Thursday, June 26, 2008

Research Institutes and the "graceful exit" problem

Here comes a thought that's been on my mind a long time. I may expand this into an article later on, but thought I would try it out on my two (or three) long-suffering readers. I warn you however that, following the lead of notable political parties in Bombay, my colleagues may attack me and break my office windows for writing these heretical words!

In India we have science research institutes like TIFR, IMSc etc where a scientist has to be (in principle) outstanding to secure a job, and gets lots of benefits on securing the said job. Speaking from the experience of my own 24 years at TIFR, it is wonderful to be in a place like this where (i) the general facilities - offices, cafeteria, housing, bus service etc - are excellent, (ii) the research facilities - offices, labs, internet etc - are excellent, (iii) the funding - for experiments in particular - is generous, (iv) the teaching requirements are minimal, (v) the bureaucracy is largely benign. Each one of these factors individually would seem utopian at many Indian universities where - to give just one example - even a clean toilet is a distant dream.

What I find puzzling is that these privileges are piled on us with no clear statement of our responsibility or duty to perform, and no consequences for failing to do so. As a TIFR faculty member, the worst that could have happened to me if I had failed to carry out high-quality research is that my promotions could have been delayed. But I would still occupy prime air-conditioned space in a prime institute, receive a decent salary, and not have to teach, while my peers in universities would continue to deal with their heavy teaching load, making visits to smelly toilets in between classes.

In brief - once I'm here I cannot be sacked, pretty much whatever I do short of setting the place on fire. Now here is a contrast. I don't have precise facts with me and speak only from what I've heard, but the Perimeter Institute in Waterloo, Canada has a different model. Scientists are hired there for their exceptional research promise (admittedly it's a theoretical-only institute). They are given wonderful facilities which (besides clean toilets, apparently considered normal in Canada) include a gym, squash court, fireplaces, free wine and cheese on Fridays, a free Blackberry (P.I. is owned by the Blackberry people) etc. They are also paid lavishly. The catch is that these persons are hired on 5-year contracts and as soon as a renewal comes up, their research is judged again. If there aren't great papers, then they are out.

Where do they go once they're out? That has a neat answer too. Every Perimeter researcher has a University affiliation somewhere in Canada. The Universities are happy to have a prestigious name on their rolls, who however doesn't teach (or teaches minimally) and presumably doesn't get paid either. However as soon as Dr Prestigious is "released" by Perimeter, he/she goes back to the University and just reverts to being a normal person with normal privileges, and a normal teaching load.

This makes sure that Universities are not starved of good people by the Research Institutes, for the latter keep only those people who are truly productive in research, and only while they are productive. By accepting a job in a Research Institute, a scientist is taking on a challenge and receiving the appropriate support and incentives to meet the challenge. But some will fail to meet the challenge despite this, and all will find their productive years coming to an end some day. That's when the graceful exit comes in and we go "back" to a University.

One argument against trying out this model in a non-theoretical context could be: how can experimentalists function on a 5-year contract? Can they be expected to set up a lab and start functioning and be judged in such a short period ? My answer is: OK, make the first review period 10 years for experimentalists. After all, the parameters of this model can be tweaked keeping the underlying idea unchanged.

I can't think of any other arguments against creating an Indian version of the Graceful Exit model. But there must be many, and I'd like my readers to point them out. If I can argue successfully against all the objections the idea would gain in strength. If I can't, I'm happy to give up on it.


Anant said...

Catch anyone vice-chancellor in our country agreeing to any such proposal. They would simply say, `why should we take someone who is rejected by these institutes? Our lectures and readers, sans toilet facilities, slog day in and day out and train students, and of course that is why they cannot produce great research, atleast comparable to those in these Institutes. Why should we take someone who has been in an air-conditioned office and still could not produce research? Besides, such a person probably does not have any teaching experience, has no patience with students, and there is no reason to take someone like that.' Hard argument to beat. Your comments?

Sunil Mukhi said...

Yes actually this is a valid concern and it had crossed my mind. But it's not a flaw of my proposal, only of the practicality of implementing it.

Basically it has to do with the prevalent tension between our research institutes and universities, which has its valid reasons rooted in history but is unnecessary and mutually destructive in the long run. Also I suppose one factor is (see my previous blogs) our low maturity level as a scientific community.

