3 Forcing

Definition (Forcing). (β„™,≀,πŸ™) is called a forcing poset or forcing if it is a partial order and πŸ™βˆˆβ„™ and πŸ™ is the largest element. Elements of β„™ are called condition.

p≀q is interpreted as

β€œp is stronger than q”

β€œq is weaker than p”

Note. Unconventional.

Alternative: β€œJerusalem convention”. Interpret p≀q as β€œq is stronger than p”.

We are not following the Jerusalem convention.

Definition (Compatible). p and q are compatible if there is r≀p,q. Otherwise, we say they are incompatible (which we write as pβŠ₯q).

Definition (Antichain). AβŠ†P is an antichain if any two distinct elements of A are incompatible.

Definition (Dense). DβŠ†β„™ is dense if βˆ€β‘p∈TODO.

Definition (Filter). FβŠ†β„™ is called a filter if

  • (a) βˆ€β‘p,q∈F,βˆƒβ‘r∈F,r≀p,q
  • (b) βˆ€β‘p∈F,βˆ€β‘qβˆˆβ„™,qβ‰₯p⟹q∈F

If F only has property (a), we call it a filter base, and then

{pβˆˆβ„™:βˆƒβ‘q∈F,q≀p}

is the filter generated from F.

Note. Filters cannot contain incompatible elements.

Definition (D-generic). If D is a family of dense sets, then F is called D-generic if F is a filter and βˆ€β‘D∈D,D∩Fβ‰ βˆ….

Theorem 3.1. Assuming that:

  • D is countable

Then there is a D-generic filter.

Proof. Let D={Dn:nβˆˆβ„•}. Pick p0∈D0 arbitrarily and define by recursion pi+1 by picking some q≀pi with q∈Di+1.

Then {pi:iβˆˆβ„•} is a filter base, so let F be the filter generated by it. Then it is D-generic by construction. β–‘

Main example

Fix any sets X and Y

Fn⁑(X,Y):={p:pΒ isΒ aΒ finiteΒ fucntionΒ withΒ dom⁑(p)βŠ†XΒ andΒ range⁑(p)βŠ†Y}.

Define p≀q⟺pβŠ‡q and πŸ™:=βˆ….

What does pβŠ₯q mean?

pβŠ₯qβŸΊβˆƒβ‘x∈X,x∈dom⁑(p)∩dom⁑(q)∧p(x)β‰ q(x).

Lemma 3.2. Assuming that:

Then ⋃⁑F is a function.

Lemma 3.3. Dx:={pβˆˆβ„™:x∈dom⁑(p)} is a dense set. If D:={Dx:x∈X}, and F is D-generic, then dom⁑(⋃⁑F)=X.

Proof. If x∈X, find p∈F∩Dx, then x∈dom⁑(p), TODO β–‘

3.1 Cohen Forcing

Example 1

β„‚:=Fn⁑(Ο‰,2).

If F is D-generic, then ⋃⁑F:Ο‰β†’2 (by above).

Lemma 3.4. Assuming that:

  • Fix f:Ο‰β†’2 and define

    Nf:={pβˆˆβ„™:βˆƒβ‘u,p(u)β‰ f(u)}.

Then ⋃⁑Fβ‰ f.

Corollary 3.5. There is no F that is Dβˆͺ{Nf|f:Ο‰β†’2}-generic.

However: if M is a countable transitive model and you consider

Nf:={Nf:f∈M},

then by TheoremΒ 3.1, there is a DβˆͺNM-generic. And for this F, TODO

Example 2

Fn⁑(X,Y)=:β„™ as before.

Ry:={pβˆˆβ„™:y∈range⁑(p)}R:={Ry:y∈Y}

Lemma 3.6. Assuming that:

Then ⋃⁑F:Xβ†’Y is a surjection.

Corollary 3.7. Assuming that:

  • |Y|>|X|

Then there is no DβˆͺR-generic.

However: if M is a countable transitive model and, e.g. TODO

Example 3

Fn⁑(XΓ—Y,2). Assume X is infinite. Consider

Ey,yβ€²:={pβˆˆβ„™:βˆƒβ‘x∈X,p(x,y)β‰ p(x,yβ€²)}.

This is dense for y≠y′.

E:={Ey,yβ€²:yβ‰ yβ€²βˆˆY}.

Lemma 3.8. Assuming that:

Then there is an injection from Y into P(X).

Proof. Fix y and define

Ay:={x∈X:(⋃F)(x,y)=1}.

Ey,yβ€² guarantees that y↦Ay is an injection. β–‘

Corollary 3.9. Assuming that:

  • |Y|>|P(Ο‰)|

Then there is no DβˆͺE-generic.

However: if M is a countable transitive model of ZFC+CH, Ξ± is countable and so a DβˆͺE-generic exists.

Recap: M a countable transitive model of ZFC (or a sufficiently large finite fragment).

β„™βˆˆM: dense / filter / D-generic.

Example. Fn⁑(Ο‰,2) produces a new function f:Ο‰β†’2. Cohen forcing.

Example. Fn⁑(X,Y) produces a surjection f:Xβ†’Y. Fn⁑(Ο‰,Y) COLAPSE of Y.

Example. Fn⁑(XΓ—Y,2) produces an injection f:Yβ†’P(X).

If M is a countable transitive model of ZFC, then DM:={DβŠ†PΒ dense:D∈M} is countable, so by Theorem, we have a DM-generic.

Definition 3.10 (P-generic over M). We say F is β„™-generic over M if it is DM-generic. These always exist if M is a countable transitive model.

GOAL: Build an extension M[G] such that MβŠ†M[G], M[G] a countable transitive model of ZFC, G∈M[G] and M[G] is minimal.

Names

Idea: Think of elements of β„™ as β€œtruth values” for the von Neumann construction.

Name⁑0β„™:=βˆ…Name⁑α+1β„™:={Ο„:Ο„βŠ†Name⁑α×ℙ}Name⁑λℙ:=⋃α<Ξ»Name⁑αℙ Then
Name⁑ℙ:=β‹ƒΞ±βˆˆOrd⁑Name⁑αℙ

is the proper class of all names.

Consider: β„™={0,1}. Then Name⁑ℙ≅⁑V.

Since this is a recursive definition using absolute concepts, being a name is absolute for transitive models:

{Ο„:MβŠ¨Ο„Β isΒ aΒ β„™-name}=Nameβ‘β„™βˆ©M.

