Merges of Smooth Classes and Their Properties

Morgan Bryant

REVIEW 4 major objections 5 minor 8 references

Merges of Smooth Classes and Their Properties

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The generic of a merge of smooth classes contains, on every infinite set definable by an existential formula in the second class's language, a reduct isomorphic to the first class's generic.

desk verdict Main theorem survives the stress test; the real problems are a missing assumption in Theorem 2.26 and an unproved Ramsey transfer in Corollary 3.22. read the letter →

arxiv 2411.10689 v1 pith:56SNECTR submitted 2024-11-16 math.LO

Morgan Bryant This is my paper · ORCID

classification math.LO MSC 03C5003C4505D10

keywords smoothclassesgenericlimitsamalgamationpropertyparallelstrongnessstructuralRamseytheoryHrushovski(EPPA)Shelah-Spencergraphs1-local

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

Two smooth classes are families of finite structures with a relation saying when one structure is a strong substructure of another, and their generics are the countable limits built by amalgamation. The paper asks what survives when two such families are merged into one class of structures carrying both languages. It proves that if the first class is closed under substructure, has disjoint amalgamation, smooth intersections, and disjoint parallel strongness, then inside the merged generic every infinite set definable by an existential formula in the second class's language is an isomorphic copy of the first class's generic. This yields self-similarity corollaries: deleting a point, taking finite unions of equivalence classes, or intersecting a family of definable sets all give back the original generic. The paper also shows atomicity transfers to merged generics, gives a counterexample showing saturation does not transfer, and proves that certain '1-local' smooth classes and their merges have the Hrushovski EPPA property, a stepping stone toward Ramsey properties.

What carries the argument

The central object is the merge $K_1 \circledast K_2$: the class of finite $L_1\cup L_2$-structures whose $L_1$-reducts lie in $K_1$ and whose $L_2$-reducts lie in $K_2$, ordered by requiring both strong relations to hold. The argument in Theorem 2.12 runs on disjoint parallel strongness (dPS) of $K_1$: it lets the proof amalgamate in $L_1$ on a universe that simultaneously carries an $L_2$-structure copied from the finite witness set $E\cup W_n$, so that a single $L^*$-structure $D$ exists with $W_n\le^* D$. Because the defining formula $\varphi$ is existential, the copied $L_2$-relations force the embedded $L_1$-amalgam to land inside $C^*$. For the EPPA half, the machinery is the 1-local property—a smooth class whose closure operator is computed pointwise—expanded into an extended structure with unary functions that encode closures, which converts smooth amalgamation into free amalgamation of extended structures.

What would settle it

Take the merge of a Shelah-Spencer class $K_\alpha$ with the Fraïssé class of finite equivalence relations, let $M^*$ be its generic, fix $a\in M^*$, and let $C^*=\{b\in M^*:E(a,b)\}$; check whether the $L_\alpha$-reduct of $C^*$ satisfies the extension property of Definition 2.4 for every finite $A\le_\alpha C^*$ and every finite $B$ with $A\le_\alpha B$—if some such $B$ cannot be embedded into $C^*|L_\alpha$ fixing $A$, then Theorem 2.12 is false.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 2.12: if $(K_1,\le_1)$ and $(K_2,\le_2)$ are smooth classes closed under substructure with disjoint amalgamation, and $K_1$ additionally has smooth intersections and disjoint parallel strongness, then for the generic $M^*$ of the merge, every infinite set $C^*=\varphi(M^*,m)$ definable by an existential $L_2$-formula has $C^*|L_1 \cong M_1$, the generic of $K_1$. The proof verifies the two defining properties of a generic for $C^*|L_1$. It builds, for any finite $A\le_1 C^*$ and any finite $B_1$ with $A\le_1 B_1$, an $L_1$-amalgam over a finite approximation $W_n\cap C^*$, then uses disjoint parallel strongness to place the needed $L_1$-structure on the same universe as the $L_2$-witnesses of the existential formula; genericity of $M^*$ embeds this combined structure back into $M^*$, and the existential formula guarantees the embedded copy of $B_1$ lies inside $C^*$. The theorem is presented as the main result of the merging study, and the paper derives its automorphism-group and self-similarity corollaries from it.

Load-bearing premise

The proof of Theorem 2.12 assumes that the $L_1$-amalgam and the copied $L_2$-witness structure can be placed on the same finite carrier set to form a single $L^*$-structure with $W_n\le^* D$, but no argument in the paper shows that the two structures are compatible on that carrier.

