betaFeedback

WorksMathematics

The metric theory of tracial matrix algebras does not converge

Abstract

We show that the theories of tracial matrix algebras and metric unitary groups do not converge with the dimension. In the main argument, we recover the canonical diagonal using spectral gap and internality. Uniform definability of the diagonal and its normalizer then transfers the distinct theories of universal sofic groups to matrix ultraproducts. The same methods show that isomorphisms preserve every internal subalgebra and give fixed finite distance formulas over six unitaries. Finally, we construct an explicit continuum family of matrix ultraproducts with distinct theories and trivial first-order fundamental groups.

Introduction

We study the complete continuous first-order theories of the matrix algebras Mn(C)M_n(\mathbb{C}), equipped with the normalized trace τn=n−1Tr⁡\tau_n = n^{-1}\operatorname{Tr} and the metric induced by ∥x∥2=τn(x∗x)1/2\lVert x\rVert_2 = \tau_n(x^*x)^{1/2} on operator-norm bounded balls. A sentence has a real value in each algebra, and two tracial algebras are elementarily equivalent if all these values agree. Farah, Hart and Sherman asked whether every sentence value converges as n→∞n \to\infty [12 Question 6.6]. To express this question in terms of ultraproducts, we use Łoś’ theorem:

φ∏UMn(C)=lim⁡UφMn(C),\varphi\prod_{\mathcal{U}} M_n(\mathbb{C}) = \lim_{\mathcal{U}} \varphi^{M_n(\mathbb{C})},

so convergence is equivalent to elementary equivalence of all nonprincipal tracial ultraproducts of this sequence. Our first result gives a negative answer.

Theorem A. There are 2ℵ02^{\aleph_0} pairwise non-elementarily equivalent tracial matrix ultraproducts of the sequence (Mn(C),τn)(M_n(\mathbb{C}),\tau_n), and the same holds for metric ultraproducts of (U(n),d2)(U(n),d_2) in the pure metric group language, where d2(u,v)=∥u−v∥2d_2(u,v)=\lVert u-v\rVert_2. Consequently, the sequences of complete continuous theories

Th⁡(Mn(C),τn),Th⁡(U(n),d2)\operatorname{Th}(M_n(\mathbb{C}),\tau_n), \qquad\operatorname{Th}(U(n),d_2)

do not converge as n→∞n \to\infty.

Purely universal and purely existential sentences do converge, since every tracial matrix ultraproduct has the universal and existential theories of the hyperfinite II1\mathrm{II}_1 factor [13 Remark 6.5(1), equation (6.2)].

To distinguish the theories, we recover the diagonal and use its unitary normalizer. The quotient by the diagonal unitaries is a universal sofic group, so uniform recovery of this quotient allows us to apply the nonconvergence result of [3] to matrix algebras. We first fix the class of ultraproducts and the notion of internality used in this recovery.

Definition 1.1. Let U\mathcal{U} be a nonprincipal ultrafilter on N\mathbb{N}, and let d=(dn)\mathbf{d}=(d_n) be a sequence of positive integers with dn→∞d_n \to\infty. We call

MU,d=∏U(Mdn(C),τdn)(1)\mathcal{M}_{\mathcal{U},\mathbf{d}}=\prod_{\mathcal{U}}(M_{d_n}(\mathbb{C}),\tau_{d_n}) \tag*{(1)}

a tracial matrix ultraproduct. Its elements are classes [xn]U[x_n]_{\mathcal{U}} of operator-norm bounded sequences, modulo sequences whose 22-norms tend to zero along U\mathcal{U}. The operations are induced coordinatewise, and τ([xn]U)=lim⁡Uτdn(xn)\tau([x_n]_{\mathcal{U}})=\lim_{\mathcal{U}}\tau_{d_n}(x_n). A unital von Neumann subalgebra A⊆MU,dA\subseteq\mathcal{M}_{\mathcal{U},\mathbf{d}} is internal if A=[An]UA=[A_n]_{\mathcal{U}} for unital ∗*-subalgebras An⊆Mdn(C)A_n\subseteq M_{d_n}(\mathbb{C}). An internal MASA is an internal abelian subalgebra that is maximal among internal abelian subalgebras.

