Wikipedia:Wikipedia Signpost/2009-06-15/Book review
Review of Cyberchiefs: Autonomy and Authority in Online Tribes
Books on the sociocultural and political impacts of the Internet have typically focused on the advances online communication makes possible or the good things in modern culture that the Internet is pushing aside. Proponents of Internet culture highlight the failings in traditional systems of power and cultural production that online communities can overcome; critics emphasize the good things in those systems that the Internet is undermining. In Cyberchiefs: Autonomy and Authority in Online Tribes, Mathieu O'Neil instead explores the ways online communities recreate the failures of offline culture.
O'Neil, a researcher who works in Australia and earned a Ph.D. in American Studies from a French university, approaches online social systems from the perspective of critical theory. Other scholars have used critical theory to explore the Internet in terms of communication and knowledge production—in the case of Wikipedia, for example, "Wikipedia, Critical Social Theory, and the Possibility of Rational Discourse" argues that Wikipedia is a close approximation of Jürgen Habermas's ideal of rational discourse—but O'Neil focuses instead on the twin concepts of autonomy and authority.
The first half of Cyberchiefs develops a conceptual basis for understanding online authority. In the second half, O'Neil presents four case studies on individual "tribes" with very different authority structures: the anarcho-primitivism website primitivism.com, the political blog community dailykos.com, the free software community of debian.org, and our own free encyclopedia project wikipedia.org (specifically English Wikipedia).
Rethinking online authority
The prevailing ideology of the Internet, O'Neil observes, is closely aligned to the philosophical and political outlook of the earlier hackers who built it:
Since utopian political solutions are no longer considered likely to occur offline, the Internet has come to embody the spirit of Utopia. In such a charmed universe everyone can have a say, from 'cyberlibertarians' who decry the influence of governments to 'cybercommunists' who believe that peer production will revolutionise both the market economy and traditional hierarchy. The primary tenet of the ideology of the Internet is that online networks are privileged sites for the flowering of freedom.[1]
The flowering of freedom is indeed an important part of the Internet's impact, but this emphasis on freedom obscures the ways that traditional forms of power, privilege and domination carry over to the online world. Early students of online sociology described the web as inherently anti-authoritarian (primarily because of its technical structure, an open network). O'Neil shows that concepts of authority and power developed by social theorists can apply to both the web in general and to specific online tribes—a term he uses to indicate that the social and political structures of online communities are largely independent of nation-states. Adapting Max Weber's tripartite classification of authority, O'Neil identifies three forms of authority that structure the social environment of online tribes:
- Hacker-charisma authority – a form of charismatic authority rooted in respect for broadly-construed "hacking" ability—the special talents and skills that are relevant to the goals of a particular tribe. The canonical example of this type of authority is project leadership in free software development communities, where those with acknowledged coding ability and an intimate understanding of a particular software system are deferred to by other programmers. In Wikipedia, hacker-charisma authority is the un-codified respect given to editors who are good at what they do (whether article writing, copy-editing, identifying sockpuppets, scripting, or some other task).
- Index-charisma authority – another form of charismatic authority, based on concepts from network theory such as centrality and preferential attachment. In a social network, having links to many others and/or very important others is itself a type of authority, in the form of a large audience; just as highly linked sites are the top results in search engines and thus attract more links, well-connected members of online tribes (especially early adopters) have index-charisma authority from being well-known. This is most obvious in Wikipedia in terms of the rising standards of adminship (which means it was much easier for early editors to become administrators), but it works in subtler ways as well—as when, upon joining large discussions, we look first to familiar and respected users.
- Sovereign authority – the analog of Weber's rational authority, which is based on norms, rules and laws (including, in the online context, laws programmed into the system). On Wikipedia, sovereign authority is invoked whenever editors use guidelines and policy to justify their actions or point out violations. The various classes of privileged editors—rollbackers, administrators, bureaucrats, etc.—embody sovereign authority, and are expected to act as enforcers of the community-created rules rather than powers unto themselves.
