Tag Archives: falsification

Book Review: The Logic of Scientific Discovery by Karl Popper

I bought this book back in the spring, but had hesitated to read it. The reason for that is I found it quite intimidating, given the thickness of the spine. I will admit that the philosophy of science is something I have dabbled in only as an interested amateur though, of course, years of scientific training and discussions as a student have enabled me to reach (if you will allow me a moment of immodesty) a reasonably sophisticated understanding of my own view of science. Now, though, it seemed appropriate that I ought to look at what others have said.

Before reading this, I was aware that Popper’s views are not universally accepted and that he shared something of a professional rivalry with Bertrand Russell. Also, as a Christian, I find it interesting that I have seen Popper referenced far more by Christian scientists, whereas Russell tends to be more favoured by atheistic scientists. Of course, this is only a trend and there are exceptions. While Russell was a well-known opponent of Christianity, I was keen to learn more about what it is in their competing philosophies that has appealed to the different sets of scientists. So, of course, I will be following this up with some reading from Russell at some point, though if you have any good suggestions as to where to begin, I would be very grateful.

The book begins in a surprisingly accessible manner. I was expected some very high level philosophy that would be difficult to understand, but the translation is very easy to follow. Where he gets a little more obscure, he brings it back down-to-earth with examples that help to put his argument in context. I would describe the argument that Popper creates as being cumulative; that is, there are lots of references to earlier sections and, in particular, definitions. So you have to concentrate or else you can find yourself reading about “singular statements” and not know what he’s talking about if you haven’t followed it earlier.

For this reason, I would not recommend reading this book over a long period of time. I think it demands to be read quite intensively in as short a time as possible in order to ensure that one may follow it all.

The main thrust of Popper’s argument is to say that theories are never verified, they can only be falsified. He dismantles the positivist point of view which led to empiricism and shows that empiricism reduces to mere psychologism. From here, he then needs to discuss the degree of falsifiability. He considers a theory to be less likely the more ways it can possibly falsified. From here, what I think he should have done would then be to talk about corroboration and how a theory stands up to attempts to falsify it. Unfortunately, he leaves this to the end and instead goes off on a rather long and tortuous tangent talking about probability.

This quite long section was the downside for me, as his discussion (and in particular, notation) was quite obscurantist, making it difficult to follow and quite oblique. From here, he moves on to talk about quantum mechanics and in particular the Heisenberg Uncertainty Principle. It has to be noted that this was written during the years that quantum theory was still being formed and, having a background in quantum mechanics myself, I found many of his ideas to be simply wrong. They are a noble attempt at getting to grips with quantum mechanics but ultimately, they have not stood up to subsequent theory and experiments. So in a weird twist, you could say that his argument in this aspect has been falsified.

This brings me to my last point. If his theory is to be thought of as a scientific theory at all, then it must play by its own rules. That is to say, there must be a set of singular statements from this theory that can, in principle at least, be subject to testing to see if they can be falsified. Such a set of statements is not presented to the reader, so I could only conclude that while Popper’s contribution is to be valued and considered, it doesn’t constitute a scientific theory. It remains an application of metaphysics.

Einstein, falsification and the spirit of science

There has been much written over the last few days in reaction to the news coming out of CERN that some neutrinos may have been observed breaking the “cosmic speed barrier.” I won’t go into too much detail here, as there is plenty to be found elsewhere on the web, though of varying qualities. One of the things that has bothered me is that many of the news sources refer to c as “the speed of light.” This is incorrect as the c which is used in relativistic notation is specifically “the speed of light in a vacuum.” Everyday experience tells us that the speed of light varies in different media. When you’ve seen a straw or stick in water appear to be bent, this is due to refraction. Refraction occurs because the light has a slower speed in water than it does in air.

My second slight gripe is over the fact that the result has not been confirmed. At the time of writing (Friday night/Saturday morning) the results are still subject to further scrutiny. For me, I wonder if the experiment which produced the result has been repeated. The best science must always be reproducible, given the same setup. This, of course, does assume invariance under time translation. Now *THAT* would shatter the scientific paradigm, if it were falsified.

There is also a slight laxity when discussing the idea of neutrinos breaking the speed barrier. The special theory of relativity doesn’t actually prohibit faster than light travel. If you actually work through the equations, what you end with is that you cannot accelerate a massive particle (that is, a particle that has mass, it’s nothing to do with it being exceptionally large) to c or beyond, from a speed which is less than c. The reason being that as it accelerates, it actually gains mass. The more mass it has, the harder it is to accelerate and you end up that as the speed approached c, the mass is unbounded (or in other words, it tends to infinity) and so cannot be accelerated any further. In theory, if a particle, at the point of its creation, travels faster than c, then it will remain faster, though it does throw out a whole host of other problems which I shan’t go into here.

So what if Einstein was wrong? He was still a hell of a lot smarter than I am. He wrote the paper “On the electrodynamics of moving bodies” when he was 26, younger than I am now. I still find it amazing at how much others achieve at a comparatively young age, and to be honest I find it quite depressing at how little I have achieved given that I am fast approaching 30.

If Einstein’s theory does need to be superseded, you can’t throw the baby out with the bathwater. You have to remember the historical context in which he worked. The Michelson-Morley experiment had falsified the idea of the ether, so something new was needed. Building on the work of Lorentz, Einstein’s view not only needed to be radically different from the prevailing Newtonian viewpoint that had dominated physics (or natural philosophy as it was originally known) for the previous few hundred years, but it also had to incorporate the well-tested Newtonian model in the vast majority of cases. In other words, for massive particles travelling at significantly less than c, the theoretical predictions between Newtonian and Einsteinian dynamics should differ by an unobservably small amount.

If Einstein now needs to be revised by another paradigm shift, any new model will also have to incorporate the results that have supported Einstein’s work over the last hundred years. Einstein’s remains a very accurate model of reality. Had it been a poor model, it might have been found out much sooner.

All this points me towards falsification. I think I have mentioned in another recent post that I am currently reading through Popper’s The Logic Of Scientific Discovery. Towards the start, he makes a very good demolition of positivism, and makes his proposal that scientific theories ought be falsifiable, rather than be built up via a kind of inductive method based on verification.

While this is my first time reading it, the general philosophy is very familiar, having been woven into the fabric of my scientific education.

All of science is provisional. The work of the great scientists represents our current and best understanding of the universe we live in. As Richard Feynman put it, “science is the belief in the ignorance of experts.” Nothing is to be deemed unquestionable, whether it be the work of Newton, Rutherford, Dalton, Darwin, Maxwell, Einstein, Feynman, Witten or Hawking. All are fair game.

To claim that to challenge any scientist or any theory is a kind “heresy” is a claim that could only be made by someone who has little to no understanding of the spirit of science. No theory of nature should be beyond question, even if we don’t have the answer. In fact, the nature of science is to state that there we don’t have the answer. All we can come up with are models of how the universe works that line up as closely as possible with reality. Of course, not everyone holds this view, as was demonstrated by Hawking & Mlodinow last year when they published The Grand Design in which they proposed a bizarre model of “model-dependent realism” though, thankfully, this seems not to have caught on. Yet I applaud them for at least challenging the spirit, which in itself is a spirit to challenge. A merry-go round of Russian dolls, perplexities and unknowns, and so the confidence of an earlier generation may be shown to be no more than hubris, and the quest for truth continues.