us unless or until we get something more definitive.
3. Museum claims over the use and reuse of images from their collection, even though often called and often understood as copyright claims, are really mostly a matter of contractual agreement and simple control over access to the unique works that the museum holds.  Because these works are (more or less) unique, the ability to make a reproduction can be tightly controlled and the museum can impose contractual conditions on access to make such a reproduction or on the reuse of a museum-supplied image.  These contractual obligations actually bind only the parties that agree to them, but in practice they are often passed to downstream users in the form of conditions on reuse that the original recipient feels bound to impose.

All of this is explained very elegantly and discussed in great detail in a superb article written on the topic by Kenneth Crews of Columbia University, which is available here.  It is a subject all librarians, in my opinion, should understand, so the article deserves a wide readership.  Dr. Crews points out both that many claims made by museums tend to stretch the copyright law beyond reasonable bounds AND that some of the claims for control and remuneration are supported by sound business practices and cultural policy considerations.  This is not a black or white issue, just one where more clarity and an open discussion of genuine needs and concerns can lead to better conditions for the reuse of artworks and images.

Which brings me to the Rijsmuseum.  Their decision to offer high-resolution images of many works in their collection available for free download is a startling example of the other side of this issue — there are good reasons, especially from the perspective of fundamental museum missions to make culture more accessible to the public, to take the opposite approach from that of some museums and support radical reuse.  I was struck by the reasoning behind allowing even commercial reuse of these high-res images:

If they want to have a Vermeer on their toilet paper, I’d rather have a high-quality image of Vermeer on toilet paper than a very bad reproduction.

Maybe the example chosen is not one involving high culture, but it illustrates quite dramatically that reuse of public domain art is inevitable, even for purposes we may deplore, and that the public is, in many cases, better served by access to good-quality reproductions than it is by complex contractual terms and absurd restrictions on even the most traditional practices (like sketching an artwork at the Art Institute of Chicago).  If we must chose between extremes, there is a great deal to recommend the Rijsmuseum approach.

As libraries, museums and archives work to digitize more and more of the public domain materials that they hold, it is worth considering how others will be able to use and reuse those images.  With so many digital humanities projects being pursued around the world, the thickets of permissions and use restrictions will grow evermore burdensome.  An example like the Rijsmuseum is a great counterweight to that burden, and a graphic reminder that it really does not have to be that way.  Of course, not every cultural institution can afford to make high-res reproductions available for free; fees are often essential to support the mission-driven activities of these collections.  So fees may be needed for the provision of high-res images, as may some restrictions on further dissemination of those images. But many of the restrictions and fees we often encounter cannot be justified in this way.  So if we start from the very open policy of the Rijsmuseum and then apply only those restrictions that have sound, policy-based foundations, we can arrive at much more supportive approaches to reuse and new creativity.  The policy adopted by the Cornell University Library’s digital collections is, in my opinion, a model of such a moderate and sensible approach.

For those beginning to explore the uncharted territory of the digital humanities, permission fees and reuse restrictions will probably continue to create nearly unnavigable thickets of complication.  But with these few counter-examples, we can see that a better approach is possible.  Libraries and the digital archives associated with them need to model the best practices that we can in hopes that the most absurd kinds of copyright overreaching will become less common and rational policies based on an accurate assertion of rights and a realistic assessment of needs will begin to dominate.

The publishers proposal — the acronym stands for Clearing House for the Open Research of the United States — is described in glowing terms on the Scholarly Kitchen website and with a bit more restraint by the Chronicle of Higher Education.  The proposal from the education associations, dubbed Shared Access Research Ecosystem, is also described by the Chronicle and is the subject of a detailed draft proposal that can be found here.

For myself, I would rather SHARE than join the CHORUS, for a number of reasons.

First, I think CHORUS is being touted, at least in what I have read, by comparing it to a straw man.  Its principle virtue seems to be that it would not cost the government as much as setting up lots of government-run repositories, clones of PubMed Central.  But it is not clear that that option is being seriously suggested by anyone.  Certainly many of us encouraged the agencies to look at the benefits of PMC for inspiration and not sacrifice those benefits in their own plans, but that does not mean that each agency must “reinvent the wheel,” no matter how successful that wheel has been.  So the principle virtue of CHORUS seems to be that it does not do what no one is suggesting be done.

The most important thing to understand about CHORUS is that it is a dark archive.  The research papers in CHORUS would not be directly accessible to anyone; they would be “illuminated” only if a “trigger event” occurred.  Routine access would, instead, be provided on the proprietary platforms of each publisher, while the CHORUS site would simply collect metadata about the openly-accessible articles and point researchers to the specific publisher platforms.

It seems to me that the CHORUS proposal is “disabled” from the start, by which I mean that it lacks three fundamental abilities.  CHORUS, at least based on the descriptions we have seen, lacks find-ability, useability and interoperability.

Perhaps the most troubling remark in the description offered on the Scholarly Kitchen blog is that “Users can search and discover papers directly from CHORUS.gov or via any integrated agency site.”  Does this mean that even the collected metadata would not be available to Google?  We know how few researchers “walk through the front door” of our research tools, so limiting discovery to the CHORUS portal or “integrated agency sites” would make these open access papers virtually invisible (which, one suspects, is the point).  As things stand now, open access papers which reside on proprietary publisher platforms are difficult to find because there is no consistency in how they can be discovered.  That is the principal reason so many COPE fund institutions will not support so-called “hybrid” open access publishing that makes a few articles open on an otherwise toll-access site.  It does not seem that CHORUS would change that unfortunate situation at all, which is probably why Heather Joseph of SPARC calls CHORUS “a restatement of the status quo.”  The public would gain very little, since the major goal of the proposal is for the publishers to cling tightly to control over the research papers that have been entrusted with.

