Is the PhD a problem?

Louis Menand’s new book, The Marketplace of Ideas: Reform and Resistance in the American University is mostly the ‘backstory’ of American higher education; it is lucid and erudite reporting as readers of his New Yorker articles would expect. He has not set out to be ‘prescriptivist’ (his word). But clearly he doesn’t like the American PhD.

To start with, it’s not clear that the PhD is fit for purpose. It can’t be a qualification for university teaching, since most PhD students are already teaching. Nor do PhDs clearly provide a contribution to scholarship, since many PhD theses are not of high quality (and probably even more are not read except by the student, his/her supervisor, and the examiners). Menand – a Professor of English at Harvard – suggests that ‘if every graduate student were required to publish a single peer-reviewed article instead of writing a thesis, the net result would probably be a plus for scholarship.’

Then there are what Menand calls the ‘humanitarian’ considerations. PhDs take a huge amount of time – though in the US they are nominally 4 years, most people take much longer. Median time to completion is 7 years in the natural sciences, 10 years in the social sciences, and 11 years in the humanities (including time out).

Then there is usually a long period of insecure and part-time employment before a permanent academic post can be found. Family formation often has to be put on hold. Student poverty endures well into people’s 30s. These things compound the emotional-psychological problems often caused by the PhD experience itself – the isolation, the not knowing if you are on the right track, or whether what you are writing is any good (I speak from personal experience). Many people decide it is not worth it and drop out (again, I speak from personal experience).

And to this Menand adds the ‘huge social inefficiency’ involved in spending so much money putting people of high intelligence into ‘programs that half will never complete and for jobs that most will not get’.

I’m not sure that things are quite as bad here, though a study 10 years ago found that only just over half of people who started a PhD in 1992 had completed eight years later. Since then the inclusion of completions in the PhD funding formula may have helped focus university attention on the issue. But I expect that things are still pretty bad, and that not many people look back nostalgically on their time as a PhD student.

High barriers to entry are bad for universities as well as students. The long and tortured process of becoming an academic must surely deter many who could make a positive contribution to academic life. Indeed, given that those with good alternative job options are perhaps the most likely to not start a too-long PhD with uncertain employment outcomes, it is the people with more wide-ranging talents who are self-selecting out.

Menand argues that the PhD obstacle contributes to the political imbalances with US academia. Perhaps the difficulty right-of-centre students have in finding a sympathetic supervisor, and the risk that political discrimination will make a tough academic job market even tougher, does entrench the left-liberal orthodoxy on campus. In any case, Menand thinks that political imbalance is a problem. ‘Liberalism needs conservatism,’ he says, ‘and orthodoxy needs heterodoxy, if only in order to keep on its toes.’

But what to do about it? American PhDs already typically have coursework, as opposed to the mostly dissertation model here, so they have already minimised the more problematic aspect of the doctorate. But perhaps a literature review submitted to examiners plus one or two published articles in a peer-reviewed journal would be a sensible alternative to the multi-tens-of-thousands-of words dissertation, though presumably the requirements would differ by discipline.

Whatever the merits of this idea, Menand’s book gives little hope that it will happen. Those who have survived the current system and control the universities have no incentive to change. Not only do they want to avoid greater competition in the academic job market, but they want to keep getting the army of PhD hopefuls to do the teaching, so the permanent academic staff can get on with their research.

20 thoughts on “Is the PhD a problem?

  1. Its not clear to me that the system is broken. From this post it isn’t clear what Menand’s complaint is either. Too many PhD drop-outs, unread theses, graduate students teaching, or academics being lefties?


  2. Menand’s book is a little frustating to write about, as there is not a conventionally-structured argument. Instead, his stance needs to be inferred from comments woven into this discussion of the general topic. But I take his main points to be that high barriers to entry to academic jobs are in the long-term harmful to universities in narrowing the potential pool of staff (not that they can’t fill their vacancies, but that they don’t necessarily get an optimal mix of people), and that they are harmful to PhD candidates. This might be justified if the barriers to entry were closely related to the work academics do, but he says they are not.


  3. I find those median time to completes difficult to believe. Half of PhDs in social sciences take more than 10 years? Of course it does happen sometimes. There is the famous case of the Phd maths student at Stanford who after 20 years still hadn’t finished because his thesis advisor wasn’t satisfied. So he shot him dead.


