Friday, May 26, 2006

Clifford Johnson and the word "theory"

Clifford Johnson has informed us that the most irritating thing he has encountered during his visit to his homeland was the phrase "special law of relativity" written on this poster with Albert Einstein in the London Museum of Science.

What does your humble correspondent think about these issues and Clifford's opinions?

Clifford has no problems

First of all, I am somewhat jealous that Clifford has the problems he has. ;-) Imagine. During a visit to one of the most important countries of this monstrous world - which is itself a shi*tty place as a classic noticed - a nearly invisible word "law" replacing the usual word "theory" is apparently the most breathtaking threat for our civilization.

The only Clifford's problem is that next Monday, his co-bloggers will effectively expel him from Cosmic Variance - or at least he will understand their remarks in this way - and he will announce it on this URL.

Let me start with some positive feedback.

I agree with Clifford that the proud scientists in 2006 are using the word "theory" as a compliment: a "theory" is a finite set of ideas, concepts, laws, and equations that is capable to describe a class of phenomena that have already been observed, and to predict features of new phenomena that may be observed in the future. A theory is the best result of a scientific or rational analysis of the world you can dream about.

Knowing how to call things vs. knowing things

But I disagree with everything else that Clifford wrote. The most important thing that Clifford does not seem to realize is that the word "theory" is just a word, not science itself. By telling someone that something should be called a "theory" instead of a "law", we don't teach her anything about science. We just teach her a piece of scientific terminology.

After all, it was us, the scientists, who has distorted the true original meaning of the word "theory", not the laymen.


In order to prove my point, I must become a linguist for a little while and offer you some etymology of the word. The word "theory" derives from "theorein" which is a Greek translation of "to look at". The ancient Greeks would attend theaters quite often and they were looking what was going on. It was not quite real but it was resembling the real world. The profession of the philosophers - the scientists of that era - was viewed as something analogous to the actors' job. They were creating something that resembled the real world but it was not quite the real world.

The ancient Greek scholars were already using the word "theorein" in a similar way as we are. An alternative deconstruction of the word splits it into "to theion" (divine things) and "orao" (I see). When you have a theory, it means that you see some divine things. Some of their theories were rather successful - especially the Euclidean geometry - but some of them were deeply flawed.

These mixed results inevitably meant that most people rightfully viewed "theories" as something provisional, as an equivalent of hypotheses or speculations favored by the wise men who prefer thinking over the reality.

Theories got better

During the last few centuries, science became a bit more reliable than it was in ancient Greece. The theories - more precisely, the scientific theories - became more trustworthy, too. But it takes some time for the ordinary people to appreciate these developments. So we should not be surprised that the word "theory" still creates skeptical emotions inside the hearts of a large percentage of the world's population.

When we talk about relativity, do we want the people to understand how science works and what it has found about spacetime, or do we prefer to convince them that the terminology that was around for millenia should suddenly be changed because a certain community - namely the scientific community - started to use the word "theory" differently?

I think that the first goal is far more noble and important - especially because the laymen are etymologically more right than the scientists if they insist that the word "theory" includes some "fantasy" in it. The fact that Clifford Johnson thinks that the problematic terminology is more important than relativity shows that Sean Carroll is probably not the only postmodernist thinker among the authors of Cosmic Variance.

Theories vs. metatheories

We usually present special relativity as a theory based on two postulates that codifies many laws about space, time, matter, and motion. In some sense, special relativity is not a theory: it is not a particular complete description of reality but rather a set of constraints that every acceptable model of reality must satisfy. One can be much more confident if she wants to say that special relativity is not a model. A model should be much more specific than special relativity.

Einstein used to divide the scientific theories to principled and constructive ones: quantum mechanics and atomic physics are examples of constructive theories because they add new ingredients and insights to describe an ever wider realm of objects and phenomena. Thermodynamics and relativity are examples of principled theories because they start with some basic assumptions - such as the postulates of special relativity or the non-existence of the perpetual motion machines - and then they derive general constraints that the world has to satisfy. Using modern jargon, principled theories are the results of the top-down theorists' work, while the constructive theories are the result of the bottom-up phenomenologists' work. Of course, eventually we want these two approaches to agree on a common description of the world.

Special relativity tells us how the measurements of time, space, velocity, mass, energy, and other quantities in different reference frames are related to each other. Special relativity has implications for electrodynamics and other particular theories - and its principles guide us and constrain us in our search for new theories - but special relativity itself cannot tell you the spectrum of atoms, the forces between particles, and many other things.