Another ability that CHORUS would lack is useability, since as far as we know, all that a researcher or other user could do with these papers is read them.  It would not, of course, facilitate sharing, teaching or reuse, even those these abilities are vital to improving the speed and quality of research in the United States.  And then there is interoperability.  If the geographically desperate archives are genuinely federated, searches across all of them, even keyword searches that are not dependent on the metadata created for each article, would be possible.  So would text and data mining across a large corpus of works.  We already know that such interoperability creates tremendous new opportunities for expanded research, collaboration, and previously impossible discoveries.  But there is no reason to believe that CHORUS would support interoperability, since the various publishers have a strong competitive interest in not allowing cross-platform activities.  Research and education, however, not only do not benefit from that competition, but are actively “disabled” by it.

On the other hand, the proposal from the universities and their libraries is for a genuinely federated system of university-based repositories.  Those repositories already exist, so if we are going to make a cost argument, it really favors SHARE.  And these repositories, unlike the publisher platforms, have a strong interest in facilitating discovery.  Also, the detailed proposal offered by these groups addresses text and data mining, semantic data, APIs for research and linked data.  All of these things make university-based research better, while they pose threats to the commercial publishers who have designed CHORUS to protect themselves, not to benefit research or the public.  So all the incentives line up between the public interest and the university-based SHARE system.

If the OSTP and the research-funding agencies take seriously all of the opportunities that were described in the comments they have solicited over the past year, it will be very obvious to them that CHORUS is singing flat, while it would be good to SHARE, just as our parents always told us.

In this brief the plaintiff publishers are attempting to respond to the arguments made in the last round of filings by Defendant Georgia State University and the “friends of the court” who argued on its behalf.  This is essentially their last shot at convincing the judges that they should reverse the decision of the trial court that largely ruled in favor of fair use.  Unfortunately for the publishers, it is not very convincing.

The core argument of this brief is one the publishers have been making, unsuccessfully, throughout the case — that the so-called “course pack” cases from two decades ago are absolutely dispositive in this case and render incorrect all of the fair use analysis done by Judge Evans in her comprehensive ruling at trial.  Some of this argument is based on misdirection, wherein the publishers argue that the judge was treating digital formats differently from print.  She did no such thing, of course; she distinguished the situation before her from the course pack cases because the facts were different.

The publisher  do address the distinction more directly, however, when they argue that GSU’s non-profit status is (for obscure reasons) not relevant to the fair use analysis.  Of course, the course pack cases explicitly rejected  fair use precisely because the defendants were commercial entities who were selling the course packs to students.  In the GSU case, on the other hand, the university is non-profit and the readings are made available at no charge.  It is absurd to say that this difference doesn’t matter.  In fact, at least one publisher I have dealt with granted gratis permission for an excerpt on e-reserves only after I assured them that no charge was made to students.  So even publishers understood that this was a crucial fact in determining the appropriateness of unlicensed uses in academia, before their lawyers told them they had better deny it.

From this basic assertion that the matter was already settled by different courts back in the 90′s, the publishers proceed to explain that the trial court was wrong on every one of the four fair use factors.  Much of the argument is very strange, twisting back on itself in an effort to obscure the clear and cogent path that Judge Evans used to arrive at her original ruling.  Its overall effect, however, is to emphasize what a sensible, careful analysis that trial court opinion offers.

It is, perhaps, worth looking at some of the general statements made in the opening summary of the brief’s argument to find some clues about how this analysis went so badly wrong.

First, of course, is this assertion that fair use is not supposed to favor non-profit educational uses.  We have heard this complaint many times over the long course of this cases, but here is last week’s version of it:

To sanction GSU’s practices under the rubric of fair use on the grounds urged by Appellees — GSU’s non-profit status and the importance of Appellants’ work as teaching tools — would undermine fundamental tenets of copyright law by effectively dedicating the works of scholarly publishers to the public domain. (emphasis added)

This is truly a staggering bit of argumentation.  First, a non-profit purpose is specifically mentioned as favoring fair use in the text of section 107 itself, so it is hard to see how it could NOT be an appropriate part of a fair use argument.  And a quick look at all of the exemplary purposes for fair use that are also mentioned in section 107 should convince any unbiased reader that social utility — the importance of education, for example — is the fundamental purpose for which fair use was created.  Indeed, courts have told us this over and over.  So the rubric suggested as inadequate is, in fact, a nice, concise statement of why the original ruling in this case was correct.  And then, the grotesque hyperbole that follows illustrates for us that the lawyers writing this brief know how far out on a limb they have climbed.  To say that a fair use finding effectively dedicates a work to the public domain is simply absurd.  I do not think the publishers could cite a single instance where fair use has had that effect, where the original became unmarketable because someone made a fair use of a portion of that work. When movie companies challenged the video recorder, for example, by saying that it would destroy the movie industry, they were making a parallel, and equally silly, claim; home video recorders, of course, did not harm the market for movies at all, even though entire films could be recorded under the Supreme Courts ruling on fair use.

And of course, we should not ignore that phrase “the works of scholarly publishers.”  Throughout this brief the publishers are not sure whether they are arguing for their own profits or on behalf of authors.  Sometime they throw the authors in to their argument, but mostly they want the court to believe that they, the publishers, actually created the works in question ex nihilo.  But none of these works were created by publishers, and their argument collapses when we realize that the fees they are seeking have virtually no role at all in incentivizing scholarly creation, which is what copyright is for. More about this in a minute.

But before we leave fair use proper, we should also examine this gem: “Fair use … imposes on the proponent of fair use the burden of demonstrating the limited nature of the unauthorized use …”  There is no such requirement in the text of the law, of course, and the Sony case about VCRs illustrates, at the highest level of U.S. legal precedent, that simply because a practice is widespread does not prevent it from being a fair use.

