Interview with Christopher Kelty: On the Culture of Free Software

On the Culture of Free Software
Interview with Christopher Kelty
By Geert Lovink

It is still rare that anthropologists study modern technology, let alone the politics of free software. The Houston-based scholar Christopher Kelty, who just moved from Rice University to UCLA, has done precisely that. Instead of observing the behavior and codes of this professional group of computer engineers, Kelty decided to map the social ideas behind free software production. Kelty’s Two Bits, The Cultural Significance of Free Software (see Duke University Press) contains of a historical reconstruction of where the ideas of ‘openness’ and freedom to change code originate. Kelty is not repeating the well-known story about the 1998 schism between the business- minded open source faction around Eric Raymond and the religious free software fighters, lead by Richard Stallman. Instead, we get a fascinating time travel, back to the pre-PC period of early computing. With the different generations of the UNIX operating systems we see how collaborative forms of writing software is taking shape—and how the ideas about ownership grow with it.

In the 1980s everything revolves around ‘open systems. For me the chapter on Conceiving Open Systems was a particular highlight. Kelty writes: “’Openness’ is precisely the kind of concept that wavers between end and means. Is openness good in itself, or is openness a means to achieve something else—and if so what? Who wants to achieve openness, and for what purpose? Is openness a goal? Or is it a means by which another goal—say ‘interoperability’ or ‘integration’ is achieved?” According to Kelty openness is an unruly concept. “While free tends towards ambiguity (free as in free speech, or free as in free beer?), open tends toward obfuscation. Everyone claims to be open, everyone has something to share, everyone agrees that being open is the obvious thing to do.”

Two Bits is accessible and a pleasure to read, but it is not particularly theoretical, nor critical for that matter. No critiques here of the inward-looking geek nature of free software, the lack of a counter economy and therefore a much larger dependency on large IT corporations for jobs and income than necessary, and the dominance of the conservative-libertarian pop ideology within open source/free software circles (see www.slashdot.org). Also, do not expect to read Levi-Strauss’ the raw and the cooked adapted to Linux. What Christopher Kelty does provide us with is an interesting first 80 pages in which he describes his wanderings through Berlin in the days of Mikro e.V. and Wizard of OS (2001), Bangalore and Boston. Out of these encounters with new media culture he filters a few concepts that are worth taking up elsewhere. The first one is ‘recursive publics’. Recursive not only points at making, maintaining and modifying, but also at the depth of the technical and legal layers. “Geeks argue about technology, but they also argue through it. They express ideas, but they also express infrastructures through which ideas can be expressed in new ways.” The second valuable concept is ‘polymaths’, described by Kelty as avowed dilettantism. This is a part of the book that does address the issue of a shared lifestyle amongst programmers. Polymathy is the ability to know a large and wide range of things. It’s what Adilkno describes as the positive side of vagueness in its Media Archive. “Polymaths must have a detailed sense of the present, and the project of the present, in order to imagine how the future might be different.” All in all, enough slacker insights to get this book and read it—supposing you’ve got an interest in the history of free software and share the collective drive to push its ideas further. What follows is an email interview with Christopher Kelty, while he was moving to set up base in Los Angeles.

GL: Some say that ‘geeks’ can be studied as an ‘alien tribe’. Much like the Australian aboriginals once,  when Cook arrived, ordinary Westerns do not really notice them and thus they continue what they have always done, unaware of the big changes ahead. Apart from a few 1990s movies and novels in which they feature, computer nerds are an invisible group. In Two Bits you decided not to emphasize the lifestyle aspect of geekness. Instead, you focused on the ideas that have been behind the early Internet, Emacs and the birth of free software, Linux and Apache, and then moving to the present with Creative Commons. Why have you chosen for an ‘anthropology of ideas’?

CK: Anthropology has pretty lousy marketing these days. Outside of the discipline our two major icons are Margaret Mead and Indiana Jones, and much of what the media expects from anthropologists is just-so stories about why humans, especially exotic humans, do the funny things they do, preferably involving sex and violence. So the appeal of the geeks-as-savages story is naturally pretty strong, and I was torn as to how to deal with that. Geeks themselves like being profiled this way and I was “anthropologist-in-residence” in start-ups in both Boston and Bangalore, and routinely introduced and paraded around as such.

