Fraenkel \(\cal N2(\aleph_{\alpha})\): Jech's Model | Back to this models page

Description: This is an extension of \(\cal N2\) in which \(A=\{a_{\gamma} : \gamma\in\omega_{\alpha}\}\); \(B\) is the corresponding set of \(\aleph_{\alpha}\) pairs of elements of \(A\); \(\cal G\)is the group of all permutations on \(A\) that leave \(B\) point-wise fixed;and \(S\) is the set of all subsets of \(A\) of cardinality less than\(\aleph_{\alpha}\)

When the book was first being written, only the following form classes were known to be true in this model:

Form Howard-Rubin Number Statement
6

\(UT(\aleph_0,\aleph_0,\aleph_0,\Bbb R)\): The union of a denumerable  family  of denumerable subsets of \({\Bbb R}\) is denumerable.

37

Lebesgue measure is countably additive.

43

\(DC(\omega)\) (DC), Principle of Dependent Choices: If \(S\)  is  a relation on a non-empty set \(A\) and \((\forall x\in A) (\exists y\in A)(x S y)\)  then there is a sequence \(a(0), a(1), a(2), \ldots\) of elements of \(A\) such that \((\forall n\in\omega)(a(n)\mathrel S a(n+1))\).  See Tarski [1948], p 96, Levy [1964], p. 136.

44

\(DC(\aleph _{1})\):  Given a relation \(R\) such that for every  subset \(Y\) of a set \(X\) with \(|Y| < \aleph_{1}\) there is an \(x \in  X\)  with \(Y \mathrel R x\), then there is a function \(f: \aleph_{1} \rightarrow  X\) such that \((\forall\beta < \aleph_{1}) (\{f(\gamma ): \gamma < b \} \mathrel R f(\beta))\).

63

\(SPI\): Weak ultrafilter principle: Every infinite set has a non-trivial ultrafilter.
Jech [1973b], p 172 prob 8.5.

87-alpha

\(DC(\aleph_{\alpha})\): Given a relation \(R\) such that for every subset \(Y\) of a set \(X\) with \(|Y|<\aleph_{\alpha}\), there is an \(x\in X\) with \(Y\mathrel R x\) then there is a function \(f:\aleph_{\alpha}\to X\) such that (\(\forall\beta < \aleph_{\alpha}\)) \(\{f(\gamma): \gamma < \beta\}\mathrel R f(\beta)\).

91

\(PW\):  The power set of a well ordered set can be well ordered.

130

\({\cal P}(\Bbb R)\) is well orderable.

135

If \(X\) is a \(T_2\) space with at least two points and \(X^{Y}\) is hereditarily metacompact then \(Y\) is  countable. (A space is metacompact if every open cover has an open point finite refinement. If \(B\) and \(B'\) are covers of a space \(X\), then \(B'\) is a refinement of \(B\) if \((\forall x\in B')(\exists y\in B)(x\subseteq y)\). \(B\) is point finite if \((\forall t\in X)\) there are only finitely many \(x\in B\) such that \(t\in x\).) van Douwen [1980]

191

\(SVC\): There is a set \(S\) such that for every set \(a\), there is an ordinal \(\alpha\) and a function from \(S\times\alpha\) onto \(a\).

273

There is a subset of \({\Bbb R}\) which is not Borel.

305

There are \(2^{\aleph_0}\) Vitali equivalence classes. (Vitali equivalence classes are equivalence classes of the real numbers under the relation \(x\equiv y\leftrightarrow(\exists q\in{\Bbb Q})(x-y=q)\).). \ac{Kanovei} \cite{1991}.

309

The Banach-Tarski Paradox: There are three finite partitions \(\{P_1,\ldots\), \(P_n\}\), \(\{Q_1,\ldots,Q_r\}\) and \(\{S_1,\ldots,S_n, T_1,\ldots,T_r\}\) of \(B^3 = \{x\in {\Bbb R}^3 : |x| \le 1\}\) such that \(P_i\) is congruent to \(S_i\) for \(1\le i\le n\) and \(Q_i\) is congruent to \(T_i\) for \(1\le i\le r\).

313

\(\Bbb Z\) (the set of integers under addition) is amenable.  (\(G\) is {\it amenable} if there is a finitely additive measure \(\mu\) on \(\cal P(G)\) such that \(\mu(G) = 1\) and \(\forall A\subseteq G, \forall g\in G\), \(\mu(gA)=\mu(A)\).)

361

In \(\Bbb R\), the union of a denumerable number of analytic sets is analytic. G. Moore [1982], pp 181 and 325.

363

There are exactly \(2^{\aleph_0}\) Borel sets in \(\Bbb R\). G. Moore [1982], p 325.

368

The set of all denumerable subsets of \(\Bbb R\) has power \(2^{\aleph_0}\).

369

If \(\Bbb R\) is partitioned into two sets, at least one of them has cardinality \(2^{\aleph_0}\).

