Tommaso Dorigo has an interesting post spinning off a description of the Hidden Dimensions program at the World Science Festival (don't bother with the comments to Tommaso's post, though). He quotes a bit in which Brian Greene and Shamit Kachru both admitted that they don't expect to see experimental evidence of extra dimensions in their lifetime, then cites a commenter saying "Why the f*** are you working on it, then?" Tommaso offers a semi-quantitative way to determine whether some long-term project is worth the risk, which is amusing.
I was reminded of this when I looked at the Dennis Overbye story about Kepler that's in this morning's links dump. Toward the end, he quotes a researcher making an interesting analogy:
The public wants to know whether there is life on other planets," [Kepler team leader William] Borucki said, noting that it could take decades. The effort to get an answer, he said, reminds him of the building of the great cathedrals in Europe, in which each generation of workers had to tell themselves that "someday it will be built."
I don't think it's trivially obvious that really long-term projects aren't worth working on, but I do think it takes a certain type of personality. I know I couldn't do it (but then I couldn't do theory, period). The cathedral analogy is probably a good way of expressing the attitude you need to have if you're going to work in those areas.
I'm curious as to how other people feel about this, so here's a poll question. If you were a new grad student about to embark on the study of Science (not necessarily physics), would you be excited to work on a research program that would not see an experimental test for a long time? Here are some possible time scales:
This is a ticky-box poll, not a radio button poll, so please choose all that apply. Also, please click the box for the final item, so I can get a reasonably accurate vote count, as the percentages are sometimes screwy on these.
I feel like you need more options after "Not in my lifetime." In my childrens' lifetime? Possibly. Not in the next 1000 years? Maybe not.
When I was a wee grad student working on theories of the interstellar medium, one of the diagnostics we could calculate was the column densities of O+5 ions, which provided sharp discrimination between theories of various kinds. It was something one could speculate on without cost to one's reputation, because with wavelengths of 103.2 and 103.8 nm, it was unobservable by the telescopes we had then (and Hubble, launched later).
Now, thanks to FUSE (and predecessors such as IMAP and HUT), we have the data. But it took 20 years.
I don't think it really works like that in science. When you are building a cathedral, you might not see it in your lifetime, but you know there is no obstacle to it being built eventually. In science, there are literally bazillions of ideas that *could* be true of the world, and your goal is to find one of the correct ones, not just to work on something because you like it. I think this applies to high energy physics as much as to the rest of science, despite what some might tell you about finding the truth uniquely through beauty or whatever. Therefore, if you can't find an experimental test then how would you know you were working on the right theory?
Add to that the fact that theoretical physicists are often extremely cavalier with mathematical rigor. They can afford to be in most circumstances because experiments will tell them if they have made an error and then mathematicians will clean up the mess for the next few decades if the theory turns out to be right. However, if you lack experiments and you lack rigor, then you really have no basis for any conclusions apart from intuition, which is really not good enough.
I am not sure these sort of questions are framed the right way. Maybe it is useful to stay away from string theory. Suppose you are a beginning student considering working on high Tc superconductivity, or quantum computing, for example. The lofty goals of these fields (say building a quantum computer, or "understanding" the pairing mechanism in the cuprates) may or may not be achievable in 20 years, in our lifetime, or ever. Nobody really knows. It may be more useful for that student to ask simpler and more concrete questions, like whether there is steady progress, a constant flow of new ideas, and a sense of vitality and intellectual satisfaction in the community. Of course, for any given field opinions about the answers to those questions will diverge, but they are much more relevant that the grand "cathedral" question.
We are working on a project of understanding where we live that has taken many lifetimes. Enjoy the milestones. Should I not labor because I will never see spires and angels?
Like Moshe, I think you aren't asking the right question. Part of it is that we don't know how long it will actually take to achieve the goal (classic example: fusion power, which has been 20-ish years from reality since before I was born). Part of it is that sometimes something comes completely out of the blue and it takes a few years before people understand it well enough to ask intelligent questions (as Rabi commented regarding the discovery of the muon: "Who ordered that?") There are entire fields such as astrophysics where many of the theories can never be experimentally proven, only tested against remote observations.
A better question to ask would be, "Can you envision an experiment which would verify or disprove your theory?" It doesn't have to be something realistically achievable, but something that you could envision doing if you had infinite time and resources available. Something like high-Tc superconductivity or the discovery of the muon, where experiment is way ahead of theory, definitely meets this test. So does astrophysics: in situ probes of a supernova explosion could verify or falsify certain theories of supernovae. Quantum computing is more ambiguous because the negative of the ultimate goal is likely unprovable, but there are steps along the way where we might someday be able to do the experiment. One of my issues with string theory is that I haven't heard anyone articulate an experimental test of the theory, and some substantial parts of it, like the "landscape" ("Define the Universe. Give 10500 examples."), probably aren't falsifiable.
As a mathematician and better than average armchair philosopher, I find this question quite strange.
A historian or philologist would never think to be interested in whether his or her work will be validated in X years, whatever X is. She or he would ask instead whether his or her work is interesting and useful to present and future historians or philologists.
One can have (intellectual) influence or impact on one's field without being right.
As an undergrad working on summer math research, and looking to pursue an academic career in mathematics, I can safely say that I am currently just happy to be doing research, regardless of whether we end up proving a particular theory anytime soon.
Of course, in mathematics it is a little different, as, in my experience, while working towards a larger problem, we often solve smaller, but still relevant, problems along the way. For instance, my research right now is tangentially related to Artin's Conjecture (the one regarding primitive roots), but I certainly don't expect us to solve it - I would be very pleased, and perhaps not altogether surprised if it were solved in my lifetime. However, we are still yielding interesting, and new, results, which is a very exciting experience.
I suppose, in math, very rarely are you ever pursuing a theory that you have no actual way of testing. Either you prove it, disprove it, or it remains a conjecture to be tackled another day. It is possible, particularly in computational number theory, to gather evidence to support claims, but even overwhelming statistical evidence needs to be taken with a very large grain of salt in mathematics.
Don't know about you, but that doggy is pretty cute, too.
I am actually here for the dog stories, but I check the baby pictures as dog stories was not an option.
As a side note most of work for 20 to 25 years on a long term project that may or may not show any progress in our life times. They are called children.