Wikipedia:Fallacy of selective sources
 | This is an essay. It contains the advice or opinions of one or more Wikipedia contributors. This page is not an encyclopedia article, nor is it one of Wikipedia's policies or guidelines, as it has not been thoroughly vetted by the community. Some essays represent widespread norms; others only represent minority viewpoints. |
 | This page in a nutshell: The argument that consensus discussions about content may only (or should primarily) consider those sources already used for citations is patently false. |
An uncommon but long-term-recurrent argument suggests that if the sources presently used for citations in an article seem to agree on something, that this constitutes some kind of finding of fact or establishment of consensus (real-world or on-wiki). That is, of course, nonsense. It is sometimes recast in other words, e.g. that the sources already cited should be given primacy over all other sources (as if they have been through some kind of formal vetting process, which they haven't); or that if the sources we're using now seem to agree on something that this is good enough, and further examination of source material is unneeded, even unwarranted.
Wikipedia editors consider the real-world consensus (scientific, historiographic, English-language-usage, etc.) as determined by a preponderance of all available relevant, modern, independent, reliable, secondary sources we can bring to a consensus discussion. If this were not true, then:
- It would not be possible to do a WP:Due and undue weight analysis of anything that all, since only the sources cited right this moment could be used to assess viewpoint balance.
- Several forms of WP:Original research could not be identified and removed, chief among them the use of cherry-picking of some carefully selected sources that seem synergistic, so as to "lead" the reader to a conclusion not actually found in those sources, while omitting sources that made it clear that this seemingly inexorable conclusion was a misinterpretation.
- Imposition of false consensus and outright false facts would become trivially easy. E.g., if the article Kesha were edited to predominantly cite entertainment-news sources that referred to her as "Ke$ha" following her old album-cover stylization, this would produce a faked appearance that her WP:Most common name in reliable sources was in fact "Ke$ha". Yet not only is that false today, it has always been false, even back when Kesha was using that stylization in her marketing.
- WP:Requested moves and other processes would be subject to gaming with trivial ease: For example, if you didn't like the fact that our article title used lower-case for something you thought should be capitalized, despite most reliable sources not capitalizing it, you could "win" by just swapping out all lower-casing sources with upper-casing ones of otherwise equal quality and pertinence, then request a move with the so-called proof that "all of our sources capitalize this".
- Our articles would be frozen in time, against WP:Wikipedia is not paper policy, failing to use newer source material to update formerly accepted but now disproved claims, or other changed facts, that we'd gotten from older source material.
Wikipedia consensus formation considers all available, valid source material. For simple matters like titling and style questions, we directly depend on aggregate results (Google Ngrams that track string-usage frequency in books over time; Google News, Google Scholar, and Google Books search results and the patterns they reveal in sources; etc.). The idea that they are to be ignored, or are second place to what just happens to be cited already in the article as of this timestamp, is not only unworkable but absurd. It bears no resemblance to how Wikipedia:Consensus is actually formed.