I have enjoyed the above posts and would like to add a couple of comments of my own.
(1) The Tom Wright piece that began the discussion interested me because it seemed to appeal to a consensus that I wished to dispute. And herewith lies the importance of the idea of consensus in scholarship, that by appealing to consensus one is making a public claim that is subject to disagreement and even refutation. When one appeals to consensus, one is doing something that is potentially risky because the appeal relies on a certain self-evident acceptance that what one has described would be accepted by one's peers. So if someone comes along and says, "No, it's not the case that the consensus is that the Gospels were all written by the 80s", the original appeal can looks pretty flaky. In other words, appeal to consensus is not something to undertake lightly. And that pressure itself exercises an influence -- it constrains the scholar not to make an appeal unless s/he thinks that the appeal will meet with widespread recognition.
(2) A related point. Michael brings up the question of polling and the like. Michael's second and third questions are "Who gets to be part of the polling sample? . . . ." and "How does one actually go about doing the polling to assess consensus? . . . ." But I am not sure that consensus can be easily and necessarily equated with "the majority view". A given consensus emerges over time and is something that is the result of the combined force of monographs by experts, the introductory level text books, websites, the passing comments in conference papers, conversations over a beer etc. I am not being facetious about the importance of the latter -- it is in the casual discussions that one begins to feel the existence of a consensus, or the lack of one. Of course there is some relationship between consensus and majority view -- I cannot imagine a consensus that is not also the majority view -- but consensus is about much more than just a count of heads.
Let me illustrate the point with Historical Jesus studies. These comments are, of course, subject to disagreement and refutation, but my reading of the current scene would be to say that the majority of scholars and graduate students hold broadly to the view that Jesus was an apocalyptic preacher (or words to that effect). But this is not the consensus view because there is a significant, vocal minority who hold to a different view, a view most commonly associated with John Dominic Crossan in particular and the Jesus Seminar in general. It's possible that in due course one or other of these views might attain to a consensus, I'd guess the apocalyptic Jesus view because I am persuaded, broadly, by those arguments, but I don't think that it could yet reasonably be called a "consensus".
(3) Another related point. It is often the case that a statement of the lack of consensus on a given point can be useful. Take, for example, the question of Thomas's knowledge of the Synoptic Gospels. Here is an area where I think it can be reasonably said that there is no consensus. And that is a useful thing to know -- I would mention something like that in my teaching so that it could give the students something of a feel of the lie of the land, or of a hotly disputed topic. Similarly, in another area that interests me, the question of Johannine knowledge of the Synoptics: there is no consensus in that area.
(4) Michael also raises the question of the expertise of those who are involved in establishing and declaring consensus,
"Who gets to be part of the polling sample? Only those working within a particular historical or theological perspective or methodological paradigm? All those who have published scholarly monographs on the subject? All scholars who have studied the subject in depth, whether they've published on it or not? And who determines what makes a "scholar" or appropriate authority on the subject?This is an area where sometimes one needs to break the question down a little. I wrote in my introductory book on The Synoptic Problem that I thought that there was a major difference between those who might be regarded as experts on the Synoptic Problem, or specialists, and the rest. Among the former group, the specialists, there is no consensus -- those who have published on the Synoptic Problem in the last generation have come from a variety of perspectives and there is real disagreement on a solution. Among the latter group, I think that there is still a broad consensus, a consensus in favour of the traditional Two-Source Theory. I mention that as an example of the difficulty of judging the niceties of the establishment of consensus, and an interesting case of disagreement among specialists that has not always filtered down to the rest.