Now I don't know how our maturity levels are supposed to improve, but the tension between research institutes and universities might actually decrease if a proposal such as I made were to be implemented, with a visionary Vice Chancellor (not an oxymoron, I hope!) and a few other visionaries involved.

Let's just imagine that a single research institute (e.g. TIFR) in conjunction with the Univ of Bombay and IIT Bombay were to strike up a graceful exit deal. The Univ and IIT would go into the deal with their eyes open and acting as equal partners, not inferiors. They would ensure that their own interests were served and that the movement of TIFR faculty to their place would be on their own terms. This could seed a change countrywide.

As for the worry about teaching experience and having no patience with students, let's not forget that it's possible to gain teaching experience and learn patience!

Gautam said...

The points you make struck me very forcefully on a recent visit to the European Molecular Biology Laboratory (EMBL) in Heidelberg. This place is, as far as I can tell, pretty close to heaven if you happen to be a molecular biologist. Research funding is very generous, the administrative and technical support is outstanding, the labs appear to lack for nothing and the students are mostly, outstanding. The interesting thing about the place is the following: Everyone - with the exception of the Director General and his deputy as well as a very tiny number of additional people - is on a contract which can run to a maximum of 9 years. No more. Scientists there are evaluated on a regular basis, of course, contracts can be shorter than the maximum allowed and you can be fird with reasonable notice. But the interesting thing is that the maximum is non-negotiable. Also
interestingly, this goes not just for scientists but for everyone else as well: administrators, secretaries, machine shop people, lab technicians .. (I would have thought this could be done for scientists, at a pinch, but machine shop technicians and repairmen
.. ?)

Why would anyone go there, since this is essentially what in the US would be called "non-tenure-track"? Well, the money for starters. I think salaries are around 1.5 times more than what a regular university position would earn. Plus the fact that funding is, for most things and even given the fact that high quality molecular biology is an expensive pursuit, generous in the extreme and the system is supportive of radical new ideas and encourages collaboration.

I don't think that those who leave EMBL after their tenure have problems with getting positions either at other labs or at universities. First, they seem to typically have a research record that is outstanding. Second, their record with funding agencies is also typically good, since they will have applied for a wide range of external funds while there and would typically have put that money to good use. For a university, hiring such a person would make total sense, since departments gain from having high-profile individuals of demonstrated creativity who can also generate their own funding in large measure. Of course, much of this is biology-specific and I don't know how such a model would translate to physics, but you get the idea.

My subversive idea of the day agrees completely with yours. Our research institutes should fulfil precisely this function - provide the initial leg-up, freedom from excessive teaching responsibility in the initial stages, enable the setting up of research labs and allow the person to gain credibility with funding agencies and, finally, ease the transition to the university set-up. But this should, ideally, apply to everyone, as in the EMBL model, otherwise it won't work. Institutes such as TIFR IMSc etc. can also then take an occasional risk in hiring, since one is not faced with the prospect of having "the candidate who shouldn't have been there" across the office for the rest of your natural life. Universities profit because they get ready-made labs and experienced researchers to head them, who are at the cutting edge of their disciplines.

I don't think this is a new idea - I heard something very like this from a former director of TIFR once upon a time. But somehow it just doesn't seem to have attracted much attention, comment or interest either way.

Rahul Basu said...

A lot of US universities work with a variant of this model. Since retirement has been banished from academia, a person can stay on the rolls indefinitely but would have to apply for grants to fund all activity. If you don't want to do this or are unable to get grants because of your lack of research output, you can offer to take on more teaching responsibilities in the same department - so instead of being associated with a different organisation like a university as in Sunil's (or PI's) model, you teach in the same familiar surroundings.

As it turns out, some of us have discussed this internally and even identified a university for IMSc and thought about writing a proposal about it (Gautam??) to be presented to our council. One main problem is the idea of a permanent secure job that is so important in India, particularly since, as you get older, your output decreases and you insecurity increases. So one could, for the purposes of selling the idea, not give the impression of farming off unproductive people but having useful synergy between the universities and institutes on an equal footing.

I really don't think this is such a difficult idea to implement. Moreover the discrepancy between a retirement age of 60 (as for institutes presently) and those of at least central universities and IITs of 65 can be addressed at least for the last 5 years, by indirectly having people in the institutes work those extra years in a suitably identified university. One can then slowly move that period to somewhere into the 50s where, people would not retire but would just spend a lot more time teaching in the universities.

Incidentally CMI, Chennai is already using a slightly distorted form of this model. They, of course, use retired people whereas Sunil's plan would envisage using somewhat younger people who are not as active. But this is just a matter of degree. Incidentally some of our younger faculty have also been teaching at CMI. Some of the IISERs and NISER are also using guest faculty from other places including institutes to teach some of their courses.

So I think the time is ripe to start selling this idea in earnest and think about how to formalise such an arrangement.

And yes, with some exceptions, who cannot be named in a public blog, one's teaching abilities can be improved with experience.

Naresh said...

I have discussed similar ideas before and one of the argument against such proposals that I DON'T BUY is the following.

India is a diverse country and in an evaluation, there may be other factors apart from professional abilities that come into play and these factors are proportional to the distance in a multi-dimensional space whose coordinates could be religion, region, language, dialect, caste, sub-caste, gender, sexual preferences, and what have you. Therefore, we should make firings more difficult.

The PROBLEMS with this argument are that you start by implicitly doubting the professionally integrity of Indians and secondly, by putting too many safeguards, you block the exit routes as well that seem so vital for research Institutes to remain competitive with the best of the world.

A way out could be that the firing process could be made simpler for some honorable premier Institutes and let this continue for 5 years and then see if this policy needs to be reviewed. So what is said about the individual researchers could be applied to the Institutes as well.


Naresh Sharma said...
This comment has been removed by the author.
Naresh Sharma said...

It should have been "professional integrity" and not "professionally integrity".


I am not very familiar with blogs and thought deleting the previous post would delete it completely. But it leaves a funny message.

One learns something new every day.

Debajyoti said...

I made a rather long (or as Rahul B. would put it, prolix) comment on this, but it seems to have disappeared.
Didn't save it either (this being only my 3rd attempt at commenting on blogs), so I guess I have to type it again !

Interesting thought, especially when
it is emanating from an influential person. And since you want to ``argue successfully against all the objections so that the idea would gain in strength'', here are a few possible bottlenecks.

First and foremost, the idea, as expressed now, is going to raise hackles in all quarters.

Much of what I am going to say may seem
undoubtedly colured by what you term ''the prevalent tension between our research institutes and universities''. While this tension is indeed destructive, you cannot just wish it away and any effort at reforms must
address this mutual antagonism. Consider my position as that of a Devil's Advocate !

1. There is an undercurrent throughout the
posting that rates the research activity
in Institutes to be qualitatively superior
to that in Univs (I include IITs amongst
the latter). This is not quite the stuff
that will endear you to Univ teachers.

And frankly, I don't think that this is
necessarily true. Take out some leading
lights, and the research output quickly
drops to pretty much the same level.

2. You also presuppose that the average
research inst guy would do a pretty decent
job of teaching.

I submit that this is a fallacious assumption.
(notwithstanding the obvious counterexamples)
And I base my statement on direct experience
of grad school in some of the bigger research

This is not just a matter of 'teaching experience
and patience', but of sheer ability (while this
assertion falls short of `how to disagree', I
hope you do appreciate that coming up with
concrete examples, and I can give innumerable
ones, is not what is called for).

Please note that I am NOT asserting that the
current Univ teachers do a good job. Far from
it. Nonetheless, why would you expect them
to accept 'hand-me-downs' ?

3. Apart from the ability to teach, would these
people even be willing teachers ? From the
exhibited reluctance to teach at Insts (even
when the average load is, say once in 10
semesters or so), it doesn't seem very likely.

When people expound on how teaching, even once
in a while, interferes with higher thinking,
it does not bode too well for a post-move
work ethic !

Once TIFR has gotten rid of them on account
of inferior research, would the Univ have the
right to terminate their job for ingerior
teaching ? Or for research below the Univ's
own standards ?

4. There is a further problem. In the Indian
scientific establishment, the pecking order
is clear. Those who can, do research while
only those who can't, teach (a second grade
profession even amongst academicians). And
this worldview is prevalent not only in a
class of succesfull professionals, but,
unfortunately, is getting rather quickly
ingrained into grad students.

