LINGUIST List 2.888

Mon 30 Dec 1991

All: LINGUIST reforms

Editor for this issue: <>


  1. The LINGUIST moderators, LINGUIST reforms: response to your suggestions

Message 1: LINGUIST reforms: response to your suggestions

Date: Mon, 30 Dec 91 13:23:44 EST
From: The LINGUIST moderators <>
Subject: LINGUIST reforms: response to your suggestions
Some months ago, in response to several serious and
semi-serious complaints about the volume of mail on
LINGUIST, we asked for your suggestions. We wanted to
know whether there was anything we could do to help with
your problem--and ours. We received a fair number of
responses, our favorite being "Don't start a discussion
ABOUT the net--it will only add to the mail."
That respondent was, of course, pointing toward the
ever-present paradox in all this, the paradox we hope
will excuse this tardy response to the suggestions: the
volume of mail has been such that we have only had time
to deal with the mail, not to deal with THE PROBLEM OF
the mail in any long-term way. For that reason too,
some of the excellent suggestions below will have to be
left until we have more time to implement them. Next
semester we'll be writing grant proposals in the hope of
securing the time, money, and staff we need to improve
LINGUIST; and those proposals will certainly benefit
from your suggestions. But, for now, we must ask you to
be content with the minor changes listed below.
We'd like to thank those who replied: Vicki Fromkin,
Ted Hansen, Lee Hartman, Peter Cole, David Powers, Alice
Harris, Jacques Guy, Ronnie Wilbur, Jon Aske, Derk
Ederveen, Mimi Klaiman, Charles Laughlin, Ron Kuzar,
Claudia Brugman, Elise Morse-Gagne, Sergio Scalise, Bert
Peeters, Philip Swann, Robert Shull, Michel Grimaud,
Nancy Belmore, Paul Chapin, Mark Johnson, Alice Freed,
Steve Harlow, Jane Edwards, Natalie Maynor, Stavros
Macrakis, Ron Hofman, Karen Jensen, Sally Thomason,
Jerry Domnick, Gorka Elordieta, Louise McNally, Rob
Stainton, Stephen Spackman and George Huttar.
Their suggestions and our responses are summarized
 a. Into two lists, one for jobs and conferences,
and one for everything else. (8 replies: 5 for, 3
 b. Into multiple lists, one for each subdiscipline
of linguistics. (14 replies: 2 for, 12 against)
The idea of splitting LINGUIST into a job/conference
list and a discussion list has an intuitive appeal in
that it would allow subscribers to focus only on what
interests them. Job-postings, certainly, are typically
of interest only to those seeking a job! But we do have
some qualms about this, since other list-owners have
told us that lists specifically aimed at job-postings
usually flounder, because those who subscribe to them
have no jobs to offer, and those who have jobs don't
subscribe to the job-lists. Unless there is some
structure which compels employers to post to the job
list, the job-list gradually fades away. We have
anecdotal evidence that job-postings on LINGUIST have
resulted in fairly dramatic increases in number of
applicants, especially in areas outside North America;
and we would be reluctant to see this LINGUIST service
impaired. However, our main problem with this idea
right now is that we don't have the time to set up a
separate list. So this is one of those potentially
valuable ideas which has been tabled for further
Splitting LINGUIST into subdisciplinary lists: we're
glad that few subscribers seem to want this, because it
runs counter to one of our main objectives in starting
LINGUIST. We felt that, as academics, most aspects of
our jobs push us toward more and more specialization.
We had hoped that, by contrast, LINGUIST would offer an
easy way to keep up with the field as a whole by
"listening in on" discussions that colleagues in other
subdisciplines find interesting. A comment made by
Elise Morse-Gagne perhaps says this best: "To me the
single most valuable result of the fact that LINGUIST is
a single list, not separated into a few to
show us that outside our areas we are all non
Most subscribers want subject headings to be transparent
and each LINGUIST issue to include only a single topic. We
have been trying to do this, and we will try harder in
future--though you'd be surprised how hard it is
sometimes to come up with a "transparent" one-word
heading for something like "Case-marked subjects of non
finite clauses in Amharic"! Transparent headings on
your submissions will help us a lot, so we'd like to ask
you not to simply hit "reply" when your message actually
changes the original topic.
About combining topics in an issue: in response to your
suggestion, we'll do less of this. It generally happens
only when a topic is winding down, so that there aren't
enough responses to fill a 200-line issue.