In addition to these forms of potentially legitimate authority, O'Neil shows that vestiges of power, privilege and symbolic violence from the broader culture, what he terms archaic force, have a dramatic impact on the web's social landscape. For example, in principle blogging is a way for anyone—no matter how qualified or unqualified, powerful or marginal—to reach a wide audience and make him or herself heard. But in practice the "A-list" bloggers that do reach large audiences are overwhelmingly social elites; "they are not only white, male and middle-class," writes O'Neil, "they are also highly educated, placing them effectively higher on the social ladder than the 'elite' mainstream journalists whose power they are supposed to be contesting."[2]
This type of pattern—those with the training and free time afforded by social privilege rise to the top—is also apparent in free software communities and on Wikipedia and other seemingly egalitarian online knowledge projects. O'Neil sees at work here "the heart of social domination [which is] making the socially constructed appear natural."[3]
Archaic force also manifests itself in received netiquette conventions and patterns of online discourse that encourage symbolic violence. Flaming and trolling are the purest expressions of archaic force; the flaming and trolling of newcomers and others who do not conform to community norms is a way of asserting power. O'Neil writes that "In general women have a deep aversion towards the kinds of adversarial exchange that men thrive on", and argues that early netiquette specifically encouraged male styles of adversarial discussion, even flaming, about intellectual and ideological matters but discouraged discussion of personal matters.[4] (We see the legacy of such netiquette on Wikipedia, where aggressive discussion is acceptable but personalizing disputes is forbidden; whether O'Neil would consider this an archaic residue of sexism is unclear, but at least one scholar of wiki communities has argued that Wikipedia-like projects have an inherent gender bias.)
The tribes
The four online communities explored in Cyberchiefs sit at different points on what O'Neil terms the space of online authority. Online primitivists—whose philosophy is fundamentally opposed to the internet and who have no interest in organizing a participatory online community—eschew both charismatic authority and sovereign authority online; primitivism.com acts as a venue for presenting the views of a small number of primitivist "anti-authorities" and extending offline debates among published primitivist thinkers through the fisking of rival essays.
The liberal American political blog Dailykos.com, like other blog communities and the blogosphere as a whole, is structured mainly by charismatic authority, with the hacker-charisma and index-charisma types both playing important roles. The initial popularity and broad political influence (i.e., index authority) of Daily Kos derived, in part, from founder Markos 'Kos' Moulitsas's talent for predicting election outcomes (a form of hacker-charisma), and in terms of community governance, what Kos says goes. In the rest of the Daily Kos community, the two forms of charismatic authority intertwined; members with trusted judgment and blogging ability are given access to the front page and select diaries (blog posts) of others to highlight, and the most popular posts are also linked automatically on the front page. However, autonomy is limited for those who would challenge the charismatic authority of Kos and his deputies, as O'Neil documents in the case of Hillary Clinton supporters during the 2008 Democratic primary; Kos and much of the community supported Barack Obama, and strident Clinton supporters were systematically marginalized and ostracized, without recourse to much in the way of codified rights or community laws to protect them.
The debian.org development community features strong elements of both popular sovereign authority and charisma authority. O'Neil considers Debian "by far the most revealing of what tribal distributed leadership would entail for the management of complex infrastructural systems", in part because "the stakes are much higher when participants can cause significant harm to the project."[5] Debian stands out among open source communities because of its well-developed governance system. Final authority rests with the development community itself, with leaders elected by the developers and major decisions decided by vote; strict merit-based gatekeeping limits entrance into the community to those with demonstrated software skills. The community structure is highly modular (paralleling the software itself), which allows a degree of autonomy for each developer even while the output of the project as a whole must be tightly coordinated. Although there are elements of index-charisma authority—long-term influence from early leaders—by-and-large, Debian is governed according to the collective will of its community, successful anarchy in action. Conflict in the Debian community often centers on the defense of honor, either against outside threats or intra-community insults; when there is a perceived affront to a developer's honor, communications can break down into flame wars, to the detriment of the community.