When the book was first being written, only the following form classes were known to be false in this model:

Form Howard-Rubin Number Statement
3  \(2m = m\): For all infinite cardinals \(m\), \(2m = m\).
15

\(KW(\infty,\infty)\) (KW), The Kinna-Wagner Selection Principle: For every  set \(M\) there is a function \(f\) such that for all \(A\in M\), if \(|A|>1\) then \(\emptyset\neq f(A)\subsetneq A\). (See Form 81(\(n\)).  

40

\(C(WO,\infty)\):  Every well orderable set of non-empty sets has a choice function. Moore, G. [1982], p 325.

45-n

If \(n\in\omega-\{0,1\}\), \(C(\infty,n)\): Every set of \(n\)-element sets has a choice function.

88

  \(C(\infty ,2)\):  Every family of pairs has a choice function.

111

\(UT(WO,2,WO)\): The union of an infinite well ordered set of 2-element sets is an infinite well ordered set.

122

\(C(WO,<\aleph_{0})\): Every well ordered set of non-empty finite sets has a choice function.

163

Every non-well-orderable set has an infinite, Dedekind finite subset.

192

\(EP\) sets: For every set \(A\) there is a projective set \(X\) and a function from \(X\) onto \(A\).

193

\(EFP\ Ab\): Every Abelian group is a homomorphic image of a free projective Abelian group.

327

\(KW(WO,<\aleph_0)\),  The Kinna-Wagner Selection Principle for a well ordered family of finite sets: For every well ordered set \(M\) of finite sets there is a function \(f\) such that for all \(A\in M\), if \(|A|>1\)  then \(\emptyset\neq f(A)\subsetneq A\). (See Form 15.)

344

If \((E_i)_{i\in I}\) is a family of non-empty sets, then there is a family \((U_i)_{i\in I}\) such that \(\forall i\in I\), \(U_i\) is an ultrafilter on \(E_i\).

Historical background: (\(\cal N2(\aleph_0)\) is \(\cal N2\).) If\(\aleph_{\alpha}\) is regular and \(\alpha>1\), then \(DC(\omega_1)\) (44) istrue in this model; and if \(\alpha>\beta\), then \(DC(\omega_{\beta})\)(87(\(\beta\))) is true. (Therefore, the Principle of Dependent Choices (DC,43) is true if \(\alpha>0\).) On the other hand, C(WO,2) (111) is false if\(\alpha > 0\) and \(C(WO,<\aleph_0)\) (122) is false for all \(\alpha\ge 0\),\(C(\infty,n)\) (45(\(n\)) and 88) is false for each \(n\ge 2\) and \(\alpha\ge0\), and \(KW(WO,<\aleph_0)\) (327) is false for all \(\alpha\ge 0\).\parSince BPI (14) implies 88, 14 is also false. Brunner [1982a] hasshown that for all \(\alpha>0\) there is a set that cannot be well orderedand does not have an infinite Dedekind finite subset (163 is false)because 44 implies 8 (\(C(\aleph_0,\infty)\)) and 8 plus 163 iff AC. Also,Form 44 implies forms 39 (\(C(\aleph_1,\infty)\)) and 41 (Every cardinal is comparable with \(\aleph_1\).). vanJech} (\cite{1973b}, p 132, prob 5) notes that if \(\alpha\ge 1\), sincethere is a non-trivial ultrafilter on \(\omega\) (70 is true) and everyinfinite set has a countably infinite subset (9 is true), the weakultrafilter principal (63) is true.Form 44 implies forms 39 (\(C(\aleph_1,\infty)\)) and 41 (Every cardinal is comparable with \(\aleph_1\).). van\ac{Douwen [1980] has shown that 39 + 41 implies 135 (If \(X\) is a\(T_2\) space with at least two points and \(X^Y\) is hereditarilymetacompact, then \(Y\) is countable.). Therefore,Form 135 is also true.\ac{Blass-1979">Douwen-1980">Jech} (\cite{1973b}, p 132, prob 5) notes that if \(\alpha\ge 1\), sincethere is a non-trivial ultrafilter on \(\omega\) (70 is true) and everyinfinite set has a countably infinite subset (9 is true), the weakultrafilter principal (63) is true.Form 44 implies forms 39 (\(C(\aleph_1,\infty)\)) and 41 (Every cardinal is comparable with \(\aleph_1\).). van\ac{Douwen [1980] has shown that 39 + 41 implies 135 (If \(X\) is a\(T_2\) space with at least two points and \(X^Y\) is hereditarilymetacompact, then \(Y\) is countable.). Therefore,Form 135 is also true.\ac{Blass [1979] has shown that 191 (There is a set \(X\) such thatfor each set \(a\) there is an ordinal \(\alpha\) and a function \(f\) mapping\(X\times\alpha\) onto \(a\).) is true in every permutation model with a setof atoms. He has also shown that 8 (\(C(\aleph_0,\infty)\)) + 191 + 193(Every Abelian group is a homomorphic image of a free projective Abeliangroup.) iff AC. Thus,Form 193 must be false if \(\aleph_{ \alpha}\) isregular and \(\alpha > 1\). (See Note 60 for definitions.) It follows fromNote 104 that the \(2m = m\) principle (3) is false in this model.

Back