Polynomial functors and opetopes

We give an elementary and direct combinatorial definition of opetopes in terms of trees, well-suited for graphical manipulation and explicit computation. To relate our definition to the classical definition, we recast the Baez-Dolan slice construction for operads in terms of polynomial monads: our opetopes appear naturally as types for polynomial monads obtained by iterating the Baez-Dolan construction, starting with the trivial monad. We show that our notion of opetope agrees with Leinster’s. Next we observe a suspension operation for opetopes, and define a notion of stable opetopes. Stable opetopes form a least fixpoint for the Baez-Dolan construction. A final section is devoted to example computations, and indicates also how the calculus of opetopes is well-suited for machine implementation.

💡 Research Summary

This paper presents a completely elementary and combinatorial definition of opetopes using rooted trees, a formulation that is both visually intuitive and amenable to explicit computation. The authors begin by recalling the notion of polynomial functors—set‑valued functors of the form (\sum_{i\in I} X^{E_i})—and the associated polynomial monads, which provide a clean algebraic framework for describing operations with multiple inputs. Within this setting, the “type” of a polynomial monad is identified with a tree whose nodes correspond to the summands (I) and whose incident edges encode the exponent sets (E_i).

The central technical contribution is a reformulation of the Baez‑Dolan slice construction for operads in terms of polynomial monads. Starting from the trivial monad (\mathbf{1}) (the monad with no operations), a single slice produces a new polynomial monad (P_1); the types of (P_1) are precisely the 1‑dimensional opetopes. By iterating the slice construction (n) times one obtains a sequence of monads (P_n), each of which carries a family of types that the authors prove are in bijection with the classical (n)-dimensional opetopes. This iterative viewpoint replaces the historically opaque “operad‑of‑operads” recursion with a transparent algebraic process: each iteration adds a new layer of tree branching, and the resulting trees encode the higher‑dimensional shapes.

A substantial part of the paper is devoted to showing that this tree‑based definition coincides with the definition given by Leinster. The authors construct explicit translations between Leinster’s globular‑set based opetopes and their own tree‑based types, demonstrating that the two constructions produce isomorphic cell structures in every dimension. The proof hinges on the observation that the polynomial monad’s arity data exactly reproduces the source‑target matching conditions of Leinster’s globular cells.

Having established equivalence, the authors introduce a suspension operation on opetopes. Given an (n)-opetope (O), its suspension (\Sigma O) is defined by adding a new root node whose children are the leaves of the tree representing (O). This operation raises the dimension by one and respects the Baez‑Dolan construction. By repeatedly applying suspension, they define stable opetopes as the least fixed point of the slice construction: the unique collection of shapes that is invariant under further slicing. They prove existence and uniqueness of this fixed point and show that it captures precisely the “opetopic sets” that appear in higher‑category theory as models of weak (\omega)-categories.

The final section showcases concrete calculations. The authors work out the trees corresponding to the 2‑dimensional and 3‑dimensional opetopes, exhibit the associated polynomial monad operations, and sketch how these structures can be encoded in functional programming languages such as Haskell or Python. They argue that the tree representation makes it straightforward to implement the “calculus of opetopes” on a computer: composition, substitution, and dimension‑raising become simple tree manipulations. Moreover, graphical visualisation of opetopes becomes a matter of drawing rooted trees with labeled edges, opening the door to interactive software for higher‑dimensional category theory.

In summary, the paper delivers four major advances: (1) a direct, tree‑based combinatorial definition of opetopes; (2) a clear algebraic reinterpretation of the Baez‑Dolan slice construction via polynomial monads; (3) a rigorous proof of equivalence with Leinster’s definition and the introduction of suspension and stable opetopes; and (4) concrete examples and a roadmap for computer implementation. By bridging abstract higher‑categorical concepts with concrete combinatorial objects, the work not only clarifies the foundations of opetopic theory but also paves the way for practical computational tools in the study of weak (\omega)-categories.