Tribal authority on Wikipedia
In the book's final case study, O'Neil examines how authority works on Wikipedia. Wikipedia governance relies primarily on charismatic authority—users deferred to because of their reputations, as talented contributors (hacker-charisma) and/or long-standing and dedicated active community members (index-charisma)—and popular sovereign authority—community-created rules and norms.
"Can people pull rank in a rankless universe", he asks?[6] The answer, of course, is yes; things like rollback rights, adminship, checkuser, and even—perhaps especially—edit count can serve as markers of authority in a social system based on constant surveillance of everyone's actions by everyone else. (In The Wikipedia Revolution, Andrew Lih compared Wikipedia to the benign street culture praised by urbanist Jane Jacobs: cities are safe when they are always under the watchful eye of residents. Others invoke a more sinister metaphor, likening Wikipedia to the Panopticon prison in which inmates never know whether they are being watched and so must behave as if they are.)
It is when surveillance breaks down that authority becomes a problem in the Wikipedia community. The Essjay controversy is the best known example of this; while claiming (falsely) to be a professor of theology, editor Essjay at times touted his supposed credentials in content disputes. But the most significant section focuses on what O'Neil terms "the Durova dust-up", the incident in which User:Durova briefly blocked User:!! as a sockpuppet based on an investigation that was not transparent to community surveillance (which led to Durova resigning her adminship). Here the dangers of both too much and too little surveillance were at work. O'Neil explains that "the incident resonated deeply with many editors, because it commingled authority and secrecy." The affront to the project's core value of openness and transparency was matched by "an equally powerful, and opposite, feeling: that some admins had been the victim of harassment and stalking because of their work for the project; that these experiences were frightening and painful; and that most of the victims were female."[7] [Clarification: O'Neil does not discuss specific instances of harassment, but refers in the preceding quote to the broader context of harassment as part of the spectrum of disruptive action, which efforts like "sock hunting" are employed to prevent.]
Charismatic and sovereign authority predominate, but archaic force is not altogether absent from Wikipedia. O'Neil singles out a Jimmy Wales quote from a 2006 New Yorker article (the one at the center of the Essjay controversy) to show how offline injustice and inequality is reinscribed in Wikipedia: "If it isn't on Google, it doesn't exist", said Wales. (O'Neil offers a wider discussion of Wikipedia in his recent essay from Le Monde diplomatique, "Wikipedia: experts are us".)
Historical factors and offline injustices—sexism, economic inequality, political geography—can clearly tilt the scales in online tribes. There are (at least so far) no online Utopias. The question for Wikipedia is, how deep is the shadow of history? How set in stone is Wikipedia's community culture, crafted as it has been by the earliest members with their peculiar outlooks and inclinations? Through the mechanism of preferential attachment in article creation and expansion and the propagation of charismatic authority, will Wikipedia always retain the mark of the early community's interests and prejudices?
O'Neil's particular analysis of Wikipedia includes some worthwhile points (and some errors and misinterpretations), but the case study only breaks the surface of the authority issue. The concepts of archaic force and the three modes of online authority are useful concepts for thinking about the community; Wikipedia authority is heterogeneous, sometimes with charismatic authority most important, sometimes with sovereign authority, and in our worst moments with archaic force deciding things.
Discuss this story
Mathieu O'Neil responded to the review in an email. His email and my reply are posted below (by permission).
Hi Sage
I read your review of my book on the WP Signpost. I really appreciated how you engaged with the work, by applying it to WP, or by making some of the concepts work in your way. I also liked how it was illustrated with those pretty pictures;-). This is by far the best review I've had to date - OK, it's probably the only one, but still.
I do have a few quibbles which I'll mention briefly.
Overall purpose of the book
I didn't just set out to show how horrible the online environment was by "uncovering domination". I tried to have both a critical approach and one that showed how people can criticize and overcome injustice. A (possibly understated) objective of the book was to draw lessons from situations in which project / community governance operates with less hierarchy, which I (once again, perhaps too implicitly) see as a positive development - otherwise I wouldn't spend so much time talking about it. For example in the Introduction I wrote "And yet: the persistence of some forms of domination should not prevent us from recognizing instances where authority really is self-directed".
