Slashdot Mirror
Reform the PhD System or Close It Down
jamie points out an opinion piece by Columbia professor Mark C. Taylor in Nature News decrying the state of PhD education in the US, calling it "broken and unsustainable." Quoting:
"The necessary changes are both curricular and institutional. One reason that many doctoral programmes do not adequately serve students is that they are overly specialized, with curricula fragmented and increasingly irrelevant to the world beyond academia. Expertise, of course, is essential to the advancement of knowledge and to society. But in far too many cases, specialization has led to areas of research so narrow that they are of interest only to other people working in the same fields, subfields or sub-subfields. Many researchers struggle to talk to colleagues in the same department, and communication across departments and disciplines can be impossible. If doctoral education is to remain viable in the twenty-first century, universities must tear down the walls that separate fields, and establish programmes that nourish cross-disciplinary investigation and communication. They must design curricula that focus on solving practical problems, such as providing clean water to a growing population. Unfortunately, significant change is unlikely to come from faculty members, who all too often remain committed to traditional approaches."
Uhh... isn't the whole point of studying for a PhD because you want to remain in academia?
"Increasingly irrelevant to the world beyond academia"
The language of number theory seemed to be an exercise in the technical until hundreds of years later we end up with encryption systems based on their very principles. How you can claim prior knowledge of what will be useful in future, I do not know.
Comment removed based on user account deletion
The competition for tenure track positions is currently insane, since the professors from previous generations have trained too many PhDs. The funding agencies reward large labs under a single PI with large grants, with the labs mostly running on graduate students and post-docs who themselves see no way out. Now we are seeing career post-doctoral positions, especially in the biomedical sciences; see the recent suggestions about making a post-doctoral position more permanent. Not everyone can be a manager (PI), so we are stuck being graduate students or post-docs. I know industry is also a home for PhDs as I am one of those happy campers, but the fact is there are too many PhDs being trained relative to the number of positions available.
Lets have a system where the professor is rewarded for doing their own research, rather than their ability to write grants and farm out the work to their subjugated minions.
He generalizes the situation in some subjects (e.g. philosophical sciences). The situation in natural sciences is different. Having a PhD in physics (and not being an idiot who does not look left or right) enables you to talk to a lot of people and understand a lot of people. And you usually get you degree in 3-5 years (after the master) and not 12. And yes, i agree with him, weed out the subjects in the PhD courses where people waste, badly supervised, their valuable lifetime and replace the PhD courses by more appropriate new topics and fields. My feeling however is that this is more a problem for the philosophical faculties than for the science faculties.
Comment removed based on user account deletion
Mark C. Taylor's PhD is in religion. What was that about providing clean water to a growing population?
One thing that might be helpful (at least from the point of view of Prof. Taylor) would be to eliminate the bullshit Ph.D.s in fields such as political science, poetry, philosophy, English literature, and so on. Seriously. I talk to these types several times a week a bar near the Arizona State University campus and it is amazing how obscure their research topics are. Indeed, I get the feeling that there are extra points awarded (in some sense) for the more bizarre and irrelevant your topic is. And you can just feel the inner sneer as they watch you try to process the title of their dissertation.
Some of these people understand that they are shouting in an echo chamber of one, and in their circle of nominal peers, that's freaking cool.
I believe that the current inflation of PhD degrees is a direct consequence of the "everyone in university" attitude. I think that there is absolutely no point in giving a BA degree to pretty much anyone who enters university, because this produces an enormous mass of mediocre MS students, which then turn into way too many PhD candidates of dubious value. I include myself in this group, as I know full well that 20 years ago I would have not been admitted in a PhD program, let alone receive a degree. A PhD nowadays is an award to persistence, not excellence.
The inflation in titles is then carried on to the job market: more and more jobs are offered to candidates who hold a PhD, where a good MS would be more than enough. However, as a poster above noted, a PhD is basically taken as a certificate of being able to work independently (which, in may cases, is hardly true).
Treating PhD students as cheap labor is not doing a favor to anyone. I would find it much more honest intellectually to offer long-term internships for BA and MS students, instead of enrolling them to receive a higher degree which on the long run is devoided of all meaning.
