Searcharxiv⌕ Search

arXiv subjects

Saharon Shelah

Publications and source records attributed to Saharon Shelah.

At least 577 records · Page 32Linked to original sources
arXiv2000

Reflexive subgroups of the Baer-Specker group and Martin's axiom

In two recent papers (math.LO/0003164 and math.LO/0003165) we answered a question raised in the book by Eklof and Mekler (p. 455, Problem 12) under the set theoretical hypothesis of diamondsuit_{aleph_1} which holds in many models of set theory, respectively of the special continuum hypothesis (CH). The objects are reflexive modules over countable principal ideal domains R, which are not fields. Following H.Bass, an R-module G is reflexive if the evaluation map s:G ---> G^{**} is an isomorphism. Here G^*=Hom(G,R) denotes the dual module of G. We proved the existence of reflexive R-modules G of infinite rank with G not cong G+R, which provide (even essentially indecomposable) counter examples to the question mentioned above. Is CH a necessary condition to find `nasty' reflexive modules? In the last part of this paper we will show (assuming the existence of supercompact cardinals) that large reflexive modules always have large summands. So at least being essentially indecomposable needs an additional set theoretic assumption. However the assumption need not be CH as shown in the first part of this paper. We will use Martin's axiom to find reflexive modules with the above decomposition which are submodules of the Baer-Specker module R^omega.

math.LO↗
arXiv2000

On our paper `Almost Free Splitter', a correction

Let R be a subring of Q and recall from math.LO/9910161 that an R-module G is a splitter if Ext_R(G,G)=0. We correct the statement of Main Theorem 1.5 in math.LO/9910161. Assuming CH any aleph_1$-free splitter of cardinality aleph_1 is free over its nucleus as shown in math.LO/9910161. Generally these modules are very close to being free as explained below. This change follows from math.LO/9910161 and is due to an incomplete proof (noticed thanks to Paul Eklof) in the first section of math.LO/9910161. Assuming the negation of CH, in Shelah [Sh:F417] (work in progress) it will be shown that under Martin's axiom these splitters are free indeed. However there are models of set theory having non-free aleph_1-free splitter of cardinality aleph_1.

math.LO↗
arXiv2000

On nice equivalence relations on 2^λ

The main question here is the possible generalization of the following theorem on ``simple'' equivalence relation on 2^omega to higher cardinals. Theorem: (1) Assume that: (a) E is a Borel 2-place relation on 2^omega, (b) E is an equivalence relation, (c) if eta, nu in 2^omega and (exists ! n)(eta(n) not= nu(n)), then eta, nu are not E --equivalent. Then there is a perfect subset of 2^omega of pairwise non E-equivalent members. (2) Instead of ``E is Borel'', ``E is analytic (or even a Borel combination of analytic relations)'' is enough. (3) If E is a Pi^1_2 relation which is an equivalence relation satisfying clauses (b)+(c) in V^Cohen, then the conclusion of (1) holds.

math.LO↗
arXiv2000

Clones on regular cardinals

We investigate the structure of the lattice of clones on an infinite set X. We first observe that ultrafilters naturally induce clones; this yields a simple proof of Rosenberg's theorem: "there are 2^2^kappa many maximal (=precomplete) clones on a set of size kappa." The clones we construct here do not contain all unary functions. We then investigate clones that do contain all unary functions. Using a strong negative partition theorem we show that for many cardinals kappa there are 2^2^kappa many such clones on a set of size kappa. Finally, we show that on a weakly compact cardinal there are exactly 2 maximal clones which contain all unary functions.

math.RA↗
arXiv2000

The failure of the uncountable non-commutative Specker Phenomenon

Higman proved in 1952 that every free group is non-commutatively slender, this is to say that if G is a free group and h is a homomorphism from the countable complete free product (X_omega Z) to G, then there exists a finite subset F of omega and a homomorphism h:*_{i in F} Z --> G such that h=h rho_F, where rho_F is the natural map from (X_{i in omega})Z to *_{i in F}Z . Corresponding to the abelian case this phenomenon was called the non-commutative Specker Phenomenon. In this paper we show that Higman's result fails if one passes from countable to uncountable. In particular, we show that for non-trivial groups G_alpha (alpha in lambda) and uncountable cardinal lambda there are 2^{2^lambda} homomorphisms from the complete free product of the G_alpha 's to the ring of integers.

math.LO↗
arXiv2000

A space with only Borel subsets

Miklos Laczkovich asked if there exists a Haussdorff (or even normal) space in which every subset is Borel yet it is not meager. The motivation of the last condition is that under MA_kappa every subspace of the reals of cardinality kappa has the property that all subsets are F_sigma, however Martin's axiom also implies that these subsets are meager. Here we answer Laczkovich' question.

math.LO↗
arXiv2000

Constructing Boolean algebras for cardinal invariants

We construct Boolean Algebras answering questions of Monk on cardinal invariants. The results are proved in ZFC (rather than giving consistency results). We deal with the existence of superatomic Boolean Algebras with ``few automorphisms'', with entangled sequences of linear orders, and with semi-ZFC examples of the non-attainment of the spread (and hL, hd).

math.LO↗
arXiv2000

More Jonsson algebras

We prove that on many inaccessible there is a Jonsson algebra, so e.g. the first regular Jonsson cardinal lambda which is lambda x omega-Mahlo. We give further restrictions on successor of singulars which are Jonsson cardinals. E.g. there is a Jonsson algebra on the successor of beth_omega. Lastly, we give further information on guessing of clubs.