Die Invariantentheorie

The phrase "special theory of relativity" has other problematic aspects. In the text above, I argued that it is really a "metatheory", not a "theory". However, the word "relativity" is controversial, too. This word has led many philosophers to scream that relativity makes our knowledge "relative", which is why it makes philosophers' vague fantasies important at the same moment. Of course that such a conclusion is unjustifiable. When Einstein realized that "relativity" was a bad choice, he wanted to replace the term by "the theory of invariants", to emphasize that the insights of the theory are based on the analysis of quantities that all observers agree upon, but it was already too late.

It is hard to find anything disturbing about the "laws of special relativity". I feel uncertain whether it is better to talk about "one law" or "many laws" (probably "many laws" is better) but I can't understand how can someone be upset by the "laws of special relativity". We use the word "laws" for many analogous insights in science in general and physics in particular: Newton's laws, Kepler's laws, laws of genetics, laws of Nature.

Some of these examples are very specific, others are very general.

No one - not even Clifford himself - protests against the word "laws" in the case of Kepler's laws because we have summarized the insights about the planetary motion as well-defined "laws". No doubt, special relativity could also be summarized in this format.

A simple Google search reveals that Clifford who dislikes the "laws of special relativity" will have to send his hate mail not only to the London Museum of Science but also to Wikipedia, Eot-Wash gravity experimenters, David Hogg from IAS Princeton, David Santo Pietro of UC Davis, Stephen Wolfram, your humble correspondent, the late David Bohm, Roger Penrose, the late Paul Dirac, Lee Smolin, and 573+184 other physicists. Happy writing, Clifford. ;-)

You obtain a comparably overwhelming set of pages and papers if you search for "laws of quantum electrodynamics" (including three Feynman's famous papers), "laws of general relativity" (including papers by Israel or Smolin), "laws of quantum chromodynamics" (including official documents of Fermilab and CERN), "laws of QCD", among many other examples of terms where the word "theory" remains dominant. The "laws of evolution" in the context of Darwin's theory returns 13,600 pages, including first books about the subject from the 19th century.

Theory vs. model

There are other examples in science where the word "theory" might have been more appropriate. Steven Weinberg has invented the modest term "the Standard Model" for a theory that describes all of particle physics that we know as of 2006. As David Gross has repeatedly argued, the term "the Standard Theory" would be more accurate because it would reflect the truly deep impact of the specific gauge theory than the narrow-minded word "model" ever can.

A "model" is a "theory" but one should always have hundreds of comparably important models (or supermodels) to use the word "model". The word "model" still indicates that you have not chosen which of the hundreds of models is the right one. That's not the case of the Almighty Standard Model.

Language in motion

But we got used to this term - the Standard Model - and we are proud about this term much like a right-winger becomes proud to be a "reactionary" when he learns what the left-wing whackos mean by this "insult". ;-) Our language is developing all the time. The meaning of the word XY may become more specialized while the meaning of UV may become more general.

These changes reflect not only random fashionable cultural trends in the society but also the actual progress in science and technology. When technology creates new gadgets, we need some new words for them and some of the new specific gadgets "recycle" some old words whose meaning was more general.

For example, the word "Windows" used to represent all kinds of transparent glassy rectangles found on the walls of buildings. Today, it is mostly used for rectangles with a red X cross in the right upper corner.

On the other hand, when science unifies things, there are no longer good reasons to use two different words for two things that became one thing. In such situations, one of the old specialized words may survive and acquire a more general meaning. For example, we only use one unit for heat and for energy because Joule taught us that these two quantities are equivalent.

Mathematics distorts the original meaning of the words even more dramatically. If a word has a certain "flavor", a mathematician is ready to hijack the word and use it for the idealized objects with the same "flavor". In mathematics (and even science), we are attributing a much more concrete meaning to verbal constructions than ordinary people do. Crackpots often fail to appreciate the delicate difference in the meaning of the words in science and in the colloquial usage.