This is the Age of the Internet. Overspecialization isn't the problem it used to be. With instant communication and email, a PhD student can be in regular contact with all the 10 people around the world who work in his particular sub-specialty if he wants to. So it doesn't matter very much if the local faculty don't know his specialty, although in practice at least the advisor ought to be qualified enough to supervise the work. Arguably, it's superior because it may lead to more inter-university collaborations.
I've not yet RTFA, but your statement seems like a sweeping generalisation and overlooks other possibilities. In various industries, a doctorate is a pre-requisite for most roles (e.g. semiconductor engineering). You might just squeeze in with a Masters degree, but more often a Ph.D. is required to even get your foot in the door. Given the erosion of standards for 'regular' degrees (B.Sc., B.Eng., etc.), the demand for a higher degree is easier to understand. You'll get more applicants as a result. What I find interesting is the ability to study a highly technical field, get a Ph.D. and then find next to no employers in that field within the country. That's what I ran into in the UK - by the time I was out of the Ph.D. grind, the relevant employers had all left for greener shores. I had to follow them. As a result, the UK has benefited very little from the expenditure on my education. Even before undertaking my Ph.D., it was very clear that working in academia was not going to be my thing. The lack of resources and funds stood in stark contrast to the facilities available for similar R&D within companies. Having to fight for funding every n years was far from appealing. This is from a UK-perspective, but I hear similar concerns from those in the US.
The current focus on "relevant research" and turning university labs into money-making operations is part of the problem. While it's couched in terms of universities "Making Money" and "Doing something useful" (as the TFA appears to want), in practice, it means that university researchers pair up with private industry, doing only the things that private industry deems important (=incremental and rarely disruptive). Grant programs amplify this trend ("What are the industry applications of this research?", "Was your last research project a financial success?"). So, if the universities are paying researchers to do private-industry research, private industry has less incentive to fund its own research. As a result, we're moving from a system where we had academics engaged in fundamental research, with often disruptive results, and a thriving private industry research community, to one where a smaller pool of public-private academics do the bidding of private industry.
Too many Ph.D.s? You bet. In the name of "solving practical problems", we've moved industry research into the universities, and killed off fundamental research.
Note that "Columbia professor Mark C. Taylor", pontificating on how research has become too specialized and non-understandable to the public at large, and "must design curricula that focus on solving practical problems, such as providing clean water to a growing population" is himself a Professor of Religion. FTA:
"Mark C. Taylor is chair of the department of religion at Columbia University in New York and the author of Crisis on Campus: A Bold Plan for Reforming Our Colleges and Universities (Knopf, 2010). e-mail:mct22@columbia.edu"
Sort of easy to predict that, in fact. Because you know what? A person doing real, cutting-edge research, developing insights that no one else ever has before in history, is almost by definition going to be non-understandable by other people -- at least until such time as their research becomes diffused and more accepted by the mainstream. The call to "nourish cross-disciplinary investigation... focus on solving practical problems" is a thinly-disguised attack on basic scientific research. It's classic short-term thinking; if you demand profit/practical solutions right now, then the basic research that develops heretofore unimaginable solutions tomorrow will not be done.
Now, there's a lot of problems with PHD employment prospects, etc. But this is pretty damned skewed by how exceptionally non-useful this guys' graduates in philosophy and religious studies are. (I say this as someone with degrees in both philosophy and STEM.) I might suggest actual solutions would include: (a) Mandatory clear information provided to prospects about career and employment prospects, so they can make their own decisions on priorities. (b) Rollback the corporate-minded administrative takeover of higher education from faculty. (c) Return most teaching positions to being full-time tenured, instead of part-time contingent faculty as we have today, etc. The "make education practical/profitable" effort has been going on for 30 years, what we have now is the result of it, and it's time to stop digging the damn hole any deeper.