TODO

Examples

βˆ…βˆˆName⁑1β„™,

Ο„p:={(βˆ…,p)}∈Name⁑2β„™

β€œThe name for a set that contains βˆ… with value p.”

Ο„pq:={(Ο„p,q)}.

β€œThe name for whatever Ο„p describes with value q”.

Interpretation

If FβŠ†β„™, we interpret a β„™-name as follows:

val⁑(Ο„,F):={val⁑(Οƒ,F):βˆƒβ‘p∈F,(Οƒ,p)βˆˆΟ„}.

Important: This is a recursive definition.

Thus: the valuation is absolute for transitive models containing Ο„ and F.

Example.

  • (1) βˆ…: clearly val⁑(βˆ…,F)=βˆ….
  • (2) Ο„p:
    val⁑(Ο„p,F):={{βˆ…}p∈Fβˆ…pβˆ‰F
  • (3) Ο„pq:
    val⁑(Ο„pq,F):={βˆ…qβˆ‰F{{βˆ…}}q∈F∧p∈F{βˆ…}q∈F∧pβˆ‰F

    The relationship between p and q affects these possibilities.

    Example: if q≀p and F is a filter, then {βˆ…} is impossible.

    Example: if q=πŸ™ and F is a non-empty filter, then βˆ… is impossible.

    Example: if pβŠ₯q and F is a filter, then {{βˆ…}} is impossible.

Definition 3.11 (Generic extension). The (generic) extension for any countable transitive model M and any FβŠ†β„™ where β„™βˆˆM is

M[F]:={val⁑(Ο„,F):Ο„βˆˆNameβ‘β„™βˆ©M}.

Obviously, M[F] is a countable set with βˆ…βˆˆM[F] (Example (1)).

Also, by definition, M[F] is transitive.

Note:

M[F]⊨Extensionality+Foundation.
Need to show

Canonical Names

Definition 3.12 (Canonical name). Let x∈M. Define by recursion the canonical name for x by

wΛ‡:={(yΛ‡,πŸ™):y∈x}.

Lemma 3.13. val⁑(xΛ‡,F)=x if πŸ™βˆˆF.

Proof. Induction. β–‘

Corollary 3.14. MβŠ†M[F] if πŸ™βˆˆF.

Alternative construction of canonical names without πŸ™ is on Example Sheet 2.

Ξ“:={(pΛ‡,p):pβˆˆβ„™}.

Lemma 3.15. val⁑(Ξ“,F)=F.

Proof. Calculate:

val⁑(Ξ“,F)={val⁑(pΛ‡,F):p∈F}={p:p∈F}(byΒ previousΒ lemma)=F β–‘

Corollary 3.16. F∈M[F].

Remark. If N is a countable transitive model with MβŠ†N and F∈N, then M[F]βŠ†N. (by absoluteness of val⁑(Ο„,F)).

Warm-up: Suppose Οƒ,Ο„βˆˆName⁑ℙ. Define

up⁑(Οƒ,Ο„):={(Οƒ,πŸ™),(Ο„,πŸ™)}.

Pairing: Then

val⁑(up⁑(Οƒ,Ο„),F))={val⁑(Οƒ,F),val⁑(Ο„,F)}.

(β€œup” stands for unordered pair).

Corollary 3.17. M[F]⊨Pairing (if πŸ™βˆˆF).

Union: If Ο„ is a name, define

uΟ„:={(Οƒβ€²,r):βˆƒβ‘Οƒ,p,q,s.t.(Οƒ,p)βˆˆΟ„,(Οƒβ€²,q)βˆˆΟƒ,r≀p,q}.

Claim: val⁑(uΟ„,F)=⋃⁑val⁑(Ο„,F) if F is a filter.

Proof. Suppose z∈val⁑(uΟ„,F).

So z=val⁑(Οƒβ€²,F) for some (Οƒβ€²,r)∈uΟ„ with r∈F.

So βˆƒβ‘Οƒ,p,q with (Οƒ,p)βˆˆΟ„, (Οƒβ€²,q)βˆˆΟƒ, r≀p,q.

So p,q∈F. Hence val⁑(Οƒ,F)∈val⁑(Ο„,F), z=val⁑(Οƒβ€²,F)∈val⁑(Οƒ,F).

So zβˆˆβ‹ƒβ‘val⁑(Ο„,F).

Conversely, suppose zβˆˆβ‹ƒβ‘val⁑(Ο„,F). Then βˆƒβ‘y,z∈y∈val⁑(Ο„,F) (zβ†’(Οƒβ€²,q)βˆˆΟƒ with q∈F, yβ†’(Οƒ,ρ)βˆˆΟ„ with p∈F).

Hence since F a filter, find r≀p,q with r∈F.

Then (Οƒβ€²,r)∈uΟ„, so z∈val⁑(uΟ„,F). β–‘

TODO

Further Recap

We proved

M[G]⊨Extensionality + Foundation + Pairing + Union.

Homework was: Think about why power set is not easy.

β€œUnion” proof was:

collect all natural candidates of names for elements and assign the natural values.

Problem: If you try to do this for power set, neither the β€œnatural candidates for names” nor the β€œnatural values” are obvious. It’ll turn out that they are obvious in the end, but that requires some assistance.

Note: Separation and Replacement are even worse.

One remaining easy axiom: AC. By well-ordering theorem, AC holds if and only if βˆ€β‘x,βˆƒβ‘Ξ±,βˆƒβ‘i,i:xβ†’Ξ± injection. If x∈M[F], then there is ΟƒβˆˆName⁑ such that x=val⁑(Οƒ,F).

Notation. I write dom⁑(Οƒ):={Ο„:βˆƒβ‘p,(Ο„,p)βˆˆΟƒ}.

Consider dom⁑(Οƒ). In M, I have i:dom⁑(Οƒ)β†’Ξ± for some ordinal Ξ±. In M[F], define

y↦min⁑{i(Ο„):val⁑(Ο„,F)=y,Ο„βˆˆdom⁑(Οƒ)}.

Call this iβˆ—. Then iβˆ—:xβ†’Ξ± is an injection. So, AC holds in M[F].

Forcing

Definition (Forcing language). Fix M a countable transitive model of ZFC, β„™βˆˆM a forcing poset.

We call the language

Lforcing:=L∈βˆͺ{Ο„:β„™-names}

the forcing language (over M).

Definition (Interpretation of forcing language). If G is a β„™-generic over M and Ο† is in the forcing language, we say

M[G]βŠ¨Ο†

if and only if

M[G],vβŠ¨Ο†

where v(Ο„):=val⁑(Ο„,G).

