Bonds and gestures pushed to one side
Like an outdated combine harvester,
And everyone young going down the long slide
To happiness …

Philip Larkin, High Windows (1967)

Another successful attempt to get a Larkin quote into my blog. High Windows (not one of my favourites, though it got the poet’s fourth collection named after it) is about the way each generation throws off the taboos of its predecessors, and envies the next generation in turn for the taboos it will throw off. It was written four months before Annus Mirabilis, one of Larkin’s most well-known poems, which has a similar theme but is one of his most misunderstood ones, if Wendy Cope’s Larkin satire “Mr Strugnell” is anything to go by. This “long slide” is a never-ending process; indeed, if you take a slide and bend it round, you get a spiral, which could be called a regress … which brings me to the actual subject of this piece.

Following the recent announcement of gravitational wave detection by the LIGO collaboration, a friend posted the following on her Facebook page:

“CALTECH found #gravitationalwaves using super-expensive technology designed to find #gravitationalwaves confirming #einsteinwasright because we all thought that #einsteinwasright but needed to find #gravitationalwaves! #scientificprogress #circularity”

The concatenated words following the “#” symbol are links to other Facebook posts, with varying degrees of relevance to the subject under discussion. Some of these “links” are actually not far short of random; the posts listed under “#circularity” seem to have very little in common, and some of them even seem to relate to a band and a brand of ear-rings!

By removing all the hashes and concatenation, and putting some rather speculative additional words at the end, we arrive at something more intelligible:

“CALTECH found gravitational waves using super-expensive technology designed to find gravitational waves, confirming Einstein was right, because we all thought that Einstein was right, but needed to find gravitational waves! Is this scientific progress or circularity?”

This was, at any rate, how I interpreted it when I read it. My response was to ask where the circularity was, but I was informed that it was obvious; admittedly that was followed by the disclaimer that it was “not necessarily epistemically or metaphysically devastating, just worthy of a flippant Facebook post” (now there are two words that definitely go together, “flippant” and “Facebook”!)… so maybe I’m wasting my time here; nevertheless I will plough on.

If the circularity is obvious, perhaps it relates to the fact that my revised version of the post includes the words

“confirming Einstein was right, because we all thought that Einstein was right”.

But that is only circular if “Einstein was right” refers to the same thing in both places. And even then, it would only be truly circular if “thought” were strengthened to “knew”. In any case, it is clear that the phrase does not mean the same thing, although to be honest I have to admit that I am not sure exactly what “we all thought that Einstein was right” means. My own interpretation of the background to the search for gravitational waves is as follows:

Einstein’s theory of general relativity explained the already-observed anomalous precession of the perihelion of Mercury, and predicted the relationship


between the angle through which starlight is bent by the Sun’s gravitational field and the distance r from the incident ray to the centre of the Sun, with G, M, c as the gravitational constant, the mass of the Sun and the speed of light. This was confirmed – somewhat controversially, by the Eddington eclipse experiment in 1919, and more definitively later on, using radio astronomy. So on that basis our confidence in the theory is boosted (“we all thought that Einstein was right”) and we expect to be able to observe another of its predictions, gravitational waves; but the search for such waves, and the discovery, are not dependent on our previous belief that Einstein “was right” about general relativity. The method used, interferometry, predated the theory of relativity by several decades at least, so does not depend on it in any way.

The reference to circularity in the context of gravitational waves reminded me of what Harry Collins said about gravitational waves in his 1985 book Changing Order. I have of course mentioned this book in a previous blog. It is a short book written by a sociologist of science, and its two main topics are a project to build a laser and an attempt to detect gravitational waves. My previous comments were about the former, which was a vehicle through which Collins introduced the concept of tacit knowledge; the latter paved the way for a discussion of what the author called “The Experimenter’s Regress”. I did not comment on this in my previous piece on Collins, because I found it almost too ridiculous for rational discussion, and was angered by the way both of these concepts had passed into the “science studies” lexicon and are now used somewhat uncritically by Collins’ disciples.

The essence of the “experimenters’ regress” can be summarised by the following extracts from Collins’ book:

“Proper working of the apparatus, parts of the apparatus and the experimenter are defined by the ability to take part in producing the proper experimental outcome [Collins p74]

“We will have no idea whether we can do it [i.e. detect gravitational waves] until we try to see if we obtain the correct outcome. But what is the correct outcome? What the correct outcome is depends upon whether there are gravity waves hitting the Earth in detectable fluxes. To find this out we must build a good gravity detector and have a look. But we won’t know if we have built a good detector until we have tried it and obtained the correct outcome! But we don’t know what the correct outcome is until … and so on ad infinitum” [Collins p84]

The idea that we don’t know what we are looking for until we have found it is anathema to anyone who has any experience in experimental science. What Collins is saying here is that general relativity predicts gravitational waves in some vague sense, but does not actually make any testable predictions about them. If that were the case, gravitational waves would be on a par with such ethereal concepts as “extra dimensions”, which are in some sense “curled up” and hence not detectable (though, since I am not an expert in extra dimensions, I will accept that they may make some sort of testable predictions).

