An article is making the rounds about how getting a PhD is easy. It was originally published a few months ago but it continues to circulate and I just saw it this week. The article, written by a recent Cambridge PhD, makes a number of dismissive statements about PhD education. These statements rub me the wrong way, so I feel I have to respond. As somebody who has completed a PhD myself, has supervised or co-supervised approximately 20 PhD students, and has been a member of countless PhD committees, I think I know a thing or two about getting a PhD. I am quite confident that getting a PhD is not easy. Even the strongest students find it at least somewhat challenging. And most find it substantially challenging.
Let’s review what a PhD actually is and what it means to have one. First, the only thing you can infer from the three letters PhD behind somebody’s name is that they have completed a PhD program. As much as people might want to infer other things, such as intelligence, ability, or knowledge about a particular subject matter, a PhD degree does not guarantee any of these things. This is by design. A PhD is an individualized apprenticeship program under the tutelage of an expert scientist or academic. So who you’re doing your PhD with matters. What you’re doing during your PhD matters also. And the people who routinely hire PhDs (such as professors hiring postdocs, staff scientists, or faculty members) know this. For example, I wouldn’t hire somebody just because they have a PhD in computational biology. That’s literally the minimum requirement for me to even look at their CV. What I’d really want to know is who have they trained with, and then of course, what research they have done. How much and what did they publish? Do their papers have impact? Are their papers written well, and are they employing state-of-the-art tools, moving the field forward, or otherwise reporting impressive work? A PhD with one contributing authorship in a minor book chapter is not the same as a PhD with six first-author publications in top journals.
Now it is true that one of the most difficult parts of completing a PhD is getting admitted into a program and lab in the first place. Once a student is in the program, most faculty members will do everything—to the extent legally and ethically permissible—to drag the student across the finish line. Nobody wants to see a student fail. And given enough encouragement and support, most students can eventually bring their PhD to completion. But I can guarantee you that the marginal students, the ones that require dragging, will definitely not say that their PhD was easy. And neither will the majority of the stronger students.
PhD training, when done correctly, adapts itself to the student’s level of competency. A marginal PhD student will objectively not write a particularly impressive thesis, whereas a top performer—the type of person who goes on to become a faculty member at one of the best universities in the country—will tackle a much bigger and more challenging problem and so will encounter many more difficulties than the weaker student. The result is that while the objective quality and quantity of work performed may differ substantially between the two students, the perceived difficulty and effort will likely be much more similar.
In fact, it is the job of the PhD advisor to ensure students are appropriately challenged. A good PhD advisor will know exactly what the strengths and weaknesses of their students are and will push them beyond their comfort zone. No matter how talented, how smart, how accomplished a PhD student is, a good advisor can (and should) push them further. Maybe this didn’t happen in the case of the Cambridge student who found his PhD too easy. Maybe he didn’t receive good training. If so, I feel sorry for the wasted opportunity. But as I said earlier, when you’re doing a PhD who you’re working with matters. Some advisors provide excellent training and others don’t. Unfortunately that’s just what it is. The world is not fair.
The reason why a PhD is not easy is because it is a long, open-ended, independent research project. Nobody coming into a PhD program has all the skills and knowledge to complete such a project. And even if they did, they’d just be expected to try something new or different or otherwise challenging. And then stuff doesn’t work. Experiments fail. Life happens. You have a child. Your parent dies. And you have to work through all of this while you’re experiencing this massive burden of self-doubt of whether you’re actually working on the right problems and whether anything will ever work out at all. Almost every PhD student eventually reaches a low point where things seem hopeless and where they wonder whether instead of going to graduate school they should have opened that bakery after all.1 And then, somehow, things start working, you do get a paper or two out the door, and before you know it you find yourself defending your PhD. And in retrospect, it may seem that it wasn’t even that hard, and that if you had to do it again you’d know exactly what to do. You could bang out a PhD in no time. And you know what: You’re right. You could. But the next student couldn’t. Because they have to go through the process themselves to understand all its subtleties and nuances and how it connects to their own personal strengths and weaknesses.
