University of California, Los Angeles
This interview was conducted in February of 2014, by e-mail, among Christopher Kelty (CK), Anne Allison, Charlie Piot (AA/CP), Ali Kenner (AK), and Timothy Elfenbein (TE).
AA/CP: Why do you think anthropologists have been slow to embrace open access? After all, the politics of OA—democratizing access, returning ownership of scholarly production to the intellectual commons—is compatible with the political orientations of most anthropologists. It is an area where political investment can make a real difference, so it is surprising that there hasn’t been a massive social movement within anthropology around OA.
CK: I think anthropologists have actually been rather enthusiastic supporters of open access—at least compared to some other disciplines in the humanities that have ignored it completely or adopted stances toward it that are even more conservative. But the question is a good one, and it does seem that the principles of OA are in line with what anthropologists value: public interest, anticorporate concentration, academic freedom, collaboration with informants, intervention in public debate. It should be low-hanging fruit. But open access, and copyright wars generally, have two major disadvantages: they are maddeningly complex and ultimately very boring. I think most folks, quite rightly, just want to focus on other things, and let professionals handle the problem of scholarly publishing. But it is precisely this indifference that has enabled the publishing industry to take greater advantage of and extract enormous profit from our generally well-meaning labor.
What’s mysterious, though, is why anyone in anthropology might have time to be opposed to open access—that would seem to be a cause for concern. But it is not academic anthropologists who are opposed—it is the people in the American Anthropological Association (AAA) who run the publishing program who have been the only really vocal opponents. They see themselves as defending the interests of anthropologists, and it seems to them that open access is a threat. But what they are defending are the interests of the scholarly publishing industry and its extractive business model. John Wiley and Sons regularly provides the AAA with a nice, large check and a handful of poorly designed, intransigent, cookie-cutter services. But what many anthropologists do not recognize is that that check comes from their own libraries (and hence has a direct effect on the money available at their home institutions) and the services are built on their own labor as editors, reviewers, copyeditors, etc., for which they are not compensated in any direct way. It’s as if Wiley had hired anthropologists as employees for no pay, then pickpocketed them, taken half the money from the billfold, and then given it back to them via the AAA. I fail to see how anyone can perceive this as a viable relationship, either with the scholarly publishing industry or with their own scholarly society.
Perhaps one should say that the AAA is just being conservative and careful. But this makes no sense: in the past they have been more than willing to subject themselves to radical experimentalism when it comes to large, profit-hungry (Wiley) and small, innovative (University of California Press, Anthrosource) firms alike—why not in the case of open access? Making Cultural Anthropology open access is at last some proof of a willingness to change.
AA/CP: Are certain disciplines better oriented (intellectually, politically) to open access than others? Why have the natural, rather than the humanistic, sciences been quicker to get on board with OA? Shouldn’t we imagine a natural fit between anthropology and OA, given our ethical and intellectual commitments, including long-standing critiques of bourgeois notions of property and ownership?
CK: To me, the question of the difference between disciplines comes down to scale, not substance. It’s worth reminding ourselves that even though we are relatively numerous (over ten thousand members in the AAA! Over twenty publications!), we are miniscule compared to the health sciences. The University of California alone probably employs twenty thousand health scientists, and Medline currently indexes over five thousand journals. Biology is similar, the physical sciences also—there are just a lot more people in those disciplines, a lot more competition, and a lot more money. But because I have spent a lot of time over the last three years working on open access within the UC system, I feel reasonably confident saying that every discipline takes all kinds—fervent supporters and bloody-minded obstructionists were delivered equally to all the disciplines by whatever loving God populated academia. But being large and being small can have their own different effects. And in fact, this mirrors innovation in scholarly publishing generally—good ideas get traction when there are large audiences and large(r) dollars, making it hard for small venues like Cultural Anthropology to experiment. But smaller disciplines can sometimes benefit from these innovations—essentially riding along for free on the risks taken by larger disciplines.
There is a tendency in anthropology—and in the humanities generally—to suggest that open access is a problem of the sciences and engineering, and that we are somehow victims of this juggernaut on the other side of our campuses. Aside from being a form of ressentiment, this claim fails to recognize that the financial problem is the same across the disciplines—the larger humanities disciplines (history, literature) are going to have more power than the smaller ones (classics, women’s studies)—and the same is true in the sciences. Making common cause with folks in the sciences and engineering is far more effective for everyone than manning some culture-war barricade that is irrelevant to the larger dynamics of the economy of publishing, research funding, and university revenues generally.
