Saturday, August 19, 2006

French Universities

I was a professor at a French university from 1997 to 2002. In France, universities are legally bound to accept any student who has obtained the "baccalaureat" degree at the end of high school (that's 80% of the population). There is no control of either quantity of quality of entering undergraduate students. The best 10% of the student population usually prefer the parallel "grandes ecoles" system and are largely absent from the university student body. The enrollment numbers are only known a few days before the start of the academic year, which leads to last-minute scrambles to hire additional lecturers and open additional sections. Tuition fees are a few hundred dollars per year, and give the student health care benefits, possible access to subsidized student housing, subsidized student cafeterias, rebates on public transportation, on movie theater tickets, etc. Some small fraction of the students enroll solely for those benefits and never appear in class or at exams, so the official enrollment numbers are not quite accurate.

About 50% of the students fail and have to leave the university after two or three uears without a degree. The students' actual knowledge is pretty reasonable by US college standards (a testimony to the quality of French education up to the high school level), but what is striking is their inertia, lack of motivation, and inability to study by themselves outside class time. At the freshman level, there can be discipline problems: students reading newspapers in class, talking aloud, throwing paper airplanes, groups standing up and chatting at the back of the lecture hall, etc. Saddled with fear of future unemployment, even conscientious students exude a feeling of helplessness and morosity.

In terms of number of hours spent in instruction in front of the students, the teaching load is similar to a 2/2 semester load in the US, although efficient organization could in principle reduce it to 2/1. At my university, at least 75% of anyone's teaching had to be at the undergraduate level (at some universities, graduate courses do not even count in the teaching load). To deal with the shortage of instructors, faculty are encouraged to teach (for a little bit of extra money) more than their legal teaching load, and going over by 15% is quite common. Refusing to do so is possible but shunned as un-cooperative. Faculty teach lectures but also (along with graduate TAs) some recitation sections and some labs, and grading is evenly distributed among the instructors. The class size is 150 at the freshman and sophomore levels and 120 at the junior and senior levels, although non-mandatory courses may have much less. There is no required textbook since that would force the students to spend money on books and thus violate the principle of free education.

Incredibly, there is no university-wide calendar. At my former university, undergraduate courses follow the semester system, with recitation sections staggered one week after the lectures and labs further staggered two weeks after the lectures; graduate courses follow the quarter system; coop students follow the "3 weeks on, 3 weeks off" calendar; and the computer engineering school has its own calendar. After the final exam, instructors are allowed several weeks to grade, which further delays committees and jurys. Thus, overall the earliest "last day of class" is in early May and the latest jury is in the first half of July, with a gradually dwindling volume of teaching-related activities between those two dates. By law, students who fail must be given a second chance in the form of a makeup exam, and so a second round of exams, grading, and jurys, happens during the first half of september, just before the start of the new semester. Overall, the only period during which all the faculty are completely free goes from July 14 to August 31. Such a short summer is a serious hindrance to research. Sabbaticals are not a right but a privilege, like some kind of local grant for which the faculty compete with one another, and the number of available sabbaticals is budget-dependent.

Classes, often located half way accross the campus, start as early as 8am and some classes end as late as 6:30pm. Luckily, they have enough classrooms to not need to schedule any classes on Saturdays.

Although professors officially are supposed to spend 1/3 of their work time on each of teaching, research and administrative activities, in practice research is largely viewed as a luxury and must fit in the intervals between lectures and meetings, along with supervision of PhD students. Grants are needed for travel, for which the university has no money. There are many opportunities for grants, usually for minuscule amounts of money . Individual grants are non-existant. In my experience, the grant proposal success rate is high: get together with a few of your research buddies, concatenate your vitas, add the requisite number of pages of text, throw in the right buzzwords, and you're in. Once you have three or four grants, you and your students' travel needs are covered.

The salary is somewhat less than half of that for a comparable position in the US, but because of universal health coverage, retirement benefits, and free education, that difference is not as great as it might seem at first.

Overall, the system is incredibly wasteful and inefficient, and local efforts for reforms are blocked by rules and regulations emanating from the Ministry of Education at the national level. To summarize, compared to a similar position in the US, the salary is half as much, the teaching load is twice as high in practice, the administrative weight is orders of magnitude heavier, and the resulting stress is unbearable.

During my fourth year there, three faculty in my department had sick leaves, for independent reasons caused by stress and exhaustion. This served as a wake-up call for me. The following winter, I successfully applied to Ecole Polytechnique (one of the "grandes ecoles"), and that was the end of my stint as a professor at a French university. I have been there, I have seen what it was like, and... I have been vanquished. To any person who might be considering such a job, I can now repeat the advice that Philippe Flajolet gave me in 1997 when I told him that I was applying: "Don't do it."

Comments page (Usual comments link does not work properly, use this instead.)



  1. For what its worth I think this is a long french tradition. If you read the bibliography book by herd granddaughter, the work conditiions that Marie Curie and Pierre were appaling. This is the work they got the noble prize for. This is after Pierre Curie had already established himself and had a reputation (outside France) as a good experimental researcher.