Editorial extensions

If this is right

The automorphism group of the merged generic satisfies $\operatorname{Aut}(M^*)\cong \operatorname{Aut}(M_1)\cap \operatorname{Aut}(M_2)$ as subgroups of the symmetric group on $M^*$.
Point removal is trivial for such generics: for any $a\in M_1$, the structure $M_1$ is isomorphic to $M_1-\{a\}$.
There is an expansion of $M_1$ by a binary equivalence relation $E$ such that every finite union of $E$-classes in $M_1$ is isomorphic to $M_1$.
For Shelah-Spencer generics, merging with the class of finite linear orders yields a family of subsets whose finite intersections are all isomorphic to the original generic $M_\alpha$.
If the two original generics are atomic, then the merged generic is atomic; in contrast, if both original classes have many minimal pairs, the merged generic is not saturated.

Reading between the lines

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

An implicit consequence is that any smooth class satisfying the hypotheses of Theorem 2.12 has a generic that is self-similar in a strong sense: the corollaries list point removal and equivalence-class unions, but the same mechanism should produce a copy of $M_1$ inside any infinite set definable in the second language, regardless of which Fraïssé class is merged in.
A testable extension is to isolate the same-universe compatibility condition that the proof of Theorem 2.12 invokes; if formalized, it would likely extend the theorem to merges where $K_1$ lacks disjoint parallel strongness, since that property is used only to place the two amalgams on one carrier.
The EPPA results suggest a route toward Ramsey properties for smooth classes: because merges of 1-local classes with the tuple-equivalence-relation class have $\le^*$-EPPA, adding linear orders to those merges and checking the Ramsey property would test whether the EPPA-to-Ramsey connection extends beyond Fraïssé classes.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Read Bryant's merges paper. The short version: the main theorem (2.12) is very likely correct, and the flagged gap in the stress test is not actually a gap. The line B*≤1α(B') follows from B1≤1B', which the dAP step explicitly gives; α is an L1-isomorphism onto its image, so strongness transfers. The same-universe issue is a non-issue because L1 and L2 are disjoint. The proof does have a typo: ρ maps B1 to {d_i}∪A, not to B*, and the notation around ρ/α needs cleaning. But the argument holds together.

What's new: Theorem 2.12 strengthens the EHN19 strong-extension result to get self-similarity of existentially L2-definable infinite subsets of the merge's generic; the corollaries (point deletion, automorphism group intersection, equivalence relation expansions, Shelah-Spencer applications) are real. The 1-local EPPA results and the KEω extension of Iva15/HKN22 to proper smooth classes are a genuine advance for the EPPA/Ramsey program.

Soft spots: Theorem 2.26 invokes Theorem 2.23, which needs smooth intersections, but neither class is assumed to have them. Easy fix: add the assumption, or prove the chain suffices without it. Corollary 3.22 asserts the Ramsey property for K⊛KLO without proving that the extended-structure Ramsey witness descends to the original class; that transfer is not automatic and needs an argument or the corollary should be softened. Proposition 2.11's construction of B2 assumes there is an L2-structure on the same universe as B1 extending A; that is not automatic for arbitrary smooth classes, though dPS may rescue it with a bit of work. These are fixable, not fatal.

The citation pattern is clean, no self-citation, and the paper relies on standard external tools (EHN19, HKN22, Iva15, KL92). If the author patches the flagged assumptions and supplies the missing Ramsey transfer, this becomes a solid paper for the smooth-class community.

Recommendation: send it to a serious referee. The main theorem is worth refereeing now; the EPPA section is worth refereeing. I would not desk-reject.

Referee Report

4 major / 5 minor

Summary. The paper studies merges of smooth classes (K1,≤1) and (K2,≤2) into a new smooth class K1⊛K2 and the model-theoretic properties of the resulting generic structure M*. The central result, Theorem 2.12, claims that if K1 has smooth intersections and dPS, then for any infinite set C* definable in M* by an existential L2-formula, the L1-reduct C*|L1 is isomorphic to the generic of K1. The paper derives several self-similarity corollaries from this theorem, proves transfer and non-transfer results for atomicity and saturation of generics, and in Section 3 proves EPPA results for a class of 1-local smooth classes and their merges, including a generalization of Ivanov's equivalence-relation construction. The paper is written in a clear expository style and builds on standard external results (EHN19, HKN22, Iva15, KL92, Bod12).

