MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I would like to know if the following two consequences of having a supercompact cardinal are orthogonal:

1) On one hand being supercompact is equivalent to being "A ineffable for all A" (Combinatorial characterization of supercompact cardinals, M. Magidor, Proc. Amer. Math. Soc. 42 (1974), 279-285).

2) On the other hand being supercompact implies having a Laver function (T. Jech, Set Theory, Theorem 20.21).

So from 1) for every "regressive" function on $P_\kappa(A)$ we have a "guessing" subset $B \subset A$. And from 2) we have a "diamond" sequence given by a function f that "guesses" every (small enough) set x in the ultrapower.

I am not sure if there is a connection between the two.

share|cite|improve this question
Compliment = מחמאה; whereas complement = משלים. :-) Your title in Hebrew would read שתי מסקנות מחמיאות מעל-קומפקטיות. Amusing, but I think you didn't mean this title but rather the other one. – Asaf Karagila Dec 5 '12 at 22:26
Can you say more precisely the particular definition of $A$-ineffable that Magidor uses? There are several different meanings of "regressive" here, and I think you intend a very strong one to get the equivalence. – Joel David Hamkins Dec 6 '12 at 0:06
up vote 3 down vote accepted

One difference is that it is an open question whether these properties are equivalent level-by-level in the supercompactness hierarchy.

Specifically, on the one hand, if $\kappa$ is $\theta$-supercompact, then one gets the ineffability property for all $A\subset P_\kappa\theta$, at the same level of supercompactness. That is, if $\kappa$ is $\theta$-supercompact, witnessed by embedding $j:V\to M$. If $\vec A=\langle A_\sigma\mid \sigma\in A\rangle$, where $A_\sigma\subset \sigma$, then consider $j(\vec A)(j''\theta)$, which is $j''B$ for some $B\subset \theta$. Almost every $\sigma$ has $A_\sigma=B\cap\sigma$, since $j''B=j(\vec A)(j''\theta)=j(B)\cap j''\theta$, and I think that this is the threading property you have in mind.

But meanwhile, on the other hand, although every supercompact cardinal $\kappa$ has a Laver function for $\theta$-supercompactness, Laver's argument is not level-by-level. Specifically, his argument shows only that every $2^{\theta^{\lt\kappa}}$-supercompact cardinal $\kappa$ has a Laver function for $\theta$-supercompactness. One needs a higher level of supercompactness initially than one gets for the Laver function.

In my paper A class of strong diamond principles I call attention to the following open question (question 5):

Question. If $\kappa$ is $\theta$-supercompact, must it have a $\theta$-supercompactness Laver function?

This question is, of course, connected with the number-of-measures problem, since if there is a Laver function for $\theta$-supercompactness, then of course there must be at least $2^\theta$ many such distinct normal fine measures on $P_\kappa\theta$, at least one for guessing each element of $H_{\theta^+}$. It is not known whether than can be exactly one, except in the case $\theta=\kappa$, which reduces to measurability.

Meanwhile, the existence of a $\theta$-supercompactness Laver function, although not yet known to follow from the $\theta$-supercompactness of $\kappa$, does not require any extra consistency strength, since as I show in the paper, if $\kappa$ is $\theta$-supercompact, then fast function forcing with $\kappa$ will add a generic Laver function for $\theta$-supercompactness.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.