How to decide between sending paper to a specialized journal or to a journal with broader audience?

Question

I work in mathematical physics. Typically, papers in my field are either published either in general math journals (Annals of Math, Transactions of AMS, Proceedings of AMS) or in specialized math physics journals (Communications in Math Physics, Reviews in Mathematical Physics, Annales Henri Poincaré, etc.) Personally, I have had papers published in both types of journals. Yet I don't have a good sense whether a particular preprint should go to a general journal or a specialized journal, so I usually just pick at random.

What are the factors one thinks about when making a decision between submitting to a specialized journal and a general journal of similar quality?

Answer 1

I tend to think that in most cases, the specialized/broad dichotomy is not very relevant. The most important point is to send it to an interested editor; if the most relevant editor for a paper happens to be at a general journal, you will often be better of sending your paper there.

I would consider two exceptions to this principle. First, top specialized journals are usually less reputed than top generalist journals, so if you get a truly impressive result, you may want to get the best of it by sending it to a top generalized journal. Second, some generalist journals will turn back papers that seems much more specialized than the average math paper (e.g. when the basic objects you study are unheard of by most mathematicians).

Answer 2

I would pick the journal with the largest audience that you feel is likely to actually read your work. That is probably the generalist journal, but not necessarily. If you have a specialist journal that all of the people in your field read, your work might get more attention that way. But generally, more general journals have larger audiences, therefore, more people who might be interested in your work, and hence also a higher selectivity rate.

Answer 3

I don't care how important the journal is or who the editors are. The question is do I read it, or do the papers of relevance turn up in it, or do the most relevant/sympathetic authors publish in it, which means there will be relevant/sympathetic reviewers. Publish in the journals where you see the most similar papers to what you are proposing, in approach, scope, length, depth, etc. It is appropriate to publish short papers on new results quickly in the more specialised transactions, and longer papers that draw together work and provide introductions to a broader audience in the more general journals. If the work is interdisciplinary it is appropriate to present it for different audiences by retargeting to a new journal. The more ground breaking the research the more important it is rather to see what journals publish novel research as opposed to bandwagon research. If none of the reviewers understand it, a reputable venue will not publish it (although often they won't admit the problem is theirs).

But in general the story goes like this...

research --> journals that publish that research --> most appropriate journal for your research --> target that specific journal

In other words I think you have it backwards - you choose the venue and then target the paper to it. I develop repositories of knowledge for my own use (somewhere between a collection of notes and thesis-like treatise). Sometimes I publish these as a Tech.Report. but the idea is not to publish on the growing body of work, but to draw on it as it grows to target papers to particular venues (workshops, conferences, transactions and journals).

I regard general journals as the least useful kind of publication, and workshops as the most useful kind of publication, with a funnel type progression to a proper journal paper (which I my field are typically 50-100 page papers whereas the other three typically have 6-12 page papers). In the internet age, the most important thing is to publish in places that provide or allow open access, and most readers will come from a search process rather than by subscription to the journal. Subscription journals are dead - they just don't know it yet, although the publishers are scrambling to get you, your employer or oterh sponsors to continue to pay them in a paid open access model.