(This is the third lecture in Stevo Todorcevic’s Forcing class, held in the fall of 2012. You can find the second lecture here. Quotes by Stevo are in dark blue; some are deep, some are funny, some are paraphrased so use your judgement.)
- We look at 4 examples of posets that collapse
to
.
- A problem of Foreman is mentioned that asks about collapsing versus adding.
- Talking about anti-Suslin trees.
Tying up loose ends from last class is the proof that . Thanks to Oswaldo for explaining it here (pdf).
On a bureaucratic note, Stevo intends this course to be a supplemental course to Alan Dow’s course in Forcing at the Fields Institute. As such the classes might just end up being Stevo answering students’ questions about Alan Dow’s course. “I am Alan Dow’s teaching assistant.” Today is one of those days, and we chat about a variety of ways of collapsing cardinals.
Collapsing 
First we recall the usual poset that collapses . This is
which is finite partial functions from
to
. You can see how a generic object will be a function from
onto the ground model
.
Theorem (Kripke). For
we have
and
.
This tells us an interesting universality property:
Corollary. If
then
completely embeds into
.
proof. Take . [QED]
Fact (Gaifman). The regular open algebra of this collapsing poset,
is countably completely generated.
“This is one way to collapse. Do you know any others?”
Example 1. Countable support product of Cohen
Exercise. Show .
. “Telling you what ‘the reals’ is is the hint.“
Problem (Foreman). Is there a (nontrivial) forcing notion which does not add reals, nor collapses cardinals?
Note. Under CH or SH there is a positive answer to this.
Theorem (Stevo). Under PFA, Every poset which adds a subset of either adds a real or collapses
.
Example 2. Shoot a club.
I will add details to this soon.
Example 3. Mild variations of Levy Collapse
Example 4. A “drastic” example of type 2.
Take a coherent Suslin Tree and force with
.
Coherent means that is a closed subset of
and
we have
is finite.
“Why can you see [that is collapsed] without proof? A branch in one coordinate is one world, a branch in the other is another world.”
Anti-Suslin
“What is a good example of an Aronszajn tree that is Baire and anti-Suslin?” A tree is anti-suslin if every uncountable set contains an uncountable antichain.
Note. “Special is on the opposite extreme of Suslin.”
Almost Example. Take that is stationary, co-stationary. Take
compact subsets ordered under end-extension
.
Note. “Writing down what it means to be anti-Suslin, you will see it is too complex to be a consequence of diamond.”
Note (by Assaf Rinot). Suppose has a Suslin subtree
. Force with it and this addds a branch and also a club
, but this must contain a club
in the ground model.
There’s an = sign missing from the Kripke theorem. Furthermore, on my screen at least, the checks look like hats, and they are displayed to the right of the base character rather than above.
LikeLike
Fixed the equal sign. The checks really are just carets, as Mathjax doesn’t seem to like accents. I’ll look into it.
LikeLike