Wikipedia: Anarchy or Hierarchy?

Introduction

This paper addresses whether Wikipedia confirms the "analytic proposition of central importance" that, for functional goods where MC=0, production without property rights--that is, anarchism--produces inherently superior products. Though Wikipedia fashions itself as an "open" community, Wikipedia is distinctly hierarchical. Not only does Wikipedia impose non-trivial costs on users that discourage participation in the Wikipedia community, it may also dissuade entry into the community in the first place. Moreover, Wikipedia is not simply an aggregation of information, but also an organization accountable to, among others, the private (often corporate) benefactors that make its non-profit project financially feasible. To the extent Wikipedia relies on these interested parties, its benefactors' expectations and preferences may exert influence over content development with exclusionary consequences. Although these aspects of Wikipedia do not necessarily undermine its qualitative superiority, they nevertheless offer reasons to question whether its qualitative superiority is a function of anarchism.

The premise is the problem. Did you think that free software projects don't have leadership, or that everyone's patches go in? That there are no property rules, thus no exclusion based on ownership, does not imply an absence of organization altogether. But unlike the State, which employs coercion to secure consent, anarchist institutions base their power on consent.

So by measuring the wrong quantity you get a problematic measurement. Yes, it requires organization to manage dozens of encyclopedias everyone can edit around the world. It requires organization to make the Linux kernel. In neither case do property rights either determine the structure or function of the organization, nor do they flow from organization's activity. What proposition are you testing: mine or a straw man's?

Wikipedia's Superiority: Should We Thank Anarchy?

"Anarchism" can have different meanings in different contexts, and this paper uses the term to describe the freedom from exclusionary rules and traditional systemic mechanisms for limiting the evolution (and therefore improvement) of a product. Although Wikipedia rejects the proposition that it is anarchistic, this is not dispositive. Instead, it is helpful to look at Wikipedia's power structure--and the barriers to openness embedded within it--to determine whether anarchism is responsible for Wikipedia's inherent superiority.

This is the reification of the false premise. The proposition you're supposed to be testing is that Wikipedia's superiority lies in the absence of property relations determining who can produce there. At least if you're testing any proposition I advanced. But now you have smuggled in some other question and are pretending it's the same.

Far from a kumbaya-esque exchange of ideas, with large numbers of users' actions organically resulting in an equilibrium of "truth," Wikipedia is largely edited, monitored and developed by a minority of "volunteer administrators" who, although elected, wield a disproportionate amount of power over the site's content. Unlike most editors who quickly enter and leave the Wikipedia community, administrators are endowed with significant powers including deleting articles, locking pages to prevent further edits and even blocking individual users from editing any Wikipedia entry. Jimmy Wales, Wikipedia's founder, admits that Wikipedia is maintained and monitored by what amounts to a Wiki elite: "A lot of people think of Wikipedia as being 10 million people each adding one sentence . . . . But really, the vast majority of the work this done by a small core community." Indeed, research suggests that content is largely determined by a small core of repeat players who are both familiar with and integrated into the Wikipedia bureaucracy, as roughly 1% of Wikipedia editors account for about 50% of edits to the site.

You might want to do a little original research by talking to people who are Wikipedia editors. Even better would be to spend a little time involved yourself. Then you would see that you're confronting here one of the basic problems of reporting, ethnography, and legal investigation: factually accurate statements offered without context can constitute absurd misrepresentation. Here each of the statements you make could be correct (though the statistics are misleading and the quote from Jimmy is very far out of context), but the effect overall is not likely to sound like reality to anyone who actually lives in the Wikipedia community. The comments below show that with respect to marginal participants; you will find it also true if you talk to official community leaders, heavily-involved editors, or the academic outreach volunteers around you. Even talking to students who write for Wikipedia in one of my other courses might teach you what you need to know in order to interpret the data you offer, in line with the context that really exists.

The Wikipedia elite's control over content raises the specter of exclusion. Because Wikipedia bureaucrats are generally users who contribute and edit a disproportionately large amount of site content, administrators may preference content that is consistent with (or at least does not question) their own edits.

Do you have any evidence whatever to offer on this "may"? Or does it mean "I have no evidence whatever to offer, but I think I'll say this anyway because it conforms to my desired view of the universe?"

The 50% of edits that the 99% of users make must therefore survive the scrutiny of a distinctly self-interested group.

This "self-interest" of the self-interested group was established by the "may" in the previous sentence. That is, users' edits "must" survive the scrutiny of the self-interest that "may" exist. Which is nothing but pure fantasy masquerading as informed reporting. The fact, however, is that everyone's edits must survive everyone else's scrutiny. So the best questions would probably involve finding out how articles really develop. Because every page has a complete history one can read, as every page in this wiki saves all its versions, it's possible to look directly at the evidence left by editing. Why didn't you pick a few Wikipedia articles in areas of interest or intellectual specialization for you and actually assess the history of the editing? That would have taught you much less than actually involving yourself in the process of producing an article, but at least it would beat making it all up as you go along.

Even if administrators are entirely "objective" decisionmakers, the potential for and perception of bureaucratic abuse and/or bias may be sufficient to discourage participation.

