Pith. sign in

REVIEW 4 major objections 4 minor 26 references

Hypermagmas and Colored Operads: Heads, Phases, and Theta Roles

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read The paper claims that the phase-based constraints of Minimalist syntax—EPP, ECP, and PIC—are not separate principles but fall out of the generator form of a single colored operad, whose single-vertex rules also make filtering simultaneous…

desk verdict The hypermagma and operad toolkit is a genuine extension of the authors' prior work, but the headline claim that EPP, ECP, and PIC are subsumed by the generator set does not hold: Proposition 6.2 under-restricts the movement generators and admits a direct PIC violation. read the letter →

arxiv 2507.06393 v1 pith:DNZ23BBQ submitted 2025-07-08 cs.CL math.QAmath.RA

classification cs.CLmath.QAmath.RA MSC 18M70
keywords coloredoperadsbudgeneratingsystemshypermagmasheadfunctionssyntacticphasesExtendedProjectionPrinciplethetarolesMinimalistMerge
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that the main structural filters of Minimalist syntax—the specifier-head-complement skeleton, the Extended Projection, and the phase-based constraints on movement—are not separate rules but the output of one algebraic device: a bud generating system for a colored operad, the same device previously used for theta roles. The authors first prove that syntactic heads force a shift from a magma to a hypermagma: merging two headed trees leaves the head of the result genuinely underdetermined, so the operation is set-valued, and this distinction is exactly what separates c-command (magmatic) from m-command (hypermagmatic). They then encode head-complement-specifier structure, modifiers, and phases as coloring rules on binary trees, and show that the Extended Projection Principle, the Empty Category Principle, and the Phase Impenetrability Condition all fall out of restrictions on the allowed generators rather than being added as extra principles. Finally, they combine the phase filter with the theta-role filter as a transduction of colored operads, showing that simultaneous filtering constrains movement analyses of passives, exceptional case marking, and double-object constructions. If correct, this gives a single formal language in which grammaticality of phase structure and theta assignment can be checked locally during Merge.

What carries the argument

The central object is the bud generating system of a colored operad: a finite set of colors (specifier s, complement z, head h, modifier m, plus functional-head and phase-head labels) and a finite set of generators that are single-vertex binary trees with one output color and two input colors. Composing generators by gluing matching output-to-input colors produces exactly the well-formed trees of the language; the single-vertex form is what makes step-by-step 'colored Merge' equivalent to after-the-fact filtering. A second central object is the hypermagma of headed syntactic objects, whose set-valued operation records the two legitimately possible head choices when two trees merge, and which carries the m-command relation.

What would settle it

A grammatical construction that demonstrably requires either movement from the interior of a completed phase, which would violate PIC, or a specifier position that is not at v*, INFL, or C, which would violate the EPP restriction, would yield a tree that the generator restrictions of Definition 4.10 and Proposition 6.2 cannot color, refuting the claim that these principles are subsumed by the generator form. Concretely, the paper's double-object analysis predicts both goal-before-theme and theme-before-goal hierarchies must be available before Externalization; a language where only one hierarchy exists and the other ordering is produced only by movement would falsify the transduction conclusion.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that heads and phases are a coloring phenomenon. A head function on a syntactic object is equivalent to a two-coloring of tree edges, and a complemented head with specifier, complement, and modifier positions is generated by single-vertex color rules of a bud generating system; the extended projection with C, INFL, and v* arises from a cartographic partial order on generator pairs. Because all generators are single-vertex, filtering by these rules can equivalently be done during structure formation as a colored Merge. The paper claims that restricting the generator set to allow specifier discharge only at v*, INFL, and C, and to allow Internal Merge only to edge-of-phase, Root, or INFL positions, makes the EPP, ECP, PIC, and the movement constraints of Remark 6.1 into properties of the operad rather than separate hypotheses. The same generator decomposition recovers the labeling algorithm, and combining phase and theta colors via a transduction selects among competing syntactic objects for passives, ECM, and double objects.

Load-bearing premise

The load-bearing premise is that the grammar's empirical content lives entirely in the chosen generator set and phase-head set, including the cartographic order C > INFL > v*, so that EPP, ECP, and PIC are assumptions built into the generators rather than conclusions derived from an independent theory.

Editorial extensions

If this is right

  • If correct, the EPP, ECP, and PIC require no dedicated principles: their content is exactly the admissible generator set.
  • Well-formedness can be checked incrementally during Merge, so structure building and filtering are computationally one process, not two.
  • Phase structure and theta assignment can be imposed simultaneously by a single transduction; the double-object analysis then predicts that both goal/theme hierarchies must exist prior to Externalization.
  • The labeling algorithm of Chomsky's projection problems is not an extra mechanism; it is read off the generator decomposition of any colored tree.
  • The magma/hypermagma split is linguistically visible: c-command lives in the magma, m-command in the hypermagma, explaining why command relations behave differently.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the generator-set viewpoint turns every grammatical constraint into a finite list of allowed color triples; a constraint that does not reduce to such a list (for instance one requiring a multi-vertex rule) would break the colored-Merge equivalence and would be a genuine counterexample to the program.
  • Editorial inference: the hypermagma formulation suggests a general template for ambiguity in grammar: wherever structure building genuinely underdetermines a choice, such as in exocentricity or head choice, the right algebraic object is set-valued, and well-formedness is a coshort section of a forgetful projection; this template could be applied to other interface filters such as focus or informat
  • Editorial inference: because the transduction colors are the product of phase and theta colors, the paper predicts that no movement can simultaneously satisfy both filters unless it lands in positions that are both s-/z-marked and theta-marked; this yields a concrete searchable prediction for raising-to-object and ECM constructions across languages.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper develops algebraic structures for Minimalist syntax. It defines head functions on syntactic objects, shows that the c-command relation is compatible with the magma structure while m-command requires a hypermagma, and then encodes specifier/head/complement structure, extended projections, modifiers, and phases using bud-generating systems for colored operads. The authors claim that the Extended Projection Principle, the Empty Category Principle, and the Phase Impenetrability Condition are subsumed into the form of the colored operad generators; that filtering after free structure formation is equivalent to filtering during structure formation via a colored Merge; and that compatibility with theta-role filtering is captured by transductions of colored operads. The framework is applied to passivization, exceptional case-marking, and double-object constructions.

Significance. If the technical problems are repaired, the paper would offer a substantial unifying algebraic treatment of several well-formedness constraints in Minimalist syntax, connecting head-complement structure, phases, and theta assignment through colored operads and bud-generating systems. The formal development in Sections 2, 3, and 5 is mostly clear, and results such as Lemmas 2.5, 3.4, 3.5, 5.1, and 5.7 are correct and well presented. The paper also makes a concrete empirical proposal in Section 8, namely that both theta hierarchies must be available before Externalization. However, the central 'subsumption' claims are currently not established: the EPP is inserted as a generator restriction, and the PIC claim is under-restricted in a way that admits direct counterexamples.

major comments (4)
  1. [§6, Proposition 6.2 / Remark 6.3] Proposition 6.2 under-restricts the Internal Merge generators, so Remark 6.3's claim that PIC follows is false as stated. In (6.2), the condition allows any pair of phase-heads ω, ω' with ω' > ω, and the generator (4.27) explicitly allows c ∈ {s, z}. Taking ω = v*, ω' = C, and c = z↓_{v*} gives a generator that moves the complement of the completed v* phase to Spec-C, which is exactly the movement from the interior of a completed phase to a higher phase that PIC forbids. The missing condition is that when ω is a phase-head, c must be the edge color s↓_ω, not the interior color z↓_ω. This condition must be added to Proposition 6.2 and Remark 6.3, or the PIC claim must be withdrawn.
  2. [§4.6, Definition 4.10 / Proposition 4.11] The EPP result is built into the generator restrictions by fiat. The three conditions of Definition 4.10 say, in effect, that functional heads occur above lexical heads and that s↓ can only be discharged at v*, INFL, or C; Proposition 4.11 then verifies that the restricted language satisfies EPP. This is a consistency check of the encoding, not a derivation of the EPP from independent structure. If 'subsumed' in the Abstract is meant in this encoding sense, the paper should say so explicitly and should not imply that the EPP has been derived from the operad formalism alone.
  3. [§5.2, Proposition 5.6 and §6] The claimed equivalence between post-hoc filtering and colored Merge is not fully proved for Internal Merge. The argument in Remark 5.4 relies on all generators being single-vertex trees, but the movement generators (4.27) involve the unit color (1, m), and the corresponding Merge step M_{T,1} falls outside the projection Π^(2) used in Proposition 5.7. Section 6 describes informally how to extend the colored Merge to this case, but no proof is given that the extended colored Merge generates exactly L(BΦ), including all Internal Merge derivations. This gap is relevant to the paper's central equivalence claim.
  4. [§8.4, Proposition 8.1] The conclusion that both theta hierarchies must be available prior to Externalization rests on an informal complexity criterion. Options that 'would require additional generators' are called 'disfavored', but no measure of generator-set complexity or notion of 'disfavored' is defined. The argument may be a reasonable proposal, but as it stands it is not a precise proof. Either provide a formal criterion for preferring one generator set over another, or present Proposition 8.1 as a supported conjecture rather than a definitive result.
minor comments (4)
  1. [§4.6, Remark 4.12] The claim that the restrictions of Definition 4.10 give the 'minimal restriction (largest subset of R_{b,ext})' that guarantees EPP is unsupported; either provide a proof or soften the wording to 'a restriction'.
  2. [§5.2 and §9] There are several typographical errors, including 'seperately' in §5.2 and 'obtaned' in §9, as well as a missing parenthesis in the sentence 'all of L(BΦ and all those can be obtained in this way'; a careful copyedit is needed.
  3. [§4.9 and §6.2] The phase-head set Ω_{h,ϕ} and the cartographic order C > INFL > v* are introduced as assumptions but are not flagged as such in the introduction; the Abstract's 'subsumed' language would be clearer if these choices were explicitly acknowledged as part of the model's empirical content.
  4. [§6.1, Proposition 6.4] The proof of Proposition 6.4 appears to assume that each T ∈ L(BΦ) has a unique decomposition into generators in RΦ, but uniqueness of the operadic decomposition is neither stated nor proved; please clarify whether uniqueness is needed and, if so, establish it.

Circularity Check

2 steps flagged · score 7.0 of 10

EPP and PIC are installed by generator restrictions and then 'proved' to follow; the operad subsumption is by construction.

  1. fitted input called prediction [§4.6, Definition 4.10 and Proposition 4.11]
    "Consider the subset Rb,ext,EPP ⊂ Rb,ext ... with the following restrictions on the generators: ... (3) ... the pair (ω,ω′) follows a partial order structure among the functional head that determines the order in which they occur in the extended projection (e.g., C > INFL > v∗). ... Proposition 4.11. ... The resulting colored operad OΩb,ext,Bb,ext,EPP and the associated language L(Bb,ext,EPP) satisfy the Extended Projection Principle."

    Definition 4.10 builds the EPP into the allowed generators: functional heads may only appear in the C > INFL > v∗ order, and the subject/specifier discharge is restricted to v∗, INFL, or C. Proposition 4.11 then 'shows' that the language satisfies the EPP by reading back exactly those restrictions. The EPP result is identical to the generator-level filter imposed as input; no independent derivation from the operad formalism is supplied.

  2. fitted input called prediction [§6, Proposition 6.2 and Remark 6.3]
    "In order to ensure that this type of movement satisfies the rules listed above, it is necessary to restrict the pairs (ω, ω′) that occur in (4.27) accordingly. ... where ω, ω′ ∈ Ω(ℓ) with ω′ > ω and either • both ω and ω′ are phase-heads • ω′ = INFL or ω′ = Root. ... In particular, the Phase Impenetrability Condition (PIC) follows from the set of generators of L(BΦ) if the generators (4.27) are restricted to satisfy the conditions of Proposition 6.2."

