*Concerns of Young Mathematicians*
                       Volume 2,  Issue 33
                         October 19, 1994
                      AMS Election Issue!!!

        An electronically distributed digest for discussions 
        of the issues of concern to mathematicians at the 
        beginning of their careers.


      PLEASE FORWARD TO ANY POTENTIALLY INTERESTED INDIVIDUALS

Please, direct submissions and questions to Frank Arlinghaus
 frank@math.ysu.edu , editor for the month of October.
Next issue: Wednesday, October 26.

Editor for September: Bob Dobrow         dobrow@cam.nist.gov 
Editor for October:   Frank Arlinghaus   frank@math.ysu.edu 
Editor for November:  Frank Sottile      sottile@math.toronto.edu 

To subscribe: Send mail to Charles Yeomans at cyeomans@s.ms.uky.edu
Back issues and other information are available via anonymous FTP to
ftp.ms.uky.edu, in pub3/mailing.lists/ymn-list.

Table of Contents
Item #    Title
------    -----
   1     Editor's notes
   2     Survey
   3     Candidates for Vice President
   4     Candidates for Board of Trustees
   5     Candidates for Member-at-Large
   6	 Candidates for Nominating Committee
   7	 Closing Credits
_______________________________________________________________
Item #1  Editor's notes: 

	This week, we present the results of our AMS candidates 
survey. I've attempted to shorten the newsletter without editing
the candidates responses (including typos), so the survey questions
are printed first, followed by the responses, organized by office.
For biographical information on the candidates, see the September 
issue of "Notices of the AMS," pages 751-761. We did not receive
responses from 13 of the 25 candidates.

YMN does not endorse candidates, but does want to provide you with 
more information on the candidates. In the future, we plan to do 
this earlier, and to include the other societies as much as time
allows.

Special credit goes to Wendy Brunzie, who designed the survey (with
some minimal advice from the other editors), sent it out to the
candidates, and collected the data.

	Our next issue will include information on the YMN Town 
Meeting in San Francisco.
_______________________________________________________________

Item #2 The Survey Questions.

1) Last year's Annual AMS-IMS-MAA Survey of new doctorates showed that of 
the 1202 awarded, 721 obtained domestic employment and about 1/3 of the 
remainder left the country. Do you believe, as these figures might suggest, 
that Ph.D.'s are being overproduced?  Is it your opinion that the AMS can 
and/or should act to reduce the number of Ph.D.'s being produced? What 
specific measures, if any, would you endorse?

2) Do you believe that young mathematicians are getting the support and 
training necessary to become successful professionals? In particular, 
fewer than 60% of new Ph.D.'s obtain academic employment. Should 
graduate education be changed to reflect that? If so, how? If not, why?

3) Would you support an endorsement of flat-rate funding for NSF grants 
as a way to increase the number of awardees, in particular beginning 
investigators? Would you support the creation of an initiative within 
AMS to ask NSF to divert more of its resources to young investigators?

4) The pages of  Concerns of Young Mathematicians are full of ideas for
improving the current system of matching applicant to position. What is 
your opinion of the following:

 	i) AMS-administered matching system for postdocs, patterned after 
the resident-matching system used by the medical profession. 

 	ii) Creating a standard format for jobs on the e-math node and 
asking employers to post periodic updates on the progress of their
searches there.

	iii)The present format of the Employment Register at the Joint 
Meetings and its effectiveness.

5) What is your opinion of this employment situation? Are there other 
methods you would suggest to resolve these difficulties?
 _____________________________________________________________
Item #3 Candidates for Vice-President   

Candidates: C. Herbert Clemens, Joseph B. Keller, Gian-Carlo Rota.

Clemens: did not respond 

Keller:

1), 2)  There is an overproduction of mathematicians trained only with the 
goal of teaching in universities.  I believe training of mathematicians 
should be changed to enable them to get more jobs outside of academia.  
This broadened training should include the study of fields in which there 
are job prospects.

3) The main remedy is to increase funding for mathematics. One way is to 
increase the NSF budget.  Another is to seek support from other government 
agencies.

4) I strongly favor systems for simplifying the process of applying
for positions and selecting applicants.  Systems analogous to that 
used in the medical profession would be some help.  Similarly, a 
standard format for announcing available positions would be helpful.

5) I believe that many mathematicians are trained too narrowly, and this limits 
the field in which they seek jobs.  The field of finance is an example of one 
in which many mathematicians are employed.  The level of some of the work they 
do is very high and the problems are interesting.  There are undoubtedly many 
other such fields in which mathematicians would be used if the employers
were shown that mathematicians could be helpful.
 
Rota:

1) I do believe that Ph.D.'s are being overproduced and their number should be
cut down by about 50%. The only way I see do to this that would not hurt
too much is to jack up the requirements for a Ph.D., that is, to demand
thesis to be more substantial and, say, at least five published papers in
refereed journals are requirements for a Ph.D.
I am opposed to closing down graduate programs at any university.

