I'd modify this a bit, since a PhD candidate is still a student, and still works under a teacher. The purpose is to learn how to produce original research, by doing it under a mentor. The mentor has some responsibility to guide the student away from making themselves so specialized and unemployable that their first research result is their last.

Now, I think that as students move up the ladder, they are afforded greater responsibility for their own careers, and greater freedom to -- frankly -- make mistakes. A PhD student should be able to figure out that a job in business might require being conversant in business. But they should also be free to master a skill that no employer cares about, with eyes-open awareness of the risk to their careers.

On the other hand, a student could also think strategically about their research project and employment prospects. I had a physics thesis project with a particular research goal, but considerable latitude as to how I could carry it out, and I steered the project through a series of obstacles that "forced" me to become demonstrably proficient in areas such as electronics and programming, which also happened to be my hobbies. In other words, I tackled a specialized problem by being a generalist.

My father gained his PhD yet was and is a generalist. He went to work in industry. Perhaps this gave me a natural role model, and a background that enabled me to manage the same transition.

M.Sc. gives you the background to produce the research, but just like in the industry, you will learn while on the job. You could be doing equivalent work outside the academia with your M.Sc., in an R&D department of some large company.

As for specialization, you must be extremely specialized to advance your field, which is a requirement for a Ph.D. The point is not to be hired because of your particular speciality, the point is to earn the highest "license to practice science".

Most PhD students don't have that choice, and that does seem like something of a problem. Realistically, most of them aren't going to have the opportunity to enter a meaningful tenure-track program and stand a real chance of becoming a full professor. Pushing a bunch of people down that track without providing any support for the ones who aren't going to make it seems pretty unreasonable.

I'm not going to argue that that means we should dilute PhDs. I'm fine with the current degree remaining as it is. But we should stop encouraging people to do PhDs just because there's a chance that they're going to get into academia. I gained a lot from my degree, but I could have been taught that without the aspects that were tied to the assumption that I'd spend the rest of my life in academia. We should work out how to provide the more generally applicable aspects without making people feel like shit for doing that and then discovering that they don't an academic position to take up.

A PhD is either a segue to a job in that field, if there is an industry (like pharmacology) or to launching a new segment of industry (as in computer technology) or a non-pecuniary pursuit of folks who are intelligent and financially independent.

see: figure 3, p7 http://www.aaup.org/sites/default/files/2011-12Economic-Stat...

[1] This data is cited above.

A "university" is not a single actor. PhD applicants generally apply to advisers or departments, not universities administrators.

Do they have the time?

My undergrad program was extremely... focused (a word the administration would no doubt use to describe it). In my entire 4.5 years I had exactly two non-engineering electives, the rest of the schedule was a full stack of "core" courses. This was not atypical at my school.

The result of which is that you create a bunch of people who are very narrowly educated. I dodged that bullet with a combination of suicidal course loads (e.g., taking a 7th course in a 6-core-course semester) and class-skipping. Neither were good solutions and I only made it out by the skin of my teeth.

The problem here is that the definition of a Ph.D. for you is extremely different from the definition of Ph.D. for most of your students. I think there's next to no one whose primary objective with a Ph.D. is to produce a piece of original research - it is always a means to another end: usually career-oriented.

There is a dramatic misalignment of what you think you're doing there and what your students think they're doing there.

The purpose for doing a Ph.D. is to produce original research.

Motivation can vary: intellectual curiosity, desire for employment (academic or otherwise), because their parents demand it, because "Dr." sounds impressive when hitting on strangers at bars... I'm fine with any of the above, as long as the student is willing to do the work.

The students might love research, or perhaps they feel as if they are being made to jump through hoops --- but, unless the students are getting their information from people who aren't actually familiar with academia, there's no misalignment and our students are indeed getting what they came here for.

I dropped out of my PhD when I realised my strongest motivation - my actual strongest motivation, no jokes - was to be a "Dr." who dressed downmarket. This is not a reason to choose a career path. I'd fallen into the PhD program by accident and it had never been a particular ambition of mine...

People in my program take Spanish. I knew one person who did creative writing. We actually have a class on consulting in a closely related department that you could manage to sell as an elective. I took some only faintly related classes in another department.

