Since I decided to become an aspiring physicist back in high school, looking for tips and advice on what to do has been a valuable resource to me. One of the first guides I read that I think is still very valuable today, is ZapperZ’s So You Want to Be a Physicist. It gives extremely practical and detailed tips on what to do, as opposed to more general pieces of advice that everybody gives to you all throughout your education. That being said, general advice is still useful as long as you take it seriously and translate it to concrete steps to take in your everyday learning. Some of the well-written guides out that I’ve either either read recently or were memorable enough to stay with me include:
- John Baez’s Advice for the Young Scientist
- Steven Weinberg’s Scientist: Four golden lessons
- Martin Schwartz’s The importance of stupidity in scientific research
- Terrence Tao’s career advice, especially the importance of learning and relearning your field. (This is meant for mathematicians but the general philosophy and mentality can be transferred easily even to experimental sciences, I think.)
I think it’s generally a good idea to start as early as possible in reading advice on what to do, as you want your later years to be preoccupied with actually learning and studying science. Being still in my first semester of graduate school, I’ve decided to try out some of the books published on the subject, which is new for me.
Peter Feibelman’s A PhD Is Not Enough: A Guide to Survival in Science is a succinct guide on navigating a career as a scientist starting from the graduate student level all the way to getting tenure as an assistant professor or junior scientist in a government or industrial lab. It includes tips on giving scientific talks and job interviews. By today’s standards, with the proliferation of many websites and blogs with all sorts of advice for the aspiring scientist, the advice is rather general and boilerplate, but there are still a lot of good pointers to take away from it.
Probably the most unique piece of advice Feibelman gives is the importance of picking good, short-term projects to work on as a postdoc instead of blindly accepting whatever your supervisor gives. Also, be aware of the big picture of what you’re doing. He also stresses how crucial project completion is in forming the perception of your success as a scientist. Thus, it is crucial to choose the right project, as important as choosing the right supervisor. This is something I’ve noticed recently, and Feibelman is the first person who explicitly and forcefully drives this point home. It’s always better to have two small completed projects than one 50% completed large project.
In giving out his advice, Feibelman does have certain strong preferences, which seem to reflect his own life journey as a long-time physicist in a national lab:
- Always try to work for an established professor (not an assistant professor or a newer person) at all stages of your life
- Go to national or industrial labs first and try to skip the assistant professor phase, even if you ultimately want an academic job
- If you’re a theorist, always find ways to impress experimentalists by talking to them and explaining your ideas in simpler, conceptual terms.
While the book is slated as a general book for scientists, it’s slightly geared towards theorists, as that is Feibelman’s area of expertise. I think the skills needed to impress and network scientific colleagues as an experimentalist are slightly different in subtle ways, based on what I’ve listened to from older and more established experimentalists so far (such as always remember to know the theories and motivations underpinning your experiment, instead of falling into the temptation of blindly doing technical lab work). Networking for experimentalists is also a different experience, I think. The book would’ve been better overall if more concrete tips were given for how to network with scientists other than in your own lab. (Feibelman just mentions shyness as a problem to be “overcome” by focusing the benefits of not being shy. I guess I shouldn’t hold that too much against him, since he’s just not a psychologist or motivational speaker.)
I would also like more details about how to gain expertise in your field and intellectually mature as a scientist. For example, how do you read papers effectively? Which papers should you read in full, which ones should be skimmed over? Should you read papers from other subfields (or even other fields)? How do you learn to become a designer of experiments, as opposed to just an executor? Maybe elucidating such details is outside of the intended scope of the book. It’s even questionable whether it’s possible to craft a general theory of “becoming a mature scientist” that is applicable to all sciences, or even all physics. However, it is these things which are sorely lacking even in today’s wealth of information on “how to become a scientist.”