Meaning of index-charismatic authority
I use this as indicating a person / actor / node's _central position_ in an online network: many others link to the node, or the node operates as a link between separated clusters. Some people might refer to this as "social capital" - the benefits received from being well-connected. Preferential attachment (i.e. early entrance in the network) is a possible and convincing explanation for how nodes acquire index-charisma but you seem to equate the two things when you define index-charisma as "another form of charismatic authority, based on the concept from network theory of preferential attachment". It is based on the social network analysis concept of network centrality or popularity (which has been theorized as deriving from preferential attachment by Barabasi, a notion which I argue obscures social domination).
Google / LMD
I am a little puzzled as to why saying "if it isn't on Google it doesn't exist" reinscribes injustice and inequality in WP - it reinscribes definitions of what constitutes acceptable content but I don't see how this discriminates against a particular group? Finally, my LMD article is not really a wider discussion of WP - rather it focuses on (and expands a little) the analysis of expertise but eschews most analysis of governance and administrative power which forms the bulk of the book's WP-chapter.
Cheers,
Mathieu
Mathieu,
Thanks for the comments!
Can I post your response on the talk page (and my reply, and we can continue the discussion there so that others can participate), or perhaps you'd like to write a more formal reply that could be published in next week's Signpost?
I didn't mean to imply that that you were merely explaining how horrible the Internet is, although now I see it kind of reads that way. My intent was to say that your book was primarily a sort of constructive critique of the Internet, rather than the more common modes of praising its best qualities or complaining about its worst: you show how the Internet's best qualities fall short and are imperfect (and thus how they can be improved upon). In the review I de-emphasized your discussion of the ways the Internet *is* already being used to criticize and overcome injustice, mainly because I think the Signpost audience for the most part takes those things for granted.
You're right, in trying to explain index-charisma concisely, I botched it. I was trying to emphasize what I found most insightful about the concept, that when one describes being socially well-connected in the language of network theory, the mechanism of preferential attachment makes it much easier to see how legitimate authority and unjust privilege are totally entangled even in the online social world. ("Social capital" seems too broad here; hacker-charisma is also a form of social capital, is it not?)
On "if it isn't on Google it doesn't exist": along the lines we discussed before in the context of your LMD piece, groups that don't get covered in online sources are already subject to social domination. Else why wouldn't they appear in online discourse, encompassing as it is? In our previous discussion, you gave examples of groups you thought got the short end of the stick in this regard.
I think your LMD piece *is* a broader look at Wikipedia. True, it focuses on expertise rather than authority, but it looks at expertise broadly, in a way that seems to me to encompass more of what goes on on Wikipedia. Your treatment of authority in the book is more particular, and you focus in on a set of examples that I would say are instructive but not representative. As I said in the review (approximately), I think your framework for thinking about authority is more valuable for Wikipedians than your particular applications of it *to* Wikipedia.
