One contributor to the comments here (JC) has pointed out that Susskind has withdrawn from the arXiv his recent paper on the stupendous Landscape of sting theory. This is pretty unusual, but often when it happens the author puts in the withdrawal statement some indication of why the paper was withdrawn, something Susskind didn’t do in this case. Another contributor to the comments (Serenus Zeitblom) points out that one should look at recent changes to Douglas’s paper on the arXiv, which is now up to its fourth version.
One feature of the arXiv is that all posted versions of papers are available, so one can compare them and see what changes the author made. The history of Douglas’s paper is quite something. The original version was posted on May 30. Susskind’s now withdrawn paper was posted on June 17, and in it he claims that Douglas’s paper showed that Susskind’s argument in an earlier (May 21) paper (which exists in three versions) was wrong. The latest version (4) of Douglas’s paper now says that earlier versions of the paper are wrong. So, one reason Susskind withdrew his paper is presumably that its claims that his earlier paper was wrong were now wrong because they were based on Douglas’s wrong paper. Got that? This all seems to me to be a new and original version of the “Not Even Wrong” phenomenon.
Some other high points of the changes in the four versions of Douglas’s paper:
1. Going from version 1 (May 30) to version 2 (June 2), he changes
“If I had to bet at the moment, I would still bet that string theory favors the low scale, for the reasons outlined above, but it is not at all obvious that this is what will come out in the end.”
“At this point, it is not at all obvious whether high or low scales will be preferred in the end.”
2. Going from version 2 (June 2) to version 3 (June 22), he adds a reference to Susskind’s June 17 paper, some criticism of it, and the sentence
“The correct assumptions could be determined from string/M theory considerations with more work, and we are optimistic that this can be done in time to make convincing predictions before LHC turns on in 2007.”
3. Going from version 3 (June 22) to version 4 (June 29), he removes the sentence above (I guess he became less optimistic last week) and announces that the argument in the previous versions was wrong.