However, what I think is most important in anthropological research today is the vibrancy with which researchers try to identify new “objects” emerging through cultural practices–not just new kinds of behavior or new organizations of people. And especially today, this includes new kinds of practices that are globally distributed. Even people who work with the Australian Aborigines (like Kimberly Christen http://www.mukurtuarchive.org/) struggle with this issue. Indeed, socio-cultural anthropologists arguably no longer study “cultures” as such, but only practices and meanings which are not easily (or violently) reduced to economics or biology.

In Two Bits, I wanted to capture why it is that a large and very diverse global population of people recognize and find affinity with each other. They do that by understanding, using and building free software, which is in turn deeply interconnected with the growth and spread of the Internet itself. So the type “geek” doesn’t come first–it is the result of adopting certain practices and habits, learning particular histories and myths, and becoming deeply committed to certain political ideals–and changing them as well. People want to know why some people become geeks and some don’t (or more often, why more men than women do), but I don’t have an answer to that. I think the fact that geeks exist, are multiplying and diversifying is hard enough to explain… why they don’t become investment bankers or firefighters is the wrong way to start asking questions about the phenomena at hand, I think.

When the mainstream media (and many ordinary people) talk about Aborigines, by contrast, they are often essentialized, either culturally or genetically, as trapped within their culture, usually as representatives of a primitive mode of life, rather than vibrant actors in a field of practices, technologies and politics. This can happen with Geeks as well, when one hypostatizes them as a “culture” preceding the advent of the practices and technologies that give their lives orientation and meaning… but it rings hollow, I think, even to geeks who enjoy such objectification.

GL: A highlight in Two Bits for me is the non-meeting you have Eric Raymond. He gets to sit next to a lady and during the dinner you do not get to speak to him. You mention a number of topics and controversies that you wanted to discuss with him. Instead, you get to talk to other people out of which a interesting collaboration grows (the Connexions project). Could you nonetheless perform your Raymond critique here?

CK: I wrote a lot of stuff before the book, arguing with Raymond (mostly in my head) and trying to figure out how to position this person who is the ultimate “principle informant” in anthropological terms–someone who has deep experience of, and tries to formulate theories and explanations about, the practices that an anthropologist wants to explain. It didn’t help that Raymond called himself an anthropologist. Indeed, it’s a good indication of the low status of the discipline—you can’t call yourself a physicist or a biologist without a lot more training, and you can go to jail if you call yourself an engineer or a lawyer and you aren’t!

Nonetheless, Raymond’s work is really very good in a certain 19th Century mode of anthropology—he is the Sir James Frazer of hacker anthropology—but the problem is that there is another 130 years of anthropology in between his style and that of today’s anthropology, which he ignores in favor of a pop evolutionary-psychology, which has almost zero status in anthropology today. So he’s a weird mix of old and new and it’s really hard to know what to do with him.

Take the popularity of the notion of a “gift economy” which almost every geek in the world can talk about with some familiarity, thanks to Raymond. This was a really good orienting idea–an “object lesson” which helped make sense of Free Software. On the one hand, this is exactly the right direction, and anthropologists inspired by or trained by Marilyn Strathern immediately grok how our concepts of exchange and person-hood are challenged by the emergence of Free Software. On the other hand, rather than take it in this direction, Raymond concocts a mix of vulgar Marxism (stadial theories of development), innate “territorialism” (shades of 1960s Robert Ardrey), and vague definitions of reputation and credit to offer a putative explanation of why Free Software works. Needless to say, I don’t think it will be remembered as an explanation–it will be remembered as a kind of geek-myth, which in some cases is what Raymond almost seems to think he is doing.

Ultimately, I left all this out of the book for just this reason: if I argue with him, I give him the status of a fellow researcher, and I don’t think either his research or his ideas merit that. Rather, I think it’s important for people to understand that Free Software includes Raymond as an *actor*, as one of the key actors in making it into the vibrant phenomenon it is, and so I include him (and Stallman and Torvalds and Perens and O’Reilly and others) as one component of five–the “movement”–which makes up the practices of Free Software. I needed to explain why Raymond existed more than I needed to explain why his explanations were off base.

GL: Rishab Ayer Ghosh and his Cooking Pot Theory would be another case. But anyway. Maybe it was a missed opportunity that you have not dwelled upon your Raymond criticism. There is no culture of debate and criticism in these circles. Look at Stallman and how hysterically he responds if you criticize him for his embarrassing lack of knowledge of political philosophy, talking about freedom this and that. We, social scientists and humanities scholars are supposed to learn Linux, know the technical basics of operating systems, but the other way around, forget it. Engineers can see whatever they want about society, and get away with it. You, Steven Weber and many others are from a new generation of FLOSS scholars that do try to push the boundaries of theory. Do you think there is a new wave of software studies in the making? In what direction would you like this field of knowledge to grow?