math.LO↗
arXiv2000

Historic forcing for Depth

We show that, consistently, for some regular cardinals theta<lambda, there exists a Boolean algebra B such that B=lambda^+ and for every subalgebra B' of B of size lambda^+ we have Depth(B')=theta.

math.LO↗
arXiv2000

Forcing for hL and hd

The present paper addresses the problem of attainment of the supremums in various equivalent definitions of hereditary density hd and hereditary Lindelof degree hL of Boolean algebras. We partially answer two problems of J. Donald Monk (Problems 50 and 54 in his book "Cardinal Invariants on Boolean Algebras), showing consistency of different attainment behaviour and proving that (for the considered variants) this is the best result we can expect.

math.LO↗
arXiv2000

Continuous images of sets of reals

We will show that, consistently, every uncountable set can be continuously mapped onto a non measure zero set, while there exists an uncountable set whose all continuous images into a Polish space are meager.

math.LO↗
arXiv2000

A Partition Theorem

We prove the following: there is a primitive recursive function f_-^*(-,-), in the three variables, such that: for every natural numbers t,n>0, and c, for any natural number k>=f^*_t(n,c) the following holds. Assume L is an alphabet with n>0 letters, M is the family of non empty subsets of {1,...,k} with =<t members and V is the set of functions from M to L, and lastly d is a c-colouring of V (i.e. a function with domain V and range with at most c members). Then there is a d-monochromatic V-line, which means that there are w included in {1,...,k}, with at least t members and a function r from {u in M: u not a subset of w} to L such that letting Y={eta in V: eta extends r and for each s=1,...,t it is constant on {u in M: u is an s-element subset of w}}, we have: the restriction of d to Y is constant (for t=1 those are the Hales Jewett numbers).

math.CO↗
arXiv2000

Some Nasty Reflexive Groups

In "Almost Free Modules, Set-theoretic Methods", Eklof and Mekler raised the question about the existence of dual abelian groups G which are not isomorphic to Z+G. Recall that G is a dual group if G ~ D^* for some group D with D^*=Hom(D,Z). The existence of such groups is not obvious because dual groups are subgroups of cartesian products Z^D and therefore have very many homomorphisms into Z. If p is such a homomorphism arising from a projection of the cartesian product, then D^* ~ ker(p)+Z. In all `classical cases' of groups D of infinite rank it turns out that D^* ~ ker(p). Is this always the case? Also note that reflexive groups G in the sense of H.Bass are dual groups because by definition the evaluation map s:G-->G^{**} is an isomorphism, hence G is the dual of G^*. Assuming the diamond axiom for aleph_1 we construct a reflexive torsion-free abelian group of cardinality aleph_1 which is not isomorphic to Z+G. The result is formulated for modules over countable principal ideal domains which are not field.

math.LO↗
arXiv2000

Decompositions of Reflexive Modules

We continue [GbSh:568] (math.LO/0003164), proving a stronger result under the special continuum hypothesis (CH). The original question of Eklof and Mekler related to dual abelian groups. We want to find a particular example of a dual group, which will provide a negative answer to the question. In order to derive a stronger and also more general result we will concentrate on reflexive modules over countable principal ideal domains R. Following H.Bass, an R-module G is reflexive if the evaluation map s:G-->G^{**} is an isomorphism. Here G^*=Hom(G,R) denotes the dual group of G. Guided by classical results the question about the existence of a reflexive R-module G of infinite rank with G not cong G+R is natural. We will use a theory of bilinear forms on free R-modules which strengthens our algebraic results in [GbSh:568] (math.LO/0003164). Moreover we want to apply a model theoretic combinatorial theorem from [Sh:e] which allows us to avoid the weak diamond principle. This has the great advantage that the used prediction principle is still similar to the diamond, but holds under CH.

math.LO↗
arXiv2000

Consistently there is no non trivial ccc forcing notion with the Sacks or Laver property

Boban Velickovic asked the following question: Is there a nontrivial forcing notion with the Sacks property which is also ccc? A ``definable'' variant of this question has been answered in [Sh:480] (math.LO/9303208): Every nontrivial Souslin forcing notion which has the Sacks property has an uncountable antichain. Here we show that it is consistent that every nontrivial forcing notion which has the Sacks property has an uncountable antichain. Independently, Velickovic has also proved the consistency of this statement.

math.LO↗
arXiv2000

Weak reflection at the successor of singular

The notion of stationary reflection is one of the most important notions of combinatorial set theory. We investigate weak reflection, which is, as the name suggests, a weak version of stationary reflection. This sort of reflection was introduced in [DjSh:545] (math.LO/9601219), where it was shown that weak reflection has applications to various guessing principles, in the sense that if there is no weak reflection, than a guessing principle holds, and an application dealing with the saturation of normal filters. Further investigations of weak reflection were carried in [CDSh:571] (math.LO/9504221) and [CuSh:596] (math.LO/9703219). While various ZFC restrictions on the one hand, and independence results on the other, were discovered about the weak reflection, the question of the relative consistency of the existence of a regular cardinal kappa such that the first cardinal weakly reflecting at kappa is a successor of singular, remained open. This paper answers that question by proving that (modulo large cardinal assumptions close to 2-hugeness) that there indeed can be such a cardinal kappa.

math.LO↗
arXiv2000

The automorphism tower problem revisited

It is well-known that the automorphism towers of infinite centreless groups of cardinality kappa terminate in less than (2^{kappa})^+ steps. But an easy counting argument shows that (2^{kappa})^+ is not the best possible bound. However, in this paper, we will show that it is impossible to find an explicit better bound using ZFC.

math.LO↗