2) I strongly believe that our present training of Ph.D.s is seriously deficient.
New Ph.D.s should be required to know more mathematics in more fields at a
deeper levels; specialization is inevitable, but we should avoid producing
mere technicians that know next to nothing outside their narrow specialty.
There should be a required "general culture" exam, possibly to be carried
out by a committee made up by mathematicians designated by the AMS.
In addition, Ph.D.s in mathematics should receive extensive training in
mathematical exposition and in explaining mathematics to non-mathematicians.
Every Ph.D. should also be requred to know at least one other science to the
degree of having published a paper (of a mathematical nature, of course)
relating to that science: economics, biology, physics or even philosophy.
We need a complete overhaul of graduate education, tightening of requirements
and strict enforcements of national requirements set by the AMS or some
similar body.

3) This is a purely academic question, since I am convinced that NSF will soon
get out of supporting mathematics, and we will have to learn to survive
on our own teaching, like in the thirties.

4) What is important is to diminish the present power of old boy networks, and
the overwhelming favor granted to mathematicians in some fields over others.
I will support any method of appointing mathematicians that is truly
fair and impartial, and that rewards competence and excellence irrespective
of fields. I am strongly opposed to anything that tends to favor 
"mainstream mathematics", an execrable expression.

5) I think that without decisive action we are headed for a catastrophe in
employment. What can be done? I have two suggestions:

1. Make sure that all mathematics is taught mathematicians. Get rid of
engineers and physicists teaching calculus, probability and what not, 
and replace them by mathematicians.

2. Sad to say, no enterprise today can survive without effective public
relations. Traditionally, mathematicians have been the worst among the
sciences in public relations. We cannot survive without a change in this
attitude.

In closing, I recommend the reading of my note "Ten rules for the survival
of a mathematics department".
_______________________________________________________________
Item #4 Candidates for Board of Trustees: 

Candidates: Frank C. Hoppensteadt, Donald E. McClure

Hoppensteadt: did not respond

McClure:

1) Yes, there currently is an imbalance between production of new Ph.D.s
and the demand for new doctorates.  This is indisputable.

As the author of the report that is cited, however, I have to quibble
a little bit with the interpretations of the data that are referred
to in the question.  What is cause and what is effect?  The fact that
178 doctoral recipients left the U.S. is not evidence of
overproduction of Ph.D.s.  As a percentage of total Ph.D. production,
this is near a historical low.  It is 15% of the total number of
Ph.D.s and 27% of the total number of non-U.S. citizen new Ph.D.s.
Compare the numbers five years earlier in 1988, when most of the 1993
degree recipients were deciding whether to pursue graduate study.  In
1988, 588 new Ph.D.s obtained domestic employment and 177 took
positions in a foreign country, the latter count essentially equal to
the 1993 count. But there were considerably fewer new doctoral
recipients in 1988.  The count of 177 leaving the U.S. represented
21% of the total number of new Ph.D.s and 41% of the non-U.S. citizen
new Ph.D.s.

Thus, if anything, the low count of those taking positions abroad in
1993 is evidence that in the world market for mathematical
scientists, opportunities are better in the U.S (in relative terms)
or pressures are considerably worse in some other parts of the world
than in the U.S.  This is no consolation for an abysmal job market,
but our community does need to keep this in perspective and recognize
the degree to which our own situation is partially affected by
conditions that are totally beyond our control and virtually
impossible to foresee.

The very serious evidence of the difficult job market is in the
numbers of new doctoral recipients who (i) hold positions at the
institution that awarded their degree and (ii) the numbers holding
part-time employment.

The AMS cannot regulate the number of new doctorates being produced.
The AMS has no power or authority over any university, nor should it.
The AMS and all professional associations serving the mathematics
community do have a responsibility to the community, however, to
provide the most timely and best possible information to the
community about factors affecting education and employment of new
Ph.D.s.  Individual departments have the ultimate responsibility to
act responsibly in decisions they make about their own levels of
Ph.D. production and about allocation of their resources to best
serve the needs of that department and its students.  The AMS bears a
responsibility to assure that departments remain cognizant of their
individual responsibility.

Any department producing new Ph.D.s has to seriously examine what
they are doing with their programs and resources.  I personally
believe that many departments should reduce the size of their Ph.D.
program and add a moderate number of new postdoctoral positions and a
small number of junior faculty positions.  This is the type of
resource allocation that is easy to readjust as market forces change.

2) In 1993, 70% of those new Ph.D.s whose employment status was known
held academic positions compared to 82% of the 1988 cohort at the
same time post-Ph.D.  The percentage is down, and this reflects the
terrible condition of the job market.