The article suggests that too many students pass their PhD defense (which the article calls viva), and it seems to suggest that more should fail. This betrays a complete misunderstanding of how PhD training is supposed to work and what the purpose of the defense is. In my opinion, to fail a student at their PhD defense is totally unacceptable. It indicates a failure of the system rather than the student. The times to be selective are (i) at admissions2 and (ii) during the qualifying exam, whose primary purpose is to fix admissions mistakes. The author of the article received his PhD from the UK, and I can’t speak to their system. In the US, every PhD program has a qualifying exam, typically 1–2 years into the PhD, where students have to demonstrate basic competency and ability to perform the work expected from a PhD. Students can and do fail qualifying exams and are then asked to leave their program. Once a student has passed the qualifying, however, the expectation is that they will successfully complete their degree. Now it can still happen that they don’t, for example because they never manage to write the required scientific publications, but then they will not be allowed to defend either. I would never let a student schedule their defense if I felt they weren’t ready and there was a risk that they could fail.3 On the other hand, if they have made it to the point where they are in a room in front of their committee defending their PhD, the process should be a formality. There should not be anything the committee discovers during the defense that it didn’t already know previously.4
The article also talks a lot about intelligence. I agree with the author that a PhD does not indicate high intelligence. I have commented on this previously. However, the entire part of the article devoted to intelligence is a massive non-sequitur because it makes the implicit assumption that if a PhD doesn’t require exceptional intelligence it consequently has to be easy. As I’ve said before, a PhD is not an IQ test, and things can be difficult without requiring exceptionally high IQ. As just one of many examples, being a competitive gymnast is extremely difficult and yet IQ is not a major requirement.5
You may object to my example of gymnastics on the grounds that it is primarily a physical rather than an intellectual pursuit. But I’d argue that there are more similarities than differences between becoming a top athlete and a top scientist. In particular, becoming a high-level gymnast depends on the three pillars of exceptional talent, extremely hard work, and outstanding coaching. Remove one of them and instead of an olympic gymnast you’ll end up with a guy who is good at doing backflips in the local park. And yet, these three pillars can to some extent compensate for each other. A less talented gymnast who works harder and has better coaching may outcompete a more talented one who doesn’t work as hard or has a mediocre coach. Things are quite similar in science. To become a top-level scientist, you certainly need the talent (of which IQ is probably one component, as are other cognitive abilities such as executive function and mental resilience), but you also need to work hard and need good coaching (i.e., a good advisor). And, harder work and/or better coaching can compensate for lack of raw talent to a substantial degree.
Finally, as much as the article decries credentialism, the author displays a strong affection for this type of mental shortcut. Asking for PhDs “to be hard” so that every PhD holder is an “elite intellectual” implies not wanting to evaluate people for their individual accomplishments. It’s a desire for a simpler world where a PhD committee has decided who is a strong scientist and who is not so you don’t have to. I’m sorry, but this is hopelessly naive. People have different strengths and weaknesses and different trajectories. Sometimes people who struggle during their PhD go on to do amazing things and other times people blaze through their PhD and then flame out and afterwards never again contribute anything of note. A PhD is a journey of personal development. Nothing more, nothing less. To find out anything else about a PhD holder, you’ll have to read their CV, read their work, and most importantly, engage with them on a personal level. Yeah, I know. What a bummer.
If none of this applies to you, if you would describe your entire PhD as simple and easy, then I suspect that either you didn’t push yourself hard enough or you had an advisor who failed to do so. Or you’re just really lucky. For some people, by luck, everything just falls into place without them even trying very hard. If that’s you, spend a moment pondering how your life might have been different if at some key junctures your luck had not been on your side. Most people are not so lucky. Many have to fight every step of the way.
As example, the main program I draw students from has single-digit acceptance rates. We typically receive between 10 and 20 applications for every slot we can fill. The admissions process is extremely competitive, and it involves a lengthy written application and several interviews. Nobody just walks into a PhD program without putting in serious effort and often years of preparation.
I did actually have one student who almost failed his defense. This is a story for another day (but see also the next footnote). Suffice to say that I would consider this case a failure of the process rather than the student.
I realize different programs are set up differently. In all the PhD programs I currently participate in, students have annual meetings with their committee. They justify and explain their research to their committee every year. And they are not allowed to schedule their official defense unless all members of the committee agree that they are ready. I have heard of and seen programs that are not structured in this way, where the committee sees the student for the first time on the day of the defense, and indeed things can go wrong there. I had one student who almost failed his defense and it was this type of setup. The student was competent, and if he had had a committee meeting prior to the defense I am certain the issue would have been resolved ahead of time and the defense would have gone smoothly.
I have no knowledge about the actual distribution of IQ scores among competitive gymnasts. Maybe they are high. But I think we can all agree that IQ is not the primary trait that is being assessed in gymnastics.
Great perspective!
I think this also highlights the differences in PhD programs between countries (but also within a single University - our lab participates in a couple of schemes that vary significantly). I completed my PhD in the UK and although the program is shorter than the USA (mine was 3 years, and it's unusual to go over 4) we still have these interim checks. If a program doesn't have one, I think it's a bad setup. We had our quals equivalent at the end of year 1 with the option to write up as an MSc if we/the committee felt it was not in our best interests to continue, and then an additional meeting at the end of the 2nd year to outline the plans for completion & submission.
The degree of difficulty in projects and administration is so inconsistent between PhDs and I completely agree that we need to take a more holistic approach to assessing competencies that are relevant to the task required. All PhDs are equal, but some are more equal than others?
Speaking as a UK professor, there are a few relevant differences between the US and the UK. Number one, the selection of students is more haphazard on the UK, as students traditionally are admitted to work with a particular professor rather than as a cohort reviewed by a single panel; particularly if they have external funding. (This is slowly changing.) Second, there is much more pressure to get students finished on a fixed timetable. Third, in my experience there are no qualifying exams. Students come in with more specialised education than is typical in the US, but the presumption is that they have completed their general education in the subject before beginning. They are typically examined on their progress at a couple of points, and they can fail these, but it is not common.