We know where leadership by an anti-intellectual "strongman" who scapegoats minorities and likes boisterous rallies goes
It would be hard to argue that group theory was relevant when it was developed. Or early number theory. Maybe you'd have liked Einstein to have given several applications for his theory of relativity (hint: it was before space flight and GPS). Or how about quantum mechanics. How about modal logic, that was merely an academic curiosity before Tony Hoare and a host of others came along and made it relevant, relevant enough for Intel to care about mathematically proving facts about their chips.
Science is a web of ideas, start pruning before you even know whether something is useful is stupid and short-sighted. Here's a thought, science can chew gum and walk at the same time. It produces relevant stuff and stuff that you will not think will ever become relevant...until it does.
This is not US-specific, it's like that in all western countries.
And it's actually meant to be that way. The academic world is the only place where fundamental research can be done, since the private sector has no interest in research that do not have direct applications.
If you want to do practical research, work as a R&D engineer in the private sector.
Also, when Einstein published his theory of general relativity, nobody expected this to ever become relevant for anything beyond pure curiosity. Well, that's because nobody thought of GPS back than.
And when he was arguing against completeness of quantum mechanics, there's no way he could have imagined that his thoughts would one day lead to quantum cryptography.
When Kepler thought about the movement of celestial bodies, he would never have guessed that his insights would one day help with weather forecast.
When Heisenberg and Schrödinger formulated the equations of quantum mechanics, they didn't think of TV sets, computers, or the internet.
The inventors of the particle accelerator thought about studying particles, not about cancer therapy.
The Tao of math: The numbers you can count are not the real numbers.
I'd like to speak on this matter as a graduating Psychology undergraduate struggling to get into a PhD program. Professor (Doctor?) Taylor raises an excellent general point. I'm not sure I agree with his entire view, as I am admittedly too short on time at the moment to read his entire article. That aside, I just wrapped up my Honors Thesis. It was an in-depth look at the state of youth suicide treatments, preventions, and interventions. My research conclusively led to one point: academia knows insane (pardon the pun) amounts about suicide itself. It has been so focused on the quest for knowledge that the focus of the science has been lost. There are few, if any, empirically supported treatments/preventions/interventions much less supported by longitudinal data. Perhaps I'm overgeneralizing, but I feel this issue has overtaken the sciences as a whole. Academia has become a self-contained system. We dig and dig and dig, research every aspect of every subject, publish it in dusty old journals that get crammed into a library shelf, and it never actually gets USED. We don't apply what we know to anything practical. Certainly this isn't universally true, or we wouldn't have seen any innovation, but I feel that it is a growing problem within academia. I look at the researchers in my department and I see loads of statistics and data produced on a daily basis. It gets crunched and analyzed, applied to a hypothesis, printed onto a poster or in a journal....and that's the end of it. It isn't actually used. My department recently churned out a rather impressive study on tattoo stigma. Long story short? It exists. Woo. Published. The data was recycled for a couple other studies, which were in turn made into posters, won a conference award, and.....then what? The data isn't used for anything! Why are Psychologists not working with advertisers or equal rights groups to implement a program to alleviate the stigma? This is just one example in a sea of millions. Anyone else feel the same?
Taylor is really just advocating a return to the Dark Ages, where monks could sit around at ponder philosophies at little expense to their feudal masters. While that might be OK if one's major concerns are debating just how many ferries dance on the head of a pin, this is not true for science. In science, mathematics, engineering and medicine, such specialized technical training is absolutely essential to even begin to understand the issues at the frontiers of science and knowledge. There is simply no way anyone can predetermine what odd fact or phenomenon will be at the heart of the next breakthrough nor learn enough fast enough not to specialize. Who would have thought that the properties of dielectric materials would spawn entire industries and revolutionize the way people communicate when they were first discovered in the 1840's? If you read the comment section of the article, Igor Litvinyuk's response was right on target.
What Taylor calls for is really a dismantling of funding for science under the ruse that it is hurtful to students. It is not at all surprising that Taylor points to the collapse of the research economy in the 1970's. Since this was precisely when the philosophy of Ronald Regan came into being, where "government is viewed as the problem" and the solution is for all power and wealth to be ever more concentrated into the hands of a few ultra-wealthy so that it can "trickle down" to the more deserving. Taylor's piece is little more than a call to return to the Dark Ages, where more and more money that otherwise might be spent on education and expanding the frontiers of knowledge that can be used to solve humanity's many pressing problems go instead toward yet another tax break for the wealth and an other special handout to the already well to do. They want to "reform" the PhD system because there are not enough jobs, by dismantling it. Same old sham, just repeated once again. One would think eventually people would be smart enough to recognize the consequences of such a disastrous philosophy being applied once again to yet another segment of our society.
They want reform because they fear the consequences of a lot of smart people sitting around thinking there has to be a better way. It is a threat that focuses attention on the real cause of the failure in the lack of jobs. Namely, that the ultra-wealthy, in whom we after nearly 40 years of the philosophy of Reaganism have consolidated virtually all the wealth and power, don't want to spend their money on advancing the frontiers of knowledge that might contribute to the solution of the myriad of problems plaguing society, they would rather spend it on themselves and upon maintaining their special, most fortunate status. Unfortunately, it is this system that is truly unsustainable, since the planet groans at the weight of billions all trying to achieve the same status. On such a planet, humanity will only survive if every job soon requires the skills inherent in a PhD. We need more PhD's not less. We need more education not less. To accomplish this we need less concentration of wealth to make it happen. We need more PhD's and fewer crusading monks who only seek a return to feudalism and a return to the Dark Ages. If you really want to solve the PhD job problem, not to mention most other societal, political and environmental problems work to end the consolidation of wealth in hands of a few not educated enough to recognize or just too comfortable not to want to recognize the danger inherent to humanity in the philosophy of Reaganism.
In Physics at least, specialisation can lead to some very useful and broadly applicable findings. Granted, sometimes completely unexpectedly.
Indeed. In the Sciences and in Engineering, a PhD is the equivalent of a "driving license" for doing research. It does not guarantee you'll be good at it, but the odds are much better than for someone lacking the qualification. It signifies that you can plan and execute long and intellectually difficult tasks in a particular field, which may include discovery of new knowledge (experiments) as well as detailed physical and mathematical analysis. It shows that you're qualified for certain types of demanding job, which are not in particularly short supply. A PhD in physics or engineering was a prerequisite for my job and for several of my colleagues, and we're in industry, not in academia.
TFA failed to delineate the subject matter, lumping all PhDs together as if physical sciences, bioscience, and engineering suffered from the same lack of utility as the humanities or social sciences. It appears that TFA really just dealt with the humanities which tend to have limited economic applicability (PhD in Religion, or in History of art, or in Etruscan statuary). In some cases they amount to little more than an expensive hobby.
Those who can make you believe absurdities can make you commit atrocities. - Voltaire
No one's delaying their release into the workplace to get a PhD so that they can make a better contribution to "the world," period. People pursue a PhD so that they can stay in academia, where they are comfortable and proficient, and make as much money in academia as an academic can.
I am working full time while obtaining my PhD. I am getting the PhD because it is teaching me to do the things that are required, and that I cannot learn elsewhere. While I have not delayed my entry into the workforce (I like money), one of the reasons that I am getting it is because I want to make a contribution to the world. Everyone has goals in life, and while some people have goals like "own box seats to the Packers", "pay for my grandchildren's college", and "backpack through Europe" others have goals like "make a difference in the world". These are what you want out of life, and I find your derision of "help the world" to be insulting.
Since academic institutions profit directly from the milling of PhD degrees
The idea that academic institutions make any money on PhD students is downright false. The fact of the matter (and I've spoken with numerous professors/advisors about this) is that "suckers pay for their PhD". This is a direct quote from Dr. Kapoor (http://www.nanovk.com/), who has had 40+ MS/PhD students. Nearly everyone obtains funding from a number of sources (I've only met one person who didn't, and they just didn't try), including:
1 - work on a grant project (if you do your dissertation on an aspect of the project)
2 - RA work (live in the dorms for free, get tuition comp'ed, and get little-$ for it)
3 - TA work
4 - the school itself
5 - their work (full time work/part time school)
6 - Work program (work pays you go go an get skills they are interested in, owe time afterwards)
7 - governmental aid program (non-loan)
8 - grant program/award (NSF or the like)
9 - outside agency help (NAACP or whatever)
10 - outside governmental involvement (foreign government sends people to America to be educated, brings them back afterwards)
Keep in mind that many of these program stack. You can sign up for RA work (free place to live and money) to have your tuition paid for (easy), get a NSF grant (not easy), work on funded projects for your major advisor (very easy), and get a bit of outside agency help (moderate). Of course you have to produce through this time.
