Tuesday, July 31, 2012

Space, time and gravitation after Einstein

To: Katie H.
From: Geoffrey Klempner
Subject: Space, time and gravitation after Einstein
Date: 5 February 2008 12:20

Dear Katie,

Thank you for your email of 23 January, with your first essay towards the ISFP Associate Award, entitled, 'Space, time gravitation and the world after Einstein.'

I take it (from the first sentence of the fifth paragraph) that this is intended as an article, and from the content I would assume that the article is written for a physics publication.

Throughout the piece you speak to an audience who are assumed to be fully conversant with the General Theory of Relativity. The aim of the article, so far as I can gather, is to describe an illuminating perspective on the theory, one which someone who knew the bare bones of the theory might not fully appreciate.

Although you mention philosophy early on, I could find much evidence of any real engagement with the philosophy of space and time, or the philosophy of physics. This is written for physicists rather than philosophers or students of philosophy.

However, I am trying to see how I can be most helpful to you. There are issues here which are worthy of being brought out, if you have a mind to do so.

Let's start with a plain assertion of fact. General Relativity is an empirical theory. As you state early on, there was a spectacular confirmation of the theory as long ago as 1919, and further confirmation has been obtained more recently. Einstein was first inspired to formulate his Special Theory of Relativity in response to the extraordinary results of the Michelson-Morley experiment, which was originally intended to measure the 'ether flow'.

What is the significance of the assertion I have just made?

Empirical propositions are logically contingent. That means that there is no logical contradiction in the supposition that an empirical proposition, P is false.

There is no logical contradiction in the supposition that no bending of light might have been observed in the solar eclipse experiment. There is no logical contradiction in the supposition that the Michelson-Morley experiment might have succeeded in measuring the ether flow.

With the benefit of hindsight, we can say that it would be extraordinary if Newton turned out to be right after all, and space and time are Newtonian not Einsteinian. What an ugly setup! Nevertheless, we have to accept that the 'best', i.e. most elegant and economical explanation is not *necessarily* true. It usually is, but not always. We don't have all the facts because we are not God. So we have to infer to the best explanation on the basis of the evidence available.

Einstein remarked that God does not play dice with the universe. He might also have said that you can't 'second guess' God. We rely on your 'God given' sense of what explanation is the best, but at the end of the day what laws of nature actually obtain is a matter of fact, not of necessity.

What complicates this picture is that the pioneers of 20th century physics often talked as if relativity could be deduced purely from a priori, epistemological considerations. This must be wrong, because you can't 'deduce' an empirical theory without making empirical assumptions.

Suppose you could deduce all the laws of physics simply through a pure mathematical derivation from the concept of symmetry. There is one, and only one set of laws that is symmetrical from every possible point of view. This idea describes an exciting research project which, to the best of my knowledge, has not yet reached any definite conclusion. Even if it did, it would still be necessary to *assume* that perfect symmetry does in fact hold in the real world.

You say at one point that Einstein's 'best idea' was that 'there is no way for an observer to distinguish in a local frame of reference between gravity and acceleration'. This led him to pose the question: how would the universe have to be if there was, in fact, no difference? We could be living in a Newtonian world where only God, observing the whole of space and time, was able to distinguish between gravity and acceleration. But we don't. The experiments show otherwise.

Important consequences follow from this. Insofar as it is possible to discover anything through philosophical inquiry -- in other words, by reasoning a priori -- we have to consider possible worlds where the laws of nature are different from what we believe them to be in the actual world.

If (a big 'if') there is room to debate the question which Newton's student Samuel Clarke and Leibniz grappled over, concerning the absolute vs the relative view of space and time, then it must be possible to do this, *bracketing* the question whether the universe is, in fact, Einsteinian or Newtonian.

One philosopher who is helpful on this point is Richard Swinburne in his book, 'Space and Time' which is still one of the best texts available. (The version I have is the 2nd edn. 1981. There might be a more recent edition.)

Supposing that there is still room for a priori inquiry into the absolute vs relativist views of space and time, how do we apply this to the current scientific theory? In what way, if any, can we raise the question of the validity of a Leibnizian-relativist view within the context of General Relativity?

I don't know of any philosopher who has done this adequately -- that is to say, to the complete satisfaction of physicists and philosophers. Swinburne doesn't go into the physics. However, it would be a very worthwhile task for you to try to find out.

All the best,

Geoffrey