Lastly, Feibelman openly admits he is from an older generation more used to snail mail than email and Facebook. Enlisting a younger scientist to write a section about personal online branding, writing CVs and resumes and using social media to increase the visibility of your research might be useful. There are other topics that may be peripheral but interesting to touch upon. For example, is doing physics outreach ever a useful thing? How do you become a better teacher, if you’re looking for jobs with significant teaching components? How do you start writing lecture notes when you’re designing courses for the first time? (I guess Feibelman never had that experience himself.)
All of that having been said, despite its shortcomings, the book is still a good, useful read for any aspiring scientist, even biologists or chemists.
I think you raise a number of good questions, but I think most of these become quite individualised, and different people will fundamentally approach research (as well as teaching and service) very differently from one another. If one’s goal is simply to find fulfilment in work, then this choose-your-own-approach is often a good one. However, there are other metrics to consider, such as maximizing career opportunities (or, more pessimistically, survival), fame, scientific impact, teaching impact, outreach impacts, etc. At the end of the day the specific answers heavily depend on your motivations, as well as idiosyncrasies in personality (which tend to vary widely with researchers). Self-motivation is a crucial part of any successful career, so determining the kinds of things which motivate you is important. Intrinsic motivations are often better than extrinsic ones, but one rarely gets to *choose* these things.
Personally I am primarily motivated by trying to have solid impacts (as in, net good contributions) in research and teaching, with some selfish or perhaps nearsighted interests in finding particular kinds of problems or approaches more interesting than others. Of course, such motivations necessitate at minimum surviving the competitive environment of academia. My personal answers to your questions are somewhat idiosyncratic.
Reading papers: I tend to skim a LOT of papers from many fields very different from my main interests, but only enough to grok basic notions of what they are doing, and neglecting the details. In contrast, I read relatively few papers in extreme detail, and most of these are related to projects I am currently involved in. Such choices will depend on how wide the literature in your area is, how trans-disciplinary your work is, and how you get and retain information from papers.
Organization: I definitely agree that completed projects, or projects with multiple achievable milestones (one typical metric here would be publications) are ideal, especially early on. Once one is established (in other words, has less fear about not having a job) more blue sky projects without such milestones are worth exploring. I keep track of many ideas I have constantly, but I’m also aware most of them I will not fully pursue because of finite time and changing interests etc. Always writing down ideas, and revising this list of ideas (sorted by different priorities) is valuable.
Teaching: There are a wide variety of approaches to teaching. Many research faculty at top places don’t care too much for it, and so spend little hard time on it. Personally I think it is important to dedicate time to reading up on modern pedagogical research, and preparing material. How one chooses material for a new course, for instance, is a nontrivial thing and highly dependent on the background of the students, and the purpose of the course (advanced elective graduate courses can typically differ greatly in style and content from mandatory undergraduate courses). Generally considering one’s audience is crucial. Most academics were in the top of their cohorts, and from the best Universities, compared to almost all students they will ever teach. Being able to reach everyone, at least to some degree, is an often-neglect aspect of this which requires substantial empathy and work. Of course there are many further resources in this direction, and similar ones for outreach kinds of things…
On designing projects: Roughly speaking, most projects can be broken (as you suggest) into designing and execution. Of course difficult or interesting things will often require iteration, but this is the gist of it. It is important to read widely and have lots of ideas, so that you can generate many new ideas. Typically most of these will be unsuitable for experiments (or have been done by others or…). Having experience executing things is important to grow in technical expertise, which is invaluable in both further execution and in determining a priori how difficult an idea is to execute, as well as over what timescale it can be completed etc. At the end of the day experience is the most important part, and this will undoubtedly come with many failures. Ensuring in a sense that these failures are *yours*, and do not influence graduate students and others, is important. Typically it is ideal to always give students starting projects which their supervisor has certainty are tractable, and then escalate the projects further and further into more exploratory things, given the aptitude and interests of the student.
Anyway sorry for the long comment, but hopefully some of that is valuable. Reflection is always an important part of this whole enterprise.