AA/CP: How, for a journal like Cultural Anthropology, to weigh the democratizing against the anti-democratizing implications of going open access? The move to OA puts even more pressure on the journal to maintain its reputation as a producer of rigorously peer-reviewed articles by people at top-tier universities. Meanwhile, 75 percent of instruction in U.S. universities (and in anthropology departments) is carried out by precarious laborers often without the time (or means) to produce the type of article that will be accepted by the journal.
CK: This is an unacknowledged elephant in the room. I was at a scholarly publishing conference (Publish or Perish 2014, University of California, Davis) recently where the issue of precarious labor, unionization, “alt-ac” positions, adjuncts, and the question of assessment and reward for non-traditional forms of faculty labor crept into every single panel, regardless of its official topic. Administrators are still rewarding faculty—tenured or not—based on publications, and most are simply counting them. Review committees fall back on publishing metrics and journal reputation, which drives scholars to be desperate for that “credentialing” article in a major journal. In reality, just over the horizon from the debates about open access, is the question of the casualization of academic labor—how is it affecting research quality? How is it changing the dynamics and demographics of what gets published? Open access could never solve this problem (indeed, Eric Kansa [2014] recently titled a wonderful short article on this topic “It’s the neoliberalism, stupid!”). What open access can do is limited: if it is successful it might staunch some of the hemorrhaging of money to the largest publishers. Right now, a significant chunk of change is leaving the system and we are not getting anything like an equivalent value in return for it. Massive open online courses (MOOCs) also threaten to suck money out of the university system, as do the growing number of administrative positions created in order to deal with the fact that there are not enough faculty to share the burden of work required for oversight, audit, increasing enrollments, and decreasing public funding. Much of what is driving this is what Yochai Benkler (forthcoming) recently referred to as the “tyranny of the margin”—the relentless competitive drive to seek out opportunities to lower costs or extract more value—primarily amongst actors just outside of the university: scholarly publishers, educational start-ups, pharmaceutical and engineering firms, etc. But if we could solve the problem of open access within the university—that is to say, prove that the economic equation of doing research, reviewing it, and making it freely available for everyone works, then we could prove that the tyranny of the margin need not operate everywhere. It is a daunting problem to be sure, and it’s something that everyone involved—administrators, tenured faculty, and precarious labor alike—is challenged to deal with.
At a practical level, going OA actually opens up new opportunities—alongside that vaunted quality that is the core of what Cultural Anthropology provides, it is now possible to experiment with other kinds of content, other forms of validation that recognize diverse kinds of work, and perhaps a clearer, more transparent discussion of the labor of editors, reviewers, and others in the production of the journal and the research. Major scholarly publishers are terrified of OA for exactly this reason: that it threatens their brand equity in both known and unforeseeable ways. On the one hand, a bit more competition and a bit less entrenched hierarchy is a good thing; on the other hand, we don’t really understand how value and quality work in academia, and so we should mess with the system as little as possible. Small steps, deliberately chosen.
AK: What is not being said about open access? If we map out the issues and perspectives that have become dominant in this conversation, what corners of the issue have been neglected? What are the possibilities and pitfalls of digital forms and infrastructure? Sometimes I wonder if OA reinforces and reifies dominant modes of thought and practice on what is valuable production in the social sciences and humanities. What is our focus on OA forcing us to give up, or how might it be detracting us from other issues?
CK: I think open access is quite conservative, to a large extent; but many of the folks in the scholarly publishing world are experimenting with more radical changes to the ecosystem—new forms of peer review, new forms of assessment and metrics, new forms of executable papers (e.g. papers with data, software, or other code in them that can be read or transformed by software). Academic blogs are also an innovation of this sort—something that often has clear value to the discipline and community, but which has no clear home on a CV or merit-review form. Similarly, I have been to a host of conferences recently with alternative formats, alternative products, and far wider accessibility than in the past. So I don’t think open access is necessarily reducing the amount of experiment in the world of publishing—but it is probably a kind of conservative mode, interested in preserving the quality and authoritativeness associated with the traditional article, but simply making it more accessible.
