I have to say, I disagree pretty heavily with the interpretation of both NPOV and NOR you are putting forward here. Let us take a concrete example -- say, German's wartime project to develop nuclear weapons/energy. What should the article say on the subject, in an ideal world? There are at least two separate scholarly narratives on the subject (one which says "the scientists purposely didn't built Hitler an atomic bomb" and one which says "the scientists just didn't get around to it for various reasons, but weren't taking a strong moral stance"), and while I think one is more popular than the other (the latter mentioned) among scholars, there is no clear understanding of consensus or even which has the more prominent scholars behind it (there are sharp and well-respected people behind both approaches, and of course a requisite middle ground where most people live). How to proceed? In my understanding of NPOV, the article should do a good job of representing both scholarly opinions and the evidences both cite. It should mention that there is a scholarly debate. It should perhaps mention a brief history of the debate, if germane (i.e. Heisenberg's misquoted letter in Jungk's book, Bohr's angry reply, leading all the way up up to the _Copenhagen_ play). There are, of course, more fringe theories all over the internet, as there are with anythign relating to Nazis, atomic bombs, or physics. These don't have any currency in mainstream sources and there is not even a consistent fringe conspiracy theory. As such, I don't think they should be represented in the article -- they are not "significant" POVs, which is to say, they are not "POVs significantly represented in the public understanding of this subject." If there were one overriding or popular conspiracy theory, it should get a mention commensurate with its representation in the larger community of knowledge (i.e., the Apollo moon landing conspiracy is a well-known phenomena in and of itself, so it gets it own article). That's MY understanding of NPOV, which reduces knowledge and truth statements to sociological statements -- who says what, how representative the statement is of other actors, etc. With this understanding of NPOV, then NOR falls into a similar spot (and almost becomes No Original Opinions as well). If a source is not accessible, it is likely not part of the mainstream understanding, however you define "mainstream." Obscure sources, by their very definition, are likely not part of this sociological understanding of NPOV. Again, concurring with Matt Brown, if I found something that I think was wrong about the mainstream historical account, I shouldn't try to use Wikipedia as the staging point for this new understanding -- I should try to publish it in a scholarly journal! After all, isn't that what their domain is for? That is, I should turn the statement into a "mainstream statement" (under a loose definition of mainstream), at which point it becomes potentially subject for Wikipedia inclusion. I don't think Matt Brown meant "reporting the consensus" to mean "creating a sense of consensus if there is none." I think he meant more what I mean when I say "mainstream" -- statements or understandings which have either enough "followers" (in a Latourian sense) or have generated enough "attention" to warrant inclusion. The "consensus in the field" on the German atomic bomb is a number of stories which disagree with each other, in this model -- it is not one of the stories over any other. NOR (or again, NOO) means, along with the idea that I can't present my own private "data", that I can't read a book on the subject, come to radically different conclusions than the author, and then try to have those (non-mainstream) conclusions represented in an encyclopedia article. We rely on other sources (journals, academic disciplines, communities of fringe believers) to develop their own "regimes of truth" in the Foucauldian sense -- we only report on the "truth statements" they generate, if their communities are "notable" enough to warrant inclusion. (Such notability, as always, will be somewhat arbitrary and will always be in flux, but that's not new to this question.) At least, that's how I understand the policies. FF On 6/7/05, steven l. rubenstein <rubenste(a)ohiou.edu> wrote: