Walt and Isabel are talking about the newest oddity in mathematics publishing: a forthcoming journal called *Rejecta Mathematica.* This will be an online journal dedicated to mathematical papers which have been *rejected* from peer-reviewed publications.

Such a journal could be a useful publication venue: papers which show that a promising technique fails or which reprove a known theorem in a not-quite-snazzy way might be worth collecting. Furthermore, it would be neat to look at a probability argument or some “entropy” bafflegab from a cdesign proponentsist and say, “This couldn’t even be published in *Rejecta!*“

## 6 Comments

One thing I didn’t say in my original post: isn’t this what, say, blogs are for? I guess I’m skeptical of starting new journals when it’s so hard to get libraries to carry the old ones. I like the idea of having a venue for these sorts of things — obviously, as I mentioned — but I’m not sure if the traditional journal system is that venue.

Good point. I don’t think we need to fret about getting libraries to carry a free, Creative-Commons-licensed online journal. However, as you say, the journal system might not be the best place to convey such information, and so resources (time and money) devoted to managing this journal would then be better spent elsewhere.

The file-drawer effect is important in empirical stuffs. Studies that fail to reject the null hypothesis usually don’t get published.

That’s a shame because they’re important. If there are two studies of some hypothesis, and one finds an effect and the other doesn’t, folks really need to know that the positive finding is contested. Perhaps someone will scrutinize the two experimental designs and find that–ha!–the positive study didn’t control for variable XYZ but the negative study did. So I could totally dig a peer-reviewed journal of (say) experimental psychology seeking out rigorously achieved, insightfully interpreted, and important

non-results.The journal vs. blag question’s interesting. In the case of null results, one argument for a journal is that publication helps out a scholar’s resume in a way that blog posts rarely do, and it would surely be good for experimenters’ objectivity if they didn’t feel their career hinged on an experiment coming out the right way.

In either context I’m not sure “journal” and “no technical review” can live happily together. If papers are skimmed lightly, that justifies tagging them as “not rubbish” in the arXiv or some interface thereto. But not publishing. Or so I think; naysaying’s easy, but tomorrow we could be welcoming our new masters at Rejecta Publications.

…hm, never mind blogs, isn’t this what the

Arxivis for?It’s certainly a

subsetof what the arXiv is for. The question to which my mind keeps returning is how we can establish a system which gives academic folks credit for papers that uphold the null hypothesis, and such. Countering the file-drawer effect could be beneficial all around. I’m rather doubtful thatRej. Math.will gain enough renown for publications therein to look good on the ol’ CV, however.For a great example of this, see the Journal of Negative Results.