There are several examples in the literature where compactness properties of a cardinal $\kappa$ imply "bad" behavior of certain generic ultrapowers with critical point $\kappa$; particularly generic ultrapowers associated with tower forcings (Woodin's stationary tower forcing is an example of a tower forcing). I will discuss instances of this phenomenon due to Burke, Foreman-Magidor, and Cox-Viale.
I will discuss the Diagonal Reflection Principle (DRP), which is a highly simultaneous form of stationary set reflection that follows from strong forcing axioms like $PFA^{+\omega_1}$. DRP can be viewed as a weaker version of the statement "there is a normal ideal with completeness $\omega_2$ whose associated poset is proper (i.e. preserves stationary subsets of $[X]^\omega$ for all $X$)". In the presence of sufficiently absolute partitions of $cof(\omega)$, such ideals yield generic embeddings $j: V \to_G ult(V,G)$ with critical point $\omega_2$ such that large portions of $j$ are visible to $V$.
I will discuss the Diagonal Reflection Principle (DRP), which is a highly simultaneous form of stationary set reflection that follows from strong forcing axioms like $PFA^{+\omega_1}$. DRP can be viewed as a weaker version of the statement "there is a normal ideal with completeness $\omega_2$ whose associated poset is proper (i.e. preserves stationary subsets of $[X]^\omega$ for all $X$)". In the presence of sufficiently absolute partitions of $cof(\omega)$, such ideals yield generic embeddings $j: V \to_G ult(V,G)$ with critical point $\omega_2$ such that large portions of $j$ are visible to $V$.
In recent years there have been striking applications of infinitary
methods in finite combinatorics. I will survey some of these methods,
notably the theory of "flag algebras" due to Razborov.
We use the extent to which the combinatorial principle "weak
square" holds to quantify the influence of Martin's Maximum
on the combinatorics of singular cardinals.
(joint work with Menachem Magidor)
Generic embeddings are a generalization of large cardinal embeddings.
The difference is that they are definied in a forcing extension,
using an object in the ground model called a precipitous ideal. An
interesting feature is that the critical point can be quite small.
In this talk I will develop some basic properties, emphasizing the
similarities with large cardinals.
Which consistent statements can be forced to be true?
It is shown that "resectionable" \Sigma_1 statements about parameters
in H_{\omega_2} which are "honestly consistent" can be forced
to be true in a stationary set preserving extension, and we also show
that a strong form of BMM, according to which all "honestly consistent"
\Sigma_1 statements about parameters in H_{\omega_2} are true,
is consistent. We also give some applications.