Significance. If Theorem 2.12 is correct, it is a substantial strengthening of the known fact that the generic of a merge of substructure-closed smooth classes expands the generic of each component: it shows that certain definable subsets of the merged generic already encode the original generic, yielding strong self-similarity phenomena for Shelah-Spencer classes and other smooth classes. The EPPA results for 1-local classes and their merges are also a useful contribution, as smooth classes with EPPA are rare and the paper gives a flexible construction via extended structures. The paper ships no code or machine-checked proofs, but the arguments are mostly self-contained and contain no fitted parameters or circular steps. The main theorem's proof, however, requires a careful repair before the stated results can be regarded as established.

major comments (4)

[§2.3.1, Theorem 2.12 proof] The proof of Theorem 2.12 is not rigorous as written because it conflates D1, D, and D*=α(D). Specifically, ρ is defined by ρ(b_i)=d_i, but B* is α({d_i})∪A, so the image of ρ is not contained in B*; the intended map is ρ(b_i)=α(d_i). The sentence 'α(ρ(B1))≤1α(B′)' then cannot be parsed, and the equality 'D*∩C*=α(W_n∪B′)∩C*' is written as though α were defined on W_n∪B′, which is a subset of D1, not of D. These are notation errors that can be repaired by working with the preimage p^{-1}(B′)⊆D and proving {d_i}∪A≤1p^{-1}(B′) from B1≤1B′, which does hold by the dAP construction; the dPS step only supplies W_n≤1D1 and B′⊆D1, and does not need to supply B′≤1D1. The proof also omits the needed condition that E be disjoint from W_n: if e_i∈W_n for some i, the map f:D2→E∪W_n is not injective. Because C* is infinite, one can choose E outside W_n (or replace W_n by an isomorphic copy avoiding E), but this must be stated explicitly.
[§2.3.3, Theorem 2.26] Theorem 2.26 invokes Theorem 2.23, but Theorem 2.23 requires (K,≤) to have smooth intersections, and Theorem 2.26 neither assumes smooth intersections for K1 and K2 nor proves that the merge (K*,≤*) has smooth intersections. The proof of Theorem 2.23 uses smooth intersections in the step where B≤M and A_i≤M imply B∩A_i≤A_i. Consequently the proof of Theorem 2.26 is incomplete; either add smooth intersections to the hypotheses or prove that they hold for the merge.
[§2.3.3, Definition 2.22] Definition 2.22 is internally inconsistent with its use: it defines (A,B) to be a ≤-minimal pair by 'for every A⊆B′⊊B, A≤B′ but A̸≤B', yet Theorem 2.23 and Theorem 2.26 construct chains of minimal pairs in which each consecutive pair satisfies A_i≤A_{i+1}. As printed, the final clause A̸≤B makes it impossible for consecutive members of such a chain to be strong extensions. The definition should presumably read 'A≤B and for every proper intermediate B′, A̸≤B′' (or the intended dual version); please correct it and re-check the argument in Theorem 2.23.
[§2.3.2, Proposition 2.21] In the proof of Proposition 2.21, the sentence 'By each of Th(M1) and Th(M2) being ω-categorical' assumes ω-categoricity of M1, but the proposition only assumes that K1 is a Fraïssé class. A Fraïssé class in a countable relational language need not have an ω-categorical limit (e.g., a language with countably many binary relations can produce a limit with infinitely many 2-types). The proposition should either add the hypothesis that M1 is ω-categorical, restrict to finite relational languages, or justify why the Fraïssé limit is ω-categorical in the present setting.

minor comments (5)

[§2.3.1, Theorem 2.12 proof] After defining D2, '∪{wi,...,wℓ}' should read '∪{w_1,...,w_ℓ}'.
[§2.3.1, Corollary 2.16] The statement '{P_i}_{i≤n} a set of n equivalence classes' should say n+1 classes if the index set is {0,...,n}.
[§2.3.1, Theorem 2.12 proof] The equality 'D*∩C*=α(Wn∪B′)∩C*' should be rewritten using the preimage of B′ in D, since α is not defined on Wn∪B′.
[§2.2.1, proof of Theorem 2.9] The cross-reference 'By Theorem 2.5' should be 'By Proposition 2.5'.
[§1, Introduction] The introduction advertises structural Ramsey theory, but Section 3 is mostly about EPPA; only Corollary 3.22 directly concerns the Ramsey property. The scope statement could be adjusted for accuracy.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Theorem 2.12 is derived from the stated dPS/smooth-intersection hypotheses and external amalgamation facts; no fitted parameter or self-citation carries the argument.