Only, as commentators pointed out, unless it isn't. For there to be such discouragement, other people would have to take your speculation about how Wikipedia works seriously, instead of finding out for themselves how it works by interacting with it, in which case they are unlikely to reach your conclusion.

Whether the Wikipedia elite actually skew content development in favor of bureaucratic viewpoints and preferences may actually be a secondary issue because their mere existence reduces users' confidence in a meaningful, good-faith review process.

What? What are these "bureaucratic viewpoints"? Wikipedia's policies are explicit. The "neutral point of view" is the viewpoint required, and the test is citation of secondary (only secondary) sources reflecting the full span of published opinion on all controverted points. Do you have any evidence of this second-order process of discouragement of potential contributors or editors resulting from your hypothetical problem of some supposed (but evidenceless) tendency of high-involvement editors to adopt or favor viewpoints unpermitted by policy?

If users come to believe their edits will stand or fall not according to their merits, but rather according to the subjective preferences of the Wikipedia elite, it follows that participation will feel restricted--and users will be excluded--by virtue of Wikipedia's hierarchy.

And your evidence that this "coming to belief" occurs on the basis of these supposed subjective preferences of the editorial elite, for which in turn there hasn't been any evidence presented yet?

The effects of Wikipedia's bureaucratization go further, however, as they present potential editors with non-trivial costs. A 2009 Wall Street Journal article explained that 49,000 editors left Wikipedia in the first quarter of 2009--10 times more than left in the first quarter of 2008--because of the "plethora of rules Wikipedia has adopted to bring order to its unruly universe -- particularly to reduce infighting among contributors about write-ups of controversial subjects and polarizing figures."

Aside from the question of sourcing, would you have any evidence that the drop in editors was because of policy changes? (The actual size of the drop, in English, was approximately 12,000, from 90,000 to 78,000, rather than the larger and quite improbable number presented by the Journal. There were also declines in Wikipedias, like the German edition, in which the same policy changes were not made.) A significant source of editorial departure, Wikimedia found, was job loss, which was somewhat more common in the world during the first quarter of 2009 than the first quarter of 2008.

To the extent new editors must understand the Wikipedia power structure, conform to a Manual of Style, adhere to templates, contend with highly motivated and more technologically powerful users, and shape their content in a way that makes its removal and/or alteration less likely, participation in the Wikipedia community involves costs that may have the same exclusionary impact that traditional property rights effect.

Not, not the same. Different. Property rights effect either an absolute exclusion or an exclusion one can only overcome if one has enough money and pays it. These barriers to entry, on the other hand, can be lowered by a process called education. The Wikipedia community conducts a great deal of education, for free, designed to help people learn to use the Wikipedia effectively in all the ways you've mentioned. My students in AmLegalHist this term are working with an academic outreach editor at Wikipedia, learning how to overcome the obstacles you mention in order to make parts of their academic work in my class accessible to others through Wikipedia. You might want to talk to one of them.

In other words, where the absence of property rights opened the door to participation in the Wikipedia community, an increasingly complex set of rules governing participation may be having the opposite effect by placing the Wikipedia savvy at a clear advantage in the editing process vis-a-vis newcomers and technology novices.

This "may be having" is also a marker for evidenceless assertion.

While these barriers are not the result of a legal regime akin to copyright, their exclusionary consequences are nevertheless comparable to property rights'. It therefore seems overly formalistic to credit anarchism for Wikipedia's qualitative superiority on the ground that its exclusionary rules, while significant, are not products of property rights.

All of this to conclude that my argument, which was not based on words but on things, is formalist? This formalism of mine, if I understand you correctly, results from my failure to observe the phenomena for which you have no evidence, as a result of which I stated a possible general rule that accounted for all the observations except the ones you made up?

Finally, Wikipedia faces operational realities and, as a result, accepts donations from private (often corporate) entities in order to finance its operations. While these donors may not exert direct influence over Wikipedia, they nevertheless impose, indirectly, their expectations and values on the organization's leaders.

"Impose"? Are you serious? Do you think that people donate money to organizations like Wikimedia without sharing its values, planning to "impose" values on the organization? That's not reality so far as I know it, and I both run a non-profit and provide some informal advice to the Wikimedia leadership and board. In the world as I know it, people give to organizations such as Wikimedia because they already share its values.

To the extent Wikipedia relies on these entities, its content faces an outer limit of its benefactors' values and/or expectations. To stay within this limit, it seems inevitable that content will be excluded notwithstanding users' preferences.

What? Why would Wikimedia ever take a donation from someone who had a condition to exclude content? And why would it ever listen to a donor who tried to censor an article after donating? Do you think that the board, elected by the community and not appointed by leadership or (therefore) donors, would permit such an outcome? Yet this quite implausible outcome has become "inevitable" simply by the process of assertion without evidence. That doesn't meet minimum standards of intellectual responsibility.

In turn, Wikipedia's "organic" evolution may become increasingly "synthetic."

Conclusion

It is beyond the scope of this Paper to address whether Wikipedia is an inherently superior product and what, if not anarchism, accounts for that superiority. More simply, this Paper argues that while Wikipedia is not governed by property rights, it is not without exclusionary rules. Taken together, Wikipedia's hierarchical power structure and barriers to participation must, at least, call into question the anarchic roots of its qualitative superiority.