You can see the tensions inherent in this in a lot of places—in the sciences, there is anxiety about reproducibility, conflict of interest, and the accessibility of both data and software code. In cultural anthropology and the humanities, the crisis is one of cultural authority, and as such has a much longer history and other dynamics at play. But the example I often use is this: the accessibility of our work is directly related to whether Wikipedia is a high-quality resource for the billions of people who now rely on it. If our work is hidden inside the university, then it will only seep out in tiny amounts to the most persistent seekers. But if it’s on display to the world with no barriers, it becomes something that not only our peers might cite, but lots of other people might come to rely on, refer to, or argue with. For many kinds of articles, that will not be relevant—but for a few, it can make the difference in how debate is conducted in public. So I think that we have a duty to maintain that authoritativeness in whatever ways we can—and it is not just peer review, it’s the hard work of the editors of the journals as well; it’s the commitment of the authors to revising and improving their work; and it’s the job of scholarly publishers to get that work out in front of people, make it easy to find, and easy to connect up with the massive network of other scholarship and ideas that we have created. If achieving open access gets in the way of maintaining or improving the quality and authoritativeness of our work, then that is a problem, and we should be on guard against it. But the fact that it is conservative won’t stop other people from experimenting with more radical forms of scholarly production.
AK: Do you think the digital publishing forms and infrastructures that emerged over the last decade have the potential to shift the culture of scholarly communication? Where do you see this happening, and what more is needed to intervene in the entrenched practices I describe above?
CK: Entrenched practices are definitely changing in some disciplines more than in others. In biology, for instance, and for some biological anthropologists and archaeologists, the changing data landscape is opening up new opportunities for collaboration, new kinds of research, different forms of demonstration and experiment, and so on. For cultural anthropology, I think the issues have more to do with how we associate with each other and how we debate and discuss our work. For a lot of graduate students today, the department and the scholarly society are only two possible forums within which debate and pedagogy around research questions are happening. I am not sure those of my generation or older are completely aware of what is happening, and the younger generation doesn’t understand what the implications are. But consider that a great deal of debate and discussion is happening on blogs, social networking sites devoted to anthropology, and in new forms of interdisciplinary spaces beyond traditional departments and societies. The forms of communication, practice, and research valorized in those new forums are not governed by the departments or the scholarly societies anymore, which used to do so simply by virtue of being the only places to gather. As a result, there is a kind of crisis of authority—the places that traditionally validate work (by admitting and advising students, by hosting conferences, workshops, giving prizes, etc.) are now supplemented by an unregulated array of competing alternatives. My sense is that established academics look on this with a mixture of patronizing interest and a vague, inarticulate terror. On the other hand, younger academics are the proverbial kids in a candy shop—excited about everything, amazed and enthused about the possibilities for change, but with perhaps little appreciation for the continuity that builds academic and disciplinary solidarity around problems and approaches. How can this situation not change the nature of scholarly communication? Indeed, how can it not change the meaning of disciplinary research itself?
It is not that things aren’t changing, or aren’t changing fast enough, but that there is the potential for a fragmentation and dissolution of the authority and quality associated with traditional practices. There is danger in assimilating this change to technology, when it is also a generational one—involving a retiring generation that saw the discipline completely transformed in a matter of decades (I am thinking of the 1970s–1980s challenges to anthropology), a middle generation dealing with the reconstruction of the discipline, and a new generation at home with not just new technology but interdisciplinarity and an increasingly naturalized precarity. Disentangling these things is not easy.
AK: Open access is commonly understood to be about use and access for readers, but at this point, these issues—access and use—seem secondary to me, subsumed by the political economy of scholarly publishing, the (undervalued) position of university libraries, and innovations in software and information science. I think most conversations about OA reproduce certain logics, interests, and concerns, but there is a lot more we can say about OA, I think, and I wonder what it says about the contemporary moment?
CK: I completely concur here—much of my activism around open access has very little to do with what I perceive to be as dead obvious: academics want their work to be more available than it is—and much more to do with the broken economy of scholarly publishing that prevents this in the first place. I frequently find myself frustrated that the conversation circulates back to petty concerns about individual academics’ rights or anxieties about copyright ownership when there is so clearly a larger system in need of diagnosis. Why aren’t we better, as anthropologists with a clear reflexive sensibility, at diagnosing the structural, organizational, and economic conditions that make open access seem necessary?
