The threshold for inclusion in Wikipedia is verifiability, not truth. By “verifiaibility” we mean that readers can check that material in Wikipedia represents significant views that have already been published by a reliable source. While verifiability is a requirement for inclusion, it is not a guarantee of inclusion. Wikipedia has other policies and guidelines that affect inclusion (especially whether specific material is included in a specific article).
Wikipedia aims to provide an encyclopedic account of all significant views, appropriately attributed and contextualized, presented in a neutral way. Some significant views make explicit claims about what is true or false, or what is fact or opinion, and we include these as views of the truth or the facts. We also avoid distinguishing between facts and opinions, because there are no universally held definitions for these terms, what some consider fact others may consider opinion, what everyone considers fact may turn out to be opinion tomorrow, and because the difference between the two is not black-and-white. Experts are often divided over what is true or false, right or wrong, good or bad, beautiful or ugly. Indeed, they are sometimes more divided than the general public believes.
Our responsibility is not to decide which views are right, but rather to represent such views accurately. We try to assess and acknowledge degrees of controversy. Some propositions are so uncontroversial that we can put them in without citation or attribution. Some things are so controversial that not only must we provide a citation, but a quotation and specific attribution. And between these two extremes may be many degrees. For example, virtually all scholars in the life and human sciences agree that human beings are descended from a single-cell organism, which makes this kind of claim about as significant as possible. But scholars are divided as to whether the gene or the species is the principal unit of evolution, and sscholars are divided as to what human behaviors can be explained by evolution — these views are significant, but only in the context of other, different views that are equally significant. Moreover, some people reject the theory of evolution altogether. Their views are not significant in relation to evolutionary science, and therefore do not belong in the article on evolution. But they are significant in other contexts, and we thus include a separate article covering the views of creationists. In order to make these distinctions, it is not enough to verify that the view exists. We must also provide an account of the context in which these views have currencty, and our accounts of these contexts must also be verifiable.
In other words, we avoid arguments over whether views are true or false and instead seek to verify that they are indeed significant, that we are giving each view due weight, that we include equally significant views, and that we are providing accurate accounts of these views. We do not decide these things based on our own opinions; we must verify them. In order to verify them, we look for reliable sources. Editors collaborating on an article must investigate the available sources and use their judgement as to which sources are most current and appropriate. But we must use verifiable sources so that readers can check for themselves the accuracy of our accounts.
In many cases, our articles provide apparently uncontroversial infromation (such as, the date of the signing of the United States' Declaration of Independence, or the date of the Norman Conquest, or the name of India's first Prime Minister). Such information does not normally require attribution to a specific source. If however an editor believes that the information is inaccurate, we are responsible for verifying its accuracy. If information appears to be inaccurate, it may be because we have misquoted a reliable source, or it may be because we are not using the most appropriate source for this particular information. In some cases, further research may reveal that the information is actually controversial, and has been the object of scholarly debate. It is important to avoid original research, but editors may decide that it is important to provide a verifiable source. In this case, editors must determine which source is most current, reliable, and appropriate.
Verifiability often requires editors to distinguish between mainstream and fringe views, and between fringe views and views that fail to meet the threshold of notability required for inclusion in an encyclopedia.
- a mainstream view is widely expressed in multiple media or venues for a notable constituency (e.g. "Christians" or "biologists" or "Shakespeare scholars")
- a fringe view is widely cited in multiple media or venues for a notable constituency, but ngatively.
- a non-notable view is rarely or never cited in multiple media or venues for a notable constituency. For example, a published book advocating a particular theory regarding Shakespeare is never cited, or cited only rarely and in passing, in mainstream journals on English literature. In these cases, a lack of appropriate reliable sources is actually informative and valuable in assessing the weight to be accorded a published view.
- sometimes, a work may be notable but is seldom cited because it addresses a very narrow issue. In such cases, other criteria may establish notability (e.g. the credentials of the author and the reputation of the publisher) but it is important to establish that the author is a recognized expert on the specific topic.
In order to make the above distinctions, editors need to verify more than the fact that something has been published in a reliable source. Editors must be able to:
- identify the appropriate and relevant field of expertise (in many cases, one or more academic disciplines)
- the mainstream journals of those fields
- the difference between a positive or negative citation.
- the difference between a central point that reflects the author's research, and a minor point or speculation. This can usually be determined by examining the larger context of a sentence, paragraph, or section of a published work.