Reply to Patrick O'Neil

Alexander Berezin berezin at MCMAIL.CIS.MCMASTER.CA
Fri Jun 9 09:17:54 EST 1995

Please see comments below Dr. O'Neil's
text - Alex Berezin

On Wed, 7 Jun 1995, Patrick O'Neil wrote:

> If money were soooo freely available and plentiful, then an investigator 
> would not have to spend so much time seeking out funding.  Your logic is 
> flawed.  I have yet to see a PI who is not worried at some point about 
> the necessity to obtain more funding because their latest grant is 
> running out.  Without the constant digging and writing, their research 
> would come to a halt for lack of graduate researchers and lack of 
> chemicals and equipment and software.  No chemicals, no research.  No 
> equipment, no research.  No software, no research.  No graduate students, 
> no research.  Postdocs are accepted only because they bring their own 
> funding and do not tap into the limited fund pool for the lab in 
> question.  Oh yeah, LOTS of easy money out why do so many 
> sweat blood every year or two in order to simply continue their work?
> > 
> > In short, it is time for us to learn how to do
> > much better quality science with CONSIDERABLY LESS
> > MONEY. 
> Perhaps this is true for SOME types of research, but certainly not all. 
> If this is the position you take, I could almost suspect that you are a
> Luddite Jeremy Rifkin type who sees budget cuts as a means of stopping
> biotech research.  Treatments and cures for diseases such as AIDs, TB,
> cancer, etc, are not going to come from scaled-back research on the cheap. 
> It take large amounts of time and a large number of interconnected
> researchers with the best equipment, or no progress will be made. 
> The other conclusion is that you could simply be a disgruntled researcher 
> who's research is presently turned down for funding.  With limited funds, 
> there HAS to be a cutoff whereby loser proposals or even good proposals, 
> nevertheless, have to be turned down.  It there was all this money 
> available that you speak of, then this wouldn't happen and just about any 

> two-bit research proposal would be funded.  There are priorities because 
> some directions of research are more critical than others in an 
> atmosphere of shortage of funds.

I am afraid that your last sentence defeats most of what you are
saying -... "some directions are more critical than others"... 
- who (and how) is going to define it ?

It is all a matter of opinion, with all the biases of subjective
judgement. And this is NOT alleviated but a peer review, but 
rather aggravated by it (mostly,  but not exclusively, due to
the anonymity and the inherent conflict of interest in the
current peer review system). 

Peer review historically showed that it is completely unable
to predict any major (or even minor) developments. The only
practical solution to it is to "hedge your bets strategy".
This does not mean that "all should get equal funding" and
some expert ranking of the researchers (NOT "proposals" !)
is, of course, appropriate. But the bottomline is that
significantly more equitable and less stressful funding
distribution system is long overdue.

  I append another earlier article which, along with
general points, describes the situation
in Canadian NSERC. There are some similarities between 
NSERC and NIH systems, but one very important difference is 
that NSERC does not have a multiple grant system (only one
operating grant per person at any time). So those who
are not funded (ca. 30 - 40 %) are not funded at any level.

Alexander A. Berezin  and  Geoffrey Hunter
(published in "Canadian Chemical News", March, 1994) 

   A widely held misconception about science it that its
quality can greatly benefit from the so called "competition
for excellence" which is externally "coordinated" by funding
agencies. Scientific and engineering research in canadian
universities is supported almost exclusively through the Natural 
Sciences and Engineering Research Council (NSERC). The basis for
the present NSERC funding philosophy and practice is the idea of
"selectivity", i.e. the policy of NOT funding all the applicants.
This is done in the name of the alleged "excellence" of research
and its "competitiveness". 

   This is reflected both in the adopted NSERC terminology (terms
like "grant SELECTION committee", "next COMPETITION") and the
explicit instructions (!) to committees to recommend a
significant fraction of applicants for non-awards ("NIL" awards).
Notwithstanding the best intentions of its designers, the present
NSERC funding system leads to a highly detrimental
effect: instead of being IDEA AND OPPORTUNITY DRIVEN which is the
true path to excellence (1), the research is GRANT DRIVEN,
GRANT LIMITED and GRANT SEEKING. The only real concern of any
applicant to NSERC is how to optimize all his/her research along
a single (!) criterium : fundability.