To return to the issue of incentives, let’s look at one more passage from the preliminary statement in this brief:

This blatant end-run of copyright law not only threatens to undermine the established legal norms that have long governed course-pack copying, but it comes at a time when Appellants and other academic publishers are investing heavily in publishing and delivering content in digital form.  These publishers cannot hope to recoup their investment if institutions like GSU are permitted to make exact digital copies of their works … without compensation to the works’ authors and publishers.

The first part of this quote is, as we have already seen, irrelevant; a fair use ruling in the GSU case would not alter the precedent created by the course pack cases, when those cases are properly understood.  But the second part is also interesting, and it echos an earlier statement, quoted from one of those course pack cases, about how publishers have “risked their capital to achieve dissemination.”  By way of reply, we should note first that dissemination is quite different now than it was nearly twenty years ago when that case was decided.  Many  Gold OA journals and other forms of digital writing can be distributed with much less expense than these traditional publishing houses claim is required.  So we are entitled to ask if it is being done efficiently by them; whether that capital the Court is asked to protect is being invested wisely.  And in any case, copyright law is not designed to support any particular business model, but to give creators more reason to create.

Given these statements, the Appeals Court would be justified, I think, in asking the plaintiffs to open the books and show how much investment is being made, how it is being spent, and how dependent that investment really is on permission fees.  The trial judge did not think those fees mattered all that much to publisher revenue.  If the plaintiffs continue to assert that their business will be ruined by fair use, they should be obligated to prove it, and also to show that the threat is systemic and not just the result of poor management.

Finally, it is worth noting that authors have now appeared in the argument.  Whereas earlier the issue was “the works of scholarly publishers,” now, in classical fashion, the authors have been belatedly recalled and tossed in to the mix.  Publishers have made pathos-filled appeals on behalf of starving authors in order to justify their own businesses since copyright began.  But academic authors are different, so this reflexive reference to authors should not go unchallenged.  In response to this argument, there are two questions the Court of Appeals should examine closely.  First, how much money actually makes its way into the hands of authors, rather than the publishers, from permission fees paid to the Copyright Clearance Center?  Second, what role, if any, do these fees play in creating the incentive that academic authors have to create scholarly monographs, recognizing that all of the books in question in this lawsuit are such monographs, rather than textbooks.  These questions are perfectly within the competence of the Court of Appeals, both because they are relevant to the second fair use factor and because the publishers’ reply brief has put them at issue.

Th Eleventh Circuit Court of Appeals is asked to swallow lots of unbelievable things in this reply brief.  Now is the time for them to sit down at the metaphorical breakfast table (harkening back to the White Queen) and demand substantive information from the plaintiffs before they finish this meal.

I have already written about some of the arguments made in support of the publishers who brought the original lawsuit and are pursuing this appeal after having lost in the district court.  Notably, we have discussed the strange argument made, essentially on behalf of the Copyright Clearance Center, which is helping to organize and finance this quixotic lawsuit, by two former Registrars of Copyright that fair use was never intended to favor non-profit educational uses.  We also noted the decision by the U.S. Justice Department NOT to get involved on behalf of the publishers.

When the briefs supporting the publishers are examined, the striking fact, to me, is that no one supports their position who does not stand to gain direct financial benefit if it is upheld.  No one is arguing that giving more money to the intermediaries who are such a drain on college and university budgets is a good idea in itself, or would be a benefit for scholarship.  The supporting groups like the American Association of University Presses (whose participation on this side is especially tragic), the Authors Guild and the “Copyright Alliance” are all looking to line their own pockets by supporting a reversal of the trial court.  Not surprising really, nor an evil intent in the abstract, but something that is well beyond the purpose and intent of the copyright law.  This is prospecting for gold on the part of the publishers, and they are trying to do it on public property.

I was interested to see that one of the parties on the amicus brief filed by the Author’s Guild was a group called the “Text and Academic Authors Association,” of which I had never heard.  Was this really an group of academic authors opposed to fair use on campus?  Well, only sort of.  The website of this oddly named group (all authors write text; I think they mean “textbook”) shows that the majority of their leadership council is made up of non-academics or retired professors, who presumably no longer need to rely on fair use for good teaching.  And the strange perspective of the group can best be judged by this article arguing that textbook prices are justified and are not too high to interfere with quality education, a perspective thoroughly debunked by nearly every study as well as by day-to-day experiences on campus.  Indeed, the only sensible way to read the article is to recognize that every “myth” it undertakes to refute represents a demonstrable fact. More on this shortly.

So now lets look at the friends of GSU and some of the arguments they are making.  In addition to the brief filed by GSU itself, there are at least five amici groups -  The Library Copyright Alliance filed a brief on behalf of the ALA, the ACRL, and the ARL, and was joined in that brief by the Electronic Frontier Foundation.  There was a joint brief from the American Council on Education, the Association of American Universities and three other higher education groups, one by the American Association of University Professors, and another from an independent group of academic authors and legal scholars.  Finally, ASERL, the Association of Southeastern Reasearch Libraries, which is the nation’s largest regional research library consortium, filed its own brief (which — full disclosure — I had a very small role in assisting with).

Note how clearly these friends of the court break down between those arguing for their own private gain versus those trying to uphold a public good.  Given all the rhetoric about copyright as primarily intended to benefit the public interest in hundreds of precedents, the 11th Circuit Court of Appeals cannot help, one would think, but notice this disparity.

Let’s look at some of the arguments from a few of these documents.

