A normal process of learning in a graduate school looks like taking courses that include a broad perspective on some subject. This also includes the instruments of control: exercises, tests, problem sets, and exams. Let's suppose this control exists only to show where a student has gaps in his knowledge.

A student takes such courses to learn things she will apply later in her work.

But in practice each course include much material that the student will never apply in her work, even in academic research.

Many notable scholars mention that they pick up necessary tools, while disliking the idea of a university that offers too much in general, so students have to disperse their efforts.

Why does academia use this push process of fixed course program and control for all students in the course, instead of a pull process, when students have their own research interests and just pick necessary tools when the need arises?

Shouldn't students in general use a selective, not linear, approach to getting necessary knowledge? It's obvious that students need some overview, but what prevents them from getting this overview as they work through the problems of their own?

  • 14
    It's true that most students never apply 90% of the material in any particular course. But it's also true that no student can predict in advance which 10% will be the most useful in the long run.
    – JeffE
    Oct 20, 2013 at 12:21
  • @JeffE Do they need to predict those 90%? Maybe they can find the right 10% exactly in the moment when they need it to progress in their research work? Think about a table of content in a book: you need not to read the entire book to understand what chapters can help you. Oct 21, 2013 at 4:01
  • 2
    In general, formal education assumes that you won't learn something unless you have to (and you take a course in it). Which works only for the ones who are not motivated... Oct 21, 2013 at 4:21

4 Answers 4


My immediate answer is they shouldn't. The problem is that there are differences between graduate systems. Aeismail has provided a good answer for one such system. Where I work each graduate student has an individual study plan and the idea is to tailor make the studies for each. this tailoring involves taking parts of a course and not for credit in the sense to go through examination. In our system all courses at graduate level is pass/fail. To understand the differences one has to look at the position of a graduate student as well.

In some systems a graduate student is more or less a bit more than an undergraduate (a "super student"). In my system a graduate student is pat of the employed body of the department and in that sense also a colleague. The point is that this involves both rights and responsibilities and further involvement in department affairs. This position has old traditions where a PhD was not awarded until much later in life, the term graduate student simply did not exist, there were, for example, intermediate teaching positions for those who had not yet completed the PhD yet. The thinking about courses is thus to ensure that the student does not waste much time on unnecessary material, just as you stated in your question.

So from this position, there is no right or wrong in terms of courses, the PhD programs differ so much between systems. The reason for the differences comes from deeper differences in the university systems and views on what the PhD student profile after completion. We had discussions about mandatory (tailor-made) course for students to provide a common base in the subject, but we have scrapped this approach because the field is simply to wide and then it will be up to each to read upon their sub-field.


Affirming the remarks in the two earlier answers, and continuing some of the themes:

At least in mathematics, although it is vitally important to have direct, sincere enthusiasm about projects and questions, inevitably these are not well-informed from a beginner. That is, projects grounded in ignorance of established ideas, while temporarily often very beneficial, simply aren't "professional-quality" projects. That's fine, temporarily, but certainly not in the long run.

So, in addition to "common culture" awareness, as mentioned earlier, there is the genuine and awkward problem that beginners may not have an inkling of what they'd like to know if only they were aware of it. Can't easily learn a thing if one isn't aware of it ... perhaps all the more crazily so if there really is a wonderful new concept that is unfortunately more than one remove from the familiar.

At the same time, I do agree with the questioner that all-too-often the "required" courses take a too-adversarial approach, too much homework, too tricky exams.

Certainly varying depending on the subject and on the program, many graduate courses are given in a not-so-adversarial fashion, with "grades" playing no serious role. Rigidity and an adversarial attitude are not universal.

But, again, yes, encountering an adversarial and coercive attitude is polarizing and unfortunate. Creates a bad attitude, and this can be a long-term ill, also.

Equally dangerous, though, is concluding too soon and too strongly that one is "ready to go", when, in fact, one is not merely missing some important things, but perhaps unaware. Sure, one should have enthusiasms, and follow them, but be absolutely sure to maintain at least one other thread in which one looks around fairly broadly, cultivating serendipitous encounters with helpful things one may have failed to imagine.

(Required courses are a clumsy approach to the latter, yes.)

  • I agree with this answer, especially the third and seventh paragraphs. The strategy of picking up necessary tools when you need them presupposes that you're aware of the existence of the tools and of the sorts of things they can be used for. Mar 26, 2016 at 23:01

You're presenting a false dichotomy: most graduate programs are not completely specified. There are a certain number of required classes—to ensure that all students have the same "core" knowledge that they will be expected to have, both as graduate student researchers as well as professionals in the field—and usually a number of elective courses, that can be chosen as the student chooses to meet the needs of their research, or to satisfy their interests.

But if you're really asking "why is there a core curriculum," it's because, as I mentioned, there are expectations of what an advanced degree holder in such fields will know. However, there are large discrepancies and variations in curricula between departments at the undergraduate level. Therefore, to "level the playing field," and make sure that all students have the expected knowledge and skills, they offer the courses that will guarantee proficiency at the required level.


This is a good question. I think the idea is to make sure, as aeismail says, that all grad students have the same core knowledge after having gone through the program. However, in practice this approach has major problems.

  1. Many disciplines are so big that it is often not clear what the core is.

  2. Additionally, some areas are changing so fast that even the core becomes outdated.

  3. Also, my experience with going through such courses is that it doing homework
    problems does not really make one understand the material as, say, working on a research problem does. Also, a homework problem might be a page of argument which is by definition well understood. This is poor preparation for research work (assuming that this is the goal) because a research paper is more than one page long, and does not consist of well-understood material. In practice one spends a lot of time when writing a research paper worrying
    about such things as graphs/figures, suitable notation, and organizing the material. These are issues that simply never come up in course work. Even an
    end of semester project in practice is not big enough to provide a suitable ground for exercise.

  4. Often different parts of the core (depending on the subject) are rather
    different from each other, and as you say, a graduate student, of necessity, is forced to specialize quite quickly in one area, and will likely never need to know about those other areas. So, in practice, much of this is time wasted.

I think part of the idea is that someone who is excessively specialized is not in a good position to cross-pollinate ideas across different disciplines. The very idea of interdisciplinary work is that ideas in one area are useful in others. But I'm not sure that forcing a set curriculum has the effect of creating a more rounded researcher. I think such adventures are best driven by curiosity and need.

Additionally, my experience is that beyond basic/core courses in a department, the more specialized courses may just reflect the (possibly non-mainstream) research interests of the faculty, and can be a real waste of time if such courses are made a requirement.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .