Finding the Walls to Breakthrough

May 6, 2008 12:31 AM

What are the modern fields in science currently under crisis?

I've been reading Thomas Kuhn, I am really curious as to if there are any areas in modern science where the models simply don't work any more. I'm mostly looking to see where I can fit my self into an emerging breakthrough

posted by Rubbstone to science & nature (23 answers total)

Physics has a long-standing problem modeling gravity as a quantum field. Good luck with that one, though.
posted by panic at 12:55 AM on May 6, 2008

Find a monopole magnet and you'll be famous. There are a decent number of open problems in math. Google Riemann hypothesis for something that'll also make you instantly famous.

Microfluidics are looking promising, but that's more applied.

Chemistry has plenty of areas where the models don't quite work, but unless you stumble on something major, you'll still be making incremental improvements.

Quantum computing will be a pretty interesting field fairly soon, hopefully. Come up with a good use for it with something that may actually get made and you could do pretty well.

I'd be looking for something that I was interested in, though, and not something that will get you famous. People that get their Nobel prizes in grad school usually get it through serendipidy, not directed research.
posted by devilsbrigade at 2:00 AM on May 6, 2008

Turbulence? The last real breakthroughs were by Kolmogorov (1941) and maybe later the emergence of chaos theory (Lorenz, 1963).

Turbulence may have gotten its bad reputation because dealing with it mathematically is one of the most notoriously thorny problems of classical physics. For a phenomenon that is literally ubiquitous, remarkably little of a quantitative nature is known about it. Richard Feynman, the great Nobel Prize-winning physicist, called turbulence "the most important unsolved problem of classical physics." Its difficulty was wittily expressed in 1932 by the British physicist, who, in an address to the British Association for the Advancement of Science, reportedly said, "I am an old man now, and when I die and go to heaven there are two matters on which I hope for enlightenment. One is quantum electrodynamics, and the other is the turbulent motion of fluids. And about the former I am rather optimistic." source
posted by swordfishtrombones at 2:13 AM on May 6, 2008 [1 favorite]

Computer Science?
posted by rongorongo at 2:22 AM on May 6, 2008

There's tons of biology where there never even was a model. That, or there's no convincing evidence. Don't listen to physicists; their field makes sense.
posted by a robot made out of meat at 4:24 AM on May 6, 2008

Riemann is allegedly a done deed.
posted by flabdablet at 4:48 AM on May 6, 2008

I have heard a professor say that people are having a hard time getting things to tumors. Something like about how twenty years ago, everybody had an idea, and these days ... not so much.
posted by Comrade_robot at 4:48 AM on May 6, 2008

A lot of geography, such as in my area of studies, geomorphology, has some pretty good models that are only effective in the places they were developed and no models for other areas.

For instance, I'm looking at river geomorphology in Australia, which for a long time used American models (where a huge majority of river research is done) and which simply don't work in other geological settings. This has lead to some poor management and planning in places around the world.
posted by Serial Killer Slumber Party at 4:51 AM on May 6, 2008

Wikipedia has a list of unsolved problems in various disciplines.
posted by futility closet at 4:59 AM on May 6, 2008 [2 favorites]

If you're looking for a crisis (in the sense of Kuhn) in physics, where you have a phenomenon that's observed experimentally and no good model to explain it, the best candidate is probably high-temperature superconductivity. Quantum gravity (as mentioned above) isn't so much a crisis, in the sense that there aren't really any experimentally observed phenomena that seem to need quantum gravity.
posted by Johnny Assay at 6:05 AM on May 6, 2008

When some mammalian tissue culture cells are grown in dishes, it is fairly easy to temporarily or stably introduce genes by a process called transfection. But efficient transfection or transduction of genes into mammalian cells in the whole animal seems to be extremely difficult. It's really an engineering problem, but solving it would allow for effective gene therapy and manipulation of adult tissues. If you could target specific cell types, it would be even more useful.

For true biology problems, I recommend Carl Woese's 2004 article (open access). He compares current biology to 19th century physics, when it seemed as though the big problems were all worked out, and points out the unresolved problems that biologists currently avoid.
posted by Jorus at 6:40 AM on May 6, 2008 [1 favorite]

I'm not sure that there is currently a scientific discipline that is in crisis in the Kuhnian sense.

A Kuhnian crisis requires that there be more than just a handful of modeling errors or lacunae in a theory. Such a crisis requires that the entire theory is found to be lacking to such a degree that it is abandoned. That is to say, for a field to be in crisis right now, it would need to have no dominant theory and (likely) be in the middle of a rather large argument as to which replacement candidate is the best one. (Although I guess one might say that QM could be seen as being in a protracted revolutionary period. Though this view raises questions about the proper scope of a theory. Is the theory in question physics, QM, or some subset of QM?)

Unfortunately it's not clear, to me, that you can predict when a field will shift from a period of normal science to a revolutionary period. For example, Lord Kelvin famously thought that classical physics was pretty much a done deal, and then Einstein came along. Of course, this is not to suggest that widespread satisfaction with a theory is an indicator of an upcoming revolution. Regarding Jorus's comment it is possible that biology needs to start all over and it is possible that the gaps will eventually filled in within the framework of the current prevalent theory.

It would seem that rather than trying to predict the field in which the next revolution will occur, it would be more fruitful (and vastly more satisfying) to pick a field that interests you greatly. The quality and originality of your work will be orders of magnitude greater if the field excites you.
posted by oddman at 7:01 AM on May 6, 2008 [3 favorites]

In biochemistry it's a ongoing problem that hybridization parameters of nucleic acids cannot be predicted reliably. While this sounds like a minor theoretical problem, it's a huge practical one, because a large number of the core technologies of molecular biology depend on DNA or RNA hybridization.
This lack of predictability becomes especially apparent in microarray experiments, which depend on gazillions of small oligo probes of different sequences which binding strength is the value that's actually measured to get the results.
posted by Lynx at 7:49 AM on May 6, 2008

Another one in biology is of course evolution, or more precise, speciation. There are several theories of how species arise, on the ecological as well as on the molecular level (allopatry vs. sympaty, cis vs. trans effects). Of course, in biology the answer often turns out to be 'a little bit of both with many exceptions', even if the problems are often discussed as if there was a real conflict between them.
posted by Lynx at 7:59 AM on May 6, 2008

I'm not sure any current scientific models is in crisis mode right now, honestly: we seem to be firmly entrenched in Normal Science. There are obvious holes in knowledge and experimentation out there, but not so great as to throw everything into crisis. I have to agree with those above: go with what you love, and you'll find meaningful work. Chasing after the next Crisis, to me, is tilting at windmills.
posted by absalom at 8:04 AM on May 6, 2008

I'd just like to chime in and say that the Riemann hypothesis, whether proved or not, is not indicative of a crisis in mathematics. It's just an unsolved problem, not a fundamental breakdown that forces us to reevaluate the way we think about things.
posted by matematichica at 8:19 AM on May 6, 2008

Reconciling gravity and relativity with quantum mechanics into a Grand Unified Theory (GUT) is the outstanding problem in physics. M-theory is the candidate with the most potential to be the GUT.

Protein folding has been an unsolved problem since the time of the Levinthal Paradox.

Our understanding of cancer may be on the verge of a serious paradigm shift if the Cancer Stem Cell Theory is confirmed. Briefly, it says that only a subset of cells are capable of producing an entire tumor, and that they divide very slowly as opposed to the typical rapidly dividing cancer cell. Being slowly dividing, they are resistant to traditional chemotherapy which targets dividing cells. Thus, traditional methods can kill essentially all of the tumor and leave the stem cells behind which leads to relapse.
posted by euphorb at 8:59 AM on May 6, 2008

If you're willing to consider economics as a science, I would offer that the equity premium puzzle as symptomatic of something fundamentally unknown.

Essentially, all of the basic models say that stocks should do better than bonds by some factor (say, x) while all the empirical data shows that stocks outperform bonds by something like 10x. Literally, there is an order of magnitude difference between model and reality. There are scores of proposed alternative models, but none has come to be widely accepted.
posted by mhum at 10:19 AM on May 6, 2008

It's not a crisis, but tropical geometry is an emerging field, with lots of breakthroughs waiting to happen.
posted by proj08 at 10:31 AM on May 6, 2008

Neuroscience and evolutionary psychology are putting pressure on the accepted models used in Anthropology and the other social sciences. I'm not sure if it's crisis mode yet, but it's close if it isn't.
posted by MillMan at 10:44 AM on May 6, 2008

Dunno about crisis, but there are many open problems in cognitive science. Broad stuff like how we integrate all the discreet information we sample from the world into a unified consciousness to more concrete problems like a neural model of variable binding.
posted by sophist at 1:35 PM on May 6, 2008

Seconding mhum, MillMan and sophist - people are hard. I'm a sociology grad student, and I sometimes feel like we're just entering the period that physics and chemistry did, oh, 150+ years ago. We're finally developing tools that don't consist of just various forms of asking people questions (we're getting into some great experimentation that I would actually consider experimentation, there's the increasing ease of neuro-measurement, agent-based modeling on beefy computers)... it's a weird time. The majority of people are still hacking away with old methods, but we're being driven to the new methods slowly.

If you're interested in the hard sciences, you're probably more of a positivist than many social scientists. That's okay. Me, too. Besides, physicists do sociology, too.
posted by McBearclaw at 4:59 PM on May 6, 2008

As mentioned above, neuroscience seems to be the ripest candidate for a grand scale revolution. For supporting evidence see this talk: Toward the First Revolution in the Mind Sciences.
posted by FissionChips at 8:45 PM on May 6, 2008