Throughout, MU,d\mathcal{M}_{\mathcal{U},\mathbf{d}} and MV,e\mathcal{M}_{\mathcal{V},\mathbf{e}} denote tracial matrix ultraproducts in this sense. To recognize the diagonal, we will use Lemma 2.1: internal MASAs are maximal abelian in the ambient algebra and are unitarily conjugate to the canonical diagonal [Ddn]U[D_{d_n}]_{\mathcal{U}}, where Dd⊆Md(C)D_d\subseteq M_d(\mathbb{C}) denotes the diagonal algebra. Isomorphisms are complex ∗*-isomorphisms; they preserve the trace because these ultraproducts are II1\mathrm{II}_1 factors.

To use this recovery for elementary equivalence, we work in the continuous language of tracial ∗*-algebras on operator-norm bounded sorts. A predicate is definable if it is a uniform limit of formulas, and a nonempty closed set is definable if its distance predicate is definable. We allow parameters when specified, use bounded real ranges, and write s−˙t=max⁡{s−t,0}s\mathbin{\dot{-}}t=\max\{s-t,0\}. See [10 Section 3.2] for quantification over definable sets. The following theorem gives a uniform way to recover an internal diagonal from finitely many parameters.

Theorem B. There are fixed finite formulas Δ(z)\Delta(z) and δ(x;z)\delta(x;z), with zz a tuple of six contractions, such that in every tracial matrix ultraproduct MU,d\mathcal{M}_{\mathcal{U},\mathbf{d}}: (1) the zero-set of Δ\Delta is nonempty; (2) every zz in this zero-set defines an internal MASA DzD_z with

δ(x;z)=dist⁡2(x,(Dz)≤1)(x∈(MU,d)≤1).\delta(x;z)=\operatorname{dist}_2(x,(D_z)_{\leq1}) \qquad(x\in(\mathcal{M}_{\mathcal{U},\mathbf{d}})_{\leq1}).

Every such MASA is unitarily conjugate to the canonical diagonal.

The diagonal cannot be the commutant of a finite tuple [30 Theorem 2.1(1)], so we recover it as the relative join of two abelian commutants with spectral gap. The spectral gap internality theorem of Liu and the authors [ALT26, Theorem B] makes these commutants internal. Uniform abelian Hilbert–Schmidt stability then ensures that their relative join is internal as well.

The finite code and conjugacy of internal MASAs will allow us to recover the theory of the pair (MU,d,D)(\mathcal{M}_{\mathcal{U},\mathbf{d}},D) from that of the ambient algebra. Here the pair is the expansion by the distance to the operator-norm unit ball of DD. We obtain the following reduction.

Theorem C. Let D=[Ddn]U⊆MU,dD=[D_{d_n}]_{\mathcal{U}}\subseteq\mathcal{M}_{\mathcal{U},\mathbf{d}} and F=[Den]V⊆MV,eF=[D_{e_n}]_{\mathcal{V}}\subseteq\mathcal{M}_{\mathcal{V},\mathbf{e}} be the canonical diagonal subalgebras of two tracial matrix ultraproducts. The following hold. (1) Every isomorphism θ:MU,d→MV,e\theta:\mathcal{M}_{\mathcal{U},\mathbf{d}}\to\mathcal{M}_{\mathcal{V},\mathbf{e}} sends DD to a unitarily conjugate of FF. (2) If MU,d≡MV,e\mathcal{M}_{\mathcal{U},\mathbf{d}}\equiv\mathcal{M}_{\mathcal{V},\mathbf{e}}, then (MU,d,D)≡(MV,e,F)(\mathcal{M}_{\mathcal{U},\mathbf{d}},D)\equiv(\mathcal{M}_{\mathcal{V},\mathbf{e}},F). (3) If MU,d≡MV,e\mathcal{M}_{\mathcal{U},\mathbf{d}}\equiv\mathcal{M}_{\mathcal{V},\mathbf{e}}, then

∏UmetSym⁡(dn)≡∏VmetSym⁡(en),\prod_{\mathcal{U}}^{\mathrm{met}}\operatorname{Sym}(d_n)\equiv\prod_{\mathcal{V}}^{\mathrm{met}}\operatorname{Sym}(e_n),

where the symmetric groups carry the normalized Hamming metric.

To obtain the last assertion, permutation matrices identify the Weyl group with the metric ultraproduct of symmetric groups. Its Hamming metric is one half the square of the quotient 22-metric. Definability lets us make this identification under elementary equivalence, and [3 Theorem 5.6 and Remark 5.7] then gives the matrix-algebra assertion of Theorem A. Reconstruction from unitary groups gives the other assertion. The argument that preserves the diagonal also extends to arbitrary internal algebras, giving the following structural conclusion.

Theorem D. Let θ:MU,d→MV,e\theta: \mathcal{M}_{\mathcal{U},d} \to\mathcal{M}_{\mathcal{V},e} be an isomorphism of tracial matrix ultraproducts. Then for every unital von Neumann subalgebra A⊆MU,dA \subseteq\mathcal{M}_{\mathcal{U},d},

A is internal⟺θ(A) is internal.A\text{ is internal}\quad\Longleftrightarrow\quad\theta(A)\text{ is internal}.

The constructions used here also give fixed finite distance formulas over four unitary parameters for internal abelian algebras and six for arbitrary internal algebras; see Proposition 5.2. They also show that, for commuting internal algebras AA, BB, both A′∩B′A' \cap B' and their relative join (A′∩B′)′(A' \cap B')' are internal (Corollary 4.4).

We next use the recovered permutation groups to distinguish an explicit family of factors from their amplifications. For a II1\mathrm{II}_{1} factor QQ, let Qt=pMk(Q)pQ^{t}=pM_{k}(Q)p, with normalized trace, where k>tk>t and (τk⊗τ)(p)=t/k(\tau_{k}\otimes\tau)(p)=t/k. Its isomorphism class depends only on t>0t>0. The first-order fundamental group

Ffo(Q)={t>0:Qt≡Q}\mathcal{F}_{\mathrm{fo}}(Q)=\{t>0:Q^{t}\equiv Q\}

contains F(Q)={t>0:Qt≅Q}\mathcal{F}(Q)=\{t>0:Q^{t}\cong Q\}; see [17 Section 4]. For the construction, put Gr=PSL⁡2(2r)G_{r}=\operatorname{PSL}_{2}(2^{r}), for r≥2r\geq2, so that ∣Gr∣=2r(22r−1)|G_{r}|=2^{r}(2^{2r}-1).

Theorem E. Fix a nonprincipal ultrafilter U\mathcal{U} on N\mathbb{N}. For each set SS of odd primes, put

rn(S)=2n∏p∈Sp≤np,QS=∏U(M∣Grn(S)∣(C),τ∣Grn(S)∣).r_{n}(S)=2^{n}\prod_{\substack{p\in S\\p\leq n}}p,\qquad Q_{S}=\prod_{\mathcal{U}}\left(M_{|G_{r_{n}(S)}|}(\mathbb{C}),\tau_{|G_{r_{n}(S)}|}\right).

Then Ffo(QS)=F(QS)={1}\mathcal{F}_{\mathrm{fo}}(Q_{S})=\mathcal{F}(Q_{S})=\{1\}. Distinct sets SS give non-elementarily equivalent factors.

To prove this theorem, we follow the left-right action method of [3 Section 5] and recover the regular actions of PSL⁡2(2r)\operatorname{PSL}_{2}(2^{r}) from their elementary properties. The recovered actions force the ambient dimension to be asymptotic to a group order. We then use element orders to detect divisibility of rr, which rules out nontrivial amplifications.

Popa asked in 2008 whether there are uncountably many nonisomorphic tracial matrix ultraproducts [11 Introduction]. For the related ultrapower problem, Ge and Hadwin proved that, under CH, all tracial ultrapowers of a fixed finite von Neumann algebra with separable predual are isomorphic [15 Theorem 3.2]. Li later suggested that matrix ultraproducts might be independent of the ultrafilter [27 p. 1287]. Under failure of CH, Farah, Hart and Sherman obtained nonisomorphic matrix ultraproducts [11 Proposition 3.3], and Farah and Shelah obtained the maximal number 2c2^{\mathfrak{c}} [14 Proposition 8.3]. Under CH, matrix ultraproducts are isomorphic exactly when they are elementarily equivalent. Theorem A distinguishes their theories in ZFC and hence answers Popa’s question and gives continuum many classes in the classification problem of [25 Problem 1]. For the model-theoretic framework and the game formulation, see [12, 13] and [9 Question 4.1].