This circles back to Charlie and Anne’s questions above about precarious academic labor, as well as to a more general question about the changing political economy of the institutions of academia. In the case of the former, OA doesn’t address the problem of the (suffer a bit of Marx here) proletarianization of the academic labor force that is happening at a large scale. In fact, in deflects interest toward a very specific component of that: intellectual property rights in our articles. What good, though, is it for adjunct faculty members to own the copyright in the articles they write when they can barely cobble together enough shamefully underpaid teaching gigs to reach the poverty line? This leaves no time for what actually makes an article valuable—time to do research, a network of people with time to read and assess each other’s work, an infrastructure that supports that activity rather than relentlessly chipping away at it, and so on. What OA can help diagnose is that at least some of the money leaving the university is going to scholarly publishers, and they are very good at finding ways to get more of it. Open access does not necessarily stop this money from leaving though, it just insists on finding ways to get a better value for it—namely, better services and guaranteed free accessibility for readers and authors alike. Libraries get this, and have gotten it for years—they are the true leaders in OA—but they aren’t very good at painting this larger political economic picture for the faculty and students they serve. We need a better diagnosis of the big picture, not just the library’s perspective on the economic harms of publishing.
Indeed, in the political economy of our institutions, open access is one issue among many—the growth and mission creep of institutional review boards, compliance offices, and other forms of audit; the (expensive) need for universities to more aggressively defend themselves from politically motivated public records requests; the growth in the administration of diversity initiatives (at least in California, thanks to Proposition 209, this has meant the proliferation of administrative programs designed to get around that law to address the admirable, but expensive, goal of increasing diversity); the increase in the number and requirements for course evaluations by students; immense time and energy spent in the accreditation and review of programs by external groups; the new fashion (and the third in my memory) for online courses, distance education or MOOCs; the administration of benefits … the list goes on and on, and each of these necessitates either the creation or expansion of the administration or more involvement by already overcommitted faculty members. What is clearly not expanding in any rational way is the research infrastructure, which grows only through federal research funds, foundations, and privately donated money. So I’m not sure people should look to open access as anything other than a solution to a specific subset of problems—but if you pay attention to it, it certainly gives you a terrifying look into the machinery of this system, and the political economy of the university today.
TE: From your experience with projects such as Connexions, ARC (the Anthropological Research on the Contemporary), and Limn, in what ways are new publishing infrastructures (platforms, editors, automated document markup, etc.) sufficient or insufficient to replace human labor? What kind of labor can be automated and what requires human judgment? How has the automation boundary been arranged or pushed in different directions in the projects you have been involved with? Do (or did) these arrangements affect the success or modifiability of the projects?
CK: It’s interesting, I’ve never really thought of these projects as a trade-off between human and automated labor—that seems very 1960s in some ways. But it is a perceptive way to put the question—what is being automated by the Internet is the labor of distribution and circulation of content, the labor of association, indexing, or correlation and to a lesser extent, the labor of promoting or marketing it. If that is true, then various forms of labor remain central: authorship, judgment, review, curation, design, evaluation. There are aspects of automaticity that are clearly of interest to anthropologists today—I am thinking of the vogue for machine learning, recommendation algorithms, and the general spread of software that automates aspects of our work, often without our recognizing that it does so—how is this changing what happens to scholarship, if at all?
Each of the three projects you mention teach something different about this. Connexions taught me that even if you build a really awesome technological system that completely blows your mind—and I still think it does this—it will mean nothing in a context where people are not looking for a mind-blowingly different way of creating what is essentially a textbook.1 If academics are content to write and teach in the way they always have, and if students are not really agitated about doing it differently, then there really isn’t any way around that inertia. Connexions, like a lot of technological projects, was too proleptic and not diagnostic enough: it imagined a world in which all sorts of problems were solved: automated markup, easily transformed documents, remixability of content, a centralized repository of freely available teaching modules. But these weren’t (and perhaps still aren’t) the problems most teachers face. Ask the folks at Hewlett, Sloan, or other foundations spending loads of money on education—it is not clear that open educational resources solved a problem that people were having, even though we can all agree how awesome the solution is.