full rationale

The main derivation is self-contained with respect to circularity. Theorem 2.12 proves the strong embedding property for C* by constructing a finite amalgam D in the merged class, embedding D into the generic M*, and showing the embedded copy of the arbitrary B1 lies inside C*; the argument uses only Definition 2.4, Definition 2.3(3), smooth intersections, and preservation of existential L2-formulas under embeddings. None of the hypotheses is defined in terms of C* or of the conclusion, and no parameter is fitted to the target isomorphism. Proposition 2.11 and Theorem 2.12 do not rest on any Bryant-authored prior result: the cited sources (KL92, EHN19, HKN22, Iva15, Bod12, etc.) are external, and the EHN19 result on reducts is explicitly said to have a different proof from Proposition 2.11. Corollaries 2.15-2.18 choose auxiliary Fraisse classes and are applications, not disguised assumptions. The only passage meriting flagging is in the proof of Theorem 2.12, where 'Notice that B∗≤1 α(B′)' is asserted rather than derived in detail; if this step is unsupported it is a correctness gap, not a circular reduction, because the desired conclusion is not an input to the construction. I therefore assign score 0.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The central claims rest on standard model theory (Fraïssé generic existence, Ryll-Nardzewski) and on specific external theorems (EHN19, HKN22, EHN21, Iva15, BS96, Gun18). No numerical free parameters are fitted; the inherited parameter α in Shelah-Spencer examples is a parameter of the ambient known class, not introduced here. Extended structures are defined constructs with a proven correspondence to the original class, not unexplained entities.

assumptions (7)

standard math Fraïssé-style generic existence: a smooth class with AP and countably many isomorphism types has a unique generic up to isomorphism.
Used as Proposition 2.5 to assert existence and uniqueness of generics, including the merged generic M*.
domain assumption EHN19 strong extension theorem: when smooth classes closed under substructure with dAP are merged, M*|Li is isomorphic to Mi.
Used as a baseline for the merging framework; the paper strengthens rather than reproves this result.
domain assumption HKN22 Theorem 3.10: any free amalgamation class of extended structures with unary functions has Γ_L-EPPA.
Main engine for the EPPA results in Section 3.
domain assumption EHN21 Theorem 1.3: extended Fraïssé classes with free amalgamation merged with KLO have the Ramsey property.
Used without statement in Corollary 3.22.
standard math Ryll-Nardzewski and ω-categoricity facts: an ω-categorical theory has a unique countable model and finitely many n-types.
Used in Proposition 2.21 to transfer N |= Th(M*) to N ≅ M*.
domain assumption BS96 and Gun18 facts on Shelah-Spencer classes: dPS, fAP, smooth intersections, many minimal pairs, and saturated generics for rational α.
Used in Example 2.25 and Corollaries 2.16-2.18; the saturation of Mα is external to this paper.
domain assumption Iva15: the class KEω has EPPA.
Stated as Theorem 3.28 and provides the base class generalized in Theorem 3.29.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Merges of Smooth Classes and Their Properties." pith.science (2026). https://pith.science/paper/56SNECTR

@misc{pith2026241110689,
  author       = {Pith},
  title        = {Pith review of: Merges of Smooth Classes and Their Properties},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/56SNECTR}},
  note         = {Machine review of arXiv:2411.10689}
}

read the original abstract

Given two Fra\"iss\'e-like classes with generic limits, we ask whether we can merge the two classes into one class with a generic limit. We study the properties of these merges and their generics, as well as their connections to structural Ramsey theory and the Hrushovski property (EPPA).

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

8 extracted references · 7 canonical work pages

[1996]

[EHN19] D. M. Evans, J. Hubička, and J. Nešetřil. Automorphism groups and Ramsey properties ofsparsegraphs .Proceedings of the London Mathematical Society,119(2):515–546,2019. [EHN21] D. M. Evans, J. Hubička, and J. Nešetřil. Ramsey properties and extending partial auto- morphisms for classes of finite structures . Fundamenta Mathematicae, 253(2): 121–153,

work page 2019
[2004]

[KR06] A

arXiv: math/0305241 [math.LO] . [KR06] A. S. Kechris and C. Rosendal. Turbulence, amalgamation and generic automorphisms of homogeneous structures ,

arXiv
[2006]