The straight answer to the question is that graduate programs should
do far more to prepare their students to be versatile in their
careers.  Some simple arithmetic, as well as NRC statistics, show
that even in the best of times around half of mathematical sciences
doctorates will eventually end up in a nonacademic position.  Here is
the argument: there are approximately 18,000 doctoral faculty
positions held by mathematical scientists in 4-year colleges and
universities in the U.S.  In the steady state, assuming a 40-year
career, there will be attrition of about 2% per year due to
retirement or death, a bit fewer than 400 positions per year.  The
doctorate-granting departments have been pumping new Ph.D.s into the
system at the average rate of 850 per year for the last 15 years.  If
there is not substantial growth of the total numbers of doctoral
faculty, then where are the others going? ---To nonacademic
positions.  Indeed, when the "flow" of doctoral faculty was tracked
in the mid-1980s, the studies showed that the net flow out of
academia was approximately equal to the attrition from death and
retirement.

Our Ph.D. programs should recognize this reality.  Major efforts are
currently being carried out by the AMS and SIAM to better inform
current doctoral faculty and students about the characteristics of
nonacademic positions held by mathematicians.  This information is
critical for addressing a cultural problem within our community.  Of
course, the emphasis and highest priority of Ph.D. programs should be
focused on training students to be scholars.  Still, the system
should value what highly trained mathematical scientists bring to
professions outside of academia.

3) If there were such an endorsement by the AMS and if I were an elected
representative, I would support it.

Personally, I do not support the suggestion for flat-rate funding.
There are several issues here, all of which should be amenable to
objective analysis.  I don't think that they have received such an
analysis, and I recommend that that be done.  I think that the
"flat-rate" plan is a rather oversimplified remedy for a complex
problem.  What are some of the complexities?  Here are a couple: the
NSF has several priorities to address, the most central of which is
to fulfill its mission in support of basic research; another is that
it has at least two constituencies to consider, the recipients of its
funding and the providers.  It is not obvious that the best way for
the NSF to fulfill its multiple obligations is to maximize the number
of awardees in mathematical sciences.  A critical problem for our
community is our almost total reliance on one funding agency; this
problem should be addressed.  I wrote at some length on this question
in the AWM Newsletter, and refer interested readers to that response,
in the interest of not revisiting the same question repeatedly.

Related to the suggestion of flat-rate funding is the decoupling of
grants from the salary of the PIs.  That is happening as a corollary
of reducing the amount of effort that is supported in NSF grants.
This reflects a sensible balancing of the NSF's priorities.

I did suggest in my statement in the AWM Newsletter that the
mathematical sciences might use mechanisms such as "research
initiation grants" that are used, for example, in certain engineering
disciplines at NSF, in order to directly address problems of
inadequate funding for some segments of the mathematics community.
Research initiation grants are designed for young investigators.
There are other funding priorities which affect the community broadly
and which might be helped by targeted funding---e.g., easier access
to travel funding for conferences, funding to facilitate
collaboration among colleagues, and research infrastructure needs.

4) i) I don't think the benefits warrant the effort of implementing a
matching system yet.  However, I am warming up to the idea.

There was a very thoughtful and helpful contribution on this issue in
CYM two weeks ago (Leonard Evens, CYM, 9/28/94).

What I think might be effectively implemented is a scheme of setting
up an exchange between doctorate-granting departments of recent new
doctorates who are stuck in a "holding pattern."  There have been
suggestions to this effect recently in CYM.

The medical resident matching system has its problems (see Roth &
Sotomayer, "Two-sided matching," Econometric Society Monographs No.
18, Cambridge University Press, 1990---reference courtesy of L.
Evens).  It also has significant benefits.  For our community, one
factor is that the number of positions affected is very very small.

The most tangible benefit that I see from such a system is for the
new doctorates who may be afforded the opportunity of experiencing a
new academic environment, instead of staying at their home
institution.  That benefit is worth something.  I do not think,
however, that anyone gains anything in terms of simplifying the work
done by candidates in applying for positions or by departments in
reviewing applicants for positions.

ii) Isn't there a fairly standard template now?  Ads should meet
minimal standards such as including a clear statement of
qualifications for the position, being honest about the areas of
priority to the hiring department, and specifying a closing date or
time when applications should be complete.

The suggestion of posting updates on searches is problematic.  It is
likely that we can make some improvements in this area.  The question
should be considered by JCEO and the lawyers (if it hasn't been
already), JCEO to come up with bright ideas on how to grease the
wheels of our employment market and the lawyers to assure that
everyone is protected.  What if a "position filled" announcement were
promulgated too early, before an ironclad written contract were in
place?  If that offer subsequently falls through, then anyone who is
turned away from the position on the basis of the announcement can
certainly feel that their chance for it has been limited by a
premature announcement.  Information about a search needs to be
provided in a very prompt manner by departments to individuals in
their pool of applicants.  Maybe there are some progress
announcements that can be broadcast more generally.

iii) The Employment Register is a very important part of the Annual
Meetings.  It is my feeling that JCEO has made great strides in
improving the mechanism of the Employment Register in the last few
years.  The utility functions of the participants---applicants and
employers---have been objectively modeled and optimized by the new
mechanism.