Unrelated to this: Gerard t’Hooft also put some resources online a long time ago, and years back I used some of this material to get a sense of things in physics (though formally I’m in a quite different field). http://www.goodtheorist.science/
Thanks for your valuable comment, Andrew. I wrote this book review four years ago when I started grad school, and have gradually found the answers to some of the questions I raised here. Some of my findings are similar to yours. Like you, I also tend to focus on reading papers closely to the project(s) I’m working on, as well as a few papers in my sub-sub-field of precision measurements for fundamental physics, although this is mostly motivated by necessity (to complete my project) or the fact that I am currently interested in my own field more than others (which is why I chose it in the first place), rather than any consciously thought-out plan for my education.
I’m encouraged to see that you seem to take teaching very seriously, to the point of reading up to modern pedagogical research. But what do you say to someone who says that they only really care about getting through to the top students in the class, because those are the few who will be engaging in future research and contribute to the field? Is trying to reaching students at every level of experience and ability a purely altruistic goal?
Finally, I do agree that experience in executing projects is crucial to be able to design good ones in the future. But do you think it is possible for someone to have tunnel vision in executing their own project, to the point that when it comes to the time for them to apply for faculty jobs and design their own research program, they don’t have the knowledge (or time) to do so? After all, it seems that a lot of new exciting research programs are born by applying the methods of one sub-field to another. There is only limited opportunity for you to develop a career which is doing basically the same kind of research as you did in grad school (although I do know that some people who end up doing that). Are the skills for designing new experiments/projects completely proportional to your expertise in executing them?
Ah, I hadn’t realized the post was so old! I’m glad to hear you’ve found some answers to these things. Despite any faux-authority in my previous answer, I am still considering these questions often myself, as the more experience we have the more we can refine our answers and approaches.
Regarding reaching more than the top students: this is fundamentally a question about what the teleology of education is. Of course from the perspective of science and progressing the frontiers of research, it is generally true that the top students will be the ones to have the largest contributions (exceptions exist, but the vast majority of good researchers will come from top students). However, for society more widely there are many reasons people will go to Universities and study areas like physics. There is a large literature on both improving outcomes for students in general, as well as more philosophical or perspective work on why we should care about inspiring, teaching, and generally encouraging everyone who comes to us to learn. Of course, the formally stated goals of education will vary widely between institutions and different parts of society, but I think many such motivations exist for caring about the students who are not at the top. In fact, the best students will generally be very auto-didactic and able to fill in gaps and pursue challenges posed to them with less explanation, whereas many others will benefit from careful explanation and effort from their teachers. I guess as in my previous answer, much of this depends on your own view of what the “purpose” of education is. For me, I can definitely see a substantial amount of value in engaging as many students as I can, hopefully in a way that is enriching for their lives. I don’t see most of them doing research degrees much less becoming scientists, but if I can help them take away some persistent value from the experience, then I will call that successful.
I think you are absolutely correct that tunnel-vision is a key obstacle to becoming an independent researcher, and that there must be a balance between technical expertise and something else (perhaps “vision”, “creativity”, or something in this direction). Generally careers which are doing basically the same kinds of things as one did in their PhD are not terribly valuable for really pushing the boundaries (at least as far as I am aware), and there are great advantages to overcoming novel challenges as one progresses in their career. Collaborations in general, and mentors in particular, can be invaluable at every career stage to help one develop these sorts of things. Networking at conferences and the like is sometimes a bit overblown, but I do think one can come away from a conference with new ideas and perspectives, and developing creativity in this way can be very fruitful. The vast majority of my work which I am most proud of are things where I met with someone from quite a different field, and we both looked at a problem carefully before rephrasing it in a tractable way. I think the human side of science is where a tremendous amount of insight can come from, and perhaps exceptional personal technical expertise is sometimes over-valued in driving the frontier. That said, too much collaboration/networking etc can be a bad thing, and I think particularly “interdisciplinary” fields suffer from a number of problems because of this. This blog post is a useful example of one of these problems: https://egtheory.wordpress.com/2019/03/15/motivatiogenesis/