The Anthropological Research on the Contemporary (ARC) research collaboratory, on the other hand, was a bizarre experiment in the limitations of the purportedly limitless new tools of social media, publishing, the Internet. It too was proleptic in its imagination of a world in which conceptual work,2 collaboration, and the pursuit of research could be radically transformed. In reality, the result of that experiment was that it is expensive, time-consuming, and hard to achieve such transformations, and that collaboration is a fragile object, difficult to maintain without the active work of the people involved. ARC mirrored other attempts to recreate a research infrastructure outside of the broken political economy of the university—blogs, social networks devoted to anthropology, etc.—but that fragility made it unsustainable outside of its otherwise stable locus at the University of California, Berkeley.
Limn was a kind of response to the fragility of Internet-mediated collaboration—and indeed on the surface it might look like a reactionary one—it is a print magazine and could appear quite conventional.3 But both the style of work it enables and the technologies it employs speak to the changing nature of how we can engage in scholarly communication today. Limn creates a growing network of authors, an editorial process that juxtaposes different forms of research, or finds conceptual connections across them and tries to display that—something that the journal model cannot do except in the case of a special issue. It communicates large research projects (the kind generally destined to appear in a book after a few years) in a short, accessible, and timely manner—not dissimilar to the reasoning behind the scientific research article, which is rarely more than a glimpse into a larger research project or experimental system. Technically, Limn takes advantage of an array of services that are free or very cheap, especially at small scale—something that mirrors the general trends coming out of IT and software innovation focused on redesigning the infrastructure of organizations. Tools for publishing print versions, for managing subscribers, for spamming people (excuse me, I mean marketing), for keeping amazingly detailed data (thank you NSA + Google), for managing a workflow, for paying and being paid—all these tools are easy and cheap today, begging the question of why scholarly publishers all insist that publishing is so expensive. It isn’t. Don’t believe anyone who tells you it is. It is, however, extremely time-consuming to do well—and if your time is money (as it is for editors, designers, photographers, illustrators, copyeditors, etc.), then it can be expensive. But scholarly publishers do not employ those people—they just expect academics to do all that labor. So Limn does pay, and we pay people a living wage, and we are always asking for more money from universities, foundations, and others to do so.
TE: From my understanding of your work and that of others on different copyright regimes in scholarship, the focus is authorship and the remixing of the products of authors. Can you say something about work for hire: the product of authors whose legal authorship is subsumed by the hiring institution? This seems to me to have a direct bearing on the current situation of free software, as you discuss in your work. Does the success of open source/access rest on a particular legal arrangement within institutions such as universities? If work for hire is an ascendant relationship between authors/coders and the institutions that fill their stomachs, does this point to an important limit of alternative copyright regimes like Creative Commons?
CK: This is a great question, but I actually think that work for hire is not ascendant—at least the narrow legal version enshrined in copyright law—at universities. To give a concrete example, consider the University of California. The work of people at UC is governed differently if patents or copyrights are concerned. In the world of patents, everything is basically conceived of as work for hire, and there is very little room for individuals to control what they invent unless they do so outside of the university. But in terms of copyright, work for hire is limited to a pretty narrow band of staff members. Faculty—both tenured and not—and academic appointees of all sorts usually own their copyrights and have the flexibility to do what they want. Students are a funny, fuzzy category, but for the most part, universities default to a position that privileges academic freedom and the rights of individual authors. So, actually: yes, OA depends on a very particular regime in place in our universities—one that has forsaken the challenge of exploiting copyright in favor of exploiting patent. Copyright is not worth the trouble for most universities.
Ironically, this state of affairs came about because in the 1970s and 1980s, universities started to insist that their presses become independent centers of profit, which they often did, if only in a minor key. As a result, copyright became a profit center for quasi-independent university presses and their competitors, independent scholarly publishers. Academics fed this system by systematically signing over their copyrights to these entities, and universities fed it indirectly by writing copyright policies that either encouraged this transfer of rights or implicitly gave up on trying to retain or negotiate those rights. So quite ironically, if universities had in the past taken a more aggressive stand and insisted that everything we did was work for hire, then they might now be in a much stronger position to insist on the availability of that work. As it stands, the fact that most academics own their own copyrights means that it is that much easier for publishers to demand that they hand them over. It is also the reason why open-access policies take the form they do today, which is to say, it has created the need for academics to collectively reserve non-exclusive rights in their publications, rather than handing them over to the publishers completely.