5. There are certain further advantages to being
at a premier inst even after your salad days
are long over. You can still claim to serve
the cause of Indian Science by being on
several committees. Indeed, you can be involved
in reforming science education or enthusing
young minds to a career in science (absolute
lack of any experience in UG/PG teaching is
no hindrance).

Furthermore, staying at an Inst often affords
you much greater clout as far as shaping of
Univ policies or making appointments go (actually
much more than what you can do if you actually
teach there)

Who, in his/her right mind, would throw away
such perks ?

6. Anant rues that not many Univ VCs would be
forward-thinking enough. While that may be so,
over the past few years, several Univs have
offered positions to some people employed
at Univs.

Exactly how many have moved voluntarily ??
Does it constitute a good portend ?

7. Just who is going to decide as to whether
a given person should continue in the research
inst or move to a Univ ? And on what basis ?

Number of papers ? Sunil, you'd lose out to
many of your colleagues from other sub-fields.

Outstanding research ? You'd be terminating
90% of TIFR :-) While people are productive,
not many are ``outstanding''.

8. The posting and some of the comments also
seem to presuppose that this is one of the
best ways for a Univ to get experienced
researchers working at the cutting edge and
the requisite labs.

Why should that be so ? Why cannot labs start
up at the Univs as a matter of course? Why
should the DAE/ISRO/DST/DBT/.... not fund
things directly ?

These are rather touchy issues ! :-)

9. At some point, you talk of TIFR-BU-IIT acting
as a triumvirate with BU-IIT as equal partners.
Would you allow your partners a say in the
hirings at TIFR ?

And finally, are we Indians at all worried about
graceful exit? It seems to me that, whether in
politics or academics, people have to be forced
to relinquish control. Fair enough ! How else
would they install their proteges ?!

By the way, all this negativism is not to say
that this idea is worthless. Far from it ! It
is just that, given the long history of
academic brahminism (recently, the Director of a
premier Institute wondered if it was worthwhile
collaborating with Univ people), any such effort
would have to thought through very carefully.

EMBL ? A very small core faculty and 6yr positions
a la CERN ? You would not be able to justify the
existence of so many research insts !!

Incidentally, have not the powers that be decided
that it is impossible to revive Univs and instead
the favourite sons should have their own toys
(also known as pseudo-Univs) ?

If you have reached thus far, you adore prolixity!

Sunil Mukhi said...

I'm happy to respond to Debajyoti's posting. I appreciate his detailed comments but must point out that there are not really 9 independent points in his posting, but only three or four that recur in various forms (prolix indeed!).

So let me try to distil the essential issues and comment on them.

(i) "There is an undercurrent throughout the
posting that rates the research activity
in Institutes to be qualitatively superior
to that in Univs (I include IITs amongst the latter). This is not quite the stuff
that will endear you to Univ teachers.

And frankly, I don't think that this is necessarily true. Take out some leading
lights, and the research output quickly drops to pretty much the same level."

(I've quoted this comment verbatim from Debajyoti as it's perfectly expressed.)

My response:
The sole justification for research institutes is that they produce, on average, better quality research than universities where there is a heavy teaching load. I do believe this is the case in India at a statistically significant level. However if it can be shown to be false then we can shut down the research institutes (i.e. convert them to universities) and my discussion will become vacuous.

As for endearing myself to anyone, I gave up on that long ago!

(ii) You also presuppose that the average research inst guy would do a pretty decent job of teaching.

My response:
I know some people considered very good, kind, patient, organised teachers whose teaching is somewhat sterile because of their own lack of research involvement. Conversely, I've heard it said by students at no less than Princeton Univ that some of their teachers are terrible in terms of the standard parameters (organisation, patience etc) but then they are all Nobel or nearly Nobel laureates and convey a sense of excitement and creativity that may be said to compensate for the other shortcomings.

(iii) Willingness to teach, willingness to move to Univs etc.

My response:
My proposal does not require willingness. It attempts to modify the deal offered by research institutes to a limited but renewable term. Whether people who are not continued at a research institute are taken by Univs, or industry, or no one is to a major extent their own problem. It's the same problem as what do people do who were never considered by a research institute in the first place.

I expect the Univs will do what they find to be in their own interest, no more and no less. There is definitely an opportunity for them here. Thinking of these people as "rejects" from a research institute sounds tasty, but is merely childish (see my blog on maturity in Indian academia) - they are no more rejects than anyone who was never hired by a research institute in the first place.

(iv) Just who is going to decide as to whether
a given person should continue in the research
inst or move to a Univ ? And on what basis ?

My response:
That is truly a silly question. (I've waited years to levy an ad hominem attack on Debajyoti!). Research institutes hire people based on certain criteria. Renewing or not renewing a contract would be based on exactly the same criteria. Simple isn't it?

I believe I've addressed all the (genuine) issues.

Anant said...

Debajyoti: I guess I must issue a corrigendum regarding the interpretation of what I wrote. It is not that VCs are not forward looking. It is an argument they would use. On the other hand, what if a University Prof. were to lecture to Research Institute directors on how to improve teaching there?! Would any VC agree that those in his or her University is really bad? Why should they say that they need an infusion of blood from Institutes? They would say that they would like good people who are young and worthy, and not those who have somehow `failed' in Research Institutes. Again, please do not subject my comments to rigorous standard of refereed publications. It is just some thoughts on subjects that are important to all of us. Best regards, Anant

Rahul Basu said...

I think this post and it's comments have really fleshed out many important issues. And while many of us have wanted to launch ad hominen attacks on Debajyoti, I think most of what he says, while prolix, actually has a lot of value - particularly since he is speaking from the other side of the fence (wonder if he would have said all this in his earlier avatar).

But I really must take issue with Sunil over his dismissal of one of Debajyoti's points - who is to decide who stays and who goes to the university across the road. Sunil says:

Research institutes hire people based on certain criteria. Renewing or not renewing a contract would be based on exactly the same criteria. Simple isn't it?

I am afraid this is not quite so simple. The so-called criteria are extremely ill-defined and worse, modified often to suit the person being considered. Perhaps TIFR's criteria are laid out in stone but I have heard several different explanations in my institute regarding hiring/promotions

* He is very good. He is working on a very deep problem which is why he doesn't have any recent publications. Not promoting him would discourage people from working on non trivial problems.

* He has no publications for 3 years - working on a difficult problem cannot be an excuse. We cannot promote/keep him.

* He is a deep scholar and one of the best physicist/mathematician of his generation, though his publication record is next to nill. Just his presence here will enhance the level of activity in this area.
We need scholars as well as active researchers.

* I realise his activity levels are low but how can we ignore the glowing recommendation letters he gets. That is one of the major inputs we use in our evaluation process.

* A famous incident here in IMSc where in a faculty meeting a serious question was raised about what the real standards of evaluation were - "those are our standards" (pointing to the series of photos that used to adorn some of the lecture halls of IMSc, of Bohr, Einstein. Pauli, Dirac, Feynman, Gell Mann, ..).

Believe me, these are actual statements by Directors and not something I have made up on an impulse.

Well, you get the point. I don't think therefore this is so simple. Anyway this is only one of the perhaps minor points in Sunil's proposal but I just want to stress that we are yet to reach a certain level of professionalism.

Debajyoti said...

I seem to have thoroughly misunderstood
the import of the proposal. Silly me !

I am a bit confused about a few other
issues though.

(i) You say you believe that the research
institutes produce, on an average,
better quality research ''at a
statistically significant level''

Were some studies done in reaching this
conclusion? If yes, could you point me
to some source? Else, could you list the
parameters for deciding on the quality.
And, also, if possible, if the conclusion
remains unchanged if one takes out of
reckoning the leading lights---as I had
done in my comment ? (After all, the Mukhis
and the Sens are not going to be asked to
leave, are they?)

(ii) I'm afraid I don't quite get the second
part of your point (ii) in its entirety.
Are the 'near .... laureates' who have
been let go by TIFR (say) expected to
``convey a sense of excitement and
creativity'' ??

(iii) Re your point (iii), if the idea is
that the non-performers will have to
leave, no matter whether anybody wants
them or not, then

(A) the Insts can eminently choose
to impose this unilaterally, and there
is no need to build any consensus
with Univs. After all, Univ jobs, when
they exist, are equally available to
ex-inst people. They just compete with
everybody else. Oh yes, they have to
prepared to take a large pay cut.

To those of you readers in Insts who
want to teach instead : there are
quite a few jobs out there (Univ Depts
and UG colleges). Please do apply and
I'm sure some of you will get the jobs
after due procedure.

(B) this also removes my worry about
whether the Univ would have to keep
some of the positions vacant.

(C) also renders irrelevant my query
about whether TIFR is going to allow
any say to BU/IIT-B as far as its own
hirings are concerned.

(iv) Rahul has already addressed this at
great length. (And Rahul, between you
and me, I think that the TIFR norms are
not any more set in stone than yours are)

Anant, I did understand your point :-)

Rahul, re your
``the other side of the fence
(wonder if he would have said all
this in his earlier avatar)''

We'll never know, will we :-)

Re the ``fence'', it is there and
ALL OF US are responsible for it
being there. The sooner we realise
it, the better it is for all of us.

Sunil Mukhi said...

Well, I'm enjoying this discussion! I count this blog as my most significant effort to date in the amount (and quality) of discussion it has raised.

I understand and appreciate the various negative/skeptical points being raised, particularly by Debajyoti. There is no answer to skepticism really. Whether to be halted by skepticism or try to press on in search of a positive result, is one of those eternal issues that can have no simple solution.

However the Universities in Canada who have hired Perimeter Institute researchers must surely have had something in mind. I don't believe they were doing charitable service. I don't even know the terms on which they've done what they've done.

Instead of the "default" skeptical option, we could actually try to study this example. We could also learn from Gautam's example of a Biology institute in Germany, which I find highly significant for the fact that experimental rather than theoretical science is involved.

Sunil Mukhi said...

I'd like to respond to Rahul B's comment. First of all, far from contradicting me, it proves my point (I'm surprised he didn't notice that).

All I had said was that the standards for retention (under my proposal) would be the same as for hiring. Rahul does not disagree but observes with eloquent sarcasm that the standards in his institute are not well-defined. (Later Debajyoti points out that the standards in TIFR are probably also not well-defined).

True enough, but where have we reached with this line of argument? I hope not to the (absurd) view that criteria for recruitment/promotion should be clearly laid out in advance to the point where a computer program could make the decision?

In my department at TIFR (hopefully I'm not revealing deep secrets) the criteria are evident to all and the only thing to discuss is whether X fulfils the criteria or not. As long as faculty members accept each others' sincerity, it is a joint effort to figure that out. I emphasise "effort", for it is not easy but rather an extremely complex and taxing job.

What I gather from Rahul B is that he, at least, is deeply skeptical of his colleagues' intentions. Fair enough, maybe they deserve his skepticism, but as I've tried to point out earlier, skepticism is a one-way street to nowhere. If you can convey to your colleagues that you are on their side and willing to sit down jointly with them to resolve an issue, it is that much easier. Statements like "X has few publications but is working on deep problems" are not nonsense. They might be TRUE about X, or they might be FALSE about X. Dismissing the statement out of hand does no good and very likely great harm.

Indeed most of the time and effort wasted in hiring/promotion meetings is because no one wants to be the first to say "let's join hands and figure this one out". And also, let's be frank, because some of our colleagues can be less than honest.
But then as long as they are on the receiving end of sarcasm and skepticism, they feel justified and it's just a lose-lose situation.

P.S. I think it's truly wonderful that someone would point to Bohr, Einstein, Pauli et al and say "these are our standards". Why would such a nice statement become the butt of sarcasm? Or is every nice statement supposed to be met with sarcasm??

Anant said...

Some points I would like to record here: standards for promotion and hiring should of course be imposed with some flexibility, discretion and also common sense. While it is true that `deep-problem vs. gross numbers of publication' debate must go on, I think the point of RB is that these are used selectively, to fix some persons who are unpopular for one reason or another, and to promote others who are somehow favoured. And I am sure he can prove these case by case, which of course one cannot put down in writing. Regarding the photos of Bohr, Einstein, etc., I think that the ones who say these things say it without any thought. If it is said by top management, it betrays a degree of stupidity, because those men lived in different times, in different places and the problems they attacked were different. I do not think any serious reviewer would use such analogies and anecdotes. I think this statement is not at all nice, and should be met with a sledge hammer and not just sarcasm.

Sunil Mukhi said...

Ananth, This is a question of interpretation. When a senior administrator points to Bohr etc and says "these are our standards" I would interpret it to mean that we subscribe to the highest standards. I'm sure it can be interpreted in other ways.