For universal sofic groups, Thomas obtained the maximal number of isomorphism types under failure of CH and realized algebraic ultraproducts of finite expander groups as centralizers [33 Theorems 1.1 and 2.2]. The authors obtained continuum many theories in ZFC in [3] and proposed the diagonal-normalizer route to matrix ultraproducts [3 Remark 5.8]. Jekel discussed this route in [22 Section 1.1]. Since full types include sentence values, nonconvergence also obstructs convergence of full types of random matrix tuples. It answers the corresponding ultrafilter-independence question for noncommutative stochastic processes as well; see [21 Question 5.6], [22 Section 1.1], and [23 Remark 1.3 and Proposition D(1)].

Our internality arguments use Popa’s coordinate commutant identities [30 Section 1.6] and conjugacy of coordinate MASAs [29 Section 2.1(a)]. The diagonal is a MASA, but its normalizer generates a proper subalgebra: tracial matrix ultraproducts have no Cartan subalgebras [29 Proposition 2.3]; see also [28 Theorem 7.3] for ultrapowers of factors.

While preparing this manuscript, we learned of the independent work of Ando and Goldbring [4]. They prove that every tracial matrix ultraproduct has trivial first-order fundamental group, which also implies nonconvergence.

The paper is organized as follows: Section 2 supplies the internality results needed to recognize the diagonal. Using these, Section 3 recovers the diagonal and its normalizer under isomorphism, and Section 4 extends preservation to all internal algebras. To pass to elementary equivalence, we make the constructions definable in Section 5. Section 6 then recovers the left-right actions from admissible tuples and uses element orders to prove Theorem E.

Spectral gap and internal algebras

We use the normalized trace τd=d−1Tr⁡\tau_d = d^{-1}\operatorname{Tr} on Md=Md(C)M_d = M_d(\mathbb{C}) and its norm ∥x∥2,d=τd(x∗x)1/2\lVert x\rVert_{2,d} = \tau_d(x^*x)^{1/2}. We write A≤1A_{\leq1} for the operator-norm unit ball of an algebra AA. Bounded functional calculus lets us choose contractive coordinate representatives for every unit-ball element of an internal algebra.

We use primes for commutants in the specified ambient matrix ultraproduct. For a finite tuple u\mathbf{u}, we write {u}′∩MU,d\{\mathbf{u}\}' \cap\mathcal{M}_{U,d} for its commutant, and we define the relative join of two subalgebras by

A∨relB=(A′∩B′)′.(2)A \mathbin{\vee_{\mathrm{rel}}} B = (A' \cap B')'. \tag*{(2)}

Thus the relative join is the smallest algebra containing AA and BB that equals its relative bicommutant.

Internal subalgebras, commutants and spectral gap

The following identities let us pass between coordinate models and commutants in the ultraproduct. They also reduce unitary conjugacy of internal MASAs to the matrix case.