Definition (Semantic forcing predicate). p⊩M,β„™Ο†: Forall G β„™-generic over M with p∈G, M[G]βŠ¨Ο†.

β€œp forces φ”

We often omit M,β„™.

Two theorems at the heart of forcing:

We are going to do the following:

Lectures 11 and 12.

Observations (under the assumption of the Forcing Theorem):

Proof of Separation in M[G]. Let Ο†(x,x1,…,xn) be an L∈-formula (not in the forcing language) and x∈M[G]. Need a name for

A:={z∈x:M[G]βŠ¨Ο†(z,pΒ―)}.

[For readability, we now drop parameters pΒ―].

Fix Οƒ such that val⁑(Οƒ,G)=x. Define

Ο±:={(Ο€,p):Ο€βˆˆdom⁑(Οƒ)∧pβŠ©Ο€βˆˆΟƒβˆ§Ο†(Ο€)}.

By the Definability Theorem, Ο± is a name in M.

Claim: val⁑(ϱ,G)=A.

  • β€œβŠ†β€ If z∈val⁑(Ο±,G), then there is (Ο€,p)∈ϱ such that z=val⁑(Ο€,G), p∈G. Since (Ο€,p)∈ϱ, we get Ο€βˆˆdom⁑(Οƒ), pβŠ©Ο€βˆˆΟƒβˆ§Ο†(Ο€). Together with p∈G, we get
    M[G]βŠ¨Ο€βˆˆΟƒβˆ§Ο†(Ο€)⏟z∈xβˆ§Ο†(z).

    Hence z∈A.

  • β€œβŠ‡β€ If z∈A, then z∈x and M[G]βŠ¨Ο†(z). So there is Ο€βˆˆdom⁑(Οƒ), z=val⁑(Ο€,G).

    By the Forcing THeorem, βˆƒβ‘p∈G, pβŠ©Ο†(Ο€). Also, there is q∈G and qβŠ©Ο€βˆˆΟƒ. Find r∈G, r≀p,q such that rβŠ©Ο€βˆˆrβˆ§Ο†(Ο€).

    Hence (Ο€,r)∈ϱ, so z=val⁑(Ο€,G)∈val⁑(Ο±,G). β–‘

Proof of power set. Example Sheet 3. β–‘

Proof of Replacement.

Since we already have Separation, it’s enough to show the following:

if Ο† is a functional formula and x∈M[G], then there is R∈M[G] such that

M[G]βŠ¨βˆ€β‘y∈x,βˆƒβ‘z∈R,Ο†(y,z,pΒ―)(βˆ—)

(as before, we suppress parameters for notational ease).

We work in M and identify a name ρ for R. Fix Οƒ such that x=val⁑(Οƒ,G). Find Ξ± large enough such that dom⁑(Οƒ)βŠ†VΞ±. Consider ψ(p,Ο€):=βˆƒβ‘ΞΌ,pβŠ©Ο†(Ο€,ΞΌ) (pβˆˆβ„™, Ο€ a name). By Definability Theorem, this is an L∈ formula.

By Levy Reflection Theorem, find πœƒ>Ξ± such that ψ is absolute between Vπœƒ and V=M. Define

ρ:={(ΞΌ,1):μ∈Vπœƒ}

and R:=val⁑(ρ,G).

Now we verify (βˆ—) holds: Fix y∈x, y=val⁑(Ο€,G). Since Ο† is functional, we know that Ο†(Ο€,ΞΌ) holds in M[G] for some ΞΌ. By Forcing Theorem, there is p∈G such that pβŠ©Ο†(Ο€,ΞΌ). So M⊨ψ(p,Ο€).

By absoluteness, VπœƒβŠ¨Οˆ(p,Ο€). This means βˆƒβ‘ΞΌβˆˆVπœƒ such that pβŠ©Ο†(Ο€,ΞΌ).

Thus: val⁑(μ,G)∈val⁑(ρ,G)=R.

β–‘

Definition (Dense below p). DβŠ†β„™ is called dense below p if βˆ€β‘q≀p,βˆƒβ‘r≀q,r∈D.

Lemma 3.18. Assuming that:

Then
  • (i) If E is dense below p, q≀p, then E is dense below q.
  • (ii) If {r:EΒ isΒ dense belowΒ r} is dense below p, then E is dense below p
  • (iii) Either G∩Eβ‰ βˆ… or βˆƒβ‘q∈G, βˆ€β‘r∈E,rβŠ₯q.
  • (iv) If p∈G, E is dense below p, then G∩Eβ‰ βˆ….

Proof. Example Sheet 3 β–‘

Definition of the syntactic forcing relation

pβŠ©βˆ—Ο†(τ¯).

Two recursions:

First β€œβŠ©βˆ—=” by recursion on Name⁑α.

Then β€œβŠ©βˆ—βˆˆβ€ without recursion.

Then the rest by recursion on formula complexity.

pβŠ©βˆ—Ο„0=Ο„1: if and only if

βˆ€β‘(Ο€0,s0)βˆˆΟ„0{q≀p:q≀s0β†’βˆƒβ‘(Ο€1,s1)βˆˆΟ„1,(q≀s1∧qβŠ©βˆ—Ο€0=Ο€1)}isΒ dense belowΒ p

and

βˆ€β‘(Ο€1,s1)βˆˆΟ„1{q≀p:q≀s1β†’βˆƒβ‘(Ο€0,s0)βˆˆΟ„0,(q≀s0∧qβŠ©βˆ—Ο€0=Ο€1)}isΒ dense belowΒ p

Remark. This is a recursion on Name⁑α.

pβŠ©βˆ—Ο„0βˆˆΟ„1: if and only if

{q≀p:βˆƒβ‘(Ο€,s)βˆˆΟ„1,(q≀s∧qβŠ©βˆ—Ο€=Ο„0)}

is dense below p.

Remark. No recursion involved, just β€œβŠ©βˆ—=”.

Recursion on complexity of formulas:

Remark. These definitions remind us of Kripke semantics for intuitionistic logic.

Lemma 3.19. The following are equivalent:

  • (i) pβŠ©βˆ—Ο†.
  • (ii) βˆ€β‘r≀p,rβŠ©βˆ—Ο†.
  • (iii) {r:rβŠ©βˆ—Ο†} is dense below ρ.

Proof. If this is true for Ο† of the form Ο„0=Ο„1 and of the form Ο„0βˆˆΟ„1, then it’s true for all formulas.

For Ο† atomic, we get that (ii) ⟹ (i) and (ii) ⟹ (iii) are trivial.

(i) ⟹ (ii) follows from Lemma 3.18(i).

(iii) ⟹ (i) follows from LemmaΒ 3.18(ii). β–‘

Strategy:

Theorem 3.20 (Syntactic Forcing Theorem). M[G]βŠ¨Ο† if and only if βˆƒβ‘p∈G,M⊨rβŠ©βˆ—Ο†.

Corollary 3.21 (Definability Theorem). pβŠ©Ο† if and only if pβŠ©βˆ—Ο†.

(p⊩MΟ†βŸΊM⊨pβŠ©βˆ—Ο†)

Corollary 3.22 (Forcing Theorem). M[G]βŠ¨Ο† if and only if βˆƒβ‘p∈G,pβŠ©Ο†.

This corollary is immediate from combining the two previously stated results.

Proof of Definability Theorem from Syntactic Forcing Theorem.

  • ⇐ This is just Syntactic Forcing Theorem.
  • β‡’ Suppose pβŠ©Ο†. By LemmaΒ 3.18, we need to show
    {r≀p:rβŠ©βˆ—Ο†}

    is dense below p.

    Suppose not. Then I find q≀p such that βˆ€β‘r≀q, rβ„βŠ©βˆ—Ο†. So qβŠ©βˆ—Β¬β‘Ο†.

    If q∈G, then p∈G. Since pβŠ©Ο†, M[G]βŠ¨Ο†. Since qβŠ©βˆ—Β¬β‘Ο†, M[G]βŠ¨Β¬β‘Ο†. Contradiction! β–‘

Proof of the Syntactic Forcing Theorem. The Theorem for = can be proved by induction on Name⁑α.

The Theorem for ∈ can be proved once the Theorem has been established for =.

Rest is induction on formula complexity.

Let’s do ¬⁑ and leave the rest to Example Sheet 3. Assume we have proved it for Ο†. We will prove it for ¬⁑φ.

  • β‡’ M[G]βŠ©Β¬β‘Ο†. Consider
    D:={p:pβŠ©βˆ—Ο†Β orΒ pβŠ©βˆ—Β¬β‘Ο†}.

    By definition of βŠ©βˆ—Β¬β‘, this is dense. Find p∈G∩D.

    By assumption, pβ„βŠ©βˆ—Ο†, so pβŠ©βˆ—Β¬β‘Ο†.

  • ⇐ Let p∈G such that pβŠ©βˆ—Β¬β‘Ο†. By definition, βˆ€β‘q≀p, qβ„βŠ©βˆ—Ο†. If M[G]β„βŠ¨Β¬β‘Ο†, thus M[G]βŠ¨Ο†. By induction hypothesis, find r∈G, rβŠ©βˆ—Ο†. So q≀r,p,q∈G.

    By Lemma, qβŠ©βˆ—Ο†, contradiction to βˆ€β‘q≀p,qβ„βŠ©βˆ—Ο†.

β–‘

Note. The Forcing Theorem is very useful! The inner details of the proof are not very important though. We won’t really discuss these details once we have proved the theorem.

Continuing the proof of Syntactic Forcing Theorem. Assume Syntactic Forcing Theorem for β€œ=” and prove if for β€œβˆˆβ€.

Want to show:

pβŠ©βˆ—Ο„0βˆˆΟ„1 if and only if

{q:βˆƒβ‘(Ο€,s)βˆˆΟ„1,(q≀s∧qβŠ©βˆ—Ο€=Ο„0)}

is dense below p.

Proof.

  • β‡’ Assume that M[G]βŠ¨Ο„0βˆˆΟ„1, i.e. val⁑(Ο„0,G)∈val⁑(Ο„1,G). Thus, there is (Ο€,s)βˆˆΟ„1 such that val⁑(Ο€,G)=val⁑(Ο„0,G) (⟺M[G]βŠ¨Ο€=Ο„0) and s∈G. So by Syntactic Forcing Theorem for =, find r∈G such that rβŠ©βˆ—Ο€=Ο„0.

    Find p≀r,s such that p∈G.

    Claim that D is dense below p. Let q∈D and pick the above (Ο€,s). Then we have q≀p≀s and qβŠ©βˆ—Ο€=Ο„0 (since q≀p≀r).

  • ⇐ Assume p∈G, pβŠ©βˆ—Ο„0βˆˆΟ„1, i.e. D is dense below p. By LemmaΒ 3.18(iv), find q∈G∩D, thus there is (Ο€,s)βˆˆΟ„1 such that q≀s, qβŠ©βˆ—Ο€=Ο„0. Since q∈G, s∈G, we have val⁑(Ο€,G)∈val⁑(Ο„1,G). By Syntactic Forcing Theorem for β€œ=”, we have val⁑(Ο€,G)=val⁑(Ο„0,G). So val⁑(Ο„0,G)∈val⁑(Ο„1,G). β–‘

Now we prove it for β€œ=”.

We prove this by induction on name rank, so assume that Syntactic Forcing Theorem for β€œ=” is true for names Ο€0,Ο€1 with smaller name rank.

Reminder: we need to show M[G]βŠ¨Ο„0=Ο„1 if and only if βˆƒβ‘p∈G, pβŠ©βˆ—Ο„0=Ο„1.

TODO: double check the below

Proof.

  • ⇐ Assume p∈G such that pβŠ©βˆ—Ο„0=Ο„1. Need to prove val⁑(Ο„0,G)=val⁑(Ο„1,G).

    We’ll show that the fact that D0 is dense below p implies val⁑(Ο„0,G)βŠ†val⁑(Ο„1,G). The other direction is the same proof but with 0 and 1 flipped.

    Proof of this: Let x∈val⁑(Ο„0,G), so find (Ο€0,s0)βˆˆΟ„0 such that s0∈G and x=val⁑(Ο€0,G). So, the corresponding D0 is dense below p. Find q≀s0,p such that q∈G. Then D0 is dense below q. So find r≀q such that r∈G∩D0. (Note that r≀q≀s0, so r≀s0).

    Since r∈D0 and r≀s0, find (Ο€1,s1)βˆˆΟ„1 such that r≀s1∧rβŠ©βˆ—Ο€0=Ο€1. By induction hypothesis, r∈G and rβŠ©βˆ—Ο€0=Ο€1, which implies x=val⁑(Ο€,v)=val⁑(Ο€1,G) (since (Ο€1,s1)βˆˆΟ„1 and s1∈G).

