There seems to be an important area that is being missed by this
disucssion.  Rather than trying to ask editors to spend more of their time
filtering out poorly written manuscripts, or asking more folks to spend
time reviewing these manuscripts, the problem should be addressed closer
to the source... the author.  There should simply be fewer badly written
manuscripts sent to journals.  The best way to accomplish this would
probably be to have the authors put more effort into ther internal review
process than they commonly do.  More effective internal review of
manuscripts would benefit almost everyone.  Authors would get quicker
turn-around time from reviewers, editors and referees would get fewer
poorly written manuscripts to put through the system. Frankly, at most
academic institutions there is not an effective internal review mechanism.

        From 1991-1994 I worked at Oak Ridge National Laboratory where
_every_ manuscript that went out for publication was required to have 2
internal reviews conducted with a "green sheet" signed by each reviewer
and sent into the editorial office prior to submission.  Similar systems
operate at other federal research facilities (NIH, CDC are two that I'm
familiar with).  I will admit that there are important flaws with the
system at ORNL: 1) finding internal reviewers that were qualified was
occasionally difficult, and 2) some reviewers were known to sign off with
relatively uncritical and sometimes unhelpful reviews.  However, the
system did force authors to have manuscripts read by others not on the
author line and did reduce the frequency with which badly written
manscripts were sent to journals.  I believe that if one were to look at
acceptance rates, the federal facilities with internal review system would
have significantly higher rates of manuscript acceptance than most
academic institutions.  The more researchers that read a paper before it
is sent to the journal, the more likey that easily fixed flaws in language
or obvious problems (generally with statistics and "first cut"
interpretation) will be caught early.  Also, internal reviewers need not
necessarilly come from the author's specific area or even department.

        From the point of view of academics, a formal system like that at
the fedral labs is almost certainly not feasible.  However, I have
reviewed manuscripts that had truely horrible language problems.
Generally, but not exclusively, these were mauscripts submitted by
graduate students or young scientists (not that I'm an "established"
scientist myself).  In such cases, it seems that the mentor of the
manuscript's author should spend considerably more time with the author
prior to the manuscript going out the door.  My writing has benefited
greatly from the efforts of my mentors.

        Thus, before a call goes out to have editors put in even greater
efforts with poorly written manuscripts, I would prefer to see mentors do
a better job of editing and teaching students how to write in the first
place.  If the mentor can't spend the time to teach a student how to get
out a manuscript (which really _is_ the product of scientific effort),
then s/he might consider if s/he is selecting the right student to bring
into the research effort.  Alternatively, maybe students with poor
writing skills should be required to spend more time taking courses with
significant writing assignments before they start submitting manuscripts
for publication; even if they have completed the officially required
coursework.

cheers,
Jeff Tyler

========================================================
Jeffrey A. Tyler, Ph.D
Great Lakes Center                    ----------------
Buffalo State College                | G = mG+(1-m)g  |
1300 Elmwood Ave.                     -:--------------
Buffalo, NY 14222                     _:______
     716/878-4329                   / o       \/|
FAX: 716/878-6644                   \_________/\|
[log in to unmask]
========================================================