I am not sure this forum is the right place for this question, but I'm pretty sure it would quickly be closed at MathOverflow.

There is a well-known mathematics journal called "International Mathematics Research Notices", published by OUP. It's almost invariably cited as "Int. Math. Res. Not. IMRN" [sic] or similar, which has always struck me as oddly redundant. The curious ubiquity of the abbreviation is doubtless due to the fact that MathSciNet provides references in that format: "JOURNAL = {Int. Math. Res. Not. IMRN}".

On the other hand, Worldcat has many listings for the journal, some of which have the double IMRN, and some of which don't. As far as I can tell, the journal's page doesn't specify a preferred abbreviated form.

Another possibly relevant factor is that the journal is usually referred to be this initialism: people call it "IMRN" in ordinary conversation, which isn't the case for any other journals I can think of, JAMS perhaps excepted if you count pronouncing it (and nobody writes "J. Am. Math. Soc. JAMS").

The question: why does MathSciNet double the journal name, and should I be doing the same? (Perhaps the underlying question is: who decides on official journal abbreviations?)

  • 1
    If there won't be any definitive answer from the math community in a few days, one may consider reasking at MO…
    – Dirk
    May 24, 2017 at 4:22
  • 1
    "which isn't the case for any other journals I can think of" - I use in conversation JFM (Journal of Fluid Mechanics), ZAMP (Zeitschrift für angewandte Mathematik und Physik, that one is a bit of a special case as the title isn't English), PRL (Physics Review Letters).
    – SPPearce
    May 24, 2017 at 8:24
  • Interesting, I don't know these! I'd just skimmed through a couple algebraic geometry bibliographies, so I've probably forgotten others. "MRL" might count. J. Reine Angew. Math. is another complicated (to me) German name, but it just becomes "Crelle". And "Math Z" is "Math Z".
    – Mark
    May 24, 2017 at 15:24
  • Another example where the journal is almost always referred to by the initials: GAFA. Outside of the mathematical sciences, PNAS is common. May 25, 2017 at 15:15

2 Answers 2


This is just speculation, but the reason could be because of the slightly complicated history of the IMRN journal.

Throughout you should remember that MathSciNet grew out of what originally was a print publication called Mathematical Reviews. It was extremely important that they have a uniform house style for referring to journals.

Originally there was a journal called International Mathematics Research Notices published by Duke Univ. Press from 1991-2001. On MathSciNet its abbreviation was Internat. Math. Res. Notices. When the journal moved publishers to Hindawi in 2002, the abbreviation became Int. Math. Res. Not. on MathSciNet. In 2005 Hindawi started a new journal called International Mathematics Research Papers, which was indexed by MathSciNet under the abbreviation IMRP Int. Math. Res. Pap.

In 2006/7, the two journals were both transferred to Oxford Univ. Press. There was some relationship between the two journals (I remember seeing it in an old author submission guideline) and the two are treated as sort of a pair. Probably in an effort to clearly distinguish the two journals, either OUP or MathSciNet decided to list the two journals as, respectively, International Mathematics Research Notices. IMRN and International Mathematics Research Papers. IMRP with abbreviations Int. Math. Res. Not. IMRN and Int. Math. Res. Pap. IMRP.

In 2009 or thereabouts, OUP merged IMRP into IMRN.

I would surmise that the "duplication" appeared in an effort to add some redundancy for better disambiguation of two journals with extremely similar titles. Now it survives as a vestigial testament to the journal's history.

  • Re: the relationship, IMRP was meant to be for longer papers than IMRN. I guess, in part, because IMRN was meant to be fast.
    – Kimball
    May 25, 2017 at 22:33
  • @Kimball: turns out you are mostly right. I dug up the journal about page on the Wayback Machine, and while IMRP is also meant for "fast" publication, they are meant for much longer articles. From the description, each issue will be only one article. May 26, 2017 at 1:21

Not a mathematician, but neither I nor any of my colleagues was ever able to paste bibliographic references without revising them. I wouldn't necessarily care about somebody else writing "Int. Math. Res. Not. IMRN"—. At least, "Int. Math. Res. Not. (IMRN)" or similar punctuation would highlight the redundancy better—but I'd avoid introducing a third form if two are standard.

The question: why does MathSciNet double the journal name, and should I be doing the same? (Perhaps the underlying question is: who decides on official journal abbreviations?)

Your style file comes with guidelines—I'd guess they don't standardize journal abbreviations, so there's no standard. Style manuals also come with guidelines, but I've never seen this matter being regulated.

  • 8
    I think this is a case where math is special - the MathSciNet database is comprehensive and almost universally used. I've always been able to paste BiBTeX entries directly from it, and use them without any revisions. Journals invariably provide BiBTeX style files that process those entries as the journal prefers - but I think any math journal where you couldn't paste from MathSciNet would find its authors really annoyed. May 24, 2017 at 1:25

You must log in to answer this question.

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