The development of the NPOV rule on the English-language Wikipedia

24 May, 2011 at 15:44 | Posted in Web 2.0, Wikipedia policy | Leave a comment
Tags: CPOV, Encyclopedias, NPOV, Wikimania, Wikipedia

Image via Wikipedia

It is hardly a secret anymore that my initial enthusiasm about Wikipedia has turned into a deep concern. This concern stems from Wikipedia’s inability to resist damaging trends in the world in general, in the business and academic worlds in particular, and most particularly in the high-tech business world (damaging, that is, in my humble opinion, though I think I’m not alone on that). Wikipedia is not commercial, but its promoters often treat it as a business or corporation that needs to yield as much profits as possible and overcome competition. I believe this approach is unintentional, resulting from the fact that we live our lives by pseudo-business models; and yet Wikipedia was meant to be different. Its drift along in global trends is perhaps not surprising but deeply disappointing.  This is just one example, and it is not directly related to the core issue which I would like to discuss in this post, so I’ll hold my horses here, and get to the main point.

A final note before doing so – My ideas here are influenced by the CPOV conference in Bangalore, India (January 2010), which is one of the most interesting conferences I have ever took part in (so far…). The products of that conference and several others are compiled in the CPOV Wikipedia reader, edited by Geert Lovink and Nathaniel Tkacz (I gave a modest contribution to this reader together with Johanna Niesyto). I’ll give reference to articles on that reader in the course of the text.

Phrasing of the neutrality principle; then and now

While preparing a presentation for this year’s Wikimania conference, to be hosted in Haifa, Israel this August, I looked into the history of Wikipedia’s NPOV rule. NPOV stands for “neutral point of view”, and the rule basically says that Wikipedia, being an encyclopedia, is obligated to unbiased presentation of any subject, particularly when describing controversies.

I went to the archived first online version of Wikipedia and found the following phrasing of the NPOV rule:

The neutral point of view attempts to present ideas and facts in such a fashion that both supporters and opponents can agree. Of course, 100% agreement is not possible; there are ideologues in the world who will not concede to any presentation other than a forceful statement of their own point of view. We can only seek a type of writing that is agreeable to essentially rational people who may differ on particular points. (November 2001)

Ten years later, the rule is phrased in much more words, but the leading paragraph says as follows:

Editing from a neutral point of view (NPOV) means representing fairly, proportionately, and as far as possible without bias, all significant views that have been published by reliable sources. All Wikipedia articles and other encyclopedic content must be written from a neutral point of view. NPOV is a fundamental principle of Wikipedia and of other Wikimedia projects. This policy is non-negotiable and all editors and articles must follow it. (May 2011)

The differences are evident, and yet one may argue that this is simply putting the same idea in different words. I am going to explain why it is not the case.

Collaborative work

The 2001 phrasing emphasizes collaborative work as the meta-policy of Wikipedia^{[note below]}. Wikipedia’s articles should be written in a way that is “agreeable to essentially rational people who may differ on particular points”. This wording brings to mind a vivid picture of people working together to find a consensual written version that would be acceptable on all. Also, like a good scientific principle (even though we are not talking about pure science here), this phrasing suggests a test by which the principle can be verified – If you are a rational person (more or less), and you feel uncomfortable reading a certain article, because it seems to be advocating for a certain opinion, then something must have gone wrong.

Of course there are problems embedded in this phrasing. First and foremost, the danger of avoiding hard truths because many people are unwilling to accept them. Even when people are willing to write about such hard truths, there is a risk that they would try to “soften” them by using unpointed words in order to account for the NPOV rule.

[Note]

Andrew Famiglietti writes:

Wikipedia’s most important content policy, NPOV, also took shape to recruit labor. Based on Nupedia’s ‘Non-bias’ content policy, NPOV was one of Wikipedia’s first policies, and early versions of NPOV quickly evolved to meet the  needs of collective labor […] Textual evidence from later versions of NPOV, as well as early Wikipedia press releases, demonstrates Larry Sanger and others saw NPOV as a key to ensuring Wikipedia would attract free contributions. [Andrew Famiglietti, CPOV Reader, “The Right to Fork – A Historical Survey of De/centralization in Wikipedia” p. 305; Nupedia = A project that preceded and eventually developed into Wikipedia – DK]

Blurring the phrasing, subduing NPOV to other principles

The new presentation of the NPOV rule has done little, if anything, to overcome the problems embedded in that 10-year-old abandoned phrasing. In fact, it represents a strong drift into a different direction altogether, not necessarily a better one (and you can tell my opinion about this by now). First of all, while it seems well-phrased, the new version is actually highly ambiguous.

Saying that “neutrality” is “without bias” is a tautology

Since “neutrality” and “bias” are antonyms, saying that “a neutral point of view [is] without bias” is a tautology, or if you will, a mere play of words. You might as well write that keeping clean is avoiding being dirty. Interestingly enough, many people fall into the trap of tautologies when they try to define something for the first time. After several attempts, they realize the problem and come up with a more meaningful definition. Here we see the a process moving in the opposite direction – the first phrasing is more meaningful than the current one.

Of course one bad choice of words doesn’t make a whole phrasing inappropriate, and we often use “redundant” dictionary definitions just to make sure that people understand us correctly. However, even if we ignore this seemingly minor fault, we still find heavy problems.

What does “proportionately” means?