There is concern about broadening participation by prospective
employers, and I think that this concern is being addressed by JCEO.

5) The situation is tragic for the highly qualified graduates who cannot
find suitable positions.  The difficult market presents dilemmas that
go beyond its impact on the people who are currently underemployed.
In particular, how can we address the problems of
under-representation of some groups in mathematics and assure career
opportunities for graduates at all levels?

To resolve these difficulties, I think that all departments should
carry out a careful self-examination of how they can best use their
resources to fulfill their department's mission and serve their
students.  Specifically, every department should:

(i) ask whether it is doing its part to reduce the number of new
Ph.D.s being produced (admit fewer students, apply high standards,
and advise students honestly);

(ii) redirect resources for a short period of time to increase
postdoctoral support. The goals of the increased postdoctoral support
should be to better prepare students and graduates already in the
stream for their eventual careers, whether academic or nonacademic,
and to give today's new Ph.D.s a real chance for the system to absorb
them into meaningful careers.
 _____________________________________________________________
Item #5 Candidates for Member-at-Large of the Council   

Candidates: Georgia M. Benkart, Carlos Castillo-Chavez, David B. A.
Epstein, Cameron M. Gordon, James M. Hyman, Benjamin A. Lotto, 
Jerrold E. Marsden, Cora Sadosky, Alice Silverberg, Mark W. Winstead

Benkart, Epstein, Hyman: did not respond

Sadosky: sent note through colleague that she was unable to respond 
because her computer was down

Castillo-Chavez:

1) Bad question?  Since nearly 50% of the Ph.D. students are foreigners then
isn't it natural that some (if not all) might leave the country?  How many
of those 500 were foreigners?

2) If this question is based on (1) then it is also a bad question.
Are mathematicians getting the training to function outside academia?
In general...NO.  Should they?  Depends on what they want to do with their
degree!

3) Wouldn't it be better to have one question per question??????????
I support a flat rate for ALL mathematicians.
I can't suppor initiatives without specifics.
However, it is clear to me that more funds should go to
SOME but NOT ALL young researchers.

4) i) Bad idea. The hiring  is done by employers.  The old boy network
controls the job decision making process.  Creating mechanisms with no real
teeth is a waste of time and resources.
ii) OK.
iii) I have not filled one in about 7 years.  So I can't comment.

5) Which employment situation?  You need to be more specific?
The end of the cold war has reduced the number of jobs and of
funds available for mathematical research.   There are new efforts to
increase the size of the pie by doing the SAME mathematics but with
different names.   The SUDDEN growth of interest in INDUSTRIAL mathematics
is a but a good example.  The fact that not every Ph.D. is getting a job in
academia and that some foreigners are going back to their countries is not
necessarily a  bad thing. 

A glance at other areas where employment is highly competitive such as
ecology will make our situation pale.

The biggest problem is that we train young people just to do academic
research (and sometimes not even that).  Some will be very successful,
others will do OK, while others will not become decent researchers at all. 
There are 3,000 USA univeristies that have jobs and they have jobs for all
these types.  If some of them learn other skills such as statistics (and do
statistical consulting) then could also get a high paying job outside
academia.  The students should be informed of the fact that getting an
academic job is not automatic and that in general is a good idea to
complete a serious masters or minor in an applied field.  The ffort has to
be SERIOUS and not just take a few courses.

Gordon:

(1) Ph.D.'s in mathematics in the US are clearly being overproduced at the 
moment, and have been for a few years. The unemployment rate among new Ph.D.'s
has been rising, and even of those finding academic employment, the level of 
the institutions at which they are taking jobs has been falling. These are the
facts of the current marketplace. We all know (with hindsight) the various
reasons for this state of affairs. What can the AMS do about it? Not much. It
can (as it has been doing) collect and make available data on Ph.D. production
and employment, so that everyone can find out what the facts are. It should, 
however, be wary of making predictions based on this data. (The current high
number of Ph.D.'s in mathematics is partly the result of predictions by the
AMS of shortages of Ph.D.'s in the 1990's, predictions that were convenient at
the time in making arguments for increased support of mathematics.) The 
situation would be helped if university departments were to reduce the numbers
of students entering their Ph.D. programs, and the AMS can certainly encourage
this. It can't create jobs, at least directly.

(2) Actually, 68% of the 1992-93 Ph.D.'s whose employment status was known took
academic positions (AMS Notices, July/August 1994); the previous year it was 
78%. Clearly non-academic employment is an option that more and more math 
Ph.D.'s are having to consider at the moment, and math departments are 
typically not well-equipped to provide information on this market. I think this
is somewhere that the AMS could help, in creating more contact between the 
academic community and the non-academic mathematical research community. As far
as changing graduate education is concerned, I think we will see a move towards
more applied math, for example. But basically the current problem is not that
students are not getting the training they want, it's that they're not getting
the jobs they want.