It's not easy, and it will add time (that's the true problem with a PhD, balancing 'Things I want To Do' and 'I want to finish') but it's doable. PhDs are also often much less "locked in" to coursework than undergraduates, especially after their first few years.

"The problem here is that the definition of a Ph.D. for you is extremely different from the definition of Ph.D. for most of your students. I think there's next to no one whose primary objective with a Ph.D. is to produce a piece of original research - it is always a means to another end: usually career-oriented."

I can't think of a single person in my program who wasn't fully aware that a PhD is about producing original research. They might struggle with doing that, but it's a known and understood requirement.

There's a career-oriented degree. It's a Masters.

The Ph.D. program, in our times at least, is pitched as a quid pro quo: you produce some original research, something worthwhile and of substance, to add to the sum of human knowledge, and in exchange they gain access to an academic career.

The problem is that there are a lot of people paying the quid, and barely anyone getting the quo. Unless you are a student of independent wealth, you are not producing this research out of the goodness of your heart with no greater, more important objective. To the vast, vast majority of people doing their Ph.D's, a Ph.D. is a career maneuver.

PhD students generally just love research. We sign up for this because it's what we enjoy doing. Companies try hard to recruit us away, but we stay in spite of that, because why would we want to do anything other than work on the one problem we've always wanted to solve?

Maybe it's different in the humanities, but every STEM PhD student I've ever met has low expectations, with the exception of one-- and she has a first authorship on a Nature paper and is working on a Science submission, so I still think she's pretty well-grounded.

Honestly, at this point, finding out that the job market is grim takes only a shred of research, and absolutely the kind of research you should do before applying.

I don't know exactly what I'm going to do when I'm done.

Granted, not everyone thinks like me. But certainly not everyone thinks as you do.

Anecdotal rebuttal: The first conversation I had with my research advisor concerned the difficulty of getting a tenure track position at a research university straight out of Ph. D. school. In her words, "that sort of thing just doesn't really happen anymore in this field (Computer Science)."

There are no delusions of academic grandeur, at least in my neck of the woods. I agree that the Ph.D. may be pursued as a career maneuver, but if your end game is a teaching position at a research university your approach to the program has to be very different (and extremely more ambitious) than someone pursuing industry or other goals. And if you're pursuing a prestigious academic position, you are quickly educated on the slim possibility of attaining it, even if you're willing to post doc.

</bitter>

I work in academia here in Spain. The Masters is usually a prerequisite for doing a PhD here.

There appears to be a dramatic misalignment between what you think a PhD is for (seemingly, to secure a job) and its actual purpose (to advance human knowledge in some area).

Next you'll be telling me that people get degrees for education rather than as a signal to employers...

That would be a perfectly sensible position if not for the fact that they kick you out of PhD school after you finish instead of promoting you.

A PhD is an apprenticeship in doing original research. It is not your first job. It does not pay well, or have any job security or benefits. Your first job is the one you get after your PhD, because they punished you by accepting your thesis, printing your degree, and kicking you out of graduate school, thus forcing you to find an actual position.

Which means that if the job prospects after PhD school are "nothing permanent involving research", then the system is indeed broken. It's broken because thousands of people spend years of our young adulthood working incredibly hard on a single project, only to be told at the end that we haven't really ever held a job, that we are not qualified for a job, and that figuring out what to do after five years of experience in the world's most difficult intellectual endeavors is our problem.

If grad school is not going to lead to, well, anything afterward, then we should stop considering it to be school at all, and just all go on strike to demand that our first real job out of undergrad pay a whole lot better for the hours it makes us work.

Promote you? Why would you expect that? A PhD is not a meal-ticket. It's just an endorsement that says you are very good at undertaking novel and independent research. There is no promise made any point that an academic position or related industry position will be waiting for you at the end. As you say, it's just an apprenticeship. Nothing more. It's up to you to figure out what, if anything, you will do with that education once you receive it.

Because that's what happens when you spend five years in one job and spearhead a large project independently.

