In the life of a scientist there are times when the word “change” takes on a particular appeal. The change I am referring to, is that of area of research. Many, perhaps most scientists spend their whole careers researching in the same, rather narrowly defined field, often investigating the same subject over the course of years, decades. There are reasons why that is the case.
Disclaimer: this post is not about change forced upon someone by adverse circumstances, e.g., loss of job or inability to secure one in the desired area of research. The change that I am describing here is self-motivated.
Any worthwhile problem is difficult. If it were not, someone else would have solved it already. Insight, progress, breakthroughs, normally only come (if ever) after a long, intricate process of trial and error, usually extended in time (we are talking years). Perseverance, no, stubbornness is an essential quality of a successful researcher. It is almost necessary to become obsessed with a particular problem, in order to get to the bottom of it. And, doubtless, a great deal of narcissism and ambition drive one’s relentless quest at being the one to uncover that puzzling riddle of Nature that no one else has been able to crack. There is no scientist who does not have, at least in the back of his/her mind, the worldwide recognition, monetary rewards and place in history that go with ground breaking discoveries…
As a consequence of that, and given the stakes and the investment of time, human resources and infrastructure that is required, it becomes extremely hard to let go of an unsolved problem. Giving up is tantamount to conceding defeat. And how would one ever know that maybe, just maybe, that last remaining idea still not yet tried out, that alternative computational approach, that difficult but in principle possible measurement, would not turn out to yield the missing piece of the puzzle ? Wouldn’t it be devastating to see someone else try that very thing, in many respects building on all the hard work carried out by us until then, and succeed in our place ?
Still, there are also times when a change of direction, field of research, or maybe even area of science, may not be a bad idea, one that might warrants careful consideration. And yes, there can be good reasons for doing it. In this post I am going to write about the potential benefits of branching out into a new field. In a follow-up post I shall explain why I personally (while generally not a “conservative”), tend to be wary of change when it comes to one’s own research activity. Notably, one has to beware of “change for its own sake”.
And the one thing that we should get straight is: a change of direction is not a requirement.
There is nothing inherently wrong in pursuing for ten, twenty, thirty years the same scientific objective. There is, in my opinion, a legitimate way of spending such a long period of time attempting to elucidate an important, outstanding fundamental problem, without ever coming to a definitive “answer” or “solution”, but still yielding a body of work that the community will find useful, possibly for reasons unrelated to the main goal of the research program, in the process contributing to the education and training of graduate students and postdocs .
However, as we all know stubbornness and persistence are not always positive qualities. There is such a thing as running out of ideas, endlessly trying the same thing, each time hitting the same wall. And it is not unreasonable to be tired of thinking of and working with the same things all the time, to want to try out something different, especially after a number of years without any tangible progress.
Trying out something new has the following possible (by no means guaranteed) payoffs:
- Breadth is always good. The knowledge of a scientific field that one acquires by actually working in it, is unmatched by anything that anyone can learn from books. And knowing more about different fields of research makes one a better teacher, grant reviewer, mentor, science officer and citizen. This is the one clear, undeniable benefit of working in different areas of research throughout one’s career. It is the single best reason why, every once in a while, I do consider to make at least an excursion into a different area, without necessarily abandoning my own. But here too, there are pitfalls — it depends on how one does it, and one may not in the end get much out of such an exercise, as I shall explain in my next post.
- There is more to science than just, for example, high-temperature superconductivity, or dark matter, or grand unification. Although it may be difficult to see that, for someone who has spent a decade working on one “seminal” problem, there are many different, interesting problems in other areas within one’s discipline, or even in different areas of science altogether, where one may (please note the emphasis) be able put to good use what learned working on a difficult problem in one’s own original problem of interest .
- There may be “strategic” benefits as well. Specifically, there is no question that access to funding is usually crucial to the continuation of a research program. The “sexiest”, “hottest” subjects, the ones that are most current and about which one reads most often on the newspaper and hears on television, are also the ones that attract more researchers. That means that competition for funding is fiercer, and one may be left out. Once the funding tap is closed shut, one is forced to scale down one’s activity significantly, and competing with better funded groups may never become possible again.
On the other hand, in other areas, perhaps less fashionable or “glamorous”, the competition may be less stiff .
 Obviously, this does not mean that a faculty at a research university is entitled to receive unlimited funding and/or relief from other duties (e.g., teaching or service) to pursue indefinitely a research program that does not produce results deemed valuable to others, and/or whose training and educational components have dwindled.
 By this I do not mean to imply that problems in other fields are “easier”. They are not, but one might make an original contribution by bringing in a different perspective, background, and possibly investigative tools, which may turn out to be effective in other contexts in which no one has yet thought of applying them. Examples of that are countless, even though, as I shall try and articulate in my next post, one ought not oversell this scenario.
 If I sound unconvincing, is because I am not convinced myself. I really do not think that this is a good reason to make a switch.