    Remark 6.1 lists the movement restrictions to be enforced, and item (3) is essentially PIC: 'IM does not move from the interior of lower phases once the phases are completed.' Proposition 6.2 then restricts the allowed (ω,ω′) pairs in the movement generator (4.27) to exactly those conforming to those rules. Remark 6.3 announces that PIC follows from that restriction. The conclusion is the input: PIC is not derived from the operad structure but installed by choosing which phase-head pairs may appear in the generator. Moreover, since (4.27) allows c ∈ {s,z}, the restriction still permits a z-interior move out of a completed phase, so the subsumption claim is also incomplete.

full rationale

The paper contains independent mathematical content that is not circular: the hypermagma formulation of head functions, the c-command/m-command compatibility statements, and the equivalence between filtering-after-formation and colored Merge are genuine constructions. The circularity is concentrated in the advertised subsumption results of §4.6 and §6. Definition 4.10 restricts the generator set to encode the EPP, and Proposition 4.11 reads that restriction back as a theorem. Similarly, Remark 6.1 states the PIC-like movement rule as a desideratum, Proposition 6.2 writes it into the allowed (ω,ω′) pairs of the movement generator (4.27), and Remark 6.3 announces that PIC follows. The empirical content thus resides in the freely chosen generator restrictions rather than in a derivation from the operad, so the central 'subsumed into the form of the colored operad generators' claim reduces by construction. No load-bearing uniqueness theorem or unverified self-citation chain was found: the citations to the authors' prior work [19] and [20] supply definitions and analogies, but the reduction above is internal to the paper's own equations. The PIC step is also under-restricted: because (4.27) lets the moved color c be either s or z, Proposition 6.2 does not exclude moving an interior z-down color from a completed lower phase to a higher phase, so a direct PIC violation is still generated. That is a correctness gap in addition to the constructional circularity. Overall score 7: partial circularity through fitted generator choices, with independent content elsewhere.

Assumptions & free parameters 4 free parameters · 6 assumptions · 2 invented entities

The model's explanatory power is supplied largely through free choices: generator sets, phase-head membership, and the cartographic partial order. The formal background (magmas, hypermagmas, colored operads) is standard mathematics; the paper's own additions are the specific encodings.

free parameters (4)
  • Generator set RΦ (phases) = Specific list in Def 4.10, 4.14 and Prop 6.2
    Chosen by hand to make EPP, PIC, ECP and IM movement come out as claimed; a different set would give different 'derived' constraints.
  • Phase-head set Ωh,ϕ = {C, v*} ∪ {INFLφ} (INFL and v excluded)
    The assignment of which functional heads are phase heads is stipulated in Definition 4.16, not derived; it is a free choice of the model.
  • Cartographic partial order on Ωh = C > INFL > v* (and variants)
    Assumed in §4.5 and §4.6 to encode the Extended Projection; cartographic models K allow arbitrary partial orders, so the specific order is a free parameter.
  • EPP specifier restrictions = s↓ only discharged at v*, INFL, C heads; s↓ω' output requires determiner/noun head; the pair (ω,ω') must satisfy ω' ≠…
    These are literally the EPP content inserted into Definition 4.10; they are chosen, not predicted.
assumptions (6)
  • domain assumption Syntactic objects are elements of the free non-associative commutative magma over lexical items and features.
    This is the SMT-Merge picture from [19]; the entire paper operates inside it.
  • domain assumption Head function as in Def 2.4 characterizes syntactic headedness; there may be exocentric objects not in Dom(h).
    Adopted from [19] and Chomsky; the hypermagma construction relies on it.
  • domain assumption Extended Projection structure: lexical heads are extended by functional heads C, INFL, v*, v, D, with specified hierarchy.
    The colored operad generators encode this hierarchy; if it is wrong, the generators are wrong.
  • ad hoc to paper At least one of the two theta hierarchies (θGO ≺ θTH or θGO ≻ θTH) is fixed before Externalization, and both are possible.
    The paper's conclusion in §8.4 depends on this being an empirical matter.
  • ad hoc to paper All relevant generators are single-vertex trees.
    The paper notes this is a peculiarity (§5.2) and it is essential for the equivalence between filtering and colored Merge.
  • ad hoc to paper The formal empty tree 1 acts as unit of the magma and can carry only color (1,m).
    Used to model IM movement; comes from [19], [20].
invented entities (2)
  • INFLφ (phase-head variant of INFL)
    purpose: Accounts for cases where INFL inherits phase-head features from C, enabling movement from Spec-INFL in the that-trace configuration.
    Introduced in §4.9 to distinguish the two behaviors of INFL; no independent evidence is offered within the paper.
  • The unit element 1 with color (1,m)
    purpose: Models Internal Merge as merging with an empty tree; used to handle movement traces.
    Inherited from [19]; a formal device with no independent empirical backing in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hypermagmas and Colored Operads: Heads, Phases, and Theta Roles." pith.science (2026). https://pith.science/paper/DNZ23BBQ