Yours in discourse, Sage
Futher discussion
Hi Sage
OK, I guess my first point has to be a question: since you say that you "botched" your analysis of index-charisma should we modify the review? I understand that this is not an encyclopedia entry so I'm not expecting that to happen but perhaps you should add a parenthesis or something indicating that the wording is not quite right and referring people to this discussion? Just a thought. Hacker-charisma would be more like cultural capital (possession of esoteric knowledge) than like social capital (possession of many valuable connections). --Mathieu O'Neil (talk) 13:40, 17 June 2009 (UTC)[reply]
Re. "If it isn't on Google": I am unsure about equating discriminating against pre-1995 marginal / subcultural forms which have flown under the radar of legitimate cultural arbiters and hence have not been digitised and hence find no corroborrating hyperlinked source when they are submitted to WP as legitimately worthy of inclusion (my original point in our earlier discussion about notability) with "social domination" on the project or in society in general which I would venture derives from class, gender, ethnicity etc. These two things strike me as quite different.. I can see how the analogy might work if someone was pointing to cases in which (for example) black art forms were not considered as notable as white ones but since that's not what I was talking about I would be more careful about saying that. But I don't really detect any archaic force at work when someone says "Your music / art / fanzine isn't notable enough because not enough trace exists of it online". --Mathieu O'Neil (talk) 13:40, 17 June 2009 (UTC)[reply]
Re. whether my Le Monde diplomatique (LMD) article is broader or not... My main observation in the book about WP in governance terms is that its lack of clear constitutional principles, proliferation of admins / rules and simultaneous embrace of slightly fuzzy "consensual" practices was stronly related to the still-strong presence of charismatic authority in the project, which can legitimately (but not democratically) make controversial and / or arbitrary decisions. Whether that is a "broader" point than the ones I make about expertise in LMD is an open question in my view.Cheers,--Mathieu O'Neil (talk) 13:40, 17 June 2009 (UTC)[reply]
"If it isn't on Google, it doesn't exist"
Mathieu O'Neil - I'm no apologist for Jimmy Wales - as I'm sure he'd be the first to confirm - but he's really gotten a bum rap on the "If it isn't on Google, it doesn't exist" pseudo-quote. I did a long blog post debunking it, Google: 1, Michael Gorman: 0, and wrote about this in a column about Britannica's blog: "This arose from misreading a very short discussion in a print article. The saying is attributed to Jimmy Wales, co-founder of Wikipedia. But he never said it with that truncated form; his own intent was to point out the weakness of complete reliance on search engines, and that there's still value in reference books." -- Seth Finkelstein (talk) 04:34, 18 June 2009 (UTC)[reply]
@ Seth Finkelstein - I accept that the quote attributed to Jimmy Wales may have been taken out of context – I trusted the source... thanks for pointing it out. It does raise the issue of Wikipedia editors possibly relying overly on online sources, particularly when deciding whether some topic or other is notable, or during deletion debates. Following the publication of my article on expertise and WP in Le Monde diplomatique, I was contacted in early May 2009 by Sage Ross who offered some comments and we had a discussion about various issues relating to the article, one of which was precisely the question of online sourcing. Since it's relevant to this discussion I thought I should include it here. Rather than rehashing the discussion I asked Sage if it would be OK to post the relevant excerpts and he agreed.
[Sage wrote:] On an unrelated note, you write later in the article that "marginal cultures which have not been digitised and uploaded run the risk of becoming invisible." I have two comments. First, Jimmy Wales' facetious comment that "If it isn’t on Google, it doesn’t exist" is not, and has never been, normative on Wikipedia; rather, it seems to me like a statement of the zeitgeist of the Internet age. Second, if marginal cultures are not covered in any digitized content, they don't run the risk of becoming invisible...they already are invisible. If they aren't on Google, that means that there are essentially no books or scholarly articles about them and that publishing institutions (including, in recent years, the Internet-connected public) have been ignoring them since the rise of electronic publishing. That's not to say that Wikipedia doesn't play a role in re-enforcing patterns of marginalization; through the "Reliable sources" guideline, in particular, it does do that. But I think it's unfair to lay that marginalization at the feet of Wikipedia, since that only causes problems when marginal cultures have already been made invisible (or rather, have never been made visible) by the forms of media that Wikipedia builds on and is built from. I would argue that Wikipedia actually levels the field for the unjustly marginalized, who are normally crowded out by the popular. There were thousands upon thousands of newspaper articles and television stories about Anna Nicole Smith; there are just a few dozen[1] corresponding Wikipedia articles. Conversely, there are Wikipedia articles for small villages with no particular claim to fame for which the only sources are census and geographical data...the invisible and marginal made visible and human-readable. In a paper encyclopedia, editors would have to find content to remove for every bit that got added, so that encyclopedia sets would not grow without bound and could still be sold to suburban families by door-to-door encyclopedia salesmen.