In the Appellees’ own brief (that’s Georgia State) we find the fundamental points that are elaborated and supported by all of the amici.  Primarily, the Appellees argue that all four of the fair use factors weigh more or less heavily in favor of fair use for short excerpts provided as course readings and accessible only by students in a particular class.  This argument is supported in the brief from ACE and the other university groups.  The GSU discussion is especially interesting on the fourth fair use factor, impact on the market.  First, it makes the point that the publishers argument about how the “easy” licensing that is supposedly available should make this factor count against fair use is circular.  As many courts have recognized, this argument would undermine virtually all fair use, which Congress could not have intended (although the publishers do), and insofar as Judge Evans accepted it in the trial court, she erred.  Then the Appellees go on to point out that that licensing market actually is neither easy nor comprehensive.  Fair use continues to be needed in this area even if one applies a market failure standard for fair use, because the “market” for licensing, and even the CCC’s blanket campus license, fail higher education far too often.  As we will see, other amici also support this point.

Two other points made in the Appellees’ brief are worth noting.  First, they argue, as I have in the past, that Judge Evans’ work-by-work analysis of fair use was the appropriate approach to the case, dictated both by the nature of fair use itself and by higher court precedents.  Second, they show that the publishers’ reliance on the principle of “media neutrality” is just a silly distraction (my words, not those in the brief).  Judge Evans simply did not apply fair use any differently in the digital realm than it would be applied to print works; she distinguished some cases that involved print, but she did so for much better reasons, reasons the publishers would like the 11th Circuit to overlook.

The Library Copyright Alliance also demonstrates, in their brief, that all four fair use factors actually favor Georgia State’s fair use argument, supporting from a slightly different perspective the argument made by GSU and by the multiple college and university associations.  The LCA goes on to argue that this fair use argument is widely-recognized in the educational community, and that GSU’s policy on copyright and fair use is consistent with widespread practices through education.  One result of that fact is a clear demonstration that upholding the trial court’s finding of fair use would not have any negative effect on scholarship.  Colleges and universities have relied on fair use in this way, even in the pre-digital world, for a long time.  Yet scholars continue to produce scholarly works at an ever greater rate, undaunted by fair uses made of those works (and they continue, unfortunately, to transfer copyright in those works to these irresponsible publishers).  Fair use supports scholarship, it does not undermine it.  And the publishers have not discovered any “new” threat to scholarly production that must be averted; they simply decided that they needed and deserved more money from academic budgets.

The LCA also develops the point about the failure of the licensing market for electronic reserves and other course readings.  A reversal in this appeal would not cause that market to grow.  It would not, that is, actually result in more money in the system to support scholarship, even if we assume that money in publishers’ coffers did support scholarship.  In fact, if the trial court’s findings of fair use were reversed, education would be harmed because fewer resources would be available for teaching, since library and university budgets cannot support the astronomical fees that publishers want to charge for licensing.  The overall effect on the production of scholarly work would be negative, which undermines the fundamental purpose of copyright law in the U.S.

The ASERL brief develops this point a bit further, partly by pointing out that the licensing market touted so highly by publishers is already harming the ability of colleges and universities to teach students.  On every campus it is easy to find stories about how the inability to get permission, either because of the prohibitive cost of licensing or because no license for the particular work was available, forced a teacher to changed his or her plans and resort to “plan B” pedagogy.  Most librarians have had to assist such faculty to find other, less optimal, resources in those situations; it is something we do well, but wish we didn’t need to do.

Related to this point is the discussion in the ASERL brief about the cost of higher education, and the role of publishers and licensing in those costs.  One statistic the brief cites shows that textbook prices have actually risen much faster in the past 30 years than tuition and fees have, which is a telling refutation of the argument made by the TAA in the article mentioned above.  And the structure of digital licensing from the CCC actually contributes to accelerating costs, because the CCC will refund the licensing fees for printed course packs that are not sold, but requires that e-reserve materials be licensed for each student in the class regardless of how many times an excerpt is actually accessed or even whether or not it is used by anyone.  Where, by the way, is their concern for “media neutrality” in that pricing policy?

Finally, I want to end this long post by quoting a passage from ASERL’s amicus brief that deserves to be remembered by everyone thinking about fair use.  In discussing the mistaken assertion by the publishers that all fair uses must be transformative and that, in any case, fair use should be extremely limited in application, ASERL reminds the Court of Appeals that, on the contrary, fair use is an integral and indispensable part of the very structure of U.S. copyright law, without which that law would arguably be unconstitutional:

Fair use is not a rarely-used “exception” to a copyright holder’s rights that should only be applied “on occasion.” [citing arguments from Appellants brief] Rather, fair use is viewed by the courts as “necessary to fulfill copyright’s very purpose.” [citing the Supreme Court in the Campbell case]… Fair use is necessary, in part, because “[t]he primary objective of copyright is not to reward the labors of authors, but [t]o promote the ‘Progress of Science’”… [citing the Constitution and the Supreme Court in the Feist case].  And for that reason, following direction from Article III of the Constitution, the District Court was correct to apply fair use “in a way that promotes the dissemination the knowledge, and not simply its creation.” [quoting Judge Evans and again citing the Supreme Court].

All of the arguments that support BOTH the creation and dissemination of knowledge line up on the side of affirming the trial court in this case.  It ought to be an easy decision for the 11th Circuit Court of Appeals.  And it is well past time for scholars and universities to rebel against so-called “scholarly” publishers who try to use the courts to undermine the best interests of research, teaching and learning in a futile attempt to improve their bottom line.

The decision from the Appeals Court tells the story of this dispute very clearly, but just to summarize, let me remind readers that Patrick Cariou took a series of photographs of Rastafarians that were published in a book, now out-of-print, called “Yes, Rasta.”  Richard Prince then made a series of appropriation art works, which included collages of the photos as well as various other alterations.  The trial court in the case decided on a summary judgment motion that these art works by Prince were copyright infringement.  Judge Deborah Batts, whose opinions we have had cause to regret in the past, held that a fair use defense for the appropriation art failed because the new work must “comment on, relate to the historical context of or critically refer back to the originals.”  When asked what his point was in these artworks, Prince told the trial court that he did not have a point, and that was very damning in Judge Batts’ eyes; for her, his works could be transformative only insofar as they were making a comment about Cariou’s work.