Turbulence, amalgamation and generic automorphisms of homogeneous structures

arXiv: math/0409567 [math.LO] . [Las07] M. Laskowski. A simpler axiomatization of the Shelah-Spencer almost sure theory .Israel Journal of Mathematics, 161: 157–186,

work page Pith review arXiv
[2012]

New Ramsey Classes from Old

arXiv: 1204.3258 [math.LO] . [BS96] J. T. Baldwin and N. Shi. Stable generic structures . Annals of Pure and Applied Logic, 79(1): 1–35,

work page Pith review arXiv
[2015]

com/doi/pdf/10.1002/malq.201400036

eprint: https://onlinelibrary.wiley. com/doi/pdf/10.1002/malq.201400036 . [KL92] D. W. Kueker and M. C. Laskowski. On generic structures . Notre Dame Journal of Formal Logic, 33(2): 175 –183,

work page doi:10.1002/malq.201400036
[2016]

Automorphism Groups of Generic Structures: Extreme Amenability and Amenability

arXiv: 1508.04628 [math.LO] . [Gun18] D. K. Gunatilleka. The theories of Baldwin-Shi hypergraphs and their atomic models ,

work page Pith review arXiv
[2018]

The theories of Baldwin-Shi hypergraphs and their atomic models

arXiv: 1803.01831 [math.LO] . [Her95] B. Herwig. Extending partial isomorphisms on finite structures . Combinatorica, 15: 365– 371,

work page Pith review arXiv
[2022]

All those EPPA classes (Strengthenings of the Herwig-Lascar theorem)

arXiv: 1902.03855 [math.CO] . [HO03] I. Hodkinson and M. Otto. Finite conformal hypergraph covers and Gaifman cliques in finite structures . The Bulletin of Symbolic Logic, 9(3): 387–405,

work page Pith review arXiv 1902

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1996]

[EHN19] D. M. Evans, J. Hubička, and J. Nešetřil. Automorphism groups and Ramsey properties ofsparsegraphs .Proceedings of the London Mathematical Society,119(2):515–546,2019. [EHN21] D. M. Evans, J. Hubička, and J. Nešetřil. Ramsey properties and extending partial auto- morphisms for classes of finite structures . Fundamenta Mathematicae, 253(2): 121–153,

work page 2019

[2] [2004]

[KR06] A

arXiv: math/0305241 [math.LO] . [KR06] A. S. Kechris and C. Rosendal. Turbulence, amalgamation and generic automorphisms of homogeneous structures ,

arXiv

[3] [2006]

Turbulence, amalgamation and generic automorphisms of homogeneous structures

arXiv: math/0409567 [math.LO] . [Las07] M. Laskowski. A simpler axiomatization of the Shelah-Spencer almost sure theory .Israel Journal of Mathematics, 161: 157–186,

work page Pith review arXiv

[4] [2012]

New Ramsey Classes from Old

arXiv: 1204.3258 [math.LO] . [BS96] J. T. Baldwin and N. Shi. Stable generic structures . Annals of Pure and Applied Logic, 79(1): 1–35,

work page Pith review arXiv

[5] [2015]

com/doi/pdf/10.1002/malq.201400036

eprint: https://onlinelibrary.wiley. com/doi/pdf/10.1002/malq.201400036 . [KL92] D. W. Kueker and M. C. Laskowski. On generic structures . Notre Dame Journal of Formal Logic, 33(2): 175 –183,

work page doi:10.1002/malq.201400036

[6] [2016]

Automorphism Groups of Generic Structures: Extreme Amenability and Amenability

arXiv: 1508.04628 [math.LO] . [Gun18] D. K. Gunatilleka. The theories of Baldwin-Shi hypergraphs and their atomic models ,

work page Pith review arXiv

[7] [2018]

The theories of Baldwin-Shi hypergraphs and their atomic models

arXiv: 1803.01831 [math.LO] . [Her95] B. Herwig. Extending partial isomorphisms on finite structures . Combinatorica, 15: 365– 371,

work page Pith review arXiv

[8] [2022]

All those EPPA classes (Strengthenings of the Herwig-Lascar theorem)

arXiv: 1902.03855 [math.CO] . [HO03] I. Hodkinson and M. Otto. Finite conformal hypergraph covers and Gaifman cliques in finite structures . The Bulletin of Symbolic Logic, 9(3): 387–405,

work page Pith review arXiv 1902