In some cases we may feel that some of these changes are too fast while others are too slow. Some of us are linguistic conservatives, some of us are linguistic progressives, and sometimes it is not clear who is actually a conservative and who is a progressive. In different national languages, various opinions about the language may also be correlated with political attitudes. For example, Czechs had to decide whether they wanted to ignore the Latin origin of the word "president" and write "prezident" much like the Russian peasants do. You can guess which spelling was the conservative spelling. ;-)

Magic words don't exist

At any rate, Clifford's dogmatic linguistic argument has no rational substance. How does Clifford justify that the word "theory" is so infinitely important?
  • What we should be doing instead is better educating the public as to the meaning of the word, making it clear that the terms “Theory of Evolution”, or “Theory of Relativity” do not imply that these are not powerful, established parts of the scientific knowledge base.
Previously, I have explained that preaching that something should be called a "theory" teaches the girls exactly as much science as if they are told that it should be called "wakalixes": namely nothing. But I want to say something stronger than just that Clifford proposes a counterproductive, vacuous focus of education: what Clifford says in between the lines is not really true.

Clifford is correct that the fact that the word "theory" is used for evolution, relativity, or strings does not really imply that these intellectual buildings are terribly suspicious. But it does not prove that they are correct either. If some laymen are making the error to associate the flavor of wild speculation with the word "theory", then Clifford is doing exactly the same error backwards: he seems to believe that the word "theory" has a magical power to declare an idea true.

Needless to say, it's a nonsense. The word "theory" is also being used for all kinds of vague, speculative, and irrational ideas - from the "theory of postmodernism" to the "theory of catastrophic global warming". The appearance of the word "theory" can't and shouldn't make any idea more convincing. The word "theory", just like the word "abracadabra" or any other word for that matter, fails to have this divine power. The laymen who don't believe special relativity are making a mistake. But they are doing a very correct thing not to change their mind just because someone uses the pompous word "theory" for this intellectual structure. The validity of a system of ideas has nothing to do with the question whether we call them a "theory" or "laws".

I can't believe that Clifford disagrees with this fundamental point.

One of the basic general messages of the science education should be that the words can't have the same power in the scientific world as they used to have in the religious world (and still have in the world of philosophers). In other words, one of the basic skills that the kids should learn at school is to be able to understand why Clifford is not right.

Moreover, scientists should accept the fact that science cannot determine what the words mean in everyday life. The word "theory" has a taste of uncertainty which might reflect the current situation in the case of string theory but it is not really right in the case of special relativity. By calling the insights of special relativity "laws", we might help the laymen to be less skeptical about these well-established insights. That's why I think that it is OK to put the word "laws" in their description of special relativity, and if there were a global campaign to promote the new term "laws of special relativity", I would subscribe to it.

Especially in the case of "c" as the universal speed limit, the word "law" has the advantage that the audiences understand that this is exactly what follows from science: the word "theory" indicates that there exists uncertainty or some more detailed microscopic clarification of the law. But such a clarification is not needed in this case: "c" is really the maximal speed. It's a law.

Law, dogmas, and theories

As the previous paragraphs indicate, I also disagree with Clifford's statements that the word "law" makes science sound more dogmatic than it is. Incidentally, this argument of Clifford strikingly contradicts his previous argument that the word "law" is bad because the laymen tend to think that science is too vague and loose - now the word is, on the contrary, too strict. Is not it strange to fight against a word by two arguments that are opposite to one another?

Are the laws of the society too constant to be used as a metaphor for the laws of physics?

In fact, I am convinced that the important laws of physics such as the laws of special relativity are much more robust and unbreakable than the U.S. constitution and other laws of the society. In the long run, the laws of the society are not dogmatic either. And they should not be. Surely, many groups of people want to declare certain laws to be permanent. Religious fundamentalists would like to declare their religious rules to be valid forever; feminists would like to prohibit the very consideration of theories - or laws - that the women are less likely to be great mathematicians.

But all reasonable people know very well that our granddaughters in 2050 will have the right to change the laws in any way they want as long as our civilization survives and as long as the world won't be controlled by some drastic totalitarian regime. I hope that they will be able to see which laws are important enough and should be kept.

The laws of special relativity are much stronger than any constitution that has ever been written because they are more or less guaranteed to hold until the end of our civilization and beyond. In this sense, science is more dogmatic than the social sciences. It is more dogmatic because the laws that science reveals are much more universally and much more eternally valid. Nature with all of Her laws is a brutal dictator, indeed. If Clifford does not like Her system, he may try to find a better Universe.

Let us hope that the children will learn how to approach the world rationally and appreciate Nature's beauty and cleverness, regardless whether they will call the insights of relativity "laws", "theories", or "zákony". In other words, let us hope that the children will be able to see what's wrong with Clifford's linguistic dogmatism and with his calls to fill science education with his terminological prayers.

And that's the memo.

No comments:

Post a Comment