  4. The satisfaction obtained for the doctoral student (at least in my case) was taking a non mainstream topic that was historically confused and half examined and ‘nailing it’. You don’t get the same buzz from a lit review and a couple of articles …


  5. I’m with Sincs on this. I have some further comments:
    1) Lots of people in places where PhDs (or Doctorates) are popular are not doing them to enter academia anyway.
    2) They need to distinguish between part-time (which many are) and full-time people and take into account time people aren’t studying. A part-timer that takes two years out to have e.g., children (not surprisingly, very common in some areas) will take 10 years, and that’s good, not some scary number (4 * 2 + 2 = 10). Reporting the median is probably much better than the mean too, as my bet is the distribution is skewed in terms of time.
    3) Some places already allow PhDs by publication.
    4) The quality expected of PhDs has been reduced in many places of the world to get around the problem of them taking too long (and not just poor universities — I seem to remember Cambridge changing their expectations from approximately “something good” to “something substantial”).
    5) In some universities in Australia you get kicked out after a certain amount of time, and, for all universities, the amount of funding they get for having PhDs declines over time. This is supposed to encourage timely completion (and possibly mediocrity).
    6) Universities don’t have to employ staff with PhDs if they don’t want to. So any argument about it as an entry requirement doesn’t make sense.
    7) In Australia, you don’t have to do a Masters before a PhD (many people don’t), which makes the PhD more worthwhile doing (because you’ll learn far more than an undergraduate degree) and reduces the time. You could get one at 25 if you try hard. This is not much worse than some country’s degrees (in Germany, for example, the Diplomabeit (or whatever it’s called), takes 6 years. In the US liberal Arts stream, you have 3 years as your initial degree and 2 years of something else.


  6. In Economics at least, the US PhD model already addresses many of the problems highlighted here. Few students write one 100,000 word dissertation, instead they write three extended essays on different aspects of one issue. Each of these essays is in publishable form. So, you will see Ph.D thesis headings such as “Essays in Financial Economics” and so on.

    Once upon a time we thought a PhD graduate had mastered all the scholarship in his/her area, as well as making an original contribution to knowledge. Nowadays there is an increasing preference for “theoretical” PhDs, which usually amount to ideological rants. This is particularly the case in subjects such as Social Work, Education, Cultural Studies, Media Studies, and so on.


  7. Opinion seems to be dividing between those who already have PhDs and academic jobs (see last paragraph of the post) and those who don’t (see paragraph 4 of the post):)

    On Conrad’s points:

    1) True – but does this for or against Menand’s argument? A massive dissertation seems even more pointless for people not headed into academia.
    2) The link at comment 4 does report registered time as well as asbolute time, which does reduce the number of years spent formally working on the PhD.
    3) True. But Menand’s point that if nothing is published how can it be a contribution to scholarship remains a good one.
    4) This sounds right – though some of the unpublished PhDs I looked at 20 years ago weren’t that great either. As one friend said to me when I was agonising over mine, there are really only two types of PhD, not good and bad, but finished and not finished.
    5) As I know, from personal experience (I received the you have six months -or whatever it was – left letter, and decided to quit rather than take time off work). This policy probably reduces the number of PhDs ever completed, but has focused universities on issues at their end that are obstacles to completion.
    6) The PhD is a de facto requirement for a permanent academic job in most disciplines, and an advantage in all. Menand’s book is partly about how academia developed as a self-regulating profession; he suggests that de facto acceptable political views may now also be a condition of entry.
    7) True. We have slightly more PhDs per 100,000 persons than the US, though less than Germany (though there I think a lot of professionals are called ‘Dr’?).


  8. Those who have survived the current system and control the universities have no incentive to change.

    Conrad and I don’t control the university system. Alas.


  9. “A massive dissertation seems even more pointless for people not headed into academia.”
    I think there’s some misconception about what most theses are these days — many are broken into parts, so it isn’t really one massive document at all (no doubt some fit the stereotype). Something like four-by-fifteen thousand is common. It also depends on what is in different chapters. If you did a social science thesis, for example, the first chapter might be qualitative (e.g., we interviewed 30 people about their families and what makes them happy…), the second quantitative (we tried to design a quantitative study based on on the important factors found in our qualitative analysis…), the third more theoretical (we tried to integrate our findings into broader theories) etc. . You could really learn a lot from that.
    I guess it also depends on what you end doing. If you are going to work on policy, for example, it’s probably not too bad. If I flick through some of the documents released by some of the semi-government research agencies like the AIFS, then I think it’s fair to say that this is not dissimilar to some of their reports. The one thing I often note from these studies, however, is how poorly they are often written — one thing that is not commonly done in a normal thesis is writing for different audiences (at least in Aus).