@misc{pith2026250706393,
  author       = {Pith},
  title        = {Pith review of: Hypermagmas and Colored Operads: Heads, Phases, and Theta Roles},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DNZ23BBQ}},
  note         = {Machine review of arXiv:2507.06393}
}
read the original abstract

We show that head functions on syntactic objects extend the magma structure to a hypermagma, with the c-command relation compatible with the magma operation and the m-command relation with the hypermagma. We then show that the structure of head and complement and specifier, additional modifier positions, and the structure of phases in the Extended Projection can be formulated as a bud generating system of a colored operad, in a form similar to the structure of theta roles. We also show that, due to the special form of the colored operad generators, the filtering of freely generated syntactic objects by these coloring rules can be equivalently formulated as a filtering in the course of structure formation via a colored Merge, which can in turn be related to the hypermagma structure. The rules on movement by Internal Merge with respect to phases, the Extended Projection Principle, Empty Category Principle, and Phase Impenetrability Condition are all subsumed into the form of the colored operad generators. Movement compatibilities between the phase structure and the theta roles assignments can then be formulated in terms of the respective colored operads and a transduction of colored operads.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

  1. [1]

    Baker, K

    M. Baker, K. Johnson, I. Roberts, Passive Arguments Raised, Linguistic. Inquiry 20 (1989), N.2, 219–251

  2. [2]

    Chomsky, Some concepts and consequences of the theory of government and binding , MIT Press, 1982

    N. Chomsky, Some concepts and consequences of the theory of government and binding , MIT Press, 1982

  3. [3]

    Evolution and Revolution in Linguistic Theory

    N. Chomsky. Bare Phrase Structure. In “Evolution and Revolution in Linguistic Theory”, edited by H. Campos and P. Kempchinsky, 51–109. Georgetown University Press, 1995

  4. [4]

    Chomsky, Problems of projection, Lingua, Vol.130 (2013) 33-49

    N. Chomsky, Problems of projection, Lingua, Vol.130 (2013) 33-49

  5. [5]

    Structures, Strategies and Beyond: Studies in honour of Adriana Belletti

    N. Chomsky, Problems of projection: Extension , in “Structures, Strategies and Beyond: Studies in honour of Adriana Belletti” (E. Di Domenico, C. Hamann, S. Matteini, Eds.) John Benjamins Publishing, 2015, pp. 1–16

  6. [6]

    Chomsky, Minimalism: Where Are We Now, and Where Can We Hope to Go , Gengo Kenkyu, 160 (2021), 1–41

    N. Chomsky, Minimalism: Where Are We Now, and Where Can We Hope to Go , Gengo Kenkyu, 160 (2021), 1–41

  7. [7]

    Chomsky, Genuine Explanation and the Strong Minimalist Thesis , Cognitive Semantics, 8 (2022) 347–365

    N. Chomsky, Genuine Explanation and the Strong Minimalist Thesis , Cognitive Semantics, 8 (2022) 347–365

  8. [8]

    Working toward the Strong Interpretation of SMT

    Noam Chomsky. Working toward the Strong Interpretation of SMT. Lecture Series Theoretical Linguistics at Keio-EMU, 2023

Show all 26 references
  1. [9]

    In A Cartesian Dream: A Geometrical Account of Syntax

    Noam Chomsky, The Miracle Creed and the Strong Minimalist Thesis. In A Cartesian Dream: A Geometrical Account of Syntax. In Honor of Andrea Moro, (M.Greco and D.Mocci, Eds.) pp. 17–40. Research in Generative Grammar Monographs. Lingbuzz Press, 2023

  2. [10]

    Rich Descriptions and Simple Explanations in Morphosyntax and Language Acquisition

    N. Chomsky, Genuine Explanation , in: (G.Bocci, D.Botteri, C.Manetti, V.Moscati, Eds.), “Rich Descriptions and Simple Explanations in Morphosyntax and Language Acquisition”, Oxford University Press, 2024, pp. 15– 44. 48 M.MARCOLLI, M.A.C.HUIJBREGTS, R.K.LARSON

  3. [11]

    Chomsky, T.D

    N. Chomsky, T.D. Seely, R.C. Berwick, S. Fong, M.A.C. Huybregts, H. Kitahara, A. McInnerney, Y. Sugimoto, Merge and the Strong Minimalist Thesis , Cambridge Elements, Cambridge University Press, 2023

  4. [12]

    Cinque, L

    G. Cinque, L. Rizzi, The Cartography of Syntactic Structures , STiL – Studies in Linguistics, 2 (2008) 43–59

  5. [13]

    Citko, Phase Theory, Cambridge University Press, 2014

    B. Citko, Phase Theory, Cambridge University Press, 2014

  6. [14]

    Diaconis, C.Y.A

    P. Diaconis, C.Y.A. Pang, A. Ram Hopf algebras and Markov chains: Two examples and a theory , J. Algebraic Combin. 39 (2014), no. 3, 527–585

  7. [15]

    Giraudo, Colored operads, series on colored operads, and combinatorial generating systems , Discrete Math

    S. Giraudo, Colored operads, series on colored operads, and combinatorial generating systems , Discrete Math. 342 (2019), no. 6, 1624–1657

  8. [16]

    Golmohamadian, M.M

    M. Golmohamadian, M.M. Zahedi, Color hypergroup and join space obtained by the vertex coloring of a graph , Filomat 31 (2017), no. 20, 6501–6513

  9. [17]

    Larson, The double object construction , LI 19, N

    R.K. Larson, The double object construction , LI 19, N. 3 (1988) 335–391

  10. [18]

    New Explorations in Chinese Theoretical Syntax: Studies in honor of Yen-Hui Audrey Li

    R.K. Larson & C. Zhang, Applied objects in Mandarin and the nature of selection , in “New Explorations in Chinese Theoretical Syntax: Studies in honor of Yen-Hui Audrey Li”, Linguistik Aktuell/Linguistics Today, 272 (2022) 357–394

  11. [19]

    Marcolli, N

    M. Marcolli, N. Chomsky, R.C. Berwick, Mathematical Structure of Syntactic Merge , MIT Press, 2025

  12. [20]

    Marcolli, R.K

    M. Marcolli, R.K. Larson, Theta Theory: Operads and Coloring , preprint, lingbuzz/008857, arXiv:2503.06091

  13. [21]

    Marty, Rˆ ole de la notion de hypergroupe dans l’´ etude de groupes non abeliens, Comptes Rendus Acad

    F. Marty, Rˆ ole de la notion de hypergroupe dans l’´ etude de groupes non abeliens, Comptes Rendus Acad. Sci. Paris 201, (1935), 636–638

  14. [22]

    May, The geometry of iterated loop spaces , Lecture Notes in Mathematics, Vol

    J.P. May, The geometry of iterated loop spaces , Lecture Notes in Mathematics, Vol. 271, Springer, 1972

  15. [23]

    Mittas, Hypergroupes canoniques, Math

    J. Mittas, Hypergroupes canoniques, Math. Balkanica 2 (1972), 165–179

  16. [24]

    Nakamura, M.L

    S. Nakamura, M.L. Reyes, Categories of hypermagmas, hypergroups, and related structures , arXiv:2304.09273

  17. [25]

    Smith, Augmented quasigroups and character algebras , Adv

    J.D.H. Smith, Augmented quasigroups and character algebras , Adv. Math. 363 (2020) 106983 [42 pages]

  18. [26]

    Yau, Colored Operads, Graduate Studies in Mathematics, 170

    D. Yau, Colored Operads, Graduate Studies in Mathematics, 170. American Mathematical Society, 2016. Department of Mathematics and Department of Computing and Mathematical Sciences, Cali- fornia Institute of Technology, CA 91125, USA Email address : matilde@caltech.edu Departme...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.