"Even assuming administrators are completely objective in their decisionmaking, the perception of and potential for bureaucratic abuse and/or bias may be sufficient to discourage participation. "

This is only true if you assume that the 99% who posts 50% of the content actually know the inner-workings of Wikipedia. I had no idea Wikipedia was structured that way so in making posts on wikipedia I've never felt any form of discouragement as a result of the structure. Further, I think that even if someone is aware that the structure is hierarchical, wouldn't it still make sense to post the information on the website anyway? Worst case scenario, it gets taken down or edited by the "wikipedia elite". But, isnt the possibility that it stays on the site worth putting it on there; I mean, it won't be on the site if you don't put it on there, so you may as well try to get it on there

-- AustinKlar - 25 Oct 2011

Perhaps it would be helpful to distinguish between editorial vs. content generation. Aaron's analysis here suggests that while a small group of editors do make the bulk of edits, the majority of content is actually authored by occasional contributors with niche areas of expertise. It is my understanding that the extensive edits done by regular editors go more to style, format, sourcing, etc., than alteration of factual content. Of course, this editorial oversight may still afford them considerable power over the community. But in support of Austin's point, the editorial oversight has not discouraged me from making (admittedly infrequent) contributions to Wikipedia's more esoteric pages. Knowing I was being held accountable by a horde of scrupulous monitors only increased the neuroticism with which I included citations ...

-- CrystalMao - 28 Oct 2011

Crystal makes a good point, it is useful to distinguish between content and edits. The Slate article I link to explains (somewhat contrary to Aaron's analysis):

"Palo Alto Research Center's Ed Chi, the scientist who determined that 1 percent of Wikipedians author half of the content, told me he originally hypothesized that the site's most energetic editors were acting as custodians. Chi guessed that these users mostly cleaned up after the people who provided the bulk of the encyclopedia's facts. In reality, he found the opposite was true (PDF). People who've made more than 10,000 edits add nearly twice as many words to Wikipedia as they delete. By contrast, those who've made fewer than 100 edits are the only group that deletes more words than it adds. A small number of people are writing the articles, it seems, while less-frequent users are given the tasks of error correction and typo fixing."

This suggests that content is in fact authored by a small contingent of users while the vast majority of us do the less substantive "clean up" work (style, format, grammar, etc.).

I agree that Wikipedia's hierarchy doesn't have categorically exclusionary effects--to the extent it forces would-be contributors to conform to a higher standard, it results in a qualitatively superior product. But, the presence of the hierarchy surely excludes some potential users who are not willing and/or able to incur the costs (neurotically citing, for instance) that come with participation. Because Wikipedia has a hierarchy that, to an extent, discourages participation (or, at the very least, disproportionately encourages a certain type of informed, legitimate participation), it seems somewhat dubious to credit anarchism with its qualitative superiority.

-- MatthewLadner - 29 Oct 2011+

Reading your piece, I have the sense that there might be definitional issues involved. I think a point motivating Eben's comments is that anarchism is not necessarily synonymous with a lack of organization. The term anarchy, like privacy, has been used with a very wide range of meanings. In your opening sentences, you equate anarchism with "production without property rights," an association Eben might have made in class at some point (or not, I'm not sure). However, I think the equation is too simplistic. What about a Stalinist society in which there are no property rights? Is this anarchist? I don't think so.

More broadly, I think the "is Wikipedia anarchist"/"what is anarchism" issue cluster is somewhat orthogonal to what I take to be your main point. As I read it, your main interest is more like "has Wikpedia become too bureaucratized/hierarchical to do its job well?" That's a question that can be, in a straightforward manner, tested against evidence. I don't think you need to get into the definitional quagmire of whether it qualifies as anarchist or not.

Instead, I think what you would need to do is define some metrics and apply them. What is that Wikipedia should be doing? How good is it at doing that thing? Is there evidence that "X or Y added procedure" has decreased the ability of Wikipedia to do the things it should be doing?

-- DevinMcDougall - 06 Dec 2011

It didn't appear to me that an issue with the paper is a "definitional issue." The author defined anarchism specifically for purposes of this paper. He said "this paper uses the term to describe the freedom from exclusionary rules and traditional systemic mechanisms for limiting the evolution (and therefore improvement) of a product." He could have chosen a word other than anarchism but that he chose that word and defined it specifically for purposes of this paper isn't an issue. The main issue is whether wikipedia's superiority as a product can be attributed to the "anarchy", as defined in this paper, i.e., a lack of property rules. So whether wikipedia is anarchist is certainly an orthogonal issue and its meant to be. You don't have to focus on whether his definition of anarchist is correct to understand the paper. I think Devin's suggestions about defining metrics and applying them might be a good idea. In the conclusion it might be good to outline some things wikipedia could do to help mitigate some of the issues described in the paper. However, given the length limitations for the paper, it might be too difficult to fully explore the issue of whether wikipedia's superiority owes itself to "anarchy" and then to explore metrics and applications.

-- AustinKlar - 08 Dec 2011