 a. When a query is made, those responding should
direct replies to the inquirer alone, who should then
summarize the responses for the list (6 for).
 b. Discard postings which largely duplicate others
already posted (4 for)
Discarding duplications seems like a very good idea, and
it's one we will attempt to implement in future.
Responding directly to the inquirer is also a very good
idea in many situations. However, interesting
discussions have sometimes been generated by queries and
their responses, and it is important for the dynamics of
a list like LINGUIST that discussion be carried out
publicly under some circumstances. As always, what is
most required is discretion on the part of all of us.
If an inquiry is made about what books are appropriate
for a course in language classification, this is best
dealt with privately, and a summary posted to the list.
But if the inquiry concerns the methodology of language
classification or a theoretical innovation, this may be
better dealt with publicly.
However, it is always a good idea for the inquirer to
provide a summary, whether the discussion was public or
Two subscribers expressed concern that, in an attempt to
decrease volume, the moderators may exercise too heavy a
hand. We can only say that we agree with them that
it is better to let individual subscribers discard
unwanted messages, and that, if err we must, we
we should do so on the side of too much mail than
too little.
Most of you are probably familiar with USENET lists,
where interaction between subscribers and board is
somewhat different than with a Listserv list like
LINGUIST. Instead of subscribing directly to a list,
and then receiving ordinary mail from that list,
interested users run a news-reading program on their
host computers, and select from the (literally hundreds)
of bulletin boards those they wish to read. These news
reading programs provide sophisticated means of reading
individual postings, and of excluding categories of
postings a user does not wish to see. Since the
messages they read are stored on some designated host
computer at their site, and this host computer handles
USENET feeds for all users, individuals need never worry
about overflowing mail-files. If they fail to read a
bulletin board, the old mail will eventually quietly
Now, USENET is an immensely useful way of getting large
quantities of information, but it has some drawbacks.
First, many sites simply do not take USENET feeds, for
these require considerable computational facilities.
Second, on USENET it is difficult to set up a list which
is well focussed on some field or discipline, because
interacting with USENET lists is easy and open to
everyone. People who are only peripherally interested
in a field commonly read USENET lists and post messages
to them. And thus USENET lists tend to be a mixture of
highly worthwhile postings and irrelevance. Sci.lang,
for some subscribers at least, exemplifies this
We, as moderators, fear that if LINGUIST
joined USENET, the volume of mail would increase
dramatically, and that much of it would be of such a
nature as to require considerable editing by us. If
this occurred with our present facilities, we would be
overwhelmed. However, it is worth pointing out
that the news-reading programs which are the
strength of USENET lists are equally available
to subscribers to Listserv lists, provided that
these lists follow a standard format, as LINGUIST
does. One solution for subscribers to LINGUIST, then,
is to use such programs on our mailings, as some
subscribers already do.
It seems, then, that we can extract the following
principles from the above discussion, and these we will
attempt to follow:
1. LINGUIST will stay one list.
2. The moderators of LINGUIST will make every effort to
make subject headings clear, and to categorize messages
consistently by topic.
3. The subscribers to LINGUIST should post replies
directly to those who make queries, except where it is
felt that the reply is of general interest.
4. Inquirers should attempt to provide summaries of
discussions or replies to queries, especially where the
discussions were private.
5. The moderators will attempt to exclude postings
which duplicate information already posted.
6. LINGUIST will not for the moment join USENET.
Thank you again for your suggestions and your interest
in the list. Thanks, too, to all of you who have sent
us holiday greetings--we hope to meet some of you at the
LSA symposium ("Computers and the Working Linguist,"
Linguistic Society of America, Jan 9). Please introduce
yourselves! ~
And may you all have a peaceful and happy new year.
--Anthony & Helen
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue