We seem to throw around "consensus" more and more as time goes on with skeptics. However it also seems that a claim of "consensus" is the one claim that does not require a reference. Should a claim of consensus require a reference to meta research which shows that there actually is a majority opinion of the appropriate scientific community?


A claim of consensus is a claim like any other. There should be evidence that the claim is true and a reference to that evidence included in the answer using the claim.

If anything, I swing the other way than your opinion that claims of consensus require more evidence... but I may be misunderstanding exactly what you are aiming for.


Absolutely, why not. However, we should be careful as to what references we allow. Usually (exceptions prove the rule) there is no “paper” proclaiming a consensus on a certain point.

I think we should accept as consensus any fact that has been repeated in a University-level text book, or if it’s visibly taken for granted in scientific publications. Otherwise a consensus may be very hard to prove in many cases.

  • 2
    I think "printed in a college textbook" is a pretty low bar to set. – Russell Steen Jun 7 '11 at 20:04
  • @Russell That’s why I explicitly didn’t say that. – Konrad Rudolph Jun 7 '11 at 20:16
  • @Konrad - I'm just not understanding your post then. It sounds like "I think we should accept as consensus any fact that has been repeated in a University-level text book" is implying that. – Russell Steen Jun 7 '11 at 20:56
  • @Russell There’s a fundamental difference between a University-level textbook and a college textbook.” The latter is written and published by institutes which don’t necessarily have any scientific credentials while the former is authored by eminent scientists and reflects the (more or less) current state of research. – Konrad Rudolph Jun 8 '11 at 7:03
  • @Konrad -- Thanks for the clarification. When I attended University we were required to use quite a few textbooks written by the professors who taught the class, and while some were eminent scientists, I don't trust them not to bias such texts towards they're own conclusions. – Russell Steen Jun 8 '11 at 17:09
  • @Russell True, there is always bias. But (at least in all text books that I used) all claims are referenced. The references aren’t necessarily directly next to the claims – often they are at the end of a chapter. – Konrad Rudolph Jun 8 '11 at 17:39
  • @Konrad -- Very good point. Given that though, I would accept a standard citation to the book as a reference. However just claiming "my textbook said so" shouldn't cut it. A reference doesn't have to be a "web link" to be a reference. – Russell Steen Jun 8 '11 at 18:06
  • @Russell Yes, but unfortunately the single reference wouldn’t (usually; unless it’s a review) show a consensus. Neither does the textbook necessarily, but from experience claims in U-level textbooks aren’t controversial (any more) and hence do represent a consensus. – Konrad Rudolph Jun 8 '11 at 18:51
  • A textbook should teach the state of the art; and if the state of the art is that there's disagreement or speculation instead of consensus then the textbook should say so, explain the evidence, and the limits of the evidence (I'm thinking here of ancient history for example: historians aren't sure of various details of the history of early Rome, for example). – ChrisW Jun 9 '11 at 14:37
  • @ChrisW My point exactly. – Konrad Rudolph Jun 9 '11 at 15:32

I think this question is fundamentally flawed since consensus, by definition (even when existant in the field), ignores anomaly and breakthrough directions.

Many arguing against consensus should return to their R&D and establish the data for or against the perceived consensus. Mainstream thinking can change suddenly with a single new ray of truth.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .