Via Deane Yang’s Facebook feed, this New York Times round table on the question of tenure, featuring weigh-ins from faculty members in education, English, religion, education again, and economics. Notice anything missing? Like, say, science, engineering, law, and medicine? I said this before but I’m cranky about this piece so I’ll say it again. The reason we need tenure in these fields is not because we’re worried about getting fired for teaching an anti-establishment line on epsilons and deltas.* It’s because universities have to compete with private employers for scientists, mathematicians, engineers, lawyers and doctors. Mark Taylor writes:
If you were the C.E.O. of a company and the board of directors said: “We want this to be the best company of its kind in the world. Hire the best people you can find and pay them whatever is required.” Would you offer anybody a contract with these terms: lifetime employment, no possibility of dismissal, regardless of performance? If you did, your company would fail and you would be looking for a new job. Why should academia be any different from every other profession?
Maybe because academia pays a lot less than many other professions? Does Taylor have any suggestions as to what alternative benefit we should offer candidates in order to make an academic job worth their while? Does he really think that, absent tenure, our board of trustees would tell our chancellor, “Hire the best people you can find and pay them whatever is required?”
Right now, tenure is what universities have instead of money. I don’t see a lot of money coming our way soon. So I think we’d better hold on to tenure.
* Although this actually happened to Cauchy!
Quomodocumque 
I can’t claim to have thought about this recently, but I once did think about this a lot and it seems to me that it would be easier to find a not-completely-ridiculous replacement for tenure for scientists and engineers than for humanities faculty. You definitely want something different from what corporations would offer, so the idea would be to offer long but finite term contracts (maybe an initial 5 year contract to replace “tenure track” followed by a renewable 10 year contract). The salary would definitely have to increase, but I’m pretty sure that it would still save universities a lot of money.
It’s just not so clear to me that this would work for philosophers and poets. Even number theorists have more options than they do.
I would also point out that one of the biggest issues is how to deliver the educational mission of a university. Using primarily tenured faculty is both unreliable in quality and very expensive. So most schools now have a substantial shadow teaching staff consisting of graduate students, adjuncts, and other contract faculty that has developed in an ad hoc fashion. The question for me is whether we can somehow create a workable orderly system that delivers both a higher quality and more consistent education to the students and high quality long term research and scholarly activity.
I’ve never understood how replacing tenure with, as Taylor suggests, multi-year contracts is supposed to save money, even without the very important issue you raise. After all, assuming salaries stay the same, unless a contract system allows you to cut the total number of faculty then it doesn’t actually cost less (it might increase productivity, though). Conceivably you could get rid of departments or specialties that are no longer “needed” (though good luck getting agreement on what those are) but the natural process of retirements and new hires also does this, at least at places where units don’t own their faculty lines. It’s not clear to me that a big institution like a university should be turning over at faster than that natural rate.
There’s at least one private college in the US that does contracts rather than tenure, but my understanding is that contracts are always renewed and so functionally there is no difference. I suspect this what would typically happen in a contract-based system, actually.
There would be other costs associated with a contract system that could well eat up any productivity gains — if contracts were frequently not renewed, that would mean a whole lot more hiring, for instance. Also, contracts might result in even more of an every-person-for-themselves attitude that would leave no one willing to work hard at things that are not directly rewarded (service) or hard to measure (teaching).
Even in mathematics, as a political as it is, I think there is some value in tenure in allowing people to pursue long-term projects, and to protect us from political pressures. Funny story: the trustees of Caltech tried to fire Linus Pauling after he received his second Nobel Prize, but were unable to get around tenure to do so. Admittedly he got wind of this and moved to Stanford, so in some sense they succeeded…
“$12,198,578 at a private institution and $9,992,888 at a public institution. To fund these expenses would require a current endowment of $3,959,743 and $3,524,426 respectively and $28,721,197 and $23,583,423.”
12198578/9992888 = 1.2207
3959743/3524426 = 1.1235
28721197/23583423 = 1.2179
Doubtless this reflects the superior return on investment obtained by private institutions, subject as they are to the discipline of the market.
Perimeter Institute in Waterloo, Ontario, was founded on the model of five year contracts renewable upon review (and they pay nearly a factor of two time the typical Canadian university), based on the business model of the private industry donors who founded the place. When the first contracts came due the term was extended to seven years, and when these came due PI converted to a tenure system. I believe that this is because of the great difficulty of carrying out the necessary review: the contribution of academics is not so easily quantified.
It is indeed quite difficult to evaluate the performance of an academic. That is a strong argument for tenure, which limits the number of serious evaluations to one.
As I understand it, tenure has been eliminated throughout the UK. As I understand
it the result is that one effectively has permanent employment following a relatively
lightweight review after only 3 or 4 years of employment. This may change over
time, but this is the short term result of eliminating tenure there.
Here at Indiana something similar has happened. Our most common non-tenure track faculty
are lecturers. However there is an up or out review for lecturers after five or six years. If they pass the review, they become senior lecturers and appear to me to have something more or less like permanent employment. Formally they are easier to fire than tenured faculty, but in practice it does not seem to happen.
Anyway, tenure’s demise seems to get announced prematurely from time to time and doesn’t
really seem that likely to occur soon. About 15 years ago, a friend predicted that tenure would
be eliminated first in California, where the elimination of tenure at the U of C would be a ballot proposition. The theory was that once it was up to a popular vote, tenure couldn’t possibly win.
It sounded plausible at the time, but it keeps not happening.
Yeah, I don’t really get this debate. If a university wants to get rid of tenure, they could just start offering two options to their new hires: salary X with tenure, and salary Y without tenure. I predict that, except for some marquee names, people wouldn’t find the Y’s being offered higher enough than X to take it. On the other hand, if they offered salary Z with a N-year contract, where N=65-age (say), I bet there would be a big range of Z’s acceptable to both the hires and the administration. If universities started doing that, then I bet everyone would be happier: the academics would be, or if they weren’t they wouldn’t have anyone to blame but themselves for taking the wrong package, and the administration would be because their departments would be younger and more active.
By the way, one thing about this debate I find especially amusing is that libertarian types are usually the ones against tenure, which is just a essentially just a contract between private parties. Imagine what they’d say if someone proposed eliminating bonuses at corporations.
The same issue is of course true in the public school system (although “tenure” is less well defined or historically defended there).
You have a situation in which your employees have agreed to work for a particular combination of salary and job security.
Unless the employees somehow care nothing at all for job security, this means that a significant reduction in job security will require an increase in salary to fill the jobs with the same caliber of employee.
(I have maintained for a long time that the overall quality of our public schools is a highly stable equilibrium that is a function of the demand level and the willingness to spend. Since the demand is more or less exogenous–the number of school-aged children at any point in time–the willingness to spend is the limiting factor.)
I didn’t read the article yet, but while mathematicians, engineers, etc may not be teaching material that is anti-establishment, we could hold political views or beliefs that may get us fired. We may publicly and vocally disagree with the dean or provost about the decisions of the institution. We may be Arthur Butz.
While Mark Taylor may be comfortable speaking his mind without tenure, I disagree that this is true in all cases and with all people. Personally, as I’m about to navigate the job market, I’m already nervous about how the smallest things will inadvertently affect my getting a job. I cannot imagine how I would not worry about the same sorts of things if I’m constantly up for contract renewals.
I understand the argument and the concern that people have about tenure, but I think that anybody who claims that poor education is the result of tenure is not looking at the whole picture. Moreover, I believe that good teaching involves the occasional bad teaching semesters. If I’m to “provide the education our students and children deserve and our country and the world require,” I would explore different teaching strategies each semester to continually find better ways to teach material. This may involve semesters in which the strategy fails dramatically. If my contract was dependent on my teaching, why wouldn’t I just muddle by with good enough?
I’m also kind of cranky with people who are cranky about tenure.
I will try to play devil’s advocate and defend the attacks on a tenure system as it currently exists. As I see it, there is a sort of “cognitive dissonance” here: while we get (or not get) tenure largely for our research, having achieved tenure we are paid largely for our teaching as research is no longer required. To improve research we as a community should encourage tenure, to improve teaching we want to discourage it. As they say, “where you stand depends on where you sit” on the issue.
To make matters more concrete, think of this theoretical but plausible example. Some university ABC has four people in field X, two of them not active in research. For historical reasons, they don’t have anyone in field Y (say, they had one but this person left 5 years ago to a better place). As it stands, no one in field Y wants to come as they realize they will not have anyone in the same field for years. This is bad for research in ABC, bad for people in field Y, bad for attracting graduate students who may want to work in field X but take classes in Y, etc. If the university could push even one of people in X out, they could in principle attract two people in Y at once, jump starting the field. So while tenure may be a good thing in general, it is definitely restricting free movement of people and stifling innovation (in case field Y is new and rapidly developing).
I say in principle it might be possible to keep the tenure system, but that would require a much greater change than the contract system. We could, in principle, “federalize” the tenure (think of CNRS in France), by awarding it essentially like an NSF grant but still “for life”. What it would mean is some kind of matching contribution to whatever the university is paying (making this net-neutral will require revamp of the current grant system as well). This would allow lots of financial instruments for people’s movement. Essentially, ABC could just fire one or two inactive people in field X and noone will be too sorry as they would presumably can find other teaching jobs and keep their NSF “tenure” money. If their teaching at ABC is fine, firing might contradict labor laws, but this system might still allow lots of flexibility to “buy out” the salary difference with whatever lesser university they end up with. In fact, if I were to advise universities what to do I would suggest following baseball major league example and start buying (or teaming up with) “triple-A” and other “minor league” universities and colleges, where they can push away people.
I think part of problem arises from the fact that professors are only rated by their research. This is, in my opinion, already biased in the first evaluation for tenure, but becomes more so for the hypothetic subsequent periodic evaluations. This is because research may be aleatory, and people do not know in advance (by definition of research) the outcome of their endeavours.
Take someone who is doing good research now and gets a tenure say at 34: there is no guarantee that he will still be doing well in ten years. If there are periodic evaluation every 7 years, as the article suggests, he may pass the first one and then fail the second, losing the job at 48. With no experience whatsoever outside the university market. What is this person supposed to do? Hardly a good perspective to attract great minds.
If we want to avoid that tenured professors stop doing their work altogether, we should measure them by goals which are obtainable for sure with enough commitment and effort. For instance, take a math professor who is not doing very well with his research.
He may compensate by giving a lot of courses, both pre- and postgraduate, in different areas. He may need to enlarge his background for this, but this is something he can do with enough work. He may mentor students and organize seminars. He may take the current big thing, say the fundamental lemma, learn enough about the Langlands program and present the new result in a series of seminars for the department. A few years ago he may have done the same thing for the Ricci flow.
These are all things that require work, but are doable with little randomness. Any person doing all the stuff I have mentioned will surely add a great value to the the department, even with little or no research on his own.
If we were able to measure these different kind of contributions, we could fire the people who are not working seriously, yet guarantee a lifetime work for all the professors who are devoted enough, regardless of any lack of inspiration for research.
The premise that “having achieved tenure we are paid largely for our teaching as research is no longer required” is not completely correct, in that salary raises (at least at my institution, but I think we’re far from alone in this) are based primarily on research accomplishments.
Igor: one complication with your scenario is that while the two people in X may not be active in research, they may well contribute significantly more toward the teaching and service needs of the department than other two people in X who are research active. Math departments at research universities strive for a uniform sort of faculty member, one very heavily focused research, but I don’t think most departments would actually function if everyone followed that model. Of course, there are some faculty who do seem to abuse tenure by not pulling their weight, but it’s a real minority, probably less than 10% of full professors. (This is one of the reasons I think would be very hard to save money with moving to contracts.)
Also, like Matt, raises in my department are heavily dependent on research performance. Several math departments I know uses a formal system of evaluating (research, teaching, service) which are then weighted 3:2:1. Since research performance is more variable and easier to measure, under such a system it probably contributes to well over half of final raise score.
Or, in other words, what Andrea says below
If you’re going to start applying business economics to universities, you would also have to involve tuition. The best universities don’t have that much of a difference in tuition from the mediocre ones (the last I checked). There is clearly more demand when your acceptance rate is ~10%, so why not charge 5 times that of mediocre universities? Tenure and not competing with industry may have something to do with this – the university knows it has “fixed” costs in terms of salaries.
The anti-tenure movement is actually an anti-intellectual movement. “The people” don’t like those pointy-headed do-nothings that inhabit all those wonderful ivory towers (especially at state institutions).
One factor that was also ignored is that many, if not most, of the science faculty have to at least partially support themselves by getting grants for their research. Tenure or not, you have to do something to get a grant and get your pay.
Igor, would you also advise universities to pay professors as much as players get in the baseball major league? I have no idea how much they make, but I suspect it is enough so they can retire after a few years and live very comfortable life. If this would be the deal, then go ahead cancel tenure.
I think it’s a real oversimplification to suggest that what universities have to offer, in competition with the private sector for talent, is only or even mostly tenure. Most people I know who want to teach at the university level are primarily motivated by the fun of teaching and being around students in that environment, the high level of intellectual challenge and freedom, and the generally lower perceived stress and workload. It’s just a better life in many ways than, say, working as an actuary in a big insurance company or practicing law or medicine in this economy. Better job security is nice but generally not a huge factor in peoples’ decisionmaking. (After all, who really thinks, going into any job, that they are likely to be fired?) I actually bet that for most people the whole tenure system is a substantial net negative. It’s a nice perk if you get it, but the process of getting there is so lengthy, unpleasant, and political that many people would opt out at the outset if they could choose a more “normal” performance evaluation system and only a “normal” level of job security. (As compared to the private sector).
Indeed I bet the main benefit of the tenure system is for universities, in the form of enforcing an “up or out” employment model that allows them to sift through large numbers of employees for the few that will turn out to be superstars, without the discomfort of firing lots of people who don’t expect it. Big law firms are built on the same model, but they don’t try to pretend that the purpose is to attract talent!
Dear Scott,
There are several differences between universities and the tenure-track system and big law firms and the associate-to-partner track. The first is that associates get extremely high salaries, whereas tenure-track assistant professors, at least in mathematics, get much more modest salaries. The second is that there are are various natural tracks for associates who don’t make partner at their big firm to follow, whereas I don’t see the analogous career paths for assistant professors who are denied tenure.
I’m not sure how this impacts your arguments; I just want to say that the analogy is very imperfect.
Indeed, at Columbia I’ve spoken up vocally against pet projects of the administration, a required first-year science course in particular. Several of the mid-level administrators hate my guts as a result. Though I wouldn’t fear for my job in any case, having tenure has certainly emboldened me. It helps create an atmosphere where faculty speak their minds instead of parroting the administration — which is one reason it’s controversial!
I agree with Mark’s assertion that there is an anti-intellectual component to the anti-tenure movement. And there is probably also an element of shear jealousy, as many people in the private and public sector work in stressful socially pathological environments where every day may be their last depending on the whim of their “psycho boss.” The cartoon here may be slightly over the top, but a large part of the working population can probably identify with it:
http://wondermark.com/632/
Anti-tenure people are part of the race-to-the-bottom pack.
I also think that tenure (especially together with a decent retirement system) adds some needed stability to a department. Although it does allow some underperforming professors to stick around, it also encourages the good people to stay and not jump from university to university looking for ever greener pastures and higher paychecks. I would not want to attend a university where everyone worked on a five year contract. I would want to know with some certainty that the inspiring professor that I finally find one year will be there the following year to continue advising me.
And yes, I do believe that there would be a lot more political firings and contract non-renewals at universities where there is no tenure.
(While my regular laptop is in for repair for the second time, I’m writing this on a slow and creaky 8 year old laptop that never needed repair and amazingly still works. Also, I’ve noticed that someone else has hijacked my old posting name “Richard” that was unique on math and science blogs for a long time.)
I think that an important point is the lack of alternatives. Attorneys can usually find another job, and nobody would doubt that a 50 years old lawyer has work experience. But a 50 years old mathematics professor has close to no experience in the world outside academia.
When you get into a new work you don’t expect to be fired, but if there is a clause that you will have to have a fixed amount of research work every 5 years, this can easily be worrying. Working is not enough to guarantee that you will obtain results in research, and the possibility of being fired at 50 with no alternatives could be the factor that makes someone ditch research altogether for a safer job.
Sure, I agree with all that. Certainly law firm associates who don’t make partner have (or at least had, until a couple years ago) better economic opportunities than the average assistant professor who doesn’t get tenure and is asked to leave the academy. I just think that the “up or out” employment system serves similar institutional purposes for universities and law firms. It makes it easier, interpersonally and legally, to essentially fire a large percentage of your workforce after a few years without any particularly tangible reason. My point is that the main purpose and effect of the tenure system may be to give universities (like law firms) the ability to screen a lot more promising young people than they can offer truly permanent jobs to. The main purpose and effect of tenure probably is not making academic jobs more appealing to new entrants than they would be in the absence of the tenure system. Or at least that’s my hypothesis. Of course the tenure system makes academic life more appealing after you have tenure, but that’s not the right question. For people considering whether to embark on an academic or some different career it is far from clear that the academy offers better job security than, say, an insurance company.