Also, getting someone through their PhD is incredibly time-consuming on behalf of the professor and organization. Although the school is compensated for the classes, they have to compensate the student for project work. Then, they get to foot the uncountable-but-still-very-real cost of advising PhD students (~2 hours/week at ~$100/hour = ~$10K/year for 4-5 years) with professor time.
Precisely why should we emphasize cross-disciplinary research? What is the evidence that this approach is better than more narrowly focussed research? I would agree that we have too many PhDs, too few jobs for them, and or too little incentive for real innovation. I would also agree that the system needs reform. I don't agree that we should all be doing cross-disciplinary research.
Are you a virtual scientist if you work on a computer?
Comment removed based on user account deletion
He has a Ph.D. in religion. The headline should read "Person with worthless Ph.D. argues that all Ph.Ds are worthless." There is plenty of room for folks with Ph.Ds that actually train them to do something.
Should men who "lie together" be killed? How about unfaithful women? It's in your bible... just saying... (okay, it would be fair to mod me -1 troll for this but I just can't resist the taunting...)
I never figured out why things went wrong
NIH started as a means to support fundamental biomedical research. It expanded from $4M in 1947 to 100M in 1957 and $1B in 1974 and $30B today. It became the way that biomedical research is funded, and dwarfs the NSF budget of $7B. Everybody wanted a piece of that pie, but it turns out to be tied up with political strings. Universities came to depend on research money that often exceed student tuition and state grants. But it's hard to justify basic science to congress - that's the whole reason NSF's budget is so much smaller than NIH - so NIH has been progressively steered towards clinical, applied, "translational" research. Other branches of science have been pushed in that direction, too, as they struggle to justify their existence next to curing heart disease and making the lame walk.
The argument for Government funded basic science used to be that we couldn't know what would come out of it, but that the simple process of discovery would result in unforeseen benefits. Society couldn't trust commercial enterprises to take such altruistic risks (although some of them did consider support of long-term, fundamental research part of good corporate citizenship or part of their own 20 year success program). Government now, at least in the US, has little foresight or capacity for long term planning. If the corporate attention span is one fiscal quarter, then the government attention span is one election cycle. So, we've sacrificed our long-term prospects for short term reward.
Don't eat the marshmallow yet.
"The idea that academic institutions make any money on PhD students is downright false." They're not training you out of the goodness of their heart. If Universities were not producing a glut of PhD candidates and graduates, how much would it cost them to hire labor to run the labs / discussion sections / classes, etc.? I didn't pay for my PhD, but I worked in my advisor's lab as an RA. I was making about $500 a month via stipend (this was years ago) and we used to figure that conservatively we were being paid about $3-5 an hour for our lab work. After getting my Master's I was teaching classes. I would teach anywhere from 40 to 325 students in one of my classes. The most I was ever paid for this was $2500 / class. Compared to the compensation package of a tenure track professor, I was a bargain! Thanks to the glut of PhD students you could get the teaching of a tenure track professor done for $15,000 / year. Face it, PhD minions are a cheap, exploitable (you don't like it, we've got 5 other applicants who would gladly take your place) labor force.
There is much more to it. Grad students have their tuition paid out of their supervisors' grants. Additionally, there is a 50 to 60% overhead on the grant, including this tuition payment as well as grad student salary and benefits. A university certainly makes more money from hiring a grad student than the student does.