A different answer to your question, however, concerns open-source software. Here, indeed, there is far more work for hire today than in the past, but this is only because corporations have embraced open source. So now, if you have a nice, cushy, paying job at Google, Facebook, or any other firm where open source is embraced, the work you do is both initiated by and owned by your employer—but because it’s open source, you can go to a different firm and work on the same thing, rather than having to reinvent the wheel. You may not be able to determine what you work on or when, which was one of the original purported advantages of open source (voluntary task choice increases engagement), but you at least can be certain that it won’t be boxed up and disappeared by your employer, and that you will get a certain kind of credit for it because your name will likely be associated with it. Again, not true in the case of patents. For cultural anthropologists this is likely irrelevant—but for biologists, physicists, engineers, and others, the ability to do the same thing with software, code, or other modular reusable research outputs is essential (we move universities all the time, we want to take our work with us), and so the liberal copyright regime in place actually facilitates that for the most part.
CK: If I had to ask myself a question about open access today, it would be this: is it working? Is it obvious that scholarship should be openly accessible, and how do we know that this is the right way to orient power and knowledge?
CK: A lot of people harbor this concern—it is actually quite easy to find critiques of the ideology of openness today, and they become more intense the closer one gets to Silicon Valley. But I find that they are never clearly articulated—they tend to devolve into accusations of libertarianism or misrecognition of some kind. But it is a concern. In science studies, for instance, it is routine to criticize the separation of science and society—anything that appears to maintain or reinsert the divide between science and politics is treated as suspicious. Openness does just this—it suggests that by making everything open and accessible, politics and science can operate properly—nothing can be deliberately hidden from view or obscured, and the open circulation of scientific knowledge will allow politics to operate without distortion, restoring balance and order to the operation of science, on one side, and society, on the other.
But to my mind, this focus on the open in open access misses the point, and I think our discussion here has really identified a different set of problems related to the political economy of the university, the state of precarity in academia, the problem of making work authoritative and high quality, etc. These things have to do with freedom, with money and power, and with the still-political process of producing knowledge. Open access is best seen not as a solution to some problem, or as a utopian desire for a perfectly open scholarly record, but as a tool for remaking the environment of scholarship in a very particular, albeit limited way. I think such politics occur through the making of university policy, the crafting of organizational relationships, the creation of software and new communications tools, as well as through critical discourse itself, and as long as open access is perceived that way—and not as some universal libertarian panacea—then I think it is changing things. What we do with those changes is up for grabs.
1. Connexions is an early attempt to apply open source to scholarly textbooks by encouraging the creation of modular open educational resources. See http://cnx.org and https://openstaxcollege.org/.
2. ARC was and remains Paul Rabinow’s project, beginning in roughly 2006, and emerging out of innovative ideas for rethinking graduate pedagogy in anthropology. It is partially documented in Rabinow 2013.
3. Limn (http://limn.it) is a scholarly magazine that is published as an open-access online publication, but also as a nicely designed print version costing between $10 and $15. It draws material from networks of experts in the social and human sciences and is intended to be timely, diverse in perspective, authoritative, well written, and beautifully designed. The focus is on contemporary problems in our global, politically interconnected, technologically intense culture: problems of infrastructure, ecological vulnerability, economic interdependence, and relentless technological invention.
Forthcoming “Peer Production and Cooperation.” In Handbook on the Economics of the Internet, edited by J. M. Bauer and M. Latzer. Cheltenham and Northampton: Edward Elgar.
2014 “It’s the Neoliberalism, Stupid: Why Instrumentalist Arguments for Open Access, Open Data, and Open Science Are Not Enough.” Impact Blog, Impact of Social Science Project, London School of Economics and Political Science, January 27. http://blogs.lse.ac.uk/impactofsocialsciences/2014/01/27/its-the-neoliberalism-stupid-kansa.
2013 The Accompaniment: Assembling the Contemporary. Chicago: University of Chicago Press.