CK: I agree… and I would much rather see Rishab’s work, and work on FOSS by anthropologists like James Leach, Bernard Krieger, Gabriella Coleman and others be valued by engineers and programmers more than the ravings of Stallman and Raymond… but I also think that’s impossible. I don’t think of the latter two as scholars at all, more as politicians or demagogues, which explains why you can’t really argue with either of them. I think the same is true in many domains, where there are a few loud voices that capture all the attention.

I would argue to the contrary, however, that there is indeed an extremely well developed culture of debate in hacker circles, once you get beyond the demagogues, and this is something Gabriella Coleman has captured well in her work. Projects like Debian and Ubuntu represent the best of that culture, I think, combining an even-increasing understanding of the political and legal issues with the technical sophistication. But that kind of debate is much less visible than the histrionics of the big men, so people miss it unless they are directly involved. Such geeks are also far less libertarian than they are often accused of being and are more likely to be practicing a form of liberal communitarianism; and they are well aware of the form of sociality they are building and promoting, even if Stallman and Raymond are not. Again, I think the accusation of libertarianism comes from listening to a few loud voices, rather than getting close to the work of the mass of people involved.

I do think there is a new wave of software studies emerging and it represents a kind of generational shift away from the quick and dirty explanations towards sustained research questions that seek not only to explain FLOSS as such, but to challenge existing theory in different disciplines—whether that’s public goods and collective action theory in political science and economics or theories of technology and culture in anthropology. Much of the earliest work on FLOSS lacked depth because it was so new and responded so quickly to the phenomenon. But with sustained attention, I think some of the deeper issues have started to become clearer. A new generation of “software studies” might be able to move beyond the logic of newness that dominates the world of IT and software; it could be a chance to identify a “longer duree” of political, economic and cultural issues of which each new generation of cool tools and “new” ideas are seen to be expressions. That might allow scholars to gain purchase on this sense of rapid change and simultaneously to become more authentically critical of the claims of each new generation of toys. That would be a real achievement.

To create a successful new field of software studies, however, requires that scholars are willing to sustain their attention and take the risk of collecting, observing, participating and reflecting over a longer period of time. When I started this project in 1999, it was about Free Software… but by the time I finished it, the project was about the cultural significance of the various practices involved and how they could be understood and related historically to more recent changes (like Wikipedia and Web 2.0), as well as much older events (like UNIX and the Open Systems debates of the 70s and 80s). I like to think that it is a more general analysis, and a better one, as a result.

GL: Open and free are two key concepts if we want to understand the significance of free software. There is a great chapter in your book on the history, the use and abuse, of the term openness. You did not write about the confusion about free and freedom. You have not deconstructed the Cult of the Free into the realm of peer to peer networks, or the debate about precarity, for instance. Why not?

CK: Well, in a way I’ve tried to do this in a different idiom–that of publics and public spheres. For me, the language of freedom and openness—and the concern with definitions, principles and the enumeration of freedoms are a small part of the phenomenon of Free Software. I repeatedly insist that what makes Free Software interesting is that whether you call it free, libre or open, whether you are with or against Stallman, as long as the other four practices are in place (sharing source code, copyleft, coordinating collaboration, open infrastructure debates), then the shouting doesn’t matter–it only matters that those vitriolic debates are conducted *in the service of* the other four components, and the phenomenon of FLOSS as such. The debates very rarely imply clear practical choices about how to do FLOSS, they are much more often about the meaning of it.

Where the ‘public sphere’ aspect is important is that I want my readers to focus on the places where these debates (about free or open) are conducted in the service of maintaining an independent, technically mediated and radically modifiable public sphere. And independent means independent of states, corporations, professions, churches and so forth. I think this is in line with the concerns over “precarity”, “casualization” and some aspects of anti-globalization. I think it relates wherever there are questions of fairness and the construction of public infrastructures that give people the freedom both to speak freely and safely, and to modify or extend those infrastructures in ways that don’t serve only the interests of constituted powers. So I would say that skepticism about both openness and freedom is certainly warranted–but I’m trying to help give researchers ways to ask whether there is anything behind that talk that might really contribute to the expansion of an authentic public sphere, rather than just being cynical about the claims

GL: The trend is clearly away from software towards a proliferation of social, cultural and political fields where the basic notions of free software, eat themselves into the issues, so to say, as memes. Do you also think that the core of the philosophy will remain the same, or will certain elements mutate, once they travel from context to context?