I may expand on this another time, but I think it's a rather special and unlovable Indian habit to seek out and hang on to unfavourable interpretations of administrators' actions. There is some historical justification because we have sometimes had very bad administrators, but now it's a vicious cycle. The more we treat our administrators with contempt, the more we will only get administrators who merit contempt. Like all cycles, this can only be broken if someone is courageous enough to opt out.

So my recommendation is: take an administrator who (in your view) merits contempt, and try treating them nicely. (That does not force you to agree with them. Merely be nice, even respectful.) Watch the specimen for some months. Does his/her behaviour start rising to your (positive) expectations?

In another test-tube, take another equally contemptible administrator and show them your contempt. Watch their behaviour sink to your (negative) expectations.

It's called back-reaction, we often neglect it in physics but we shouldn't neglect it in life.

Anant said...

Sunil: thanks for the comments. Going back to the subject of evaluations, etc., and away from administrators, I really feel that the best way to avoid all these personality issues, excuses that can be used one way or another depending on likes and dislikes, is one of having several international and national referees. Clearly, once those views are in, a candidates list of publications are in, and a nuanced view of empowered committees that can smooth out the issues of `deep-thinking vs. publicaton numbers', then the outcome would be fair. In marginal cases, a candidate could be given a year or two to shape up. Then I think many of these problems will go away. Unfortunately, this has digressed very far away from the original mandate of this post, but the subject kept cropping up.

Debajyoti said...

Now, where did you see the skepticism
in my second post ?

If you can implement your point (iii) on
your own, the problem is solved ! And rest
assured, those of the retrenched faculty
who SHOULD definitely be lapped up by Univs,
WOULD be. And even if the Univ people are
recalcitrant, you shouldn't worry too much.
The appointments at the larger Univs/IITs
are decided on by Research Inst people !!
At my own Univ, e.g, it has been mostly
your pals :-)

Honestly, I would love to have answers
to the questions I raised. Simply because
these are practical bottlenecks in
implementing Inst-Univ partnerships.

And I agree that the Canadian Univs are
not being charitable. That is precisely
why I had asked if you are willing to let
your hypothetical partners (BU & IIT-B)
some say in faculty appointemts at TIFR.

Whiler the relevance of such questions
are determined by the parameters of your
framework, I had hoped that you would
see the necessity for having some answers.

Finally, there is no isomorphism between
the Canadian/European models and the situation
in India. They don't have to contend with
the 'fence' (as Rahul B calls it).

Sunil Mukhi said...

Anant: I agree totally.

Debajyoti: You asked where did I see the skepticism in your second post! Would "in every word" be an answer??

However in the latest post you've brought into focus a question that is begging to be addressed. Is it true that there's no "fence" between universities and research institutes in Canada/Europe? And if not, why not?

I truly don't know the answer but to start a discussion, I can make two comments.

One point (suggesting that there is a fence) is that in France there is a divide between CNRS and regular University faculty, both of whom work in the same places but with different salaries and duties. My impression is that this does generate resentment. It would be nice to know how much (and also whether CNRS people have a "graceful exit"?).

The other point (suggesting that there's not so much of a fence) is this: if you step into the Physics Dept of a good US university (even Stony Brook will do) and then compare the experience with stepping into the Physics Dept of a government-funded research institute (Los Alamos or Brookhaven), there's really little difference and if there is any, it could be that the former is more pleasant (of course, both have clean toilets). And being a faculty member at Los Alamos is not seen as some fantastic privilege. The rule seems to be that at a research institute you gain something but lose something else.

Such a balance is conspicuously lacking in the Indian system.

Debajyoti said...

Sunil says :
" Debajyoti: You asked where did I see
the skepticism in your second post!
Would "in every word" be an answer??''

It would be, but a wrong one, I'm
afraid. Unless you term the expression
of any doubts as skepticism.

In that particular mail, I just asked some
pointed questions, namely

(i) The basis (parameters) for the
oft-repeated assertion that research
institutes produce better quality research ?

(ii) Whether you expect people who are
FORCED OUT, to ``convey a sense of
excitement and creativity'' ? (your words)

(iii) On the need to have a formal agreement/
structure with the Univs if Res Insts would
anyway be prepared to cull unwanted stock.
(Why increase bureaucratic bottlenecks?)

(iv) In case of equal partnerships, what,
if any, role would the Univ play in faculty
selection at the Res Inst ?

I would have imagined that these would figure
amongst the very first questions that any VC
or a Univ HoD would ask.

sunder and sonati said...

What I say may be naive, to some extent, but then I have the advantage of being out of academia, and hence perhaps can offer a fresh perspective.

One way to tackle the Graceful Exit issue could be for the retirement age in the research institutions to be say 45 or 50. Then the exits are for everyone, and there is a substantial amount of time to work in a university/college after that.

Another point I would like to make is that even the Research institutions like TIFR and MatScience should open their doors to youngsters even in the theoretical sciences. Why not a B.Sc. degree by research? I do not buy the argument that the work is too theoretical for 17-18 year olds. Many children younger than that indulge in a lot of theoretical work.

The need to include such young people would also help the older faculty in divers ways, I feel.

Not being "good teachers" whatever that means is also a bit of a lame excuse: I think that if a person is doing good work (experimental or theoretical) then he is bound to be an inspiration to the students and hence a good teacher; he doesn't have to be able to lecture well, or teach well at the blackboard to communicate his subject to the students.

We in India are used to the idea of Teacher and Student as on two sides of a divide and perhaps cannot conceive of young and older co-researchers working together. Perhaps that is the way to strike out.

Debajyoti said...

On Sunil's comments to the existence
of fences between universities and
research institutes in India and elsewhere:

I am not very conversant with the system
in France. It is logical that there would
be some resentment if two sets of ``cohabiting''
people have different working condns.

Compare with the situation in Italy.
INFN-supported people work at Univs. They too,
do not have teaching responsibilities (or
at least as much as the direct Univ faculty).
But most of them desperately try to, simply
because being able to teach at the Univ is
considered prestigious.

Similar situations abound in several North
European countries.

Some key issues (not the only ones) here :

1. The working condns are not grossly
different (a point that you too make)

2. The prestige associated with teaching.

3. The absence of a 'big brother'-ly
attitude (on the part of either set).

I dare say the two situations are quite

Gautam said...

Just a final comment on what I understand to be the situation in France regarding research and university positions in the general context of how things are done elsewhere. From what I know, and I'd appreciate comments from those who can correct me, at least in physics, the pure-research jobs in the CNRS carry both prestige and mobility. Someone who holds a CNRS position can choose to take it up at any lab. I believe these are also considered more prestigious, at some level, than the university system. However, university positions get more money, which is one reason why several high-profile people in the CNRS position also hold or teach in (the difference is not too clear to me) university positions. Also, there is much closer contact with students if one holds a university position and this is a huge positive, especially if one has a research lab. The downside: a fairly
substantial amount of teaching vs. virtually no teaching in the research positions.

Also, as I understand, things are liable to change in the french system, with Sarkozy planning to bring in far more autonomy and funding for the universities, have closer interactions between research labs and the universities as well as have some American-style institutions, such as a French "MIT". Not everyone is happy with these proposals, at least not with all of them.

As I said, this is my understanding of how things are and I'd appreciate corrections/clarifications. I tend to think of the french system as being fairly close to the Indian system: some distinction between the pure research institutions and the universities regarding prestige, a much larger distinction between Paris-based institutions vs "provincial" institutions which mirrors, roughly the urban-rural divide here, an understanding that the research-education divide is a bad thing and some recent steps to set things right which appear to be well-intentioned from some points of view.

jpolchinski said...
This comment has been removed by the author.
jpolchinski said...

Hi Sunil, nice blog. I haven't been wasting quite enough time on the internet so this will be useful, and will broaden my cultural knowledge as well (what, for example, is a page 3 party?).

Ironically, Perimeter has just adopted a standard tenure system, in contrast to their original idea of term appointments, just as the first appointments would have come up for renewal. I am not sure what drove them to this, whether competition, or the difficulty of making these assessments.
Also, the backup appointments at the universities were problematic for various reasons mentioned by other posters: it is very nontrivial for a university to make an open-ended and unconditional agreement to take someone on any time in the next several decades, and any university of quality will be very reluctant to do so. My own belief is that the problem with term appointments is that it lessens the commitment of the faculty to the institution; if you tell me that I have a term appointment, I immediately start thinking about my next move. I like the point of view that the faculty _are_ the institution, and this requires tenure.