The Standard Aspect of Dialectical Logic

Dialectical logic is the logic of dialectical processes. The goal of dialectical logic is to introduce dynamic notions into logical computational systems. The fundamental notions of proposition and truth-value in standard logic are subsumed by the notions of process and flow in dialectical logic. Dialectical logic has a standard aspect, which can be defined in terms of the “local cartesian closure” of subtypes. The standard aspect of dialectical logic provides a natural program semantics which incorporates Hoare’s precondition/postcondition semantics and extends the standard Kripke semantics of dynamic logic. The goal of the standard aspect of dialectical logic is to unify the logic of small-scale and large-scale programming.

💡 Research Summary

The paper “The Standard Aspect of Dialectical Logic” proposes a unified logical framework that subsumes both dynamic (modal) logic and Hoare‑style pre‑/post‑condition reasoning under a single categorical structure called the “standard aspect” of dialectical logic. The authors begin by re‑interpreting the basic notions of proposition and truth‑value as “process” and “flow”. A dynamic system is modeled as a collection of arrows (terms) each equipped with a source and a target type, representing state transitions. Nondeterminism among terms is captured by a preorder (⊑) on terms, and sequential composition of processes is expressed by a tensor product (⊗) that is associative and monotone on both sides.

The central algebraic structure is a biposet: a category whose hom‑sets are partially ordered sets and whose composition respects the order. Within a biposet, an adjoint pair (y r ⊣ s ⇁ x) is defined by a unit inequality (y ⊑ r⊗s) and a counit inequality (s⊗r ⊑ x). Such pairs encode functional terms; the right (or left) adjoint, when it exists, is unique, giving a precise order‑theoretic notion of a function. Functional terms thus become the building blocks for program specifications.

Two complementary internal structures are introduced:

• Comonoids (subtypes) for each type x are endomorphisms u : x → x satisfying a part‑of‑x axiom (u ⊑ x) and idempotence (u⊗u = u). The set Ω(x) of comonoids forms a poset that can be interpreted as a state space indexed by x. The interior operation (u⊗v)◦ yields the meet (u∧v) in Ω(x), mirroring the affirmation modality of linear logic.

• Monoids (dual to comonoids) for each type x are endomorphisms m : x → x satisfying reflexivity (x ⊑ m) and idempotence (m⊗m = m). The set ✶(x) of monoids carries a join operation; the closure (p •) gives the least monoid containing p, analogous to the consideration modality of linear logic.

When the inclusion Ω(x) ↪ P