Here is an excerpt from the interesting interview *A conversation with S. R. S. Varadhan*, by Ofer Zeitouni. published in Statistical Science 33 (2018), no. 1, 126–137, which should not be confused with another interesting interview with exactly the same title, by Rajendra Bhatia, published in The Mathematical Intelligencer 30 (2008), no. 2, 24–42.

**4. ON PUBLISHING **

**Ofer:** I wanted to ask you about publications. When one looks at your publication record, a good deal of it is in CPAM (Communications in Pure and Applied Mathematics, the Courant journal), and another striking fact is that almost none (or maybe none at all) is in “big” journals. This does not seem to be an accident… Did you never want to submit to these journals?

**Raghu:** As a probabilist I wanted to publish in a place that probabilists would read. So I wanted my work to appear in probability journals – as long as it appeared in reasonable probability journals, I did not care where it appeared.

**Ofer:** It seems that there has been a big change – it used to be that probability papers rarely appeared in those journals, lately it is more common.

**Raghu:** I never thought of it. For me, CPAM was natural because it was our journal. Every once in a while I would submit elsewhere for a change – few to CMP (Communications in Mathematical Physics) or to the Annals of Probability and PTRF (Probability Theory and Related Fields). The earliest publications are in Indian journals, such as Sankhya. To me, if you have a good paper and it comes out in a journal that people look at, I am happy – it doesn’t really matter where you publish it.

**Ofer:** And that brings me to another question. Do you think people still look at journals, in the days of arXiv? Are journals important anymore?

**Raghu:** Indeed, there are no preprints anymore… Journals are important for deans – for promotions, for the institutions. This is what the dean wants – what is the number of your publications, where have you published, what is the impact of the journal; the department needs to put these things together for a promotion to be approved.

**Ofer:** That’s a somewhat cynical view of the publication world…

**Raghu:** That’s the only reason we now have journals.

**Ofer:** So scientifically, their role is over?

**Raghu:** I think that what would make more sense scientifically is to put your paper on the arXiv, have reviewers read the paper and give their seal of approval. This could be anonymous – there could be a group that requests reviews for articles they deem important, the reviewers read and make their comments, find mistakes etc, and eventually approve. There is no reason for journals – there is no reason for universities to spend millions of dollars to make publishers rich. Publishers have an important role in publishing books, not journals. They don’t make that much money out of it, however.

**Ofer:** You do have a long experience as editor, both at CPAM and in the Annals of Probability. How does that align with what you just said about journals?

**Raghu:** 20 years at CPAM… Well, so long as the system is there, you have to play according to the rules.

I had the chance to discuss a few times with SRS Varadhan… and that’s basically how I wrote my best papers 🙂 He is really one the most influential and brilliant mathematicians in the world.

Yet, one cannot say that all of his papers are written with much editorial accuracy. So, I also think he does not target the top journals because he is too lazy to write down things in the style required on such journals.

As for the fact that commercial editors have become a ballast -not a tool- for mathematical sciences… well, one does not need the wit of Varadhan to remark it. It’s just pretty obvious. How to practically solve the matter, that’s a more interesting question.

Regarding the first point, Varadhan published in PTRF, AoP, etc, but not in Annals of Maths, Acta, etc. Actually the level of PTRF and AoP in terms of editorial accuracy is not below the one of AoM and Acta! For CPAM, it is also questionable even if what you say is more plausible. In the mean time, regularly, very well known people do publish non accurate papers in top prestige journals.

Regarding the second point, I have discussed with several “senior” mathematicians about publishing. I can assure you that they do not have the same opinion! I like very much the one of Varadhan, notably on the possible role of arXiv. The system does not change presently for mathematical publication mainly because of massive conservatism and lack of audacity and vision of seniors in charge of mathematical journals. Also I am sorry to disagree with you: yes we need the wit and charisma of all the Varadhans for this matter!

PS: By the way, why staying anonymous?