References

  1. [1]D. Akhtiamov and A. Dogon, On uniform Hilbert Schmidt stability of groups, Proc. Amer. Math. Soc. 150 (2022), no. 4, 1799–1809.DOI
  2. [3]V. Alekseev and A. Thom, On non-isomorphic universal sofic groups, arXiv:2406.06741 (2024), version 3 (2026).arxiv.org/abs/2406.06741
  3. [4]H. Ando and I. Goldbring, Matrix ultraproducts have trivial fundamental group, arXiv:2611.18328 (2026).arxiv.org/abs/2611.18328
  4. [6]O. Becker and M. Chapman, Stability of approximate group actions: uniform and probabilistic, J. Eur. Math. Soc. 25 (2023), no. 9, 3599–3632.arxiv.org/abs/2005.06652
  5. [7]I. Ben Yaacov, A. Berenstein, C. W. Henson and A. Usvyatsov, Model theory for metric structures, in Model Theory with Applications to Algebra and Analysis, Vol. II, London Math. Soc. Lecture Note Ser. 350, Cambridge University Press, Cambridge, 2008, 315–427.
  6. [9]I. Farah, Games in the matrix: Director’s cut, arXiv:2511.07456 (2025).arxiv.org/abs/2511.07456
  7. [10]I. Farah, B. Hart, M. Lupini, L. Robert, A. Tikuisis, A. Vignati and W. Winter, Model theory of C*-algebras, Mem. Amer. Math. Soc. 271 (2021), no. 1324, viii+127 pp.
  8. [11]I. Farah, B. Hart and D. Sherman, Model theory of operator algebras I: stability, Bull. Lond. Math. Soc. 45 (2013), no. 4, 825–838.DOI
  9. [12]I. Farah, B. Hart and D. Sherman, Model theory of operator algebras II: model theory, Israel J. Math. 201 (2014), no. 1, 477–505.arxiv.org/abs/1004.0741
  10. [13]I. Farah, B. Hart and D. Sherman, Model theory of operator algebras III: elementary equivalence and II₁ factors, Bull. Lond. Math. Soc. 46 (2014), no. 3, 609–628.
  11. [14]I. Farah and S. Shelah, A dichotomy for the number of ultrapowers, J. Math. Logic 10 (2010), nos. 1–2, 45–81.arxiv.org/abs/0912.0406
  12. [15]L. Ge and D. Hadwin, Ultraproducts of C*-algebras, in Recent Advances in Operator Theory and Related Topics, Operator Theory: Advances and Applications 127, Birkhäuser, Basel, 2001, 305–326.DOI
  13. [16]I. Goldbring, Spectral gap and definability, in J. Iovino (ed.), Beyond First Order Model Theory, Volume II, Chapman & Hall/CRC, Boca Raton, FL, 2023, 103–138.arxiv.org/abs/1805.02752
  14. [17]I. Goldbring and B. Hart, A survey on the model theory of tracial von Neumann algebras, in I. Goldbring (ed.), Model Theory of Operator Algebras, De Gruyter Series in Logic and Its Applications 11, De Gruyter, Berlin, 2023, 133–157; arXiv:2210.15508.arxiv.org/abs/2210.15508
  15. [18]I. Goldbring, B. Hart and H. Towsner, Explicit sentences distinguishing McDuff’s II₁ factors, Israel J. Math. 227 (2018), no. 1, 365–377.DOI
  16. [19]I. Goldbring and T. Sinclair, Games and elementary equivalence of II₁-factors, Pacific J. Math. 278 (2015), no. 1, 103–118.DOI
  17. [20]V. Iyer, S. Jain, S. Jordan and R. D. Somma, Efficient quantum circuits for high-dimensional representations of SU(n) and Ramanujan quantum expanders, preprint (2026), arXiv:2602.15180v1.arxiv.org/abs/2602.15180
  18. [21]D. Jekel, Free probability and model theory of tracial W*-algebras, in I. Goldbring (ed.), Model Theory of Operator Algebras, De Gruyter Series in Logic and Its Applications 11, De Gruyter, Berlin, 2023, 215–268.DOI
  19. [22]D. Jekel, Information geometry for types in the large-n limit of random matrices, Comm. Math. Phys. 406 (2025), no. 11, article 272, 67 pp.arxiv.org/abs/2501.00703
  20. [23]D. Jekel, Free information geometry and the model theory of noncommutative stochastic processes, arXiv:2604.12212 (2026).arxiv.org/abs/2604.12212
  21. [24]M. Kassabov, A. Lubotzky and N. Nikolov, Finite simple groups as expanders, Proc. Natl. Acad. Sci. USA 103 (2006), no. 16, 6116–6119.pubmed.ncbi.nlm.nih.gov/16601101
  22. [25]S. Kunnawalkam Elayavalli, 50 Open Problems: Ultraproduct II₁ factors, arXiv:2511.20377 (2025).
  23. [26]G. Kun and A. Thom, Inapproximability of actions and Kazhdan’s property (T), arXiv:1901.03963 (2019), version 3 (2026).arxiv.org/abs/1901.03963
  24. [27]W. Li, On ultraproducts of operator algebras, Sci. China Ser. A 48 (2005), no. 9, 1284–1295.DOI
  25. [28]S. Popa, Orthogonal pairs of *-subalgebras in finite von Neumann algebras, J. Operator Theory 9 (1983), no. 2, 253–268.
  26. [29]S. Popa, A II₁ factor approach to the Kadison–Singer problem, Comm. Math. Phys. 332 (2014), no. 1, 379–414.
  27. [30]S. Popa, Independence properties in subalgebras of ultraproduct II₁ factors, J. Funct. Anal. 266 (2014), no. 9, 5818–5846.DOI
  28. [31]S. Popa, A. M. Sinclair and R. R. Smith, Perturbations of subalgebras of type II₁ factors, J. Funct. Anal. 213 (2004), no. 2, 346–379; corrigendum, J. Funct. Anal. 235 (2006), 355–356.DOI
  29. [33]S. Thomas, On the number of universal sofic groups, Proc. Amer. Math. Soc. 138 (2010), no. 7, 2585–2590.DOI

Paper details

vlnc:W_74fvk48n
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Large language models

View statement
Looking for related papers…
Contents