One thing that a sensible tenure process might do is give people a chance to try an academic career early but with the understanding that either you will get a lifetime job or the academy will spit you out while you are still young enough to do something else. That would ensure that we don’t end up with a lot of fired 50 year old math professors, which as Andrea points out is a lousy position to be in. But does the present system even do that particularly well? By the time you get a phD, do some postdoc work in math, land a tenure track job, and then work up to your tenure review how old are you? At least 35? That’s pretty late to start a new career too. (Although of course not nearly as late as 50).
I don’t have strong feelings about tenure overall either way. I’m just not convinced it’s accurate to say that the tenure system is necessary to attract talented people into academic work, or even that it on balance serves that function. To really think about that question clearly, you need a vision of what university hiring and retention practices would be like if there was no such thing as the tenure process. Would job security become as bad for even senior professors as it is now for junior professors and adjuncts? Or would it be more like the typical private sector job for both junior and senior faculty alike, where of course you can be fired but that’s not the expectation? It depends on how universities would react to the elimination of the current system. If they reacted by being even more careful up front about who they hire into faculty positions, with the expectation that the job lasts indefinitely unless you screw up or the business tanks (the typical private sector model), then that job might be even more appealing to talented people than entry level academic jobs in the current system are. Just even harder to get.
My point is that the main purpose and effect of the tenure system may be to give universities (like law firms) the ability to screen a lot more promising young people than they can offer truly permanent jobs to.
There are certainly exceptions at the very top, but most universities tenure a substantial majority of their tenure-track assistant professors. Its a little hard to measure because some people leave before coming up for tenure (sometimes because they know they’re unlikely to get it, sometimes not), but I think at a top public university like Illinois tenure rates in math would be something like 80-90%.
According to wikipedia, in the 19th and early 20th centuries faculty served at the pleasure of a universities board of trustees, though a de facto system of tenure existed. The current system of tenure arose after various cases of trustees, legislators, donors, and university administrators trying, and sometimes succeeding, in firing faculty for various unpopular views (e.g. opposing US participation in WW I). One important case, apparently, was the 1903 dismissal of mathematician G. B. Halsted by the University of Texas after 19 years of service. These cases were apparently pretty rare, and overall the system functioned pretty much as it does today.
The tenure system as we know it today was mostly put in place during late 40s and 50s, in part as a side benefit by universities desperate to recruit faculty to teach all the returning GIs.
Thanks Nathan for that refreshing dose of historical research. Even if it did just come from reading Wikipedia. (Nathan’s summary is better written than the article,
so don’t bother.) I think it does indicate one thing: this is all about expectations. Reminds me of a story:
Shortly after I started my postdoc at Yale, I was visiting my sister in SF. We went
out to dinner one night with about 8-10 of her friends. Being a good sister, she bragged about how I was working at Yale. I was a bit embarrassed and tempered her enthusiasm by saying it was only a temporary job, not tenure track. One of her friends asked how long temporary meant and I said 3 years. They then went around the table telling about the longest they had ever stayed in a single job. None of them had ever stayed in a job for more than 18-20 months, so they all found “lack of tenure” a really perplexing negative about a job. For the record, they were almost all MBA’s from Berkeley working in high tech and mostly web related industries.
I will say that I do prefer our level of stability to theirs. But expectations can be different and as Nathan points out, have been different in the academe in the past.
And probably they will be different in the future as well. Though I do share Nathan’s skepticism about most current suggestions for eliminating tenure. All such plans will, of course, improve quality while cutting costs. I’ll believe that when I see it.
Did you also inquire as to how long these friends have lived in the Bay Area? I’ll bet that for many of them it’s been more than three years (depending on how old they are and how long ago they graduated from Berkeley, of course). Now imagine that you’re at a speed-dating session in New Haven and you ask all of your dates whether or not it would be possible for them to stay in New Haven for more than three years if their personal life inclined them to do so. I would guess that more than half would say that it is possible. But that’s not your answer, is it? Given that in our profession changing jobs almost invariably requires changing cities, tenure means that eventually you can offer a (would-be) significant other some measure of permanence.
Thanks Nathan for that refreshing dose of historical research. Even if it did just come from reading Wikipedia.
Hey, I added the bit about faculty opposing US participation in WW I from a second source! (There as a prof at Columbia who was dismissed for this).