CK: Since I don’t think the philosophy is at the “core” I suspect it will not remain the same at all. What has occupied my attention is what happens when the *practices* of free software are adopted more as templates for action than as memes, and then are modified based on pragmatic concerns. So Creative Commons modulated the notion of a copyleft license, but in an attempt to be all things to all people, they also created a new problem–multiple conflicting licenses and debates about the meaning of “non-commercial” or “third world” or “sampling.” The Connexions project modulated the meaning of “source code” to include textbooks, but in doing so encountered (and has not quite solved) the problem that educators don’t write or share textbooks the way programmers do code. These modulations are interesting in themselves for what they can tell us about different domains (e.g. how film works or doesn’t as a collaboration, how music can be pulled apart, recombined and re-valued), but the bigger question, I suggest, is whether in modulating these components, the people and practices involved maintain any hope of expanding or strengthening a public sphere that provides an autonomous space for material and discursive experimentation, even if such practices are not on their surface explicitly Political (with a capital P).

So to answer your question, I think the modulation of the “philosophy” of free software will continue. The world of open educational resources has a much different approach to understanding the relationship between freedom and the tools of thought; groups like Autonomo.us are modulating the principles of Free Software to deal with web services; and perhaps the clearest case are the debates within various “free culture” movements about whether the philosophy is too software-centric, and what freedom means with respect to other cultural materials. Certainly within anthropology there is massive suspicion of projects like Creative Commons and its imperial approach to defining cultural freedom– but this is, as I say, just one component of the changing landscape—it’s also important to pay attention to whether and where the other practices are replicated—licenses, definitions of open infrastructure, tools and schemes for coordination and collaboration, the definition of what objects can be shared, etc. The modulation of the philosophy of free software is part of the more general process of these practices being adopted and transformed—and not the driver of those changes.

GL: How do you look at the Oekonux debates in 2002-2003, the current activities of www.keimform.de, the P2P foundation and theoretical work of Adam Arvidsson, Michel Bauwens and others? What do you make of such practical and theoretical efforts to bring together the principles of free software and peer-to-peer production? Do we have we an economic turn ahead of us? Would this be a very European idea or do see similar tendencies in the USA? Some say that it is really urgent that the FLOSS efforts focus on cell phones and RFID tags. In which direction would you like to see research and activism go?

CK: I think this is a huge question, far beyond what I tried to do in the book. In some ways, I see this as the next iteration of social science questioning after the “information economy” or “network society”– ethical economies, creative capitalism, germ-forms, peer production (Benkler), and p2p societies are grand socio-economic diagnoses, and as such, crucial for debating how to analyze and make sense of the changes we are seeing. I don’t think it is particularly European, but in the U.S. it is more likely associated with things like von Hippel’s “User-driven Innovation”, Henry Chesborough’s “open innovation” and other work in management and innovation studies. Scholars in those domains in the US are often less aware of the socio-political and activist concerns that I think are much more on the surface in Europe, much more philosophically grounded in cases like Oekonux and P2P Foundation. By contrast, groups like Indymedia or Riseup.net represent a more radical genealogy in the US and abroad, which is the subject of Jeffrey Juris’ recent book (Networking Futures). So there are obviously different ways to tell the story of this confluence of ideas.

One way to understand my position vis-a-vis these debates is that I have started from the assumption that the practices involved in the creation of Free Software (and the Internet as well) which emerged in the 1980s and 1990s are at the core of the changes we are seeing—and not general economic or cultural ideologies, which I see instead as effects of changing practices. So for me, Wikipedia and Facebook are not examples of the same thing that Free Software is an example of (peer production or creative capitalism or user-driven innovation etc.) but *derivatives* of the practices that coalesced so productively in Free Software. And Free Software is also not original in this sense, but drawn from the modulation of UNIX in the 1970s, the open systems debates in the 1980s. I think it is important, for instance, to understand the role of telecommunications regulation and anti-trust politics in the US and Europe in the 1980s to understand why Free Software gained a foothold in the 1990s. I’d be less likely to attribute the emergence of Free Software to a new stage of history than I would to a detailed working out of a previous structure of legal and economic practices. In this, I think I’m in partial sympathy with the Oekonux and P2P Foundation projects because I think “critique of political economy” in the strictest sense of the term is what is needed here.

