This is the second part of a series exploring Arthur Eddington’s philosophy of science. The first part can be found here.
——————————–
After making the case for a priori methods in physics as a way of extracting certainties about our observations, Eddington outlines some examples of such ideas that were the impetus for modern physics.
I would emphasize the question “What is it we really observe?”… . Because as soon as we ask the question, the classical scheme of physics is a punctured bubble; and we start on a path of revolution of which perhaps the end is not yet in sight. (p 31)
He offers two destructive examples. Destructive, because they makes claims on what is not possible to observe.
Relativity
The foundation of the special theory of relativity… was the realization that “velocity of the aether” is unobservable. (p 32)
The discovery … was that the assertion that the aether has a relative velocity of 80 km/s turns out on scrutiny not to specify any observational procedure at all. The ingenuity of the experimenter is not called upon; for he cannot be asked to devise a practicable procedure equivalent to a procedure which has never been specified. (p 33)
This rings true with Einstein’s own account of his discovery of Special Relativity. He denied every having heard of the Michelson Morley experiment (which failed to detect the aether) prior to his initial publications, arriving at his conclusion by finding a paradox in the concept of ‘simultaneity’, and considering asymmetries in electromagnetism.
Eddington says as much later when he remarks that:
The unobservability of distant simultaneity is essentially the same principle as the unobservability of aether velocity, but it is free from the possibly ambiguous phraseology associated with the old aether hypotheses. (p 37)
Quantum Theory
In quantum statistics, the distinguishability of the particles affects the outcome of future observations.
For instance, part of the mathematics of quantum statistics includes quantities such as ‘A is to the west of B’, where A & B are two distinguished particles. From this mathematical model, we can make very good predictions about the future behaviors of these particles.
Contrastingly, if all the available data about A and B are identical, the particles are indistinguishable. This means that ‘A is to the west of B’ is undefined – we don’t know which is A and which is B! In these cases, we need a different mathematics. These models, such as for Bose-Einstein condensates can lead to some very counterintuitive predictions. More strangely, these predictions are very accurate!
Eddington remarks that many people find this difficult to believe.
It is naturally objected that the particles cannot be affected by our inability to distinguish them, and it is absurd to suppose that they modify their behavior on that account. (p 36)
As he made plain at the start of the book, Eddington’s philosophical view includes the idea that our observations are shaped by our intellectual and sensory equipment. So, he agrees with the objection, but point out that our physics doesn’t describe the objective particles, as is often assumed.
The objective particles are unconcerned with our inability to distinguish them; but they are equally unconcerned with the behavior which we attribute to them partly as a consequence of our failure to distinguish them. It is this observable behavior, and not the objective behavior, that we are concerned with. (p 36)
It is a very interesting attempt to find an intuitive philosophical explanation for the seemingly counter-intuitive results of quantum theory.
——————————–
These two examples, Eddington claims, are ‘epistemological principles’. They tell us – with certainty – that ’distinguishable protons’ and ‘simultaneous events’ are unobservable because the ideas themselves are self-contradictory.
Since the discrimination of unobservables depends on a study of the procedure of obtaining observational knowledge, or alleged observational knowledge, and not on a study of the results of carrying out the procedure, it comes under scientific epistemology; and a principle of unobservability, such as the special relativity principle, the uncertainty principle, or the modified mechanics of indistinguishable particles, is an epistemological principle. Such principles have an altogether different status from physical hypotheses, though they lead to the same kind of practical consequences. (p 38)
And again, in summary:
Unobservability of a quantity arises from a logical contradiction in the definition which professes to specify the procedure for observing it. (p 40)
——————————–
Anticipating that the reader may find a disparity between the initial claims and the ideas presented so far, Eddington admits that he is yet to get to the a priori laws of physics he claims are possible.
By developing the consequences of this unobservability, we can deduce laws of nature which had previously been discovered or suggested empirically, and thereby transfer them from an a posteriori to an a priori basis; but there is seemingly little as yet to support the outlook which I have called selective subjectivism. (p 40)
I am intrigued to see where this is going. My initial suspicion is that these so-called ‘epistemological’ considerations are really dependent upon physical laws.
I’ll post more when I’ve digested the rest of his book.
