Without Experiment, There Is No Theory

A lot of people have been blogging and Twittering about this subway map of science, which puts various branches of science in the place of the lines on the London Underground map, showing connections between them. It's a huge graphic, but a kind of cool image.

I do, however, have a problem with it, which is illustrated by the key to the lines shown at right. The category of physics is presented as "Theoretical Physics and Quantum Mechanics." I have no problem with the quantum part, as quantum mechanics is one of the greatest intellectual achievements in human history. I do have a problem with the theoretical part, though.

"Theoretical Physics" is not a synonym for "Physics." It's at most half the story, and probably less, because without experiment, there is no theory. Without Michael Faraday, who had no great aptitude for mathematics, there would be no Maxwell's Equations. without Ernest Rutherford's scattering experiments, there would be no Bohr model for hydrogen (Bohr was actually working for Rutherford at the time). Without Heinrich Hertz and Robert Millikan, Einstein wouldn't've gotten the Nobel for his model of the photoelectric effect, which was in many ways his most revolutionary contribution to physics. Without Albert Michelson and Edward Morley and their attempt to measure the aether, relativity might've taken a very different form. And so on.

Now, the map includes a lot of these people, so this is mostly a problem of labeling, rather than the gigantic oversight I thought at first (Faraday's name is on the opposite side of the physics line from all the others, so I initially thought he'd been left out entirely). But you encounter this sort of thing all the time-- people speak of theoretical physics as if it were the whole of physics, and that's nowhere near being true. Modern theoretical physics could not exist without experimental physics, and to the extent that there is a crisis in physics today (as you often hear), it's a crisis that comes from an excess of theory and an under-supply of experiment.

You can't have one of these without the other. Any map of the path to modern science needs experimentalists just as much, if not more, than it needs theorists.

More like this

Well said, Chad. I often wonder what "thought experiment" is actually as advertised. Einstein stood on the shoulders of many Giants (not the NY Giants, because they look to be sloppy terrible this year).

I'm glad to see you mentioned Faraday. Einstein had pictures of but 3 giants in his study: Newton, Maxwell, and ... Faraday. What? No Gauss?

And Einstein's great Theory of Special Relativity is sweet because it rests on but 2 simple postulates, one concerning the speed of light. Without experiment, we wouldn't know that light even had a finite speed.

Well said, except I think the final, apparently exasperated ", if not more," is not quite as good. Surely the point of experiment in Natural Science is that it is systematized experience. If experience is not tied into some choice of theoretical system, it's not experiment, at least not in the Natural Science sense. The symbiosis of chicken and egg can't be solved by saying that the egg is more important than the chicken. You say something much like this in your post, but the final slip, while understandable, breaks your usual easy-going spell.

By Peter Morgan (not verified) on 31 Aug 2010 #permalink

"If experience is not tied into some choice of theoretical system, it's not experiment, at least not in the Natural Science sense."

Experiments don't need any theory, only language to describe conditions and results.

And it's certainly not a chicken and egg situation, experiments are much more important then theory. Without experiments theorists turn into impotent philosophers unable to decide anything. It's experiments that marked the birth of science, since it's experiments that are the foundation of scientific method.

A theory is just a tool used to simplify and generalize the knowledge obtained by doing experiments.

Without experiments to test it any theory is completely useless, without a theory experiments still give us valuable and useful knowledge, even if we are unable to generalize it.

In certain scientific fields, like molecular biology for example, the amount of theory is negligible compared to experiments since outcomes of those experiments cannot be generalized much beyond the conditions in which they were performed.

Yes, theory is important but it's nowhere near as important as experiments, not even close.

kea: I'm a (solid state) theorist, so I'm gonna go with "please don't fire us; we're people too". :)

On the topic, it's (almost) entirely true that without experiment there'd be no theory. (The "almost" comes from the fact that people would still make theories, but they'd not necessarily be on anything beyond the realm of pure philosophy and angels dancing on pinheads.)

Much more critically, however, theory divorced from experiment *isn't science*. It may be thoroughly logical and sound within the mathematical framework from which it's derived, but it's not science.

"theory is important but it's nowhere near as important as experiments, not even close."

Theory's the other half of the scientific process. It creates hypotheses to test. Sure, stuff gets discovered without a theoretical framework, but it's reduced compared to guided experiment. Plus, without an adequate theoretical framework, all one has is raw experience: "in 99% of trials, we did X and saw Y" Sure it's useful information, but it's not nearly as useful as it can be with adequate theoretical work.

Theory guides and explains experiment, experiment guides and (in)validates theory, yin and yang, eh?

uh - both experiment and theory need math, and if the math is inadequate to the task of making sense of experiment, it can be as bad as theory running amok with no experimental constraints. Nature knows more math than we do.

By joel rice (not verified) on 01 Sep 2010 #permalink

"Theoretical Physics" existed before science did - consider Aristotle's Physics. It's the experiments that make physics science; and chemistry, and biology.

Quite. That's precisely what Freeman Dyson means when he talks about the "craftsmanship of science" in his book "The Sun, the Genome and the Internet". He cites the work of Peter Galison, who unlike Thomas Kuhn, believes that scientific revolutions have been mainly driven by new techniques and experiments and not ideas and concepts. That's certainly true in chemistry and biology and it's also true in physics to a much bigger extent than what people believe.

Joseph, I am also a theorist, but an unemployed particle theorist who is not exactly fond of string theory.

Of course there's theory without experiment. Don't you read hep-th? ;-) Problem is there's no way to tell whether a theory is useful or not without experiment.

I don't think though the "crisis" that some parts of physics suffer from can be as easily explained as "lack of experiment." It's also a "lack of boldness" or maybe a "lack of creativity," which I believe is fostered by too short-term contracts and projects. It is true though that in the present situation progress is more likely to come from experimental input, so keep on looking.

I just found your blog, it's a joy!
I do theory/computation with strong interaction with experimentalists. One very prominent professor (a theorist) said that the world needs no more than 1 theorist for every 10 experimentalists. I tend to agree. Experiment is the basis of physics.

I think Chad's reaction here is to what many experimental physicists (like myself) see as a major popular misinterpretation about 'physics'. Most laypersons think that physics is all about determining the basic constituents of matter using bigger colliders like that thing in Switzerland. In fact, that is just one part of a much larger field in which basic physics research is done.

On top of this fact, because theoretical particle physicists get the vast majority of public notoriety and exposure, it seems the public believes that these people are representative of what a 'physicist' should be.

Like Chad, I think that, culturally, experimental physics is very undervalued. Unlike Chad, I do not think our reaction to this cultural undervaluing is to undervalue the role of theory in guiding and explaining experiments. I'm not outright saying that he undervalues theory in any respect as a scientist, but this post leads one down a slippery slope...

What you say is truth.

However, if you want to be famous or win a Nobel Prize you have much better odds if you are a theorist - and that is who largely made the map.

By Doggie Dog (not verified) on 20 Sep 2010 #permalink