(3) Yes. Of course, the catch here is "as a way to increase the number of 
awardees". The NSF summer grants are already in practice approximating more and
more to flat-rate funding, and the numbers funded are not increasing. And it is
very difficult for a young investigator to "break into" the system. A high
priority for the AMS should be to lobby the NSF for more funds for mathematics,
and for the summer grant program and young investigators in particular. We all
know that mathematics is inexpensive, and it just doesn't make sense to save a 
relatively small amount of money by not supporting talented and dedicated 
people who have important contributions to make. 

(4) (ii) would be helpful, and (iii) could probably be improved. I'm not really
in favor of (i). Certainly one wants to make the job-seeking process as 
efficient as possible, but the main problem is that there are more Ph.D.'s than
jobs.

(5) The current employment situation for math Ph.D.'s is bleak. For the past 
few years there has been a gap between expectations and jobs (expectations
which were reasonable on entering the Ph.D. pipeline), although presumably
expectations have now caught "down". It will be a while before the numbers 
match again, because of Ph.D.'s still in the pipeline and young postdocs on
short term appointments. However, I believe the situation will improve (a
prediction!), although of course this is little consolation to those currently
un- or under-employed. There will be a lot of retirements in the near future,
and although not all the vacated positions will be filled, some will be. One
thing that the AMS could do to try to minimize the adverse effects of 
fluctuations in the market is to campaign for more NSF postdocs in math; the
number is pathetically small at the moment. But in general, the AMS should be
aggressively leading the mathematical community in carrying to the political
and academic establishments, and the public at large, the message of the 
importance of mathematics, to ensure the discipline's fair share of funding,
and to ensure that math departments are not regarded as second-class citizens
by university administrations.

Lotto:

1) It does seem that math PhDs are being overproduced, at least for the market
that currently employs them.  One solution is to reduce the number of PhDs
that are produced.  Some schools are already taking this approach without
prompting from an external source like the AMS.  For example, UC Berkeley
is planning a first year class of 25 for this fall, which is significantly
smaller than, say, 10 years ago, when the first year class had on the order
of 70 students.  I don't think the AMS should take specific measures to
reduce the number of PhDs produced, since I believe that natural forces
will take care of that (at least, data from the last bad job market of the
70s suggests that this is the case).  In addition, reducing the number of
entering students won't have any effect on the number of PhDs graduating in
the next several years.  What the AMS can and should do, in my opinion, is
focus on expanding the employment possibilities for those people who do
graduate with PhDs.  More about this in the answer to the next question.

2) It is absolutely true that graduate students need to be made aware of
employment possibilities other than academia and that graduate programs
need to offer training and guidance for students that choose to seek a
nonacademic career.  The first step towards this would be a massive
consciousness raising effort on the part of the whole profession to learn
what opportunities there are for PhD mathematicians outside of academia.
Potential nonacademic employers could be invited to AMS meetings to give
talks about their employment needs.  Ties between academic institutions and
industry could be encouraged.  Perhaps some sort of intership programs
could be established, with graduate students working in industry during
summers.
 
3) I am in favor of increasing the number of people who receive support from
the NSF, and especially increasing the percentage of young mathematicians
who are so supported.  There are several ways this could be done.  One is
some sort of flat-rate scheme.  Another is a salary cap on summer salaries.
A third is to insist that senior mathematicians include funds for
postdoctoral positions in their grant proposals (perhaps limiting summer
salary amounts for those who don't support postdocs).  All of these
possibilities would have the desired effect of increasing support for
worthy young mathematicians.

4) I'm not really sure how I feel about a matching system, since I very little
about the way things are done in the medical world and alternatives.  It
does seem to me that there is a great deal of inefficiency in the way
employers and applicants find each other, so I would be in favor of looking
at this system to determine whether it would be useful in the mathematical
world.  On the second issue, I strongly believe that the use of standard
formats for job advertisements and applications would make the hiring
process a lot more efficient for both employers and applicants.  On the
third issue, I believe that the Employment Register has some positive
features, but with the declining number of employers and the increasing
number of applicants taking part in the Register, it is becoming less and
less useful (although it still gives applicants a chance to look at
potential employers that they might not otherwise see and employers,
especially small, less well known places, a chance to promote themselves to
candidates who might not otherwise have been interested).

5) I strongly believe that the long term effects of the current employment
situation could be devistating to the mathematical profession.  My biggest
fear is that young mathematicians with a tremendous amount of potential
will either leave the profession or become so discouraged with finding new
temporary jobs every year that their talents will be wasted.  We have to
find ways of using this talent until the job market improves.  There are
many suggestions that I and others have made.  These suggestions need to be
seriously examined and implemented in a timely fashion so that the damage
that this job market is causing to us and our profession is minimized.

