The Berkson effect shows that two independent diseases, A and B, become negatively correlated if they are confined within the walls of a hospital. We explain that, simply by adding a third disease, C, the negative correlation may flip into a positive one, and we identify the point where this happens. That leads to a necessary and sufficient condition for a positive as well as a negative correlation between A and B. We further explain that a flip from negative to (...) positive is impossible if C is independent of A, of B, and of the disjunction of A and B: with these three independences in place, the Berkson effect remains in force. However, if only two of the three independences hold, the effect is not guaranteed. (shrink)

Given a regular cardinal κ such that κ<κ=κ (e.g., if the generalized continuum hypothesis holds), we develop a proof system for classical infinitary logic that includes heterogeneous quantification (i.e., infinite alternating sequences of quantifiers) within the language Lκ+,κ, where there are conjunctions and disjunctions of at most κ many formulas and quantification (including the heterogeneous one) is applied to less than κ many variables. This type of quantification is interpreted in Set using the usual second-order formulation in terms of strategies (...) for games, and the axioms are based on a stronger variant of the axiom of determinacy for game semantics. With respect to this axiom system we prove the soundness and completeness theorem with respect to a class of set-valued structures that we call well determined. We also investigate intuitionistic systems with heterogeneous quantifiers for Lκ+,κ,κ (when only conjunctions of less than κ many formulas are allowed), and prove analogously a completeness theorem with respect to well-determined structures in categories in general, in κ-Grothendieck topoi in particular, and, when κ<κ=κ, also in Kripke models. Finally, we consider an extension of our system in which heterogeneous quantification with bounded quantifiers is expressible, and extend our completeness results to that case. (shrink)

We prove that, given a planar embedding of a graph in the sphere, the expansion of the graph structure by predicates encoding vertex separation by simple graph cycles is dp-minimal. This provides a rich natural class of examples of unstable, dp-minimal, and also monadically NIP theories. We also show how to infer the existence of a distal expansion of the theory of the Farey graph.

We give a proof of the Marker–Steinhorn theorem which fills a gap in previous proofs of the result.

We introduce Broad Infinity, a new set-theoretic axiom scheme based on the slogan “Every time we construct a new element, we gain a new arity.” It says that three-dimensional trees whose growth is controlled by a specified class function form a set. Such trees are called broad numbers. Assuming AC (the axiom of choice) or at least the weak version known as WISC (weakly initial set of covers), we show that Broad Infinity is equivalent to Mahlo’s principle, which says that (...) the class of all regular limit ordinals is stationary. Assuming AC or WISC, Broad Infinity also yields a convenient principle for generating a subset of a class using a “rubric” (family of rules); this directly gives the existence of Grothendieck universes, without requiring a detour via ordinals. In the absence of choice, Broad Infinity implies that the derivations of elements from a rubric form a set; this yields the existence of Tarski-style universes. Additionally, we reveal a pattern of resemblance between “wide” principles, that are provable in ZFC, and “broad” principles, that go beyond ZFC. Note: this paper uses a base theory that is weaker than ZF but includes classical first-order logic and Replacement. (shrink)

We furnish a core-logical development of the Gödel numbering framework that allows metamathematicians to attain limitative results about arithmetical truth without incorporating a genuine truth predicate into the language in a way that would lead to semantic closure. We show how Tarski’s celebrated theorem on the arithmetical undefinability of arithmetical truth can be established using only core logic in both the object language and the metalanguage. We do so at a high level of abstraction, by augmenting the usual first-order language (...) of arithmetic with a primitive predicate Tr and then showing how it cannot be a truth predicate for the augmented language. McGee established an important result about consistent theories that are in the language of arithmetic augmented by such a “truth predicate” Tr and that use Gödel numbering to refer to expressions of the augmented language. Given the nature of his sought result, he was forced to use classical reasoning at the meta level. He did so, however, on the additional and tacit presupposition that the arithmetical theories in question (in the object language) would be closed under classical logic. That left open the dialectical possibility that a constructivist (or intuitionist) could claim not to be discomfited by the results, even if they were to “give a pass” on the unavoidably classical reasoning at the meta level. In this study we “constructivize” McGee’s result, by presuming only core logic for the object language. This shows that the perplexity induced by McGee’s result will confront the constructivist (or intuitionist) as well. (shrink)

It has recently been shown that fairly strong axiom systems such as ACA0 cannot prove that the antichain with three elements is a better-quasi-order (bqo). In the present paper, we give a complete characterization of the finite partial orders that are provably bqo in such axiom systems. The result will also be extended to infinite orders. As an application, we derive that a version of the minimal bad array lemma is weak over ACA0. In sharp contrast, a recent result shows (...) that the same version is equivalent to Π21-comprehension over the stronger base theory ATR0. (shrink)

A common response to the paradoxes of vagueness and truth is to introduce the truth-values “neither true nor false” or “both true and false” (or both). However, this infamously runs into trouble with higher-order vagueness or the revenge paradox. This, and other considerations, suggest iterating “both” and “neither”: as in “neither true nor neither true nor false.” We present a novel explication of iterating “both” and “neither.” Unlike previous approaches, each iteration will change the logic, and the logic in the (...) limit of iteration is an extension of paraconsistent quantum logic. Surprisingly, we obtain the same limit logic if we use (a) both and neither, (b) only neither, or (c) only neither applied to comparable truth-values. These results promise new and fruitful replies to the paradoxes of vagueness and truth. (The paper allows for modular reading: for example, half of it is an appendix studying involutive lattices to prove the results.). (shrink)