You've successfully proven my point: it's an apprenticeship. The problem is, apprenticeships are supposed to be followed by a career as a master practitioner of the craft you apprenticed in. If there are thousands of apprenticeships in science each year, and no permanent jobs as scientists, then the guild structure is unbalanced and the guild is responsible.

So go be a master. Make your own way. A PhD is not a promise of a salaried future at an academic institution.

People who were planning to make their own way got bachelors degrees, people who planned to teach got Masters degrees, and people who planned to research got PhDs.

Now, you can of course keep repeating "make your own way, you dumb ignoramus who doesn't understand market imperatives!" until the cows come home, but all that means is that you consider market imperatives to override guild structure.

Which they have! But a descriptive statement about how the system has come to work is not a normative statement about how it should be made to function. Certainly I would say the guild structure is malfunctioning when job qualifications have gone from "produce one masterpiece (a PhD) over a few years and you can get a job" to "produce two to three masterpieces (a PhD and one or two post-docs) over a dozen years, and then you're considered just about qualified to apply for a permanent job".

By the way, we're certainly not talking about my personal plans, since at this point I simply don't anticipate any currently-known career structure actually lasting. Even if nothing revolutionary actually happens on the political, economic, or environmental fronts, the continued subjugation of all institutions and enterprises to short-term market imperatives (the ones you support ;-)) means that job tenures longer than a few years probably just won't ever be available to me. I'm living in a world built by people who think Thomas Friedman has the right idea.

EDIT: It's important to note that if we really believed in fully and entirely replacing the Guild of Science with a scientific labor market subject only to market imperatives, it would mean taking a lot of funding from the university/lab administrators and tenured PI's, and using it to create a less-strictly-pyramidal career structure more like the tech sector has: bachelor's degree, junior-level researcher for a number of years at full pay and benefits, senior researcher at full pay and benefits with real job security, management and leadership.

In a real market system, you don't get to work people to the bone for years at a time for hourly wages less than those at Wal-Mart solely on the promise that they will someday have the opportunity to gain a 1/3 chance at a real job in their field. In a real market system, that is called an unpaid internship, it's illegal, and it ends in your workers unionizing and suing you.

Why shouldn't someone be allowed to enroll in a PhD program if they want to conduct research in the field they are interested in?

At the same time as dramatically expanding the number of higher-level positions, yes. The pyramid must be made into an ordinary tower.

Too bad, so sad.

True this. Almost every university I've dealt with loves for their math and science Ph.D. students to take a business class or two. But requiring it would be outside the point of a Ph.D. Honestly, most Ph.D. programs have a couple core classes, then they are pretty flexible about what classes you take as long as you can make up an excuse.

Additionally, one friend getting a Ph.D. really wanted a job in science journalism. He found time to intern for a newspaper doing the journalism while still a candidate. Plenty of others build up a portfolio just blogging on the side. If you come out of a Ph.D. wanting to be a writer, it's your fault if you don't have a well put-together portfolio.

I learned to how to write and to enjoy writing in my PhD program. Writing and communicating research ideas and results coherently is actually one of the primary things a PhD does...

It starts with writing and business. And of course you'll need a course on teaching as you'll be teaching. A course in research tools like statistics and designing surveys will help you avoid some potholes. You'll need to reference things properly and avoid plagiarism, better have some training for that. Everyone in this field uses (Matlab/Excel/Latex/MS Word/SPSS/R/NumPy/Fortran/EndNote/BibTeX) so we'd better train you to use that. Effective study strategies will be vital, as will effective note-taking skills. You'll need to know about time and project management, better have a course on that. Of course you'll need to know how to use the library, research databases, and how to make document supply requests. Writing papers and working out which journals and conferences to target is important of course. Presentation skills and public speaking are vital in all manner of jobs. Not to mention career planning, networking and professional development. And you can't be a public academic these days without managing your online presence. Oh, your project involves spending money? You'll need training in managing a budget and working the purchasing procedure...

If you want to work at a hedge fund, a masters in financial math is a lot more efficient than a Phd in Physics.

Tell that crap to the tourists, the purpose of the Ph.D. is afaict to get the student a high paying job that requires a Ph.D. (because existing phd's say you need one), and the professor tenure and grants.