   The net result of this system is that truly innovative
research is often suppressed by the censorship of the peer-review
process (2). The present NSERC policies encourage prolific
productivity of routine (but easily publishable) results along
well established mainstream research directions. Peer-reviewers
are invariably drawn from the scientific establishement. They
will be supportive of the established (their) projects rather
than truly innovative projects; innovative projects are by
definition not established (3). How supportive was the scientific
establishment when Boltzmann presented statistical mechanics ? 
   Stressing the very idea of "competition" is based on the
illegitimate transfer of a business model to science. This is a
case when a model is used beyond its actual range of validity.
The so called "competition for excellence" has long ago passed
all reasonable limits needed for a healthy stimulation and 
turned into a ferocious rat-race and Darwinian fight for survival
based on a principle of confirmity to the mainstream. Often
really innovative research can only be maintained by its careful
concealment behind the mainstream facade. This dilemma is
especially acute for many interdisciplinary studies and for the 
research which challenges the accepted paradigmas and the
established dogmas.

   While it is undeniable that many NSERC-supported projects are
of a very high calibre, they became so largely IN SPITE of the
system rather than because of it. Their continuing support does
not offset the highly damaging implications of "NIL-awards" for
the morale and research ethics of the entire university
community. Also, a NIL-award to a researcher has a devastating
effect on his/her graduate students, many of whom will consider
dropping a research career altogether. This means a potential
loss of the most valuable of all resources - a human talent.

   In terms of canadian research output and international
competitiveness, NIL awards to active researchers leave a
significant fraction of highly (and often uniquely !) trained
scientists FUNCTIONALLY UNEMPLOYED (even though they are paid
salaries TO DO RESEARCH !). Our (very concervative) estimate is
that at least a thousand (!) full-time faculty members in science
and engineering departments in canadian universities have no
external funding whatsoever. On the other hand, a significant
number of well established mainstream research groups
("departmental empires"), often with little real innovation, are
clearly OVERfunded. Furthermore, the overselling the notion
of research "underfunding" is in the interest of such
super-groups experienced in the game of Grantsmenship.

   It is a very common stand for almost any group, including the
research community, to attribute all their problems to the
underfunding. "Just give us more money and everything will be
OK". It is always easier to blame somebody else than to look
inside your own household - this is the reason why the
underfunding mythology is so universally attractive and popular.
However, despite that Canada indeed falls behind some other
developed contries in terms of its total R & D expenedure, the
crux of the problem is NOT so much in the bulk underfunding as in
the MISMANAGEMENT of the available resources. Contrary to what
may seem obvouis, under the present funding system "more money"
from the government (even if lobbing for extra funding will
succeed !) will EXACERBATE rather than solve the problem, as
almost all gains of new public money will go to the already well
funded groups and NOT to NIL-funded researchers. This is a well
known "Matthew effect" in science "give-to-those-who- have-and-
take-from-those-who-haven't") (4). 

   We believe that the real roots of major flaws of present NSERC
system lie in its UNDEMOCRATIC nature. Presently the membership
renewal in committees is NOT discussed publicly and no electorial
process is in place. Instead, we have an oligarchic system in
which "committees are simply designated by previuos committees".
Apart from some scattered letters in the public press from
individual researchers, there is no sound democratic feedback
mechanism to NSERC from the entire university research community.

   The ineviatble result of any oligarchic structure is that it
proliferates for its own sake. In NSERC case the consequences are
the overblown and overcomplicated (and resource-draining) funding
structure of many dozens of discipline and program committees. To
justify their very existence the multiple NSERC committees
require unnecessary lengthly proposals and multistaged process 
of "proposals evaluations". The latter process is de facto
largely consists of a second peer-review of already peer-reviewed
(!) published papers.

   Present NSERC trend to even more tighter peer-review "quality
control", even greater "selectivity" in funding (more NIL-awards)
is a step in a precisely THE OPPOSITE direction to what is
required to forster the real excellence and innovation.
Paradoxically it may sound, but agencies like NSERC need LESS (!)
(and not more !) expertise to improve their operations. The
bottomline performance of a complex decision-making system (like
NSERC) is NOT a linear function of the overall "expertise" it
has. In fact, it is an inverted U-curve with a maximum (optimum)
beyond which the system LOOSES its efficiency. This is a known
effect of an over-controlled system - too many strings damage the
adaptability. Like with vitamines, the overdose turns stimulation
into a poison. In our opinion NSERC presently suffers from a
severe "OVERexpertisation".

   To alleviate the damaging aspects of present NSERC functioning
for the canadian university system, canadian economic
competitiveness and better management of financial and human

1. Numerous (now over 40) "grant selection committees" should be
amalgamated to just a few. Their present activity is largely in
"peer-reviewing" of proposals which are almost invariably based
on already peer-reviewed published papers. There is no need to do
peer-review twice. This simply imposes an unfair "double
taxation" on the ideas, work and time of the researchers.  
2. Out of 3 present NSERC criteria ("excellence of the
applicant", "excellence of proposals" and "need for funds") only
1st and 3rd should be left. "Excellence of proposals" is largely 
a Red Herring. For all practical purposes, the presently used
1-page form (NSERC form 180: "intent to apply") is FULLY
SUFFICIENT IN ITSELF, i.e. as a rule no "longer" proposals should
be written AT ALL. This will not only save many truckloads of
paper, but millions of hours of a highly qualified professional
labor (at $ 30 per hour at cheapest !) to write AND read the
typewritten compilations of already published papers.
(Longer proposals can be left as optional only for some special
cases, e.g. for group grants in high-energy physics, or for the
first-time applicants yet without published papers). 

3. The rat-race terminology (grant "selection" ; NSERC
"competition") should be eliminated from the documents and actual
policies. Its continuous use is harmful for the morale of the
entire community and sends a damagingly wrong message, especially
to young scientists, forcing many of them out of profession
and/or out of the country. Science can not and should not 
operate by the rules of beauty contests and wrestling games. ALL
university-based researchers whose active status can be sensibly
demonstrated, should be funded at some (basic) level using a
SLIDING FUNDING SCALE rather than NIL-awards (5). These basic
awards (we suggest to call them RBMG -  Research Base Maintenance
Grants) may not be great but they should cover such fundamental
expenses as any serious researcher has: publication and reprint
charges, conference travel, computing and software, electronic
networking, etc. The gradaute student support can be much more
efficiently met through the direct grants to the departments
where the students are being trained. We also note, that personal
research expenses which professors squize from their personal 
salaries are NOT TAX DEDUCTABLE !
 In any case:  NIL awards to ACTIVE researchers should not be
tolerated. This practice is based on an ill-conceived philosophy
of the alleged efficiency of a rat-race "competition" in science
when only peer-review defined "excellence" is to be rewarded. In
reality, NIL-awards amount to wasteful and irresponsible
mismanagement of the scientific and intellectual resources of
this country.

4. It is imperative to obtain the views of the scientific
community on whether NSERC officials should be elected by all
those eligible for funding, and if found to be so, the electorial
process should be instituted. Nomination to NSERC bodies
(including the President) should be discussed PUBLICLY and in
advance, perhaps through a special bulletin. The candidates
should provide their platforms  and be open to public questioning
and criticism before they are elected to the office. They should
be regularly publicly accountable during their entire term in the
office. It is also critically important that the minority and
dissentering views are duly represented.

To conclude, contrary to a misleading similarity, the terms
"competitiveness" and "competition" are quite different. The real
competitiveness of research comes from open opportunities and NOT
from the enforced "competition" which in many cases directly
detrimental for the very spirit of the reasearch. We believe that
a wide and open public dialogue on the above issues is highly 
desirable for the strengthening of the economic efficiencey,
international competitiveness and social responsibility of the
canadian research enterprise. 
Alexander A. Berezin,                          
Department of Engineering Physics, McMaster University, Hamilton,
Ontario, L8S 4L7; (905) 525-9140 ext. 24546
Geoffrey Hunter, 
Chemistry Department, York University, North York, Toronto,
Ontario, M3J 1P3; (416) 736-5306 email: FS300022 at SOL.YORKU.CA    
1. A.K. Vijh, Canadian Chemical News, 42 (# 10), 14 (1990).
2. A.A. Berezin, American Journal of Physics, 57, 392 (1989).
3. L. Hocker, Physics Today, 46 (#8), 13 (1993).
4. R.K. Merton, Science, 159, 56 (1968).
5. D.R. Forsdyke, Nature, 312, 587 (1984).

> Patrick

More information about the Bioforum mailing list