Monday, May 6, 2013

Possibility of simultaneous and retro causation

To: Mark S.
From: Geoffrey Klempner
Subject: Possibility of simultaneous and retro causation
Date: 14th January 2010 13:36

Dear Mark,

Thank you for your email of 4 January, with your essay for the University of London BA Metaphysics module, in response to the question, 'Can effects ever precede or be simultaneous with their causes?'

This is an impressive piece of work, both in terms of the research you have done into the questions of simultaneous and retro-causation, and the persuasive way you develop your case.

I will deal with simultaneous causation first, as this seems the easier of the two scenarios.

The key question here is why anyone would have objections to simultaneous causation, given that this is something we observe all the time.

In discussing Hume, you offer two thoughts: first, 'regularity theory... is incompatible with simultaneous causation', and Hume's theory of time, according to which time is 'divided into smallest parts with a 'next instant' following each part.'

Although you make a point of investigating the metaphysical possibility of different kinds of causation rather than nomological possibility, the view of temporal instants as forming a continuum is an empirical claim. It follows that in some possible world (which according to Huemer and Kovitz is the actual world) simultaneous causation is possible. That's all you need for your case.

But what about possible worlds where simultaneous causation is (nomologically) impossible? Imagine a universe which 'ticks', like a clock. We place a ball on a cushion, and watch very carefully in slow motion. At each tick the downward 'movement' (so-called) of the ball corresponds with an 'increase' in the indentation of the cushion. Clearly, this won't do as an illustration of what Hume had in mind. Then what did he think? How is Hume's picture essentially different from the 'ticking' universe? It wasn't clear from what you say.

On the question of 'regularity theory', I don't see why this is obviously incompatible with simultaneous causation. Let's put the ball on the cushion again. A priori, as Hume would say, I don't know that when I put a ball on a cushion (or, a ball of this particular composition on a cushion of this particular fabric) the ball won't merge with the cushion, or that the cushion won't shatter like a piece of china. Try the experiment as many times as it takes to convince. Ergo, ball causes indentation on cushion.

I guess your idea is that a statement of regularity will always take the form, 'A at t1 is always followed by B at t2'. This is true of the ball and cushion. However, in the ball and cushion case, it is ALSO true that B (indentation) occurs at t1. Where's the problem? I just don't see it.

With regard to retrocausation, I agree with your strategy of discounting views of time/ causation according to which retrocausation is impossible, and focusing on the case (B-series view of time, block universe) where it at least looks, prima facie possible.

However, if we are in the game of looking for views of time/ causation according to which retrocausation is not automatically ruled out, then it is at least worth considering whether we can construct a view of the universe where retrocausation is a built-in possibility (never mind the physics).

I wrote something about this once, in the context of theories of time travel. (See my Afterword to David Gerrold 'The Man Who Folded Himself'

Consider the following scenario. I planned to take a bath before going out this afternoon, but discover that I forgot to turn on the water heater. So I tap my time belt, go back 40 minutes in time, and switch the water heater on, then return to the present. Hey presto, the water is nice and hot, and I take my bath.

You cannot deny that this is a possible experience. All we have to do now is construct a model of the universe around this which does not lead to any logical contradiction.

The most patent logical contradiction, given any one of various currently accepted views of the universe, is that at 12.20 am exactly the water heater was not on, and at 12.20 am exactly the water heater was on.

David Gerrold's solution (which some sci-fi aficionados regard as cheating, supposedly because it fails to grapple with the deeper mysteries of time travel) is to hold that by tapping my time belt, I create -- or 'hop' to -- another universe. (You can never hop back, only forwards.)

One objection would be that so far as the time sequence is concerned, the event of switching on the water heater is in the future relative to my tapping the time belt, even if the new universe I hop to (or create) has local clocks showing the time to be 40 minutes ago. That raises the question of what we want out of retrocausation. It would certainly satisfy me, if I could make up for lapses in memory by means of a quick 'visit' to the 'past'.

There are at least a family of cases where it would make sense to talk about retro causation (without blushing), even if it isn't exactly the retro causation envisaged in the original question. In this respect, the question is rather similar to questions about 'what we would say if' various science fiction possibilities in relation to personal identity were realized.

All the best,