How many copies? (Of, y'know, obscure linguistics books.)

[Daniel]

In my research recently I've been using relatively obscure sources for very specific details, even relative to Linguistics in general. Y'know the type of book-- the one that probably hasn't been checked out from the library in the last few decades.

And sometimes the library at my university doesn't have a copy, which is fine because I can get it through inter-library loan. But I've been thinking recently...

Just how many copies are printed of those really obscure books that aren't even purchased by most major libraries for universities with Linguistics departments? I can't imagine too many individuals go out of their way to buy a personal copy. So are there lots of "published" books out there with 5-10 copies printed? Or even just 100?

Let's say I publish a book based on my upcoming dissertation. How many copies would likely be printed (assuming it's not a major best-seller)?

I'm just curious. Publishing is publishing, and generally those who want the information can get a copy somehow.

Beyond that, what about the publishers? Do they really go through all of the effort of publishing for just printing a dozen copies?

Any statistics available for academic publishing in general or especially linguistics in particular?

[Corybobory]

Hm, I have absolutely no idea.  Publishing on demand though, isn't that becoming more popular now that self-publishing is becomming common?  If you published your PhD and then it was sold through amazon, do you think maybe they would print it after someone bought it?

[Daniel]

Yes, and I think it's very possible there might be only 5 copies ever printed.

Certainly if you're reasonably successful you'll get it purchased by libraries-- so you can head toward 100 copies. But somehow that just doesn't seem like much.

This applies especially to older, print-only materials. Open access online materials are quite different (if perhaps less profitable).

I'm thinking mostly about those books that have 2-3 listings on worldcat.org and are "out of stock" on Amazon. I had a book sent by inter-library loan from Sweden (it was a copy personally donated by the author), for all I know the only copy in the world (yet cited a few times), and another that I eventually got scans of from the University of Hawaii, which I'm pretty sure was one of three copies in the world (the others in Indonesia and Papua New Guinea [where it was 'published']).

[Daniel]

This link suggests most books sell a few hundred copies:
http://beforeitsnews.com/international/2013/07/the-average-number-of-copies-that-most-books-sell-is-a-few-hundred-2461808.html

Of course it's not about academic publishing, but that still gives a figure as somewhere to start. It feels to me like a few hundred copies would be pretty lucky for academic publishing, aside from textbooks.


More relevant information:
http://www-personal.umich.edu/~jaylemke/guide-bk.htm
(Pretty good article, and it looks like that's specifically for sociolinguistics.)

This one has some good info (in the "Academic" section):
http://press.uchicago.edu/Misc/Chicago/288447.html
It suggests that for an obscure topic a top-selling book might sell 400 copies. So a not-so-top-selling book would be much less than that.


Anyway, I think I now have a rough idea of the situation.

All of this seems to suggest that open access, especially for longer works, is the best plan for the future.