<CPOV> Something I've written about the development of the NPOV principle on en-wp

[Joseph Reagle]

On Wednesday, May 25, 2011, Dror Kamir wrote:
> I'm going to read it of course (as I said, I still have a lot of work to 
> do with this "thesis" I am going to present), but just to clarify 
> something - I wasn't thinking too much about the rationality issue, in 
> fact. What bothered me more is the undermining of the "meta-principle" 
> that NPOV is a desired byproduct of collaborative work (to be more 
> exact, collaborative work is the meta-principle which caters to the 
> three basic principles of WP, NPOV included) and the transition of NPOV 
> from being independent, even dominating principle to its subduing to the 
> "verifiability" principle (which is not exactly verifiability rule imho).

I am not sure I agree with his characterization, as I have been arguing the opposite quite some time myself. That is, neutrality is best not used to describe the product of collaboration, but a necessary ingredient to collaboration. Rather than describing the encyclopedia, it describes the approach Wikipedians should take with one another relative to knowledge claims.

> The latter shift is dramatic, first of all because it puts WP under the 
> somewhat post-modernistic approach that there is no truth but rather 
> people's talks about truth. When I went to "Wikimania 2006" in Boston I 
> had a strong feeling that WP rejected this notion and tried to return to 
> the modernist "search the truth" approach.

Coincidentally/interestingly, it was at Wikimania 2006 that I first made this argument [1,2].

[1]: http://reagle.org/joseph/2005/06/neutrality.html
[2]: http://reagle.org/joseph/Talks/2006/0806-wikipedia-neutral.html

That said, I wouldn't characterize Wikipedia's epistemology as necessarily post-modern -- where the real problem is they use the "truths" where "perspectives" would do -- since Wales and Sanger are both rooted in an objectivist philosophical worldview. As I quote Wales and the book:

[[ 
Surely you will agree that there are _more_ or _less_ accurate, objective, fair, [un]biased ways of putting things. We should simply strive to eliminate all the problems that we can, and remain constantly open to sensible revisions. Will this be perfect? Of course not. But it is all we can do \*and\* it is the least we can do.... if you are trying to say that someone, somewhere will always accuse us of bias, I'm sure you're right. But we should nonetheless try our best to be objective. It doesn't strike me as particularly difficult. We will want to present a broad consensus of mainstream thought.... This does mean that sometimes we will be wrong! All the top scholars in some field will say X, but 50 years from now, we will know more, and X will seem a quaint and old-fashioned opinion. O.k., fine. But still, X is a respectable and valid opinion today, as it is formed in careful consideration of all the available evidence with the greatest care possible. That's the best we can do. And, as I say, that's also the least we can do. \acite{Wales2000b}
]]

> Indeed the changes in the pseudo-"verifiability" rule gave people with 
> knowledge of academic norms and access to (Western universities-based) 
> academic libraries a huge advantage. Perhaps this another explanation 
> why issues related to Africa (for example) became even harder to write 
> about, as Heather Ford and Mark Graham showed in their articles.

I do think that verifiability and no original research privilege people with Internet access -- not necessarily academics since many high-quality online sources are accessible yet even high-quality but print-only sources are not liked as much since they cannot be easily verified. I think it can also lead to problems outside of the Western/online cultural context. A couple of years ago someone was telling me how much difficulty they were having in writing articles about free software in China because -- since there was nothing for them to cite in Chinese really -- their work was construed as original research.