Marsden:

1) Before drawing any conclusions about these numbers, I would certainly
gather and examine corresponding data from earlier years to see what the
corresponding figures were. The figures need to also be more detailed:
how many PhD's got a job they were satisfied with? Is  this number 
deteriorating significantly. Certainly the community's impression is that things
are much worse. The AMS cannot dictate any quota's to Universities, so
one probably has to settle for some advice and discussion at, for example
the annual meeting for chairs, and a report from a committee that studies the
problem. One has to also keep in mind that there are economic pressures which 
are already cutting PhD production; for example at Berkeley, the number of PhD 
graduate candidates is already gone down significantly and the primary reason is
economic together with some smaller concious decision that the job market does 
not justify so many PhD's.

2) I am convinced that our graduate education system needs some improvement.
I  believe that most graduate programs do not adequately  train students in how 
to use mathematics. This is a cultural and inherited phenomenon that will be 
very hard to change and it will take time, but I think we should try.  Some of 
this, such as the use of computers is naturally changing, but many other 
aspects, even a cultural knowledge of how mathematics is used in engineering, 
physics, chemistry, finance, etc needs to be taught as well. Mathematicians at 
the PhD really are useful in industry, and econimic reality is that many of them
will end up there --- we have to convince mathematicians that they need to be 
trained accordingly. The AMS can play a role in this by continually studying the
problem and to continue to properly advise the mathematics community about it 
with specific advice about how to change programs. The recent national policy 
statements by the AMS are a good first step, but now this needs to be put into 
concrete action.

3) Exactly these points have had ongoing discussion within the AMS and within the 
NSF for many years already. There are of course lots of conflicting points of 
view and the situation is truly complex. For example one has to be careful not 
to shoot oneself in the foot: one reason flat rate initiatives were not enacted 
in the past was basically because if mathematics does it alone within NSF, then 
people in other sciences who strongly oppose such strategies will perceive this 
as " mathematicians do not need so much money after all" (this is actual 
experience talking here) and we are in serious danger of having our budget cut. 
The proverbial "pie" to cut up does not exist; the size of the pie in this case 
is dependent on how it is cut! On the other hand, through further 
interdisciplinary and cooperative efforts with the other sciences (educating 
both them and us), I believe that some of the goals mentioned can be achieved; 
we just have to proceed cautiously.  Certainly the concerns are legitimate ones 
and we need to work towards the goal of supporting more younger people. 

4), 5)   I think that the best way for younger
people to handle the employment situation is to understand how it works
and for the profession and the AMS to deal with this reality and to try to 
fix the parts of it that are judget to be bad. What are the realities?
It is a myth that Universities possess SPECIFIC qualifications they require for 
applicants
(eg, skills required in homological algebra, coupled with a good knowledge
of Maple for teaching calculus on computers) and that the best qualified
applicant with these skills will get the job. Certainly at research Universities
research
counts most (the person perceived to be the most gifted researcher -- and this 
may depend
on the judges -- may get the job) and at teaching Universities teaching counts 
most (perhaps the one with the highest teaching evaluations), with many
Universities falling in the middle of these extremes. In industry, matching may 
be even more difficult because of the lack of specific training in the area of 
concern to the industry.
In many Universities the hiring committee gets many hundreds of applications for
each job and one cannot reasonably expect any hiring committee to systematically
go through this and objectively pick out the most qualified candidate. 
Given this reality, good advice is for the applicant to get to know faculty 
members
at as many Universities as they can. Going to the AMS and SIAM meetings (and ASK
to speak in special sessions!) can be one good strategy. If faculty know you 
they are more
likely to bring your application out of the pile of hundreds to the attention of
the
hiring committee.  

The very act of meeting potential employers, trying to get invited to lecture 
(in a special session, or to give a contributed talk) and encouraging potential 
employers  to hear your lecture might be  the most important thing to do, at 
least if one is going for a reasearch type of University. I would certainly look
for ways that the matching and contact procedures can be stramlined and 
improved, but I would also encourage perhaps even with higher priority, 
educating students about the process and how it really works. 

Silverberg:

  I found your questions interesting and thought-provoking.
I'll try to address most of your questions, either directly
or indirectly. I can't answer all of them because the job
market problem is a complicated one, and I don't have all
the answers. I'm willing to listen, and my views will evolve.
  First, I think that a mathematics Ph.D. is a wonderful
thing to have. With a background in mathematics, you can do
anything; you don't have to be a professor. 
  Having an advanced degree is becoming a requirement in more 
and more professions. In our increasingly high-tech world,
there's going to be nothing better than having a math degree.
One thing we must change is the idea that people with Ph.D.'s 
are failures if they don't become professors.
  That said, there are many things that can be done to improve
conditions for those who decide on academic careers.
  Accurate information on the state of the job market should
be widely publicized so people can make informed choices about
their careers and their graduate educations.
  Universities and thesis advisors should take greater
responsibility for advising and mentoring current and former
students.
  I support flat-rate funding for NSF grants as a way to
increase the number of awardees.
  I've always thought that a matching system for postdocs,
modeled after the system used by the medical profession, is
a great idea, and I think the AMS should initiate and oversee
such a project. This is a case where we shouldn't be re-inventing
the wheel -- we can use solutions other professions have come
up with.
  Other countries and other academic disciplines have been
facing these problems for many more years than has the U.S.
mathematical community. We can look to them for some creative
ways to solve these problems.