In Thursday’s decision, the Appeals Court told us, and Judge Batts, that this was not the right standard for assessing transformation for the purposes of fair use.

By the way, in her original injunction, which was vacated by the Appeals Court, Judge Batts had given Cariou the right to destroy Prince’s allegedly infringing works.  I wrote about this with some outrage two years ago, so it is worth noting that, to their credit, Cariou’s counsel told that Appeals Court early on that they opposed destruction of the art, even though they wanted it to be held to be infringing.

The Second Circuit begins its opinion by pointing out, in clear and forceful language, that copyright is not intended to give authors or other creators “absolute ownership” in their works, as if by natural right.  Instead, the Court notes, copyright is designed to stimulate creativity and progress in arts and sciences.  This is not new, but placed as it is in the opinion, it strongly reinforces the point that fair use is part of the structure of copyright, not an oddity or a mere exception for extraordinary situations.  Without fair use, copyright fails in its Constitutional purpose.

As for the correct standard for deciding if a work has a transformative purpose, the Second Circuit wants a broader rule than that articulated by the trial judge.  Transformation can exist even without direct comment on the original, whenever the original work is altered with “new expression, meaning, or message” (quoting the Supreme Court in the Campbell case).   The new work can be transformative if it “superseded the object of the original creation” by offering “new information, new aesthetics, new insights and understandings.”

Significantly, and especially important given Prince’s refusal to assign a “point” to his work, the court wants us to look at transformation from the perspective of the viewer, not the creator:

Prince’s work could be transformative even without commenting on Cariou’s work or on culture, and even without Prince’s stated intention to do so.  Rather than confining our inquiry to Prince’s explanations of his artworks, we instead examine how the artworks may “reasonably be perceived” in order to assess their transformative nature.

For me personally this is very reassuring.  One of the ways I frequently tell students, faculty and librarians to try to decide if a proposed use is tranformative is to advise them to ask themselves three questions.  First, will the “quotation” of the original help me make my point?  Second, will it help my reader/viewer get the point?  Finally, did I use no more than necessary to make my point? These questions, by the way, are borrowed from LA attorney Dean Cheley, thanks to a panel we were both on at the 2012 ALA Annual conference.  The part of the decision that reinforces these questions is the reference to what will help readers and viewers perceive the new message or new aesthetic.  And while my “clients” usually do have a point to make, it is encouraging to see that fair use supports even appropriation art for its own sake.

Finally, the Second Circuit is very careful in its discussion of potential markets when assessing a transformative fair use.  It is not enough, the Court says, that argue the new work could have been licensed, and thus assert that any fair use harms that hypothetical market for a license.  Instead, the Court reminds us that the question is whether the alleged infringer has “usurped” a market in which the target audience and the nature of the content is the same as for the original.  Where the audience for the new work is different, and there is nothing to suggest that the original rights holder would have thought to exploit that different audience or communicate the new message, aesthetic or insight to them, the entirely speculative possibility of a license will not undermine fair use.

In the end, the Appeals Court finds that 25 of the challenged art works are fair use.  It remands only five of them back to the District Court for a new decision applying the correct standard.  The chances are good, I would think, that this will never actually get back to the trial court, because some settlement, based on a licensing fee in regard to those five, is probably in everyone’s best interest.  But regardless of what happens about those five art works, we now have a very strong decision about transformative fair use for appropriation art out of the Second Circuit.  Although not all fair uses, as we know, have to be transformative, this kind of decision helps lend support to many of the most creative forms that teaching and learning take on our campuses.

Coincidentally, the story involves a Duke professor, Dr. Mark Goodacre of Duke Divinity School, who is an active and engaging blogger on the general topic of the New Testament and early Christian literature.  Several years ago, Mark wrote a blog post outlining a possible approach to a long-standing problem in the interpretation of one of the non-canonical gospels, the Gospel of Peter.  By Mark’s own admission it was a casual piece of writing, as many academic blogs are (he calls it “random jottings,” but that is probably excessively modest).  Recently, however, an entirely non-casual peer-reviewed article critiquing Mark’s blog post has been published in a highly-regarded journal in the field.

Mark tells his own story, and links to the relevant documents, in this later blog post.  He also raises some interesting questions about the etiquette of the situation, which are discussed at length in the comments to the post.  The entire discussion is worth reading, but I want to make a specific comment about how it relates to those intersections of scholarly communications and information literacy.

One of the things that information literacy librarians spend a lot of time teaching about is the set of “signals” by which the scholarly authority of a particular work is measured.  Everything from the presence of footnotes to a notation that the article has been peer-reviewed can help students determine where on the continuum of authority a particular work they discover belongs.  Also, students learn from librarians and others how to “backtrack” from one article to find those conversation partners whose combined contributions help form a complete and coherent view of any particular issue.

Both of these basic skills are undermined, to some extent, by situations such as the one Mark recounts.  For one thing, it is perfectly possible that he could remove or revise his original blog post.  Now Mark is a scholar and a very astute blogger, so I would expect him to acknowledge and explain any subsequent changes he might make to that post.  But the possibility certainly exists, for this set of writings or for others, that the scholarly works under discussion could change or even disappear.  That possibility presents those who teach about research skills a new challenge — to explain and help students account for the potential impermanence of the scholarly record.  And even if they remain intact and unchanged, the challenge of helping students understand that a peer-reviewed work might be based on one that was never peer-reviewed, and consider what impact that possibility would have on their judgments about authority, persists.