[Mathieu responded:] Re. your second and related point, I did not say that J. Wales' comment was normative, but it does encapsulate a certain reliance on a handy http page to link to buttress one's point. I think there might be a "loop effect" deriving from Google's ranking of WP as well. This is a complicated issue, which also depends on whether one does believe that everything which could have been digitised has been (I don't), or on where one stands on the notability question, or perhaps even on what constitutes original research in an encyclopedia which has no space limitations. So, while I appreciate your point that WP may level the playing field in some respects, I would have to reserve judgement until some more definite form of empirical evidence has been produced.
[Sage reponded:] Following up a bit on marginalization and sources...
You say:
> So, while I appreciate your point that WP may level the playing field in some > respects, I would have to reserve judgement until some more definite form of > empirical evidence has been produced.
Fair enough. But you seem willing to judge Wikipedia for marginalizing topics that don't have digital sources, without presenting definite empirical evidence and despite the fact that, in both policy and practice, Wikipedia encourages the use of print sources (including ones that are not available online). Online sources are treated as a convenience for readers, but there is no hesitation to use offline sources when they are superior.
Do you have particular cultures in mind that you see as being marginalized by Wikipedia's Verifiability requirements?
[Mathieu responded:] In general, I would venture that marginal / underground / subcultural / counter-cultural events, people and artefacts that existed before 1995 would not necessarily have been comprehensively digitised. I had some anecdotal evidence of this when I created an article for a by no means insignificant art / cultural group active 1988-1993 which was judged not notable because no online sources were available. The point being, the group in question definitely made an impact on the cultural scene at the time but the sources which document this (art or music magazines, exhibition catalogues, concert flyers, fanzines, radio shows, etc) are not online. Now, I didn't even know about the whole AFD process then; and the art / cultural group was active in Paris, while I created an entry on WP-en. So things might have turned out differently on the French WP... people might have known about it or been more receptive or whatever...
[Sage responded:] Thanks. It's true that Wikipedia creates threshold for inclusion that re-enforces existing patterns of marginalization. It sounds like what you've run up against has more to do with Wikipedia's definition of Original Research than Verifiability; offline magazines, and possibly exhibition catalogs as well, would be considered Reliable Sources on Wikipedia that could be used to establish at topic's Notability. Of course, Wikipedia is only the most prominent example of a whole class of wiki venues that follow the model Wikipedia created but often have different social structures; some of these are open to a wider array of content, and in a both a cultural and technical sense they exist because of Wikipedia. So I think I understand where you're coming from now, but I still think it's misleading to blame Wikipedia for the shortcomings of cultural institutions whose role it is do the kinds of things that on Wikipedia are called "Original Research". (The problems with that definition are interesting; there is a gap that exists for some areas of knowledge between what is allowed on Wikipedia and what is considered original enough to merit publication elsewhere, e.g., in terms of the analysis of literature.)
The other problem you ran into, perhaps, is the opaque complexity of the way Wikipedia works, so that your work was pushed out because it didn't conform to Wikipedians' expectations even though it might, in principle, have been made into something stable. That kind of thing is a big problem, and one that the community is constantly struggling with.
[Discussion ended] --Mathieu O'Neil (talk) 02:57, 19 June 2009 (UTC)[reply]
Authority as a factor in diffusion/adoption of ideas in online environments
I found the review of Cyberchiefs: Autonomy and Authority in Online Tribes interesting. Although I've not read the book yet I found inspiration for some tangential work I'm doing for my own doctorate, updating classic work of the Diffusion of Innovations within complex organisations in which communities of practice via online communication is becoming the norm. I am exploring some of the factors (including authority) which may be significant, but wonder whether the book has also examined power relationships and homophilly along with the general and context specific psychological antecedents which influence ways in which people decide to participate in public welfare/non profit endeavors. I am looking for/at evidence sources which illustrate that behaviours are different in these contexts than in those where financial reward is the driving force. The review has made me seriously consider buying the book (or at least hassling my institutions library to get a copy).— Rod talk 16:17, 18 June 2009 (UTC)[reply]
Good review