    Pretty amazing.

  2. Hi Claire,

    your post so reminds me of the situation in Slovakia.. With old folks way past their best years having a solid majority in the faculty senate, no way to fire them and hire someone who actually does some work.. even good students not showing up for lectures because the side job they have is much more lucrative than sitting in class.. professors not bothering to assign homeworks because grading them is too much work.. Just the salaries are much lower than even in France.

    Being young and enthusiastic, I still dream of going back one day :)

  3. I found this post quite surprising considering the number of top-notch French scientists and mathematicians out there.

    Perhaps all of them attended the Ecoles? What is the basis for admission into the Ecoles?

  4. Let's not forget too that in France quite a bit of research is done in public research centers (CNRS, INRIA) with no need to teach.

  5. I've heard also in the US exist universities with high teaching load etc.

  6. Top-notch researchers are more often found in the research-only positions such as CNRS. "Grandes Ecoles" admit students based on merit, with a competitive entrance exam, which gives those schools control over what and how many students they admit. It's a completely different atmosphere, and it is much easier to do research at those institutions.

  7. and yet even after two tries Galois failed to get into the Ecole Polytechnique...I guess they did a great job of controlling the quality of admitted students..

  8. This comment has been removed by a blog administrator.

  9. I am a senior mathematican with tenure in a major French university. I feel that Claire's description of the French system is quite biased and very much US-centered. It is certainly a pity she was so unhappy in France, but fortunately some top notch American mathematicians seem to do it quite well in France. As another anonymous comment mentions, how could French Math be so strong if the system is so daunting. So a lot of exageration in Claires's post!

  10. Anonymous senior mathematician:
    I don't think that your example contradicts Claire's argument. Everybody will agree that top-notch people in top-notch institutions will perform well, [ironic]even in such a barbaric country as France[/ironic].

    I think that Claire's argument was more that on average, the teaching and administrative loads are heavier in France. Having known several instances of both systems, I would agree with that, and Claire gives for this a reasonable explanation: the large number of students in first year, their lack of motivation, the format of the system which don't require the students to work by themselves.

    Claire said that "the emperor is walking naked in the streets", and you answered "No, he still has his underwear"... ;)

  11. Speaking of "average," the situation at the average north American university is perhaps somewhat different (worse) from the one where Claire is.

  12. The work conditions were very good at my other positions in France (CNRS, Ecole Polytechnique), and I think would be quite competitive with positions abroad. What I described was the situation in the Computer Science department at a very good university, as I experienced it.

  13. It would be too long to comment in detail on Claire's post, which is quite stimulating (I am myself a computer science professor at a French university, and I know quite well what she talks about).

    I want just to stress the fact that what Claire is describing is the typical situation for a Computer Science professor, not necessarily for university professors in general.

    The actual stress and load that we have to endure while being in Computer Science is definitely much higher than in other disciplines (be it mathematics, physics or the like), and I would really welcome an objective analysis of the reasons why the tenants of (relatively) new disciplines like ours need to pay such a high price in countries with an accessible higher education system, like many european ones.

  14. I remember when I teached, besides students reading newspapers, I had students talking over the phone, playing dices, hearing walkmen so loud I heared the music myself. There are many reasons for that atmosphere. It is possible that a higher registration fee could make student more concerned to their study, however for example I could not have studied if the fees would have been much higher. You must be really poor to fall in the category of those who don't have to pay the fee.

  15. As I myself had an assistant and later a full professor position in French universities, I simply applaud to Claire's post, it is right on target. One of the comments above, coming from someone who apparently has a position in France, says it is an "exageration". First, on the contrary, I saw a lot of restraint in Claire's post. For anyone willing to do research, it is simply a nightmare, and it is hard to talk calmly about this situation. In my case, I was lucky enough to later get a position at a public laboratory, otherwise, the fallback solution was to apply to the US. Coming back to the aforementioned comment about exageration, I am not surprised: one thing I observed is that we have bred a whole generation of PhDs and now assistant professors or full professors who have seen their advisors totally swamped in administrative and teaching duties. As a result, those now willing to apply to such positions expect nothing else (if not like it); those willing to do research, either apply to one of the public laboratories (and to fight a misconception: it is a small fraction of all french researchers, especially in CS, so it is by no means a solution), or go to industry.
    Now, to complement Claire's post, I would like to go further into the reasons. I believe it boils down to "evaluation". The french government gives positions to universities (yes, it is all controled by the central state) essentially based on the numbers of students at universities. As a result, the game is to have as many students as possible; not only you cannot select them when they apply, but departments tend to let lots of poor students pass; I systematically had to apply a significant multiplying factor to all my grades otherwise 10% or less would have passed. The other part of the game is to create semi-redundant curricula in order to "prove" that you need many positions and to attract more students. As a result, many courses have to be given multiple times in different format, which is absurd. Because of this lack of selectivity, as Claire pointed out, students are lazy, never work, are not motivated. When I talk to American colleagues, I can see they don't really believe me; but anyone who has interacted with both an American PhD from a decent university and a French PhD even from a (supposedly) top university would understand. It is hard to imagine they can change once they go to industry; I suspect this whole situation will have far-reaching consequences for France in the years to come.
    Only unpopular measures can stop the tide: evaluate of universities by including a strong research factor, not only the number of students; and the ripple effects may change the whole attitude: departments then faculty would be similarly evaluated; there would be no reason not to sack students who do not have the minimum required level, and students would be working more, only the motivated ones would remain, etc. So before trying the high tuition fees approach, I would much prefer to try the high selectivity approach. I guess I am still French in prefering to reward capacities than money.

  16. As a professor at a French University, I thank Claire for her comments but let's imagine ways for improving the situation. The chinese ranking plays a positive role but also shows the following paradox: Claire's former University is ranked way before all the Grandes Ecoles, which points to a big paradox. Some Universities open to all students rank better than very selective Grandes Ecoles! The reason is that they have joint laboratories with CNRS, INRIA, INSERM and so on.
    As for the tuition fees, a global European solution should be encouraged, one way to go around the paralyzed system. The LMD reform, which sets a standard in the european curriculum was a step in the right direction.

  17. A few figures may shed some light to Claire feeling.

    - The 80% myth. The enrolment at Universities in the first year is roughly 31% of the candidate population (62 % of the candidate population gets the baccalaureat diploma, and 47,8% of them enrol in universities, including faculties of technologies –IUT). The same computation for year 2004 is consistent with the figures by Unesco, of an overall higher education enrolment ratio equal to 46.5%, much lower than US (80%), and lower than the average in Western Europe (sources: and Thus, there is a severe selection process in France, but it occurs in primary and secondary education.

    - Social origin.
    80% of the students in Ecole Polytechnique come from upper-middle class or higher (source Only 32% of university students (and much less in the first year) are in this case (source

    As a professor in the same University as Claire previously was, I agree with Claire on the fact that the teaching hours are definitly too long, the administrative burden is high and not a career-enhancer, and a lot of things could be better organized. Thus, in addition to being raised in an at least moderatly affluent family and being bright, the students of polytechnique get a happy professoral staff, while the poor must be content with an overworked one.

    But I strongly disagree with the largely negative appreciation she has of university students. Yes, some of them do not have a true intellectual interest, and seek a diploma at minimal effort only in order to get a job. A small flow comes to University with research as a perspective (this flow may become even smaller since president Chirac called for enrolment in grandes ecoles preparatoty classes of all baccalaureat holders with good grades). And all possible configurations in–between. But this has been true through history, starting from Plato’s time. And yes, the ambiance is not exacly academia. However, if we (researchers) do not consider these students worth our interest, maybe the next Ministry of education could finally consider that he could staff universities with more full-time teachers. Would it be more efficient? Probably not: most of the French R&D is already performed by engineers, educated in the engineering schools (“grandes écoles”), with not so good results.

    Finally, let me suggest some readings about the perennial French discussion about Universities versus “grandes école”.

    - Balzac “Le curé de Village” (1839). Includes a harsh attack against the « grandes écoles » system

    - Bourdieu “La reproduction” (1970)

  18. After reading Claire's experience, i was quite disapointed of the content. The thinking is so American... Of course the US universities are well-ranked, of course they do have lots of educational means but, honestly, can we really enjoy a high level class like one may find in France...Yes! But,only in the best and well-known US institutions.And the access has its price...
    In France, we may have neither smart classrooms nor facilies for students but you do have a strong level of Education. In France, you never get an A at a test or homework like it is so easy to have in US faculties. Students don't have the time to go to circles (which do not really exist in fact) because we must stay at our desk to study hard.

    I studied in France, in Japan, in the UK, and finally, in the US. My conclusion is no doubt that France is the best. This country may not be well-ranked but the standards used to set up rankings are so unadapted to the French reality.

    One can find as the Grandes Ecoles as well. IEP, Les Mines, Polytechnique, Centrale, ENA, ENS have such a high level that it is very hard for foreign students to follow classes.
    But when it comes to Law, Medecine, Letters, Philosophy etc Universities are the best and the only available on the Education scene for these types of studies.

  19. Several people commented that my post was US-centered: indeed, the only two systems I know from the inside are the French system and the US system, hence my comparison. I certainly did not mean to imply that the US system was the yardstick against which everything must be compared (it, too, has its own unique flaws).

    One anonymous commenter got statistical data to challenge my claims. It seems to me from that 42% is the number of students who *obtain* a higher-level diploma: since during the first two years after high school the attrition rate in universities is close to 50% as I remember, so that close to half of first-year university students leave without any university diploma, these numbers do not seem to contradict mine.

    I was interested in reading comments about the reasons, or suggestions of remedies. In my post I tried to stick to facts, that is, to what I observed directly.