Let us explore this idea of searching for an “outcome” whose identity we are entirely ignorant of. Here is a poem by John Updike:

Neutrinos they are very small.
They have no charge and have no mass
And do not interact at all.
The earth is just a silly ball
To them, through which they simply pass,
Like dustmaids down a drafty hall
Or photons through a sheet of glass.

It goes on a bit more, but hopefully you’ll get the drift. Now, John Updike was a novelist and poet, not a physicist, so we can excuse him for not getting it quite right, and no doubt that third line was just too tempting to resist … but of course a particle that “does not interact at all” would not be detectable, and so would not be a very sensible thing to postulate. This is certainly how Wolfgang Pauli felt when he postulated the existence of the neutrino in 1930; he said he had done a terrible thing, proposing a theory that could not be tested. It was a desperate measure, put forward to account for an apparent violation of the conservation of mass/energy in the beta decay of the neutron. In this decay, a neutron turns into a proton and an electron is emitted; the neutron is slightly heavier than the proton, and the mass difference is sufficient for the creation of an electron with kinetic energy, which duly speeds away from the nucleus in which it was born. But these electrons were known to have variable kinetic energies, which at their maximum value could account fully for the mass difference, but in other cases there appeared to be a loss of mass, and hence of energy.

However, even Pauli must have realised that there was some chance that a decay in which a particle came into existence could be reversed, with the absorption of that particle by another. It’s true that the early theory of beta decay due to Fermi featured a “4-particle vertex”, with the decaying neutron producing a proton, an electron and a neutrino; the reverse of that might naively be supposed to require the coincidence of the latter three particles to make a neutron, the probability of which would be extremely small. But when the theory of the weak interaction was developed, the 4-particle vertex became a 3-particle vertex, and the decay proceeded in two stages:


The inverse interaction is shown below. The anti-neutrino interacts with just one other particle – a proton in a nucleus – via a W boson, emitting a positron, and turning the proton into a neutron.


The possibility of such an interaction – in which the absorption of a neutrino effectively transmutes an element into one with one fewer proton and one more neutron – paved the way for the discovery of the neutrino by Reines and Cowan in 1956.

Now, if there really had been no theory predicting an interaction between neutrinos and other particles, any attempt to detect them would have looked a bit like Collins’ caricature of the search for gravitational waves. But of course we already know that gravitational waves do interact – by definition they are perturbations in spacetime, affecting the metric of spacetime and hence the way other particles, such as photons, propagate. So we build a very sensitive measuring device – an interferometer – and look for any such perturbations. Or rather, we build two interferometers, 3000 km apart. If a disturbance is detected simultaneously in both devices, it rules out any local, terrestrial source and hence points to events a long way away.

I sometimes think that the problems Collins and others have with concepts such as this may stem from a particular way of thinking about scientific phenomena, namely in a discrete, “all or nothing” sort of framework in which such phenomena are regarded as either there or not there – detected or not detected, producing “correct” or “incorrect” outcomes. But in reality, what we are doing almost all the time in modern physics is to measure things, and the measurements can take any rational values.

This discrete language even finds its way into the mouths of scientists occasionally; a perfect example is the experiment I did my PhD on, which was often described as “searching for the electric dipole moment of the neutron”, as though it could be “found” or “not found”, although the same people who described it thus knew very well that the experiment only returned a measurement and an associated probability, so that we could say, in a vague sort of way, that the dipole moment was probably no bigger than a certain value; but we could never say with certainty that it was zero, so, in that sense, we should not talk about whether or not it exists, because such questions are untestable, just like Pauli’s original neutrino hypothesis.

So my version of that Facebook post would go something like this:

“CALTECH measured perturbations in spacetime consistent with a theory of gravitational waves, using super-expensive technology designed to measure such perturbations, confirming Einstein’s prediction that large-scale fluctuations in gravitational energy could propagate to remote locations. These phenomena were expected because other predictions of general relativity had already been verified. A failure to detect them would, however, have cast doubt on the theory.”

Not as catchy as the original, perhaps, but possibly closer to the mark …