I offer this anecdote as concrete evidence that the changing system of scholarly communications compels all librarians, and especially those who teach information literacy, to remain aware of what we might call the “socioeconomic” structure of information.  That is, the conditions — social, economic, legal and technological — under which different forms and types of information are created and disseminated.  We are witnessing, I believe, a radical disaggregation of scholarship, as new formats, new business models and diverse levels of accessibility become the norm for some disciplines and for many scholars.  Even in a humanities field like Early Christian Literature, which still preserves many of its traditional modes of communication, this splintering of once solid lines has its effect.  And for our students, whose entire information-seeking lives will be lived in an environment where technology, copyright and licenses control what they can find and what they can do with what they find, education on these matters is no longer optional.

It is precisely these changes, and the ever-more-pressing need to take them into account, that the ACRL White Paper is intended to document and encourage.  It deserves attention from the library community precisely because we cannot ignore the revolution in scholarly communications itself.

So let me explain how this happened and what lessons can be gleaned from my experience.

The story began when I gave a talk at the 2012 conference of NASIG, the North American Serials Interest Group.  Let me say at the start that no one at or representing NASIG did anything wrong in this encounter and that whatever misunderstanding or lack of information existed was entirely my fault.  NASIG provided me with an interesting and engaged audience of librarians, which was all I could ask.  In any case, I signed an agreement, as a “Vision” speaker (kind of ironic), allowing my talk to be mechanically recorded and also agreeing that a human “recorder” would write up what I said for an article for The Serials Librarian.  In due time, that reporter sent me a copy of the article and I agreed that it was a good representation of the talk I had given, ready to be published.  Not until the article was published did I realize that The Serials Librarian was a Taylor & Francis journal, and to the best of my recollection I never signed a copyright transfer agreement with T & F.  At least, I can find in my saved e-mail the agreement to publish in The Serials Librarian but not a CTA.

Again, neither NASIG nor the article author did anything wrong; they sought and obtained all the necessary authorizations from me.  It may well also be the case that the recorder who wrote up the article signed a CTA with Taylor & Francis, which she would have been entirely entitled to do.  But as I say, to the best of my knowledge I did not, and the lessons I take from this incident are premised on that recollection.

So the first lesson is obvious — be careful what you sign.  More careful than I was.  I should have determined who the publisher was and made an intentional decision before I signed that agreement about what would be done with the article that resulted from my talk.  It is quite likely that I would have agreed even after that small bit of research, since the article was actually written by someone else (as, I suppose, a derivative work from my original talk), and I had no further plans to use it in any way.  What I often tell authors is to consider the agreement they are presented with in light of their own plans and hopes for their work, and transfer or license rights in a way consistent with those plans.  If the agreements allow one to meet those goals, well and good; if they do not, negotiation is called for.  The decision should rest with the author.  In the experience I had, I did not make that decision in an informed way, and that, rather than the ultimate result, was the problem.

The second lesson from this experience is that authors choose journals, not publishers.  When I read over the agreement with NASIG, The Serials Librarian seemed like a proper venue for the article resulting from my talk, and I failed to inquire further.  Although I should have done, I did not look into the publisher’s identity because for me at that moment, as for many academic authors, it simply didn’t matter.  The first step in getting academic authors to pay attention to the rights they transfer or retain is helping them realize that not all publishers are alike in this matter, and that they do need some awareness of who is who.

Next, my little story provides an opportunity to remind readers about the issue of joint authorship.  Joint authors are very common, of course, in academia.  Once mostly found in the STM fields, digital humanities projects are now making joint authors out of folks from many different departments.  Joint authorship arises, of course, whenever two or more people each contribute original expression with the intent of creating a unified work.  In the case of my talk, my original expression was fixed in the PowerPoint slides and notes that I had made.  Recorder Susan Davis then created a derivative work from that original, adding a great deal of her own original expression.  Once I had indicated my assent to that process, she and I became joint authors.  Like all joint authors, we each hold an equal and undivided share in the copyright, and are each entitled to exercise the exclusive rights granted by copyright, subject only to a duty to account to each other for any profits (which I don’t expect, in this case).  Because of this situation, if Susan signed a copyright transfer agreement for publication of the article, she was perfectly entitled to do so.  And because of the potential that fact has to create misunderstandings and surprises for other joint authors, it illustrates how important it is in general that joint authors agree in advance, whenever possible, about how their shared work will be used, licensed and made public.

Finally there is this point — if I am correct that I never signed a copyright transfer, and assuming, for the sake of illustration, that Susan did, then Taylor and Francis and I are now joint holders of the copyright in this article.  One thing that means is that I can continue to exercise all the rights as a copyright owner — I could post the article to the web if I wanted to, for example — without consent from T & F.  So when publishers tell us that they need to be the exclusive holder of copyright in every item that they publish, it is important to realize that that may be an aspiration, but it is not a necessity.  In fact, my experience is only one of a large number of scenarios under which publishers routinely publish articles for which they are not the exclusive rights holders.  As we seek to reform the scholarly publishing system, partly by encouraging academic authors to pay better attention than I did in regard to this article, this fact is an important piece of information to remember.

Let’s start with an insightful article from the recent issue of Nature that contained several pieces about open access.  The one that caught my attention is “Open Access: The true cost of science publishing.”  The author, Richard Van Noorden, provides a wealth of detail, and a very even-handed analysis, about the varying cost of publishing an academic article. He is hampered, unfortunately, by on-going secrecy on the topic. Neither Nature, which is publishing the article, nor PLoS would not talk with him about actual costs.  Nevertheless, there is a great deal of information here, and it all points to the conclusion that logic alone would have suggested — open access publishing, especially by non-profit entities, is much the more efficient way to disseminate scholarship.

One thing Van Noorden is able to show very clearly is that almost all open access publication charges are lower than the average per-article revenue that traditional publishers earn.  The difference can be as much as between a $300 cost per OA article and the average $5000 revenue per toll-access one.  The difference can be accounted for in one of two ways –large corporate profit margins or inefficient publishing methods.  Whichever is the case, however, it is clear the open access is the better option.  These lower costs are among the many reasons that open access provides a much greater benefit to academia than the traditional, pre-Internet system can.