  • β‡’ Assume val⁑(Ο„0,G)=val⁑(Ο„1,G). Consider the set D:={r:rβŠ©βˆ—Ο„0=Ο„1Β orΒ Ξ¦r0Β Β βˆƒβ‘(Ο€0,s0)βˆˆΟ„0(r≀s0βˆ§βˆ€β‘(Ο€1,s1)βˆˆΟ„1,βˆ€β‘q((q≀s1∧qβŠ©βˆ—Ο€0=Ο€1)β†’qβŠ₯r)Ξ¦r1Β Β βˆƒβ‘(Ο€1,s1)βˆˆΟ„1(r≀s1βˆ§βˆ€β‘(Ο€0,s0)βˆˆΟ„0,βˆ€β‘q((q≀s0∧qβŠ©βˆ—Ο€0=Ο€1)β†’qβŠ₯r)}

    Claim: D is dense. If pβŠ©βˆ—, then p∈D, so nothing to show. If pβ„βŠ©βˆ—Ο„0=Ο„1, then there is some D0βˆ•D1 that fails to be dense below p.

    We’ll show: if D0 is not dense below p, then we can find r≀p such that Ξ¦r0 holds.

    (Other proof: D1 not dense below p β†’Ξ¦r1 is flipping 0 and 1).

    Suppose pβ„βŠ©βˆ—Ο„0=Ο„1 and there is (Ο€0,s0)βˆˆΟ„0 such that D0 is not dense below p. This means: there is r≀p such that

    βˆ€β‘q≀r(q≀s0βˆ§βˆ€β‘(Ο€1,s1)βˆˆΟ„1¬⁑(q≀s1∧qβŠ©βˆ—Ο€0=Ο€1).

    So, we get

    r≀s0βˆ§βˆ€β‘(Ο€1,s1)βˆˆΟ„1

    for any q that satisfies q≀s1∧qβŠ©βˆ—Ο€0=Ο€1. It can’t be compatible with r (finishes the proof of the claim that D is dense).

    Summary: We now have that D is dense.

    D={r:rβŠ©βˆ—Ο„0=Ο„1Β orΒ Ξ¦r0Β orΒ Ξ¦1r}.

    Claim 2: If r∈G, then neither Φr0 nor Φr1 holds (again, just do Φr0 and then flip 0 and 1 for Φr1).

    If Ξ¦r0 holds, then find (Ο€0,s0)βˆˆΟ„0 such that r≀s0 and the incompatibility statement holds. r∈Gβ†’s0∈G, so val⁑(Ο€0,G)∈val⁑(Ο„0,G). If (Ο€1,s1)βˆˆΟ„1 such that TODO

    Put everything together: since D is dense by Claim 1, find r∈D∩G. Therefore, by Claim 2, Φr0, Φr1 do not hold.

    Thus rβŠ©βˆ—Ο„0=Ο„1. β–‘

Summary: If G is Fn⁑(ω×℡2M,2)-generic, then M[G]⊨ZFC+there is an injection from β„΅2M into P(Ο‰).

This implies M[G]⊨ZFC+2β„΅0β‰₯β„΅2 if we have β„΅1M=β„΅M[G], β„΅2M=β„΅2M[G]. This is the goal for today’s lecture.

Note that our proof above is not good enough for the statement (βˆ—) from Lecture 8:

For all TβŠ†ZFC finite, there exists Tβˆ—βŠ†ZFC finite such that if M is a countable transitive model of Tβˆ—, then there is NβŠ‡M countable transitive model of Tβˆ—+¬⁑CH.

For this, we need to look more carefully at the proof of the GMT:

M⊨ZFC⟹M[G]⊨ZFC.

The proof proceeds AXIOM BY AXIOM and thus for each Ο†βˆˆZFC, we find finite SΟ† such that M⊨SΟ† implies M[G]βŠ¨Ο†.

Let SβŠ†ZFC finite such that S proves that all relevant notions (name, value, …) are well-defined and absolute. Then for TβŠ†ZFC finite, define

Tβˆ—:=Sβˆͺβ‹ƒΟ†βˆˆTSΟ†.

Then, the proof shows (βˆ—).

Let’s prove ¬⁑CH in M[G].

Definition (Preserves cardinals). We say β„™ preserves cardinals if βˆ€β‘G β„™-generic over M, β€œΞΊ is a cardinal” is absolute between M and M[G].

Definition 3.23 (Countable chain condition). We say β„™ has the countable chain condition (c.c.c.) if every antichain (note the anti!) in β„™ is countable.

Theorem 3.24. Assuming that:

Then β„™ preserves cardinals.

Theorem 3.25. Fn⁑(ω×℡2M,2) has the countable chain condition.

Corollary 3.26. Assuming: - G is Fn⁑(ω×℡2M,2)-generic over M, then

M[G]⊨ZFC+2β„΅0β‰₯β„΅2.

Lemma 3.27. Assuming that:

Then there is F∈M such that βˆ€β‘x∈X,F(x)βŠ†Y, βˆ€β‘x∈X,f(x)∈F(x) and MβŠ¨βˆ€β‘x∈X,F(x)Β isΒ countable.

Proof of TheoremΒ 3.24. Suppose M⊨κisΒ aΒ cardinal, M[G]⊨κ isΒ notΒ aΒ cardinal, so there is Ξ»<ΞΊ and f∈M[G], f:Ξ»β†’ΞΊ, f is a surjection. Apply the lemma to get F.

Define R:=⋃⁑α<Ξ»F(Ξ±).

Since βˆ€β‘x,f(x)∈F(x), R=ΞΊ.

But M⊨|R|=β„΅0β‹…Ξ»=Ξ»<ΞΊ. But then M thinks that ΞΊ is not a cardinal, contradiction. β–‘

Now we prove the lemma:

Proof. Let F(x):={y∈Y:βˆƒβ‘pβˆˆβ„™,pβŠ©Ο„(xΛ‡)=yΛ‡}.

Fix Ο„ a name for f.

By Definability Theorem , F∈M.

Then βˆ€β‘x∈X,F(x)βŠ†Y follows from definition.

βˆ€β‘x∈X,f(x)∈F(x) follows from Forcing Theorem.

Let’s look at M⊨F(x) is countable: If y∈F(x), let py be such that pyβŠ©Ο„(xΛ‡)=yΛ‡.

If yβ‰ yβ€², then pyβŠ₯pyβ€². Thus

{py:y∈F(x)}

is an antichain.

