Skip to main content
Log in

Reducibility of n-ary semigroups: from quasitriviality towards idempotency

  • Original Paper
  • Published:
Beiträge zur Algebra und Geometrie / Contributions to Algebra and Geometry Aims and scope Submit manuscript

Abstract

Let X be a nonempty set. Denote by \({\mathcal {F}}^n_k\) the class of associative operations \(F:X^n\rightarrow X\) satisfying the condition \(F(x_1,\ldots ,x_n)\in \{x_1,\ldots ,x_n\}\) whenever at least k of the elements \(x_1,\ldots ,x_n\) are equal to each other. The elements of \({\mathcal {F}}^n_1\) are said to be quasitrivial and those of \({\mathcal {F}}^n_n\) are said to be idempotent. We show that \({\mathcal {F}}^n_1=\cdots ={\mathcal {F}}^n_{n-2}\subseteq {\mathcal {F}}^n_{n-1}\subseteq {\mathcal {F}}^n_n\) and we give conditions on the set X for the last inclusions to be strict. The class \({\mathcal {F}}^n_1\) was recently characterized by Couceiro and Devillet (Algebra Universalis 80:19, 2019), who showed that its elements are reducible to binary associative operations. However, some elements of \({\mathcal {F}}^n_n\) are not reducible. In this paper, we characterize the class \({\mathcal {F}}^n_{n-1}{\setminus }{\mathcal {F}}^n_1\) and show that its elements are reducible. We give a full description of the corresponding reductions and show how each of them is built from a quasitrivial semigroup and an Abelian group whose exponent divides \(n-1\).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  • Ackerman, N.L.: A characterization of quasitrivial \(n\)-semigroups. To appear in Algebra Universalis. http://people.math.harvard.edu/%7enate/papers/ (2020)

  • Couceiro, M., Devillet, J.: Every quasitrivial \(n\)-ary semigroup is reducible to a semigroup. Algebra Universalis 80(4), 19 (2019)

    Article  MathSciNet  Google Scholar 

  • Couceiro, M., Devillet, J., Marichal, J.-L.: Quasitrivial semigroups: characterizations and enumerations. Semigr. Forum 98(3), 472–498 (2019)

    Article  MathSciNet  Google Scholar 

  • Devillet, J., Kiss, G., Marichal, J.-L.: Characterizations of quasitrivial symmetric nondecreasing associative operations. Semigr. Forum 98(1), 154–171 (2019)

    Article  MathSciNet  Google Scholar 

  • Dudek, W.A.: Idempotents in \(n\)-ary semigroups. Southeast Asian Bull. Math. 25, 97–104 (2001)

    Article  MathSciNet  Google Scholar 

  • Dudek, W.A., Głazek, K.: Around the Hosszu–Głuskin theorem for \(n\)-ary groups. Discrete Math. 308, 4861–4876 (2008)

    Article  MathSciNet  Google Scholar 

  • Dudek, W.A., Mukhin, V.V.: On \(n\)-ary semigroups with adjoint neutral element. Quasigr. Relat. Syst. 14, 163–168 (2006)

    MathSciNet  MATH  Google Scholar 

  • Dörnte, W.: Untersuchungen über einen verallgemeinerten Gruppenbegriff. Math. Z 29, 1–19 (1928)

    Article  Google Scholar 

  • Kiss, G., Somlai, G.: Associative idempotent nondecreasing functions are reducible. Semigr. Forum 98(1), 140–153 (2019)

    Article  MathSciNet  Google Scholar 

  • Lehtonen, E., Starke, F.: On associative operations on commutative integral domains. Semigr. Forum 100, 910–915 (2020)

    Article  MathSciNet  Google Scholar 

  • Länger, H.: The free algebra in the variety generated by quasi-trivial semigroups. Semigr. Forum 20(1), 151–156 (1980)

    Article  Google