Subject: Why is there something rather than nothing?
Date: Thu, 9 Aug 2001 16:14:27 -0700
From: "Brian Holtz" <Brian.Holtz@sun.com>
To: <brian.holtz@sun.com>, <pgxf@lsil.com>

"Paul Filseth" <pgxf@lsil.com> wrote [in alt.atheism.moderated] :

> > > "Why is there something instead of nothing?" cannot be explained
> > > correctly, because any answer you might offer qualifies as
> > > "something" and is therefore part of what you were supposed to
> > > explain.
> >
> > Only if you assume that the explanatory relation can never
> > be reflexive (A explains A) or symmetric (A explains B and
> > B explains A).
>
>      That's not an assumption; it follows from what it means to
> explain something.

Not necessarily. Nozick writes that "a small literature exists
that attempts to formulate precise conditions whereby circular
explanations are excluded" and cites work by Hempel and others.

> > In his _Philosophical Investigations_ Nozick explores the alternative
> > idea of "explanatory self-subsumption" and gives a hypothetical
> > example:
> > P says: any lawlike statement have characteristic C is true.
> > Let us imagine this is our deepest law[...] Next we face the
> > question of why P holds true, and we notice that P itself has
> > characteristics C. [..] Our question is not whether such
> > self-subsumption as an instance of itself can constitute
> > a proof, but whether it can constitute an explanation. [..]
> > <snip> If a brute fact is something that cannot be explained by
> > anything, then a self-subsumable principle isn't a brute fact; <snip>
>
>      To me, that looks an awful lot like Nozick just stuck in his
> conclusion as a premise.

He asked whether self-subsumption could reduce the
brute-fact quality of a deepest law, and considered two possible
senses of "brute fact". Under one sense (which you quoted) it
does reduce it, while under the other it doesn't.  Still,
non-self-subsuming deepest laws would be considered brute under
both senses, and so self-subsumption indeed seems to reduce (but
perhaps not eliminate) this brute-fact quality.

> > But I'm
> > not so ready to completely dismiss the whole concept of necessary
> > existence.  In particular, I wonder if logical possibility itself
> > exists necessarily?  If nothing existed instead something, would
> > it therefore be true that nothing can even possibly exist?
> >
> > If the answer is no, then there may be an answer to the Big Why
> > ("why is there something rather than nothing?"). The answer might be:
> > Nothing exists except logical possibility, which necessarily exists,
> > and our perception of material existence is an epiphenomenon of our
> > being logical subcomponents of a logically possible universe.
>
>      Well, let's start by not being "deceived by grammar".  I don't
> know what it means to say logical possibility exists.  The formal way
> to talk about existence is to phrase it as quantification over some
> set of properties.  You don't say "God exists."; you say "There exists
> an x such that ((x created the universe) and (x is intelligent) and
> (For all y, if (y created the universe) then (y = x)))."  Existence
> statements that can't be rephrased that way are category errors.  Can
> you put "Logical possibility exists necessarily" into that form?  What
> are the properties of logical possibility?

So how would we say nothing exists in 1st-order PC?  I guess in
2nd-order PC it might be: (for all x)(for all P)(! Px).

But to deal with possibility, don't we need modal logic and its
operators M (it-is-possible-that) and L (it-is-necessary-that)?
The modal logical system S5 is built on the extra assumption that
Mp -> LMp: if a proposition is possible, its being possible is a
necessary truth. This assumption sounds like what I meant above,
but it is an axiom rather than a theorem.

Hmm, I guess I need to learn more about modal logic:
http://plato.stanford.edu/entries/logic-modal/

--
Brian.Holtz@sun.com
Knowledge is dangerous. Take a risk:
http://humanknowledge.net