By countable chain condition, it’s countalbe. So F(x) was countable. β–‘

TODO

Proof of Theorem 3.25. Actually, we prove Fn⁑(X,Y) has countable chain condition whenever Y is countable.

Ξ”-systems form Example Sheet 3 Q33 are also called quasi-disjoint families.

(A family of finite sets D is called a Ξ”-system if there is a finite set R (called the root of the Ξ”-system) such that for all D,Dβ€²βˆˆD, if Dβ‰ Dβ€² then D∩Dβ€²=R).

Ξ”-system lemma: Any countable family of finite sets contains an uncountable Ξ”-system.

Take any AβŠ†β„™ uncountable and prove that it’s not an antichain.

If p∈A, then dom⁑(p)βŠ†X finite. Consider

S:={dom⁑(p):p∈A}.

That’s an uncountable family of finite sets, so by Ξ”-system lemma, find Ξ”-system DβŠ†S uncountable.

Let rβŠ†X finite be the root of D.

Since Y is countable, there are only countably many functions q:rβ†’Y. Since D is uncountable, by pigeonhole principle, there are p,q such that dom⁑(p),dom⁑(q)∈D and p|r=q|r. But since dom⁑(p)∩dom⁑(q)=r, p and q are compatible.

So A is not an antichain. β–‘

We got M[G]⊨2β„΅0β‰₯β„΅2.

Next time: What is the size of 2β„΅0 in M[G]?

Remember: LST Example Sheet 4: ZFC⊒2β„΅0β‰ β„΅Ο‰.

What if β„΅2 was one of the forbidden values?

Remark. Obtaining M[G]⊨2β„΅0=β„΅2 cannot be quite as general as this proof: if M⊨2β„΅0>β„΅2, then this will remain true in M[G].

Remark. If G is Fn⁑(ω×℡αM,2)-generic over M, then

M[G]⊨2β„΅0β‰₯β„΅Ξ±.

Previous lecture: Suppose M a countable transitive model of ZFC.

TODO

Question: (βˆ—) What are the possible values for 2β„΅0?

Mentioned last lecture: not all values are possible. In particular, 2β„΅0β‰ β„΅Ο‰.

Definition (Cofinal). CβŠ†ΞΊ is cofinal (= unbounded) if βˆ€β‘Ξ»<ΞΊ,βˆƒβ‘Ξ³βˆˆC,Ξ³β‰₯Ξ». We can then define:

cf⁑κ:={|C|:C is cofinal}.

Example 3.28. cf⁑℡1=β„΅1, cf⁑℡ω=β„΅0.

Lemma 3.29 (KΕ‘nig’s Lemma). ΞΊcf⁑κ>ΞΊ.

Then LST Example Sheet 4 Q10 is the special case ΞΊ=β„΅Ο‰, cf⁑κ=β„΅0 of KΕ‘nig’s Lemma.

Consequence for 2cf⁑κ: (2cf⁑κ)cf⁑κ=2cf⁑κ, thus 2cf⁑κ≠κ.

Preview of answer to (βˆ—): Every value not prohibited by KΕ‘nig’s Lemma is possible.

Now: β„΅2!

Definition. If β„™ is any forcing, let

OL⁑:={An:nβˆˆΟ‰}

be any Ο‰-sequence of antichains in β„™.

Let

Ο„OL⁑:={(ň,p):p∈An}.

We call these nice names.

If |β„™|=ΞΊ and β„™ has countable chain condition, then there are at most ΞΊβ„΅0 many antichains and thus at most (ΞΊβ„΅0)β„΅0=ΞΊβ„΅0β‹…β„΅0=ΞΊβ„΅0 many Ο‰-sequences of antichains and thus nice names.

Theorem 3.30. Assuming that:

  • M[G]⊨xβŠ†Ο‰

Then there is a nice name Ο„ such that val⁑(Ο„,G)=x.

Note. The Theorem does not need any assumptions about β„™.

Proof. Start with M such that val⁑(ΞΌ,G)=x (possibly not nice). Fix nβˆˆΟ‰. Fix either a well-ordering of β„™ (possibly using AC to get one) and build a maximal antichain An such that βˆ€β‘p∈An, p⊩ň∈μ.

TODO.

Claim: val⁑(ΞΌ,G)=val⁑(Ο„OL⁑,G).

βŠ‡: If n∈val⁑(Ο„OL,G), then there is p∈G such that (ň,p)βˆˆΟ„OL⁑, and then p∈An, so p⊩ň∈μ. So n∈val⁑(ΞΌ,G).

βŠ†: If n∈val⁑(ΞΌ,G), then by Forcing Theorem, get q∈G such that q⊩ň∈μ.

Subclaim: An∩Gβ‰ βˆ…. Indeed, by our lemma on compatibility ( Example Sheet 3), get qβ€²βˆˆG such that qβ€²βŠ₯p for all p∈An. Find r≀q,qβ€². Then r⊩ň∈μ. But that is in contradiction to An being maximal.

So find p∈G∩An. By definition, (ň,p)βˆˆΟ„OL⁑ and p∈G. So n∈val⁑(Ο„OL⁑,G). β–‘

Corollary 3.31. Assuming that:

Then M[G]⊨2β„΅0≀λ.

Proof. Follows directly from:

  • (a) Theorem.
  • (b) Calculation of the number of nice names.
β–‘
Main Application

If β„™=Fn⁑(ω×℡2M,2), then |β„™|=β„΅2M.

Calculate in M, β„΅2β„΅0.

Hausdorff’s Formula:

β„΅Ξ±+1β„΅Ξ²=β„΅Ξ±+1β‹…β„΅Ξ±β„΅Ξ².

So in particular:

β„΅1β„΅0=β„΅1β‹…β„΅0β„΅0=2β„΅0β„΅2β„΅0=β„΅2β‹…β„΅1β„΅0=β„΅2β‹…2

So β„΅2β„΅0=max⁑(β„΅2,2β„΅0).

By this calculation, if M⊨2β„΅0≀℡2, then M[G]⊨2β„΅0≀℡2.

Corollary 3.32. If M⊨CH, then M[G]⊨2β„΅0=β„΅2.

Remark. This proof also shows that if

M⊨2β„΅0≀℡n

and G is β„™-generic over M where β„™=Fn⁑(ω×℡2M,2), then

M[G]⊨2β„΅0≀℡n.

Corollary: If M⊨CH, then M[G]⊨2β„΅0=β„΅n.

Remark. What happens at β„΅Ο‰?

β„™:=Fn⁑(ω×℡ωM,2).

