[FOM] how would a physicist know that we are not living in a Skolem hull?

Dmytro Taranovsky dmytro at mit.edu
Thu Feb 7 19:17:19 EST 2013


On 02/04/2013 02:40 PM, Paul Elliott wrote:
> Suppose a physicist is using a mathematical theory saying
> that there are an uncountable number of points in space.
>
> In view of Skolem's result, what would such a statement mean really?

Current physical theories are not meant to be perfect.  The use of a 
mathematical model does not mean that the model is an exact literal 
description of the physical world, but only that it makes good 
predictions.  At this time, we do not know whether at Planck scale, 
space is continuous or discrete (or even whether points in space are 
physically meaningful at Planck scale).  Even if space is found to be 
discrete, many models will continue to use continuous space as a very 
useful simplifying assumption.  Using Skolem hulls would only complicate 
model formalism.

The metaphysical Skolem hull question is best addressed using Occam's 
razor:  Simpler explanations are preferred to more complicated ones, and 
it is simpler to claim that directed distance can be any real number as 
opposed to say any real number definable in second order arithmetic.  
Also, Skolem hulls are arbitrary because they depend on the 
expressiveness of the language in which they are defined, and, in the 
absence of uniformization, on the well-ordering.

> What experiment could convince us about the number of points in space being countable/uncountable?
An enumeration of points that is linked to experimental results by a 
general theory would be empirical evidence for countability of space 
points (assuming that using a different set of points would make the 
theory contrived).  Conversely, failure to find such enumeration would 
be evidence of continuous space (especially when physical theories are 
naturally described in terms of a continuum).  Other evidence is 
possible as well (including whether physical laws are recursive).  
However, if true, uncountability of space is not directly testable by 
finite observers, and ascertaining it uses philosophical reasoning.

Sincerely,
Dmytro Taranovsky


More information about the FOM mailing list