If "No one knows everything, and you don't have to", then how do you know where to start research?
I read an article named 'Four Golden Lessons' by Steven Weinberg in Nature. It was written by Nobel-winning physicist Steven Weinberg. Here he points out four golden lessons for people who are studying physics and aspire to be a physicist. The first lesson, as noted, is "No one knows everything, and you don't have to".
He wants to put that you don't need to know everything about the field you want research in. You can pick up the knowledge in your journey towards solving that problem. As a matter of fact, he puts that it is paralysing for one who tries to accumulate all knowledge before starting research and was very much encouraged by his advisors to do so as he was, at first, anxious.
If this way is ultimately correct, then how is one supposed to know at what point to start research as there are many levels of knowledge and understanding. I am afraid that what if my understanding lacks too much or it has started to paralyse me.
At which level of knowledge should I start research?
And to generalise it more, what is an appropriate level to start research in physics so that one is not paralysed and also does not lack sufficient of knowledge?
I have also wondered if one should start learning with a research problem only and pick up the right knowledge in the journey of solving it. But it may be wrong so I have not yet accepted that approach.
As you know my main question is just about when to start research. Usually people start during PhD but when reading this article Weinberg was encouraged to pursue research by his advisors when he just entered graduate school which is pretty early in the PhD. Same kind of thing happened with Dr. Gerard 't Hooft and Dr. Richard Feynman who started their research as grad students with their advisors i.e. Feynman and Wheeler and 't Hooft and Martinus Veltman. Moreover, 't Hooft won a Nobel for this research later.
From my experience I have seen a lot more examples who started research in grad school. But many grad students don't do research too! And many turn out to be cranks. So right knowledge and timing are important.
3 Answers
As you know my main question is just about when to start research. Usually people start after PhD
This seems like a complete misconception to me. A PhD is a research degree and one would typically only be awarded it after having produced original research. That is, the dissertation should contain results from original research. Not sure where you're located, but that is certainly the standard expectation in fields like physics; see Wikipedia for example.
but when reading this article Weinberg was encouraged to pursue research by his advisors when he just entered graduate school which is pretty early than PhD. Same kind of thing happened with Dr. Gerardus t'Hooft and Dr. Richard Feynman who started their research as grad students with their advisors i.e. Feynman and Wheeler and t'Hooft and Martinus Veltman.
Yes, that is commonplace. Some portion of people also get involved in research during their undergraduate studies, but the more major individual research projects typically start during one's pursuit of a doctorate. Note that physics gradschool in the US typically means doctoral programs. In contrast, master's programs can be entirely focused on coursework, though many include a thesis component.
The main question is at which level of knowledge should I start the research?
Ask your professors what they did to get a better sense of local expectations and opportunities. But also, take Weinberg's advice seriously. There's no realistic way to accumulate all physics knowledge, so you will have to start somewhere, at some point. The good news is that you learn more quickly by trying and failing than by putting something off until next month. In particular, by starting you will much more quickly figure out what you still need to learn to make progress on what you're working on. Or whether you like the process in the first place... (None of this is unique to research. It applies equally well to DIY home improvements and whatever else.) The tricky part is really to find a project that is valuable and tractable enough for you to make progress. This is where a research mentor, especially your advisor, should come in.
The "appropriate" level to begin research is too variable to have a meaningful answer. You begin when you are ready. No, not very helpful. A more meaningful question would be what you should do to begin at the level you are currently at. For some that will be more (guided) study. For some that will be finding (or being given) a possible problem that "looks" promising. For most, it will probably be reading fairly recent papers and trying to gain a deep insight into both the motivation for the question and the techniques used in answering it.
And, for most, it will require a backward search from any such paper to the papers that led to it.
I'll note that different people have different levels of "natural understanding", up to which learning is easy and after which learning is hard (very hard) work.
But in your case, I suggest that rather than asking such hypotheticals, you find a suitable advisor in a subfield that interests you and following their guidance, both in what to read and in what problems might be open to you.
Expect that there will be failures: dead ends. Not every question is answerable, especially at the current state of knowledge. Expect that you aren't a Feynman. Few of us are. Know that you don't need to be a Feynman to make meaningful contributions. But the way to start, is to start.
Get feedback on your early attempts. Don't get discouraged, and don't burn out. Repetition with feedback is the path you should probably follow as it is the most common one for entry to research (and expertise in most things). That is what doctoral programs are about: provide an environment in which guidance and feedback are natural from those who have been there and (hopefully) have done that.
In your readings and study, search for insight, not just facts. Why, why, why is this important? How, how, how did the authors come to a solution. There are few if any shortcuts.
I have quite a bit of experience in academia both as a former student and as an advisor. One of the most common reason for failing at research is tied to the attempt at gaining too much knowledge before even choosing a research problem.
As a corollary, I think one should start research as soon as a research question is interesting enough for them.
Trying to learn everything that might be useful would be similar to preparing for a hike by packing a 5 ton backpack with everything from climbing equipment to steel wire coils to make bridges between mountains. It just isn't practical. What is practical is to know you go on the mountain, you need boots, water, etc. and just go, and if things look too hard, hike back to the base camp and prepare better next time.
When to start research, in my experience, is the soonest possible. Some kids start in high school, making robots or doing code, some in undergraduate, and some during Phd.
The experience accumulated from every project, including the failed ones, is valuable. You learn how to approach a subject you're clueless about, how to face seemingly intractable problems, how to budget time, whom to ask for help, or collaboration, etc.
The way of graduate schools is to get one started with supervised research. That is someone would take part of the responsibility for choosing and steering the research project. However, I believe that to be a phase one needs to pass, as unsupervised research is both more fun and more motivating. Lacking the supervision constraint, has many pitfalls, but the sooner one starts working unsupervised, the sooner one gets acquainted with them and can avoid most.
Have a similar question?
Ask the community →