Peer reviewing

[Martin Haspelmath]

Bill Croft wrote:
> But the main value for "revise and resubmit" is that one doesn't know 
> how much an author really will revise the manuscript. Not 
> infrequently, I receive "revised" manuscripts which had significant 
> problems where the author has merely added a few footnotes to the 
> original submission. In those cases, I do feel that I have wasted my 
> precious time, as Lachlan puts it, and I will recommend rejection.
What Bill describes as "the main value" of R&R is the main problem, in 
my view.

In the cases mentioned above, the author probably limited herself to 
adding a few footnotes because she simply didn't agree with the reviewer 
that "the manuscript had significant problems". And often the author is 
right, not the reviewer. Reviewers are not more knowledgeable than 
authors; in fact, they generally know much less about the paper's topic 
than the author.

But predicting whether the editor will overrule the reviewers or not is 
very difficult, so should the author resubmit? This is extremely tricky, 
and I think many papers are delayed because the author is at a loss what 
to do: Follow a reviewer's proposals she is unhappy with, or try a 
different journal?

So I think a new approach that only has "accept" and "reject" would make 
everybody's lives easier.

Martin