By general theory, M[G]⊨2β„΅0β‰₯β„΅Ο‰, but KΕ‘nig’s Lemma gives 2β„΅0β‰₯β„΅Ο‰+1.

What about the lower bound?

Our theorem and counting of nice names yields

M[G]⊨2β„΅0≀℡ω℡0⏟>β„΅Ο‰.

If M⊨GCH, then

β„΅Ο‰β„΅0≀℡ω℡ω=β„΅Ο‰+1.

So by KΕ‘nig’s Lemma, β„΅Ο‰β„΅0=β„΅Ο‰+1.

Therefore, if M⊨GCH, then M[G]⊨2β„΅0=β„΅Ο‰+1.

TODO

First limit cardinal that is a possible value of 2β„΅0 is β„΅Ο‰1.

Clearly, β„™=Fn⁑(ω×℡ω1M,2) adds injection from β„΅Ο‰1M into P(Ο‰). So M[G]⊨2β„΅0β‰₯β„΅Ο‰1.

Count nice names: |β„™|β„΅0=β„΅Ο‰1β„΅0.

If for all Ξ±<Ο‰1, β„΅Ξ±β„΅0≀℡ω1 (βˆ—), then β„΅Ο‰1β„΅0=β„΅Ο‰1.

β„΅Ο‰1β„΅0=β„΅Ο‰1={f|f:Ο‰β†’β„΅Ο‰1}⏟=:X=⋃α<Ο‰1{f|f:Ο‰β†’β„΅Ξ±}

(since Ο‰1 has no cofinal Ο‰-sequence).

Thus |X|≀℡1β‹…β„΅Ο‰1=β„΅Ο‰1 (by assumption (βˆ—)).

Thus M[G]⊨2β„΅0=β„΅Ο‰1.

TODO

What about 2β„΅1?

More on nice names:

Definition (lambda-nice names). Generalise β€œnice names” to Ξ»-nice names:

Let

OL⁑={Aα:α<λ}

be a family of Ξ» many maximal β„™-antichains:

Ο„OL⁑:={(Ξ±Λ‡,p):p∈AΞ±}

for α∈λ.

These are names for subsets of Ξ».

Observe that our theorem β€œevery AβŠ†Ξ» in M[G] has a Ξ»-nice name” still goes through.

If ΞΊ is an M-cardinal such that every antichain of β„™ has size ≀κ. [On Example Sheet 3, this is called the ΞΊ+-chain condition.]

Let ΞΌ:=|β„™| (in M). Then

(ΞΌΞΊ)Ξ»=ΞΌΞΊβ‹…Ξ»

is an upper bound on the number of Ξ»-nice names.

Thus

M[G]⊨2λ≀(ΞΌΞΊβ‹…Ξ»)M.

Question: Forcing with β„™:=Fn⁑(ω×℡2,2) and calculate 2β„΅1. Assume M⊨GCH. Then:

ΞΌ=|β„™|=β„΅2Ξ»=β„΅1Ξ»=β„΅0

(since β„™ has countable chain condition). So

M[G]⊨2λ≀℡2β„΅1β‹…β„΅0=(β„΅2β„΅1)M.

Calculate (β„΅2β„΅1)M:

β„΅2β„΅1=Hausdorff’sΒ formulaβ„΅2β‹…2β„΅1=GCHβ„΅2β‹…β„΅2=β„΅2.

Together:

M[G]⊨2β„΅1=β„΅2=2β„΅0.

Question: Is it possible to get PSA, i.e.

βˆ€β‘ΞΊ,Ξ»,Β Β ΞΊ<Ξ»β†’2ΞΊ<2Ξ»

without CH.

In particular, can we get

2β„΅0=β„΅22β„΅1=β„΅3

First idea: Force with Fn⁑(β„΅1Γ—β„΅3,2).

Unfortunately,

M[G]⊨2β„΅0=β„΅3.

Interpret the generic object G as fΞ±β†’β„΅1β†’2 for Ξ±<β„΅3.

Define gΞ±:=fΞ±|Ο‰.

Claim: For Ξ±β‰ Ξ±β€², gΞ±β‰ gΞ±β€². This is since

DΞ±,Ξ±β€²:={p:βˆƒβ‘uβˆˆΟ‰,gΞ±(u)β‰ gΞ±β€²(u)}

is still dense.

So, forcing with Fn⁑(β„΅1Γ—β„΅3,2) gives the same situation as forcing with Fn⁑(ω×℡3,2) for 2β„΅0, 2β„΅1.

Idea:

Fn⁑(X,Y,ΞΊ):={p:dom⁑(p)βŠ†X,range⁑(p)βŠ†Y,|p|≀κ}.

Thus Fn⁑(X,Y)=Fn⁑(X,Y,β„΅0).

Consider

β„™:=Fn⁑(β„΅1Γ—β„΅3,2,β„΅1).

Properties:

First goal: What about preserving cardinals?

Clearly, Fn⁑(β„΅1Γ—β„΅3,2,β„΅1) does not have the countable chain condition anymore.

With example (38) (on Example Sheet 3), we need to figure out the chain condition of Fn⁑(β„΅1Γ—β„΅3,2,β„΅1).

We need a Ξ”-system lemma for this: If Ξ» is regular (cf⁑λ=Ξ») and OL⁑ is a family of sets of size <Ξ» of size Ξ»+. Then there is a Ξ”-system DβŠ†OL⁑ of size Ξ»+.

This Ξ”-system lemma gives with the same proof as before:

Fn⁑(β„΅1Γ—β„΅3,2,β„΅1) has the β„΅2M-chain condition.

So: Fn⁑(β„΅1Γ—β„΅3,2,β„΅1) preserves cardinals β‰₯β„΅2M.

Closure

Definition (lambda-closed). A forcing β„™ is called Ξ»-closed if any family {pΞ±:Ξ±<Ξ³} for Ξ³<Ξ» that is a descending chain:

α<β⟹pβ<pα

there is q such that q≀pΞ± for all Ξ±<Ξ³.

Example. Fn⁑(β„΅1Γ—β„΅3,2,β„΅1) is β„΅1-cloesd.

[If {pΞ±} is a descending chain, then ⋃⁑α<Ξ³pΞ± is a condition in Fn⁑(β„΅1Γ—β„΅3,2,β„΅1).]

Theorem 3.33. Assuming that:

Then
P(κ)∩M=P(κ)∩M[G].

