home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!wupost!sdd.hp.com!spool.mu.edu!agate!ames!data.nas.nasa.gov!wk223.nas.nasa.gov!asimov
- From: asimov@wk223.nas.nasa.gov (Daniel A. Asimov)
- Subject: Re: The Continuum Hypothesis: Must it be {True or False}, or Not?
- References: <1992Dec11.162239.8548@cadkey.com> <1992Dec14.200024.6435@nas.nasa.gov> <1992Dec24.034938.11339@smsc.sony.com>
- Sender: news@nas.nasa.gov (News Administrator)
- Organization: NAS, NASA Ames Research Center, Moffett Field, California
- Date: Thu, 24 Dec 92 17:59:30 GMT
- Message-ID: <1992Dec24.175930.12890@nas.nasa.gov>
- Lines: 61
-
- In article <1992Dec24.034938.11339@smsc.sony.com> markc@smsc.sony.com (Mark Corscadden) writes:
- >In article <1992Dec14.200024.6435@nas.nasa.gov> asimov@wk223.nas.nasa.gov (Daniel A. Asimov) writes:
- >> Do people buy this, that FLT must be {true or false}, regardless of
- >> whether it is provably true or false (i.e., decidable)?
- >>Dan Asimov
- >
- >In email, Dan suggested using the phrase "having determinate truth-value"
- >to say that a proposition must be {true or false}, regardless of whether
- >it is provably true or false.
- >
- >I'll shorten this, simply saying that such a proposition is "determinate".
- >
- >To answer Dan's question, I buy - beyond any doubt - that FLT is
- >determinate. On the other hand I can't convince myself that CH (the
- >Continuum Hypothesis) is determinate, but at the same time I can't
- >find any necessary reason to believe that CH is indeterminate either.
- >
- > [...]
- >Mark Corscadden
- -----------------------------------------------------------------------------------
-
- I would like to comment on what I perceive to be the difference between
- the likelihood of FLT being determinate versus the likelihood of CH being
- determinate.
-
- Time.
-
- That is, how much "time" would it take to check either of these propositions?
-
- In the case of FLT, we could completely check its truthfulness in countable time:
- enumerate all candidates a,b,c,n, and do the arithmetic. After countable time is over,
- we know the answer. (Here the "unit" of time is the time it takes to calculate
- a^n + b^n - c^n and decide whether it is equal to 0.)
-
- In the case of CH, there are a lot more things to check: We can imagine considering
- all possible infinite subsets S of the reals R--all 2^c of them. For each such S, we
- can imagine all possible mappings f: S -> R, and all possible mappings g: S -> Z from
- S to the integers. And we can imagine checking whether f or g is bijective.
- If, for each S, either an f: S -> R or a g: S -> Z is bijective, then of course
- CH is true; otherwise it's false.
-
- An unthinkable number of things to check. However, the whole process could be broken
- down into tiny, easily-imaginable steps: LOTS of them. The set of all steps to carry
- out could be well-ordered. Then the steps could be (tediously) carried out in a very
- large cardinal (or shall we say ordinal?) amount of time.
-
- So IMHO, the salient difference between the likelihood of FLT's being determinate,
- and CH's being determinate, is that CH would simply take a *lot longer* to check.
- But since I can clearly conceive of carrying out either check, I personally believe
- that both of them are determinate.
-
- (Although in the case of CH, given the chance I would definitely delegate the
- responsibility of carrying out the check. (:-)>)
-
- Dan Asimov
- Mail Stop T045-1
- NASA Ames Research Center
- Moffett Field, CA 94035-1000
-
- asimov@nas.nasa.gov
- (415) 604-4799
-