Inspite of this documented good news about OA, however, the article ends on a discouraging note, or perhaps it is better to say a note of frustration.  Open access is obviously growing every year, but it is not growing as quickly, except where it is mandated, as it’s obvious superiority would suggest.  So at the end, the article leaves us to speculate on the incentives faculty authors have for choosing, or not choosing, OA.

And that brings me back to the “open” in Massively Open Online Courses.  The growing popularity of MOOCs, and their potential, parallel to that of open access itself, to revolutionize higher education, is a new and powerful incentive for scholarly authors to rethink access to their publications.

The fundamental driver behind the growth of MOOCs is the desire to expand the scope of our educational mission and to reach a global community of students we could not otherwise serve.  Seen in that light, the “open” in MOOC is key.  Part of our commitment as institutions participating in MOOCs is to try very hard not to erect financial barriers to participation in these courses.  We resist the normal urge to require textbook purchase, for example.  Our instructors are encourage to recommend but not require books for purchase (with the result, BTW, that sales for the merely-recommended books nearly always skyrocket). But this commitment to keeping the courses open for students also means that we look for an increasing amount of open content for teaching.

When our instructors want to provide readings for students taking a MOOC, we generally pursue one of two options.  Either we negotiate with publishers, who are slowly figuring out the marketing advantage they gain by allowing small excerpts of books and textbooks to be made available freely, or we look for OA content.  Unfortunately, the negotiation option is slow and labor-intensive; often we must explain the purpose and the conditions over and over again, to ever-shifting groups of officials, before we can get a decision.  So open access is ever more important, because more efficient, for our MOOC instructors and their students.

One story will illustrate this growing interest in open access.  A faculty member who was recently preparing to teach his first MOOC wanted his students to be able to read several of his own articles.  When we asked his publisher for permission on his behalf, it was denied.  A rude awakening for our professor, but also an opportunity to talk about open access.  As it turned out, all of the articles were published in journals that allowed the author to deposit his final manuscripts, and this author had them all.  So we uploaded those post-prints, and he had persistent, no-cost links to provide to the 80,000 students who were registered for his course.  An eye-opener for the author, a missed opportunity for the publisher, and a small triumph for our OA repository.  Enough of a triumph that this professor has begun asking colleagues if they could deposit post-prints of their own articles in the repositories at their institutions so that he can use those for his MOOC students as well.

So when we are counting up incentives for open access publishing, whether Gold or Green, lets remember that the massive opportunity that is represented by MOOCs is also a new reason to embrace open access.

The case involves the digital music re-sale service ReDigi, through which subscribers can sell their digital music files through a complex process that is designed to ensure that the original owner absolutely cannot keep or access a copy of the music file once it has transferred to another subscriber who has purchased it.  From the trial court’s description of the process, it is very clear that ReDigi was trying to comply with a reasonable interpretation of the doctrine of first sale for the digital age.  Most of their system was intended to enforce a “forward and delete” version of first sale that would seem to embody the spirit of what that doctrine is trying to accomplish.  Unfortunately, Judge Richard Sullivan of the Southern District of New York found that ReDigi had moved too far beyond an obsessively close reading of the letter of the law.

Judge Sullivan’s order is a close analysis of the wording of the doctrine of first sale, found in section 109 of the Copyright Act (Title 17 of the U.S. Code) and a very “physical” understanding of the digital world.  The question he is addressing is “whether a digital music file, lawfully made and purchased, may be resold by its owner through ReDigi under the first sale doctrine.”  The judge answers that question with a resounding no.  His decision is based first, on the conclusion that the reproduction right in copyright is implicated whenever the copyrighted work is embodied in a new “material object” and that each movement of a digital file from one server location to another is a new material object.  “It is simply impossible,” Judge Sullivan writes, “that the same ‘material object’ can be transferred over the Internet.”  Since each movement from server to server, or from one segment of a server to another, is a reproduction, the Judge holds that the files sold over ReDigi are unlawful reproductions.  First sale, furthermore, cannot apply to such copies.  He is very explicit about this: “it is therefore impossible for the user to sell her ‘particular’ phonorecord on ReDigi… the first sale defense is limited to material items, like records, that the copyright owner put into the stream of commerce.”  There is no first sale — none at all — on the Internet, even without licensing restrictions on transfers.

The Judge does address very briefly the argument that this attempt to impose the analysis from the analog world onto the digital environment would have damaging implications for all kinds of consumer digital products, since reproductions are made all the time in ordinary computer maintenance activities and when content is moved from one device to another.  But he dismisses that concern as a “red herring” because “other doctrines or defenses” protect such activities “almost certainly.”  But he does not say which doctrines or defenses, and the fact that he has to resort to this vague reliance on the idea that the obvious negative outcomes from his decision will not come to pass shows that the decision is problematic from a policy perspective.

Indeed, it seems to me that Judge Sullivan is legally correct in his analysis, and thereby demonstrates that legal accuracy is not sufficient for this rapidly changing technological environment.  I think even he recognizes that the ruling he is obligated to make is not satisfactory; in regard to the policy implications of what he is doing, he falls back on that old judicial claim that he is not allowed to legislate from the bench, and that Congress must fix the problem he is creating.

Congress should take up Judge Sullivan’s invitation.  It seems clear, in light of Kirtsaeng, that Congress will be asked to modify first sale this term.  They should resist the publisher’s pressure to “fix” the Supreme Court’s sensible decision to leave consumers in no worse position than they have been in since 1908, but they can still readjust first sale to make it work for consumer sales of digital materials on the Internet.  They should consider Justice Kagan’s suggestion, in her concurrence in Kirtsaeng, that Congress fix the result from the Quality King case that read first sale as a limitation on the importation restrictions in section 602 of the Copyright Act. And they should recognize that the principle of “you bought it, you own it” is foundational to commerce, and in the absence of a negotiated agreement to the contrary, it should apply to the digital world.  So Congress has a chance to look at the very reasonable business model that ReDigi put in place and find ways to authorize it under a modified doctrine of first sale.  Unfortunately such reform, if it were to happen, would come too late for ReDigi and its customers.