Winstead:

1) I think there is little doubt that there is an overproduction problem.
An AMS task force concluded that there is no immmediate hope of the 
situation changing, at least on the job supply front.  While there is 
little the AMS can realistically expect to do to increase real, permanent 
(e.g. tenure-track) positions, we can take some small steps to reduce the 
production, or at least cap production.  I propose that all mathematical 
science departments actively track their PhDs for ten years after 
graduation, and supply applicants with the results.  This will allow 
applicants to make informed choices, and, I believe, would lead to 
graduate students with more realistic expectations.  It would probably 
also lead to a reduction in PhD production in the long run.
	The AMS should continue to implement the recommendations of the 
task force I previously mentioned.  Their report is available on the 
e-math gopher.

2) The situation is worse than you describe.  While more than half the 
new Ph.D's get academic jobs, how many of those are like what they have 
previously been exposed to?  Many positions are at liberal arts colleges 
or other four year/teaching emphasis colleges.  How much preparation do 
these PhDs receive?  A lot of schools advertise for people trained in the 
use of computers in the classroom.  How many graduate students receive 
even passing exposure to computer based education?  And how many PhDs are 
prepared to teach a broad variety of courses, as they might have to if 
they are in a department of less than 10?

The AMS should encourage each department awarding PhDs to make an honest 
self examination of their programs, with emphasis on looking on where 
their recent graduates have ended up.  The AMS should help create and 
promote programs which help departments more realistically prepare graduate 
students for what awaits them.  Examples:  sabbatical programs for 
faculty in industry (so faculty can become better advisors), encouraging 
faculty to do leaves at teaching colleges, summer employment programs in 
industry, etc.  While the work to be done is at the department level, the 
AMS can act to create the necessary resources to get the job done.

3) YES! Also, I think we should work to make more funds available for 
conferences and travel, especially for junior mathematicians.

4) i) There are big differences between mathematics and medicine, and between 
the AMS and the AMA.  Len Evens recently taught me more about how the
medical matching program came about and works.  The more I study the
situation, the more I doubt that we should be looking to the
resident-matching system as a model.  This is a mathematical society
problem, not a medical society one.

Now, I do believe that we should experiment with some sort of matching
program. Of the 1993 graduates, more than a third of those employed at 
Ph.D producing departments were at the department awarding their degrees.
I think we could explore the feasibility of a matching program by 
starting with departments which are employing their own graduates in this 
way.  Each year the program is successful, it could be expanded the 
next. 
ii) Sounds good to me! It should be mentioned that YMN offered to print 
status reports for schools. As far as I remember, no one took us up on the 
offer.
iii) With the various different ways all the individual employers handle or 
wish to handle their searches, I don't know how we could make everyone happy.
One way to experiment is to have job seekers and potential job seekers 
submit a resume for a book to be published in October, then modify the 
format at the Joint Meetings to give schools more control over who they 
interview (a book of late additions could be sent out in December).
This may create more contacts between employers and potential employees.

5) In short, the situation is bad.  I have been living it, so I know.  I 
proposed starting and cofounded YMN as a way to search for ways to 
resolve these difficulties, or at least alleviate the pain.  Certainly, 
others at this point are going to propose the creation of postdocs, 
industrial and/or academic, but all this really does is put a bandaid on 
the problem. It all boils down to reducing Ph.D production and/or 
creating jobs.  I don't believe departments are going to cut back, since 
it hurts them to do so in many ways.  We can be honest with graduate 
students and graduate school applicants, and in doing so , we may see a 
reduction in production.
   To create jobs, 1) the AMS should work to improve its efforts to lobby 
the government at the local, state and federal level better, in order to 
preserve (and create, if possible) public sector jobs in mathematical 
research and education, 2) the AMS can and should do a better job of 
selling the value of mathematics to the general public in order to increase 
public support and to preserve and create private sector jobs.

A separate issue is what to do about the schools who year after year look
for a tenure track position, and, not liking their options, hire someone
to cover the teaching load for one year so they can try again the next. 
These schools are looking for super teachers and researchers, but do
nothing to help develop the talent pool.  When they hire someone for a one
year position, that person ends devoting a great deal of time looking for
their next position, and as a result does not have sufficient time to do
research or to dedicate themselves to improving their teaching.  The
situation would become a lot more tolerable if more mathematical
sciences departments would act less selfishly.  I don't know what more the
AMS can do to discourage one year positions; I am open to suggestions.
_____________________________________________________________
Item #6 Candidates for Nominating Committee   

Candidates: Jerry L. Bona, Ingrid Daubechies, William James Lewis,
Rogers Newman, Stephen D. Smith, Susan G. Williams.