On the other hand, I’m skeptical that theorizing a new kind of economy will make any difference to the kinds of persistent inequalities and injustices already present in actually existing markets. For example, a colleague of mine Robert Foster, has just published a great book about Coca Cola’s role in the global economy (Coca-Globalization). Many of the things he describes about how Coca Cola interacts with its customers, encourages them to innovate and draws them into the “experience” of Coca Cola share a great deal with the explanations offered by the “user innovation” people. The difference of course is that Coca Cola is, well, evil. Identifying why it’s not the same thing for Coca Cola or Apple to engage in “peer production” as it is for Wikipedia seems to me to be the most difficult question. Similarly, for me it was important to identify the core practices of free software in order to distinguish what Apple and Microsoft were doing from what real free software projects are doing. That’s why I turned to the problem of publics and public spheres and their independence from constituted forms of power, rather than to the theory of public goods, or a revived Marxism. I don’t think they are incompatible, but I’m a pragmatist at the core: I want to see whether such theories help make sense of, and potentially transform, concrete realities of practice.

GL: As you may have noticed, there is no Web 2.0 platform for activists. Indymedia is more or less dead (at least, the English/international edition). Activism and social networks do not seem to match that well. The problem of transparency for police and other services of these platforms plays an important role in this. On the other hand, social movements have always been prime examples of networks that can scale very well, if the circumstances are right. Do you also see a problem here? The social seem to have gone technical, and it is questionable if we can just make a romantic move back in such an instance?

CK: I don’t think of any of the web 2.0 platforms as being particularly true to the principles of free software. Wikipedia yes, and a few projects such as Shay David’s Kaltura are explicit about their commitment, even as they struggle with solvency and sustainability, to say nothing of profitability. But Facebook, MySpace, Friendster, Ning, and so forth all lack some component that leads, in my terms, to the creation or expansion of a recursive public. I would like to think that this concept helps explain, in part, why activists might shy away from such platforms, insofar as we are talking about activist publics whose commitments are to an independent and legitimately powerful civil society whose discussions and deliberations have real effect on the constituted forms of power they address. The technical commitment of such publics is essential, however, because, yes, we cannot go back to a world without the technical infrastructures, new modes of expression and circulation that have been created. We are, in some ways, condemned to address the technical as a political problem. Rising ‘above’ such details into the realm of principles may clarify things, but only if such a move can be tested in the concrete and complex skein of the contemporary operating systems of our world.

GL: Would it be possible to identify ‘kernels’ of conceptual hegemony in projects like Debian and Ubuntu that are not corporate and conservative in nature? How can we open an intellectual dialogue about this? In the case of Web 2.0 we see again the importance of (collaborative) meme construction? Just think of all this talk of ‘swarms’. How to regain the confidence to build up a counter-hegemonic discourse? Is your concept of the ‘recursive publics’ offering a way out here?

CK: In the cases of Debian and Ubuntu, there is a strong core of people and practices, well developed, exquisitely argued and widely implemented that I would characterize as “pure” free software. Insofar as my characterization of the practices of free software as a kind of ideal type has a real expression of those ideal features, Debian and Ubuntu are probably the best exemplars. But just certifying these projects as pure is meaningless. The concept of a recursive public was my way of articulating the significance of these pure forms, not just the conditions of their existence. And that significance is 1) that they treat technical infrastructure and decisions about its design as political through and through, as far down the “recursive” stack of technical layers as possible and 2) they do so in order to maintain the possibility not only of an authentic public sphere that they inhabit, but the possibility of the emergence of publics oppositional to themselves, and to those that emerge, and so on. Whether or not people take advantage of these publics to develop counter-hegemonic discourses and new political powers is uncertain, it’s not implied by the form of the technology, but it is enabled by it.

Free Software provides a radical form of openness which is, perhaps, a very American way of constituting a public (suspicious of the state and corporations, obsessed with ideas of balance and fairness, and a weird mix of individualism and populism). The question I think it raises is whether, as a politics it has a content. Free Software as it exists has an insanely refined focus on form over political content (and this is the source of the suspicion about the dominance of the technical). But the question is: is this focus on form itself a particular kind of political content? At some level yes, but it is one that is open to, and maybe even encourages people to challenge it. It is a way of saying: if this is a (for instance) “libertarian” form, it is one that you are allowed to change–so make it less libertarian if you believe that will make it better. It says nothing, however, about whether people will have the power to do that, which is its weakest feature, its inability to incorporate the concrete fact that history has led us to this point.

Christopher Kelty, Two Bits, The Cultural Significance of Free Software, Duke University Press, 2008