In thinking about the Kirtsaeng case — the decision is here, there is a Chronicle of Higher Education story about the ruling here, and Kenny Crews of Columbia offers a detailed analysis of the decision here – I have come to realize that the same issue of whether or not copyright should behave like other forms of property was in play in this litigation.  In short, by trying to enforce a “geographical” reading of the doctrine of first sale, publishers were attempting to create a “super” property right that would gave them a level of control that other property owners do not get.

On a trip to Turkey two years ago, I purchased a silver and onyx ring while in Cappadocia, where onyx is mined in large quantity.  My wife and I also purchased a carpet for our dining room.  Both items could have been purchased in the U.S., but both were less expensive in Kaymakli than they would have been had we done our shopping in Raleigh.

I mention this small shopping spree to make two points.  First, the sale of these goods at lower prices in other countries than are available in the U.S. does not, apparently, make it impossible for U.S. merchants to sell at higher prices to shoppers here at home.  Price discrimination, as it is called when vendors adjust their price structure to take account of local market conditions, does not depend, apparently, on an absolute prohibition on importation or cross-border sales.  Second, now that we own the ring and the carpet, we are free to do with them what we like.  No one can tell us where to place the carpet, or to what events I may or may not wear my ring.  And we can resell either item if we wish.

What publishers wanted from the Supreme Court was an unprecedented level of control that no other property owner gets — the right to control the use of a copy that was manufactured in another country (i.e. whether or not it could be lent) and to control any resale of that copy.

Let’s think back for a minute to the Costco v. Omega case from a few years ago that also dealt with unauthorized importation of copyrighted goods.

First, it is interesting that that case involved copyright at all, since the goods in question were watches, which are not copyrightable subject matter.  Omega was able to bring the suit only by registering a small emblem on the back of the watches for U.S copyright protection.  They had to do this, of course, because simply being the owner of a batch of watches would not have given them control over the importation of legally purchased watches to the U.S.  They resorted to this ploy to take advantage of the unique feature of copyright that they believed gave them more control than “mere” property ownership did.  They were trying to exploit the “super” property features of copyright.

Second, even though Omega got a tie from the Supreme Court that left in place a favorable decision for them from the Ninth Circuit, they were ultimately unable to take advantage of that quasi-victory.  On remand, the District Court granted summary judgment to Costco because, it held, Omega was trying to misuse its copyright to prevent perfectly legal importation.  In spite of that ruling (which is under appeal), I am pretty confident that Omega still sells watches for less in South America than it does in North America.  Again, an absolute ban on importation is not a prerequisite to price discrimination; while there is always some “leakage” of “grey market” goods, price discrimination works well enough that all kinds of businesses that sell different types of property depend on it anyway.

So the outraged threats that are being heard from publishers about how U.S. market prices will now have to be charged for copies sold in the developing world are simply ridiculous.  One extremely vocal advocate for the publishing position puts the claim this way, in a comment to the Chronicle story linked above:

The only practical effect of the decision will be to stop the practice of publishers licensing the sale of U.S. works for sale in foreign countries in cheaper editions, thus greatly inhibiting the flow of knowledge to underdeveloped countries.

If academic publishers were really to do this, it would be a crime against their self-declared mission of making knowledge available.  It would also be bad business; a self-defeating fit of pique that would cost them a lot of money.  But no, I am confident that publishers will still sell books at prices adjusted for market conditions, unless they are even worse businesspeople than I think they are.  Perhaps it will be necessary to adjust prices to account for that leakage which the Supreme Court has said cannot be choked off entirely, but I actually suspect that that small loss is already built in to decisions about pricing.

One way to think about what was being considered in the Kirtsaeng case is to look at it as a tariff — an attempt to guarantee extra, post-sale income from goods when they were imported into the United States by forcing purchasers to license certain uses of those goods they had already bought.  The U.S. disfavors tariffs, and where they do exist the money is paid to the government, which is trying to protect specific domestic industries.  If Kirtsaeng had gone the other way, however, there would have been, in effect, a tariff on importing books and films that would have been an entirely private benefit.  From this perspective, as from that of property rights as a general notion, what was being sought by the publishers in Kirtsaeng was unprecedented, as well as unwise.

[By the way, Mike Masnick of TechDirt also uses this analogy with tariffs in one of his comments on the case, and his post is an indication that the publishers are already beginning to pull strings to try and get Congress to give them the extraordinary benefit that the Supreme Court has just denied them.]

One argument that the publishers have made and continue to make is that the “parade of horribles” that was predicted by libraries and many others in the Court would not have actually come about; they frequently say that what they were asking for was actually the state of the law for the past thirty years, and things have run pretty smoothly up till now.  But to make that claim is to beg the question of why the case had to be brought at all.  Libraries did not sue over first sale; neither did Supap Kirtsaeng.  It was publishers who decided that they needed to go to court because, obviously, they wanted to change the conditions that actually have been in place up till now.  Publishers were seeking a “new deal,” a super-property right that is unprecedented in any other market place.  And what libraries “won” (remembering that no library was a party to the case) was simply the right to proceed as we have been for many years.  I have no doubt that if the lower courts had been upheld in this case, publishers would begin to demand “public lending fees” from libraries whenever a book was printed in another country, and would have moved operations offshore to increase the situations in which they could demand such a fee (as the Second Circuit Court of Appeals acknowledged was a likely outcome).  It is an overstatement to call this a victory for libraries; it was merely a successful defense of what we have done for many years, which, it turns out, is something that our courts really value and appreciate.