Bona, Daubechies, Newman, Williams: did not respond

Lewis:

1)  I cannot support the idea that Ph.D.'s are being overproduced. 
Most of us study mathematics because we enjoy mathematics and we would
not want to have that opportunity denied. At the same time, those of 
us with some opportunity to influence the way new Ph.D.'s are educated
and/or to influence the job market must be very sensitive to the fact 
that there is an inadequate supply of jobs for these very same new 
Ph.D.'s who love mathematics and want the opportunity to build their 
career around mathematics. Thus, both the AMS and individual 
departments can take steps to improve the job market.

Among the ideas that I support are:
     a) efforts by the AMS to support new posdoc programs funded in 
part by federal funds;
     b) efforts by departments such as mine to increase the number of 
positions (postdoctoral or permanent) for new Ph.D.'s and perhaps at 
the same time, reducing the number of funded GTA positions; 
     c) efforts by the AMS to assist departments in forming links with
industry and research labs, leading to new employment opportunities 
for mathematicians and statisticians;
     d) efforts by departments to increase the employment 
opportunities for their graduates by altering the education received 
by their graduates. 

2) Graduate education should change. Some, but not all, departments 
should make it a priority to prepare their graduates for successful 
careers in industry and for careers in institutions where teaching is 
the primary responsibility.

3) I am led to believe that current NSF policy already makes some 
effort to increase the number of awardees, especially beginning 
investigators. Current practice often restricts the size of grants, 
especially the amount of summer support for the majority of senior 
investigators, while offering full summer support to beginning 
investigators. 

     Expressed just as it is in the question above, I would not 
support an initiative within AMS to ask NSF to divert more of its 
resources to young investigators. NSF must keep as its clear priority 
the support of the best mathematics with an understanding that part of
that effort must be to insure the renewal of the discipline. This 
latter role justifies support for graduate students, postdocs and 
young investigators. How NSF determines the proper mix of its 
different funding strategies should be based on careful thought and 
at present I have no information that would lead me to criticize the 
current balance.

4) i) The idea intrigues me both as a member of an AMS Policy Committee
and as a department chair.
ii) While possibly helpful administratively, this does nothing to 
increase the number of jobs. I would support this.
iii)  I know that more than one committee has worried about this issue 
and has sought to improve the Employment Register. I really don't have
any new ideas to offer at this time.

5) What interests me most is urging mathematicians to fight to 
preserve full-time positions at their institutions. 

Smith:  

1) Yes too many.  Yes reduce number.  
Method always will boil down to supervisor judgment--is this
person strong enough to command a job in the ACTUAL pool.
Unfortunately, supervisor vanity will typically overcome judgment.

2) For research, they are getting the training and support in
their education.  It is the overproduction for research 
that is the problem.
Certainly graduate education should ALSO emphasize the
preponderance of teaching-based positions, and industrial jobs.
How? realism in courses/programs offered.

3) Absolutely.  But of course it has to apply to EVERYBODY's salary.
There will DEFINITELY be opposition from some of the higher-paid.
Note that since my support is from NSA, I am already in effect
participating in a variant---namely 1-month summer support
as opposed to 2 months.

4) I support i) and ii).  I have found iii) an ineffective
compromise.

5) More realism about CYCLES in the academic business.  It is
always hard to get people to believe this at the time,
no matter how many past cycles can be exhibited.  (Compare
Texas and the oil business, in boom times no one ever
considers the possiblity of the coming bust.  Or California,
a recession could never happen there, right?)
 ________________________________________________________________
Item #7 Closing Credits

The Young Mathematicians' Network is administered by:

Charles Yeomans          cyeomans@s.ms.uky.edu 
Mark Winstead            mwwinst@gcr.com 
Frank Sottile            sottile@math.toronto.edu 
Vic Perera               vperera@silver.ucs.indiana.edu 
Franklin Mendivil        mendivil@math.gatech.edu 
Kevin Madigan            madigan@math.nwu.edu 
Steve Kennedy            skennedy@mathcs.carleton.edu 
Matt Hudelson            hudelson@math.washington.edu 
Bob Dobrow               dobrow@cam.nist.gov 
Lyle Cochran             address change pending 
Neil Calkin              calkin@math.gatech.edu 
Wendy Brunzie            brunzie@mathfs.math.montana.edu 
Curtis Bennett           cbennet@andy.bgsu.edu 
Frank Arlinghaus         frank@math.ysu.edu 
Jeff Adams               adams@bright.uoregon.edu 
Edward Aboufadel         aboufade@scus1.ctstateu.edu 
_______________________________________________________________
End of Journal -- Next week:  The Discussion Continues