Spacetime Metric: Which signature is better?

[LarryS]

It seems that, in general, mathematicians prefer the (-,+,+,+) signature for the Minkowski spacetime metric while most physicists prefer the (+,-,-,-) signature. Is there any evidence that Nature actually prefers one over the other?

As usual, thanks in advance.

 

[PAllen]

Nature cannot have a preference since no physics is changed. The only possible answer, IMO, is the one that you will make fewer mistakes with.

 

[Meir Achuz]

I think physicists prefer +--- for several reasons.
Time goes forward, space can go in either direction.
The four vector (E;p,p,p) has the length^2: E ^2-p^2=m^2. -m^2 would be awkward.

 

[Ben Niehoff]

I think pretty much anyone who works in dimensions other than 4 prefers (-+++...). There is a great benefit to not having to remember exactly how many minus signs you have to think about when working with Levi-Cività tensors, for example.

 

[Orodruin]

The question was what Nature prefers, not what physicists or mathematicians prefer. PAllen nailed it in the first sentence:

Nature cannot have a preference since no physics is changed.

The thread could have been locked at that point.

 

[PeterDonis]

The thread could have been locked at that point.

And now it is.