Monday, July 06, 2009

Against philosophy 2009

Sean Carroll and Steve Hsu were provoked by a quote from philosopher Paul Feyerabend who was once upset that Feynman, Schwinger, and their contemporaries were philosophically shallow primitives (even though they may have been brighter than Einstein and his contemporaries, he thought).

Feyerabend decided the the professional isolation of philosophers - an attitude advocated by the recipient of his letter, a philosophical Big Cheese from Berkeley - was the culprit.

Physics vs philosophy: the divorce

Physicists share common ancestors with the philosophers: we may say that physics has evolved out of philosophy.

The two disciplines still share the interest in questions that transcend our everyday lives. At the same moment, natural sciences have also adopted new attitudes - the scientific method - that dramatically differ from some attitudes of the philosophers. Galileo Galilei was the main founding father of the scientific method. That's why Galileo's life represented a key split between physics and philosophy. It shouldn't be shocking that Feyerabend himself was supporting the Church in its struggle against Galileo (fortunately, Feyerabend came too late to make a difference). Starting with Galileo, people calling themselves "philosophers of science" were mostly members of the camp whose goal was to advocate scientifically wrong statements about the philosophy of science.

When I was a teenager, I was influenced by giants such as Albert Einstein in particular - but also by people like Richard Feynman. Their behaviors differed. On the other hand, I don't think that their opinions about the basic philosophy of science were too different. In fact, pretty much all good modern physicists are building upon the same (or almost the same) philosophy of science. In this no-nonsense philosophy, the scientific method and particular insights and arguments are important while the philosophical fads are less important.

One of the sources of the apparent difference between Einstein and Feynman - to pick these two examples - was that Einstein lived in an old era when similar ingenious people could still be respected as physicists and philosophers.

Feynman's life already belonged to the modern age in which philosophy and physics were separated: the deliberately anti-scientific philosophers had already hijacked the whole discipline of philosophy and all the relevant trademarks. That's why Einstein used to be respected as a philosopher while Feynman was not. This difference doesn't mean that Feynman was a worse philosopher of science than Einstein: it only reflects the 20th century philosophers' ability to chase genuine scientists out of their discipline.

In reality, of course, Feynman understood e.g. philosophy of quantum mechanics more properly than Einstein did - but this difference largely stems from Feynman's relative youth. Feynman was born in the 20th century so he was no longer brainwashed by the 19th century misconceptions. In fact, Feynman's superior understanding of quantum mechanics relatively to Einstein's was about a better physics education, not about some inherent differences in their approaches to the philosophy of science.

Steven Weinberg's essay

Steven Weinberg described this topic very intelligently in a chapter of his Dreams of a Final Theory. It was aptly called
Against Philosophy (click).
Weinberg mentions the common roots of sciences and philosophy. He agrees that the physicists always have to start somewhere, with some assumptions, but philosophy doesn't provide us with any systematic method to determine the "right" starting point. And he even says that many philosophers of science actually realize this fact.

While a particular lucky philosophical paradigm may be helpful to organize a collection of insights efficiently - and it may be a beacon of progress for quite some time - it is virtually guaranteed that such a paradigm will outlive its usefulness. As soon as it happens, the paradigm becomes a dogma that inhibits further scientific progress.

After all, philosophical paradigms are just fads that are being propagated by irrational processes: by the very unscientific definition of philosophy, these paradigms are never scientifically tested. So there can't exist a good rational reason why they should be helpful to make further progress.

Quite typically, philosophical attitudes "overshoot" in one direction or another. It is the very point of these philosophies to "overshoot". They always want to summarize the world in a few simple verbal principles. Without such oversimplified principles, no philosopher could ever become famous.

But compact verbal principles simply cannot be accurate in this real world because this real world follows the physical laws that are formulated mathematically, not linguistically. Wigner has observed the "unreasonable effectiveness of mathematics" in dealing with the real world. And in the same way, physicists have noticed the "unreasonable inefficiency of philosophy" in achieving similar tasks. Verbally phrased principles are pretty much guaranteed to be inaccurate oversimplifications. And because philosophers become famous if they can offer popular oversimplifications, it is pretty much guaranteed that famous philosophies will be largely counterproductive as tools in science.

Weinberg provides us with lots of examples how various philosophies stubbornly wanted the world to be more materialistic than it is or less materialistic than it is in reality. Other philosophies wanted to introduce too many redundant auxiliary concepts into physics while complementary philosophies wanted to prevent physicists from using any auxiliary, directly "untestable" concepts that are nevertheless critical for good science, as we have learned so many times.

Once again, it may be said that philosophies typically want you to think in an extreme way. Collections of insights may occasionally obey certain extreme philosophical constraints but all of science never does. Only the scientific method can tell us what the range of validity of various philosophical paradigms is.

Unfortunately, the world's crackpots - including some of their leaders such as Peter Woit and Lee Smolin - can never understand that their philosophical preconceptions trying to dictate a priori what science should look like may be wrong and are wrong.

Physics cannot respect Smolin's opinions that it shouldn't be based on deep and accurate mathematics; it ignores his desire for the laws to respect his own version of the background independence; Nature cannot respect Woit's or anti-quantum zealots' opinion that physics must avoid all concepts that are too hard for their small brains to visualize them or to improve their everyday life; science disagrees with Woit's dumb illusions that correct physical theories are obliged to guarantee that physicists must always have sufficient resources to test these theories (or even cheaply or soon). Nature has no such obligations. Even if these people managed to write one correct sentence - which is as likely as in the case of the proverbial monkey writers - the method how they arrived to such a sentence is scientifically flawed.

During the most recent centuries, philosophy has hurt science many times. But the postmodern philosophies that have been contaminating the intellectual landscape of broader science (and other landscapes) in the last two decades have arguably been the most carcinogenic ones. Weinberg dedicated the last pages of his essay to them.

1. I agree with you, Lubos, modulo the polemic on Smolin and Woit which I regard as gratuitous.

My collaborator, Patrick Frank, and I published a piece in Free Inquiry in 2006 titled "Science is not Philosophy," which echoes many of your and Steven Weinberg's points.

Thanks for a thoughtful article.

T.H. Ray

2. I agree with most of what you've stated here, except that I think being well read in such philosophies can only help us to distinguish if we are going down the right path or not - i.e. to know what already has worked and what has not. One could also argue that what you have said about a philosophy holding back science once its assumptions have become outdated, can also be said for any physical theory that seems to have been working very well for a long time - at some point we will probably need to move beyond it, and that is never an easy step to take.

Also, just because 20th cent. philosophers were anti-science does not mean 21st cent. scientists need be anti-philosophy; well-roundedness is generally a virtue.

With regard to the "string wars" (I do loathe to use that phrase), you certainly cannot argue that experimental falsifiability is NOT required for science - at some point, presumably ST will be able to make predictions, otherwise what use is it? You can, however, argue against being hasty in saying a theory about a regime that experiment cannot even probe yet is "not even wrong."