  10. I agree that the PhD system is somewhat broken as it is so alien to what academics actually do, reducing its value as a training period. That said, pretty much every PhD student I know is trying to put out at least 2 peer-reviewed papers during their time doing a PhD.

    Secondly, it’s effectively impossible to get a real teaching job (tutoring on the side hardly counts) at Uni’s. Indeed at some unis they are now giving retirement packages to all staff who dont have PhDs. So it is a requirement for any kind of career (if not having a great Post-doc also)

    I think the basics of knowledge and finance are the biggest problems for recruiting and keeping PhD students. Many are intimidated by the idea of 100’000 words over 3 years, when in fact PhD’s are generally around 80-85 these days, and as conrad says, are broken up into a lot of different sections. As for keeping, finance is key. I know only one person who is doing a PhD who doesn’t have a scholarship, and even then 20k a year on an APA pretty much forces you to teach/research/work retail to meet the bills. So its not a surprise so many leave.

    Some unis are introducing PhD’s for past publication, so the move to a set of peer-reviewed articles, & higher funding would be a good incentive for students & deliver a better outcome for the country.

    Though, there’s also something to be said for keeping it hard (and I say this as a PhD student). It’s the highest possible education outcome, only some ought to go for it, let alone achieve it. But being a little bit more career relevant/less financially punishing would be nice.


  11. the PhD system is somewhat broken as it is so alien to what academics actually do

    No. That’s not correct. Anyway with two papers a year and some tutoring experience you should be able to get a job.


  12. I agree with Sinclair again. At the end of a PhD you should have learnt enough so that you’re an expert at something and should also mean that you can do research and have enough skills so that so that you can write and get articles into journals (or equivalent). That’s certainly a part of most academic’s jobs in Australia. Doing a PhD will help you a lot doing research, writing grants, and teaching in the future. Submitting to good journals will give you an idea of how hard it really is to get good things published and submitting to crud-kicker journals will give you an idea of the opposite.
    Incidentally, if you can’t hack it whilst doing a PhD (I’m not saying you can’t incidentally), there’s no way you’ll be able to hack it whilst writing grants, teaching, and doing administrivia. Also, if you want to do a shorter PhD than the 60,000 word benchmark, just do something with some reasonable mathematics or mathematical modeling in it.


  13. The main reason for my non-completion was that I picked the wrong topic rather than the structure of the degree, but I think a paper-focused structure would have suited me better, giving me a tighter focus and hopefully giving me useful referee feedback (though later experience showed that referee feedback is of very mixed quality and use).


  14. “(in Germany, for example, the Diplomabeit (or whatever it’s called)”

    The Diplom officially took five years (the Diplomarbeit was the thesis), although your study time wasn’t regulated so you could take thirty years if you wanted. It’s more or less been replaced by the combination of a bachelors and masters degrees: at my university (Universität des Saarlandes) there’s still some people finishing off Diploms, and they’re still used in a minority of fields, but most people are doing Bachelors/Masters. Doctorates seem more common in Germany than in Australia, but not even all doctors here have the title “Dr. med.” (which isn’t a PhD-equivalent title, which would apparently be “dr. sci. med.” or some such).

    I’m still not at PhD level yet, so I can’t contribute much to the discussion, but I know that in Finnland a PhD thesis in psychology consists of four peer-reviewed articles published in well-regarded psychology journals with a bit of decoration to tell the complete story.


  15. My biggest complaint about the PhD is the unrealistic expections that beginning PhD students tend to have about how having a PhD will improve their job prospects. In my experience, most PhD students don’t realise just how little their degree will matter in places which don’t have large numbers of PhD’s already employed.


  16. Ken – I think people emphasise the wrong skill set when they’re doing a PhD. They tend to worry more about the ‘original contribution’ and less about all the other skills.


  17. The USA phd has coursework and exposes you to several research methodologies. The Australian phd is one theses, you get exposure to one subject and one methodology. If the USA system is bad the Australian system is really really bad.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s