The other interesting word in this first paragraph of the contemporary NPOV rule is “proportionately”. This word, as far as I know, was introduced in order to account for “fringe ideas” and “conspiracy theories”. There is a just claim that Wikipedia, or any other encyclopedic source for that matter, should not present any far-fetched idea published or brought up by someone. Of course many things that we accept today as truths started as far-fetched ideas, but as long as we cannot be sure, the best we can do is to go along the mainstream view, with a pinch of healthy criticism and a watchful eye on new discoveries. Also, there is a just claim that Wikipedia should give precedence to majority views, because they are more likely to be true, especially when we talk about communities of researchers and experts in a specific field (if you take your car to a garage, and nine out of ten mechanics there tell you the problem is with the electricity, rather than the fuel system, then the problem might be with the fuel system, but it is still much more likely to lie in the electricity mechanism).

Playing down the collaborative work method by calling for “proportionality”

However, the current phrasing of the NPOV rule does not say “keep rational”, like the 2001 version did. It doesn’t say, if this idea surprises most people, or treated with total reject by knowledgeable people, then Wikipedia can do without it. I know these two suggestions sound rather blur themselves, but we can think of a more specific rule: Before mentioning an idea or theory, look for articles that criticize it, and see whether at least part of it is widely accepted. In fact, the word “proportionately” is not only equivocal in this context, it also leaves out the notion of collaborative work and its contribution to neutrality. It calls upon people to play down views that seem to them non-conventional, instead of trying to find words and phrases that would cover them as well, preferably with the help of people who actually hold these views.

Subduing neutrality to “reliable sources”

Perhaps the most dramatic change in the course of Wikipedia’s first decade is the fact that NPOV is no longer the first rule of editing. It is not even equal in importance to other rules, even though it is still presented as such. In fact the current phrasing of the principle subject it to the rule of “verifiability”. Now, one may ask where is the problem with that. After all, we want neutrality, but not on the expanse of truthfulness. I would have agreed to such claim, unless the current phrasing of the “verifiability” rule did not talk about truth.

The threshold for inclusion in Wikipedia is verifiability, not truth—whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true. (Wikipedia:Verifiability; May 2011)

Unsurprisingly, this rule too had a different versions, that changed in to the current one more or less at same time in which the NPOV rule assumed its current shape. Here are two examples, shortly before the most recent major revision:

The goal of Wikipedia is to become a complete and accurate encyclopedia. Verifiability is an important tool to achieve accuracy, so we strongly encourage you to check your facts.  We also aim to be informative and neutral. (May 2005)

Wikipedia should only publish material that is verifiable and is not original research.  The goal of Wikipedia is to become a complete and reliable encyclopedia. Verifiability is the key to becoming a reliable resource, so editors should cite credible sources so that their edits can be easily verified. (August 2005)

As you can see, summer 2005, about four years after the creation of the English-language Wikipedia, saw the rise of the “verifiability” rule, and its detachment from its original purpose – to support truthful and accurate presentation. At the same time, “neutrality” lost power and become subject to the rule of “verifiability”, because the NPOV rule now states that neutrality should account for “all significant views that have been published by reliable sources” (emphasis added), namely only those that passed the “verifiability” threshold, which doesn’t really talk about verifiability, but more about bringing references, which is not exactly the same thing. In fact, this phrasing now calls upon Wikipedia’s editors to describe the discourse about reality rather than reality itself , in strong contrast the principles that underlay Wikipedia when it came into being.

It should not come as a surprise to learn that in the current state of affairs Wikipedia’s editors are not discussing issues, but rather debating about them. Furthermore, no one bother to ask what is actually out there, but rather, can you find a source that so says, and once such a source is brought forth, there start a debate whether it can be considered a reliable source (it is not it reliability that is usually contested, but rather the possibility to include it in the “reliable source” category).

The terminology dilemma

Things gets even more problematic when terminology is debated rather than facts. People can often agree about what is there, but disagree profoundly over how to name it. In this case, assuming you don’t have an editor-in-chief or a binding style guide, working collaboratively toward consensual terminology is crucial. Instead, the “verifiability” rule (or “providing reference rule” as it should be called) prevails, and send people to argue about which source carry more weight, hence should be used for borrowing terminology. The fact that source can be reliable and use biased terminology at the same time is ignored because the rules assume that anything brought by “reliable sources” is valid.

Truth and domination

In their introduction to the CPOV Reader I mentioned above, Geert Lovink and Nathaniel Tkacz say,

Living in the shadow of decades of postmodern,‘deconstructive’ thought, claims to neutrality, however qualified and reconfigured, still make us shudder. Humanities and social science scholars and generations of artists and activists  have been trained to be deeply suspicious of such claims. We look to truth’s power, not its enlightenment.

Has Wikipedia’s editors simply got “cold feet” and returned to the safe haven of “let’s not talk about truth”? Perhaps, but there is another issue that must be taken seriously, as Mathieu O’Neil describes:

Means of domination are not limited to the crude use of blocking tools. In fact, such measures are less effective than more subtle means relying on superior project knowledge. The easiest way to defeat an opponent is to assert that their views are not authoritatively backed up by a proper source, that they are violating the sacrosanct WP:NPOV (Neutral Point of View) or WP:RS (Reliable Sources) rules. By extension, all references to editorial, stylistic, and behavioral policies and guidelines serve as battle weapons. Every single action having to do with the project seems to be distilled  into a handy WP:SLOGAN, whipped out at the slightest provocation [Mathieu O’Neil, “Wikipedia and Authority”, CPOV Reader]

As George Orwell showed in his Animal Farm, changing the principles on the wall makes these means of domination even more effective. And yet, Wikipedia is supposed to be different.