Corollary 3.34. Forcing with Fn⁑(β„΅1Γ—β„΅3,2,β„΅1)

  • (a)
    Does not change P(Ο‰).
  • (b)
    Therefore preserves β„΅1 (see Example Sheet 1 and the relation between codes for countable well-orders and preserving β„΅1).

Summary: Forcing with Fn⁑(β„΅1Γ—β„΅3,2,β„΅1) over a model of GCH gives M[G] with:

Hence 2β„΅1=β„΅3.





Forcing

Property

Preservation

Arithmetic





Fn⁑(ω×℡2,2)

countable chain condition

all cardinals

2β„΅0=β„΅2, 2β„΅1=β„΅2





Fn⁑(ω×℡3,2)

countable chain condition

all cardinals

2β„΅0=β„΅3, 2β„΅1=β„΅3, 2β„΅2=β„΅3





Fn⁑(β„΅1Γ—β„΅3,2,β„΅1)

β„΅1-closed.

If M⊨CH, then β„΅2-chain condition

β„΅1 preserved. (Closure lemma [not yet proved])

ΞΊβ‰₯β„΅2 preserved

If M⊨CH, then 2β„΅0=β„΅1, 2β„΅1=β„΅3.





Note GCHβ†’PSA, but our model of ¬⁑CH fails PSA.

Question: Can we have β„΅1<2β„΅0<2β„΅1?

Closure lemma:

Theorem. If β„™ is Ξ»-closed and ΞΊ<Ξ», then P(ΞΊ)∩M=P(ΞΊ)∩M[G].

Ξ»-closed: every descending sequence of length <Ξ» has a lower bound.

Proof. Let f∈M[G], f:ΞΊβ†’2 and assume towards contradiction that fβˆ‰M. β†’ fβˆ‰B.

B:={f∈M|f:Mβ†’2}.

Let Ο„ be a name for f.

By Forcing Theorem, there is p∈G such that pβŠ©Ο„:ΞΊΛ‡β†’2Λ‡βˆ§Ο„βˆ‰BΛ‡.

Construct a ΞΊ-sequence of conditions pΞ± TODO

TODO

The sequence {pΞ±:α≀κ} is defined in M (by Definability Theorem), so we can define

g(Ξ±)=1:⟺pΞ±+1βŠ©Ο„(Ξ±Λ‡)=1Λ‡.

Then g∈M.

But now pΞΊβŠ©Ο„(Ξ±Λ‡)=1Λ‡ or pΞΊβŠ©Ο„(Ξ±Λ‡)=0Λ‡ for all Ξ±.

uso pkβŠ©Ο„=Η§. Hence pΞΊβŠ©Ο„βˆˆBΛ‡.

But pκ≀pβŠ©Ο„βˆ‰BΛ‡. Contradiction. β–‘

Note that while Fn⁑(X,Y,β„΅1) is always β„΅1-closed, the chain condition depended on the value of β„΅1β„΅0.

The partial order Fn⁑(β„΅1Γ—β„΅3,2,β„΅1) has in general the (2β„΅0)+-chain condition.

CH implies (2β„΅0)+=β„΅2, so all cardinals are preserved.

However, if 2β„΅0>β„΅1, then there is a gap and we do not know whether TODO.

If M⊨2β„΅0=β„΅2, does

Fn⁑(β„΅1MΓ—β„΅3M,2,β„΅1M)

preserve β„΅2M?

Answer: Fn⁑(Ξ»+Γ—ΞΊ,2,Ξ»+) always adds a surjection from Ξ»+ to 2λ∩M. (βˆ—)

Application: If Ξ»=β„΅2 and M=2β„΅0=β„΅2, then Fn⁑(β„΅1Γ—ΞΊ,2,β„΅1M adds a surjection from β„΅1M onto P(β„΅0)∩M, i.e. β„΅2M. So |β„΅2M|=β„΅1M[G].

[Proof of (βˆ—). ] A generic for β„™ β€œis” a map f:Ξ»+Γ—ΞΊβ†’2.

Define h:Ξ»+β†’2Ξ» by

h(α)(β)=1:⟺f(α,β)=1.

Claim: h is a surjection onto 2λ∩M.

If g∈M, g:Ξ»β†’2, consider

Dg:={p|βˆƒβ‘Ξ±<Ξ»+,βˆ€β‘Ξ²<Ξ»,p(Ξ±,Ξ²)=1⟺g(Ξ²)=1}.

This is dense, and thus g∈range⁑(h). β–‘

Back to our question: Can we get β„΅1<2β„΅0<2β„΅1?

Start with M⊨GCH. Consider

β„™:=Fn⁑(β„΅1Γ—β„΅3,2,β„΅1)∩M

and let G be β„™-generic over M. Consider

β„š:=Fn⁑(ω×℡2,2)∩M[G],

and let H be β„š-generic over M[G].

Claim: M[G][H]βŠ¨β„΅1<2β„΅0<2β„΅1.

(M[G][H] is a forcing iteration).

This proves the claim.

Important: The order of forcings matters!

Suppose M⊨GCH.

β„šβ€²:=Fn⁑(ω×℡2,2)∩M.

H β„šβ€²-generic over M.

β„™β€²:=Fn⁑(β„΅1Γ—β„΅3,2,β„΅1)∩M[H].

G β„™β€²-generic over M[H].

Consider M[H][G]. Then

M[H]⊨2β„΅0=β„΅2=2β„΅1.

But that means that forcing with β„™β€² will collapse β„΅2M=β„΅2M[H].

Since β„™β€² is β„΅1-closed,

P(Ο‰)∩M[H]=P(Ο‰)∩M[H][G].

In particular, M[H][G]⊨CH. So, this order does not achieve what we want.

Final Remark on Forcing CH

Assume M⊨2β„΅0=β„΅2.

Question: Can you obtain M[G]⊨CH?

The natural forcing would be

β„™:=Fn⁑(Ο‰,β„΅1M).

This collapses β„΅1M; it does not have the countable chain condition, but since it has size β„΅1M, it has the β„΅2M-chain condition, so all cardinals β‰₯β„΅2M are preserved.

Clearly therefore:

M[G]⊨|P(Ο‰)∩M|=β„΅2M=β„΅1M[G].

But: is |P(Ο‰)∩M|=|P(Ο‰)∩M[G]|?

Nice names: gives upper bound of

(β„΅1M)β„΅1Mβ‹…β„΅0=(2β„΅1M)M.

That’s not surprising, since any AβŠ†β„΅1 in M becomes a new subset of Ο‰ in M[G] via the new bijection between Ο‰ and β„΅1M.