No. It is a problem with her advisor and the other mentors she has had at her school. Professors (and senior graduate students) should be teaching students that you are learning many things as part of your Ph.D. training: How to do research on a problem of interest, how to find a problem of interest, how to write a paper backing up your research findings, how to give a presentation of your research findings. These are all important. But they aren't the end.
You also need to be learning: How to explain your research to experts in your subject (department), how to explain your research to others in your discipline (college) [ We called this the "elevator talk" ], how to explain your research to those without a complete background (other graduate students), how to explain your research to students majoring in your department, how to explain your research to john q. public [ We called this the "airplane talk" ]. How to find funding and how to make a report on the results from funding.
Finally, and not every Ph.D. program gets this by a long shot, you need to be getting practice in teaching subjects both in your specialty and adjacent to it to students.
People can be narrow specialists in their research and still accomplish all of these things. I work with many of them, and I went to school with many others.
You don't have to understand the fundamentals to make use of the peculiar, but repeatable observations.
That "limb" you are going out on is called "ignorance". The components engineers "stick together" were invented by physicists based on discoveries arising from a deep understanding of quantum theory. Without quantum theory we would never have thought most of solid state electronics possible, much less been able to hit upon just the right combination of materials and dopants to create working transistor junctions.
Columbus discovered America by a combination of chance and ignorance (he had the diameter of the Earth wrong) but America is really really big and hard to miss. The quantum phenomena that solid state electronics depend on are subtle and hard to detect, much less design for.
Blasphemy is a human right. Blasphemophobia kills.
Also, when Einstein published his theory of general relativity, nobody expected this to ever become relevant for anything beyond pure curiosity.
Sure. It also served to explain some of the huge problems that held up progress in physics of that time.
No. You are thinking of special relativity.
When Kepler thought about the movement of celestial bodies, he would never have guessed that his insights would one day help with weather forecast.
But that did have obvious applications to navigation and time keeping.
No. The existing epicycle theory was unsatisfactory as explanation, but hugely successful as description. Besides the fact that I strongly doubt that the planets ever served a big role in navigation. But since you claim obvious applications, you surely can explain them, right?
When Heisenberg and SchrÃdinger formulated the equations of quantum mechanics, they didn't think of TV sets, computers, or the internet.
But that did have obvious applications such as basic electronic components (such as vacuum tubes), medicine (X-rays, fluorescent dyes), and more powerful microscopes.
Vacuum tubes don't need any quantum mechanics. They can completely be described with classical electrodynamics.
X-rays were found before even the old quantum mechanics of Planck, Bohr and Sommerfeld was formulated, much less the modern quantum mechanics by Heisenberg and Schrödinger. Indeed, even the first medical applications of X rays was before 1900, long before Schrödinger's and Heisenberg's formulation of quantum mechanics.
The inventors of the particle accelerator thought about studying particles, not about cancer therapy.
Maybe. That was a pretty near future application of the technology. Besides testing fundamental physical theories is rather important, don't you think?
Of course I consider testing fundamental physical theories rather important. It doesn't however give the immediate, obvious payoff for humanity which is demanded
I see the usual flaw, blatant mischaracterization of the past in order to rationalize poor science practices in the present.
I don't think I mis-characterize the past, and definitely not intentionally. And I didn't mention at all any science practices in the present, poor or other, so you'd better speak abou which poor science practices you accuse me to rationalize /I'm not aware of rationalizing any; I wasn't speaking about science practices at all, but only on the wrongness of the concept that every research must have obvious applications).
I just have two questions to ask you. First, is there a way to distinguish the quality of research as in its benefit to humanity or particular subgroups? Second, if there is, then why not fund the better quality at the expense of the poor quality research?
First: There is a way to distinguish quality research from non-quality research, but that is not about "its benefits to humanity or particular subgroups". Quality research is research which is likely to lead to new insights (which may or may not turn out to be useful for something), while non-quality research doesn't deliver much insight.
Second, I'm all for funding better quality research. I'm not for making the funding dependent solely on the expected usefulness of the results. Not that I'm against funding research for things which have immediate use; that would be silly. But something not having immediate or obvious use should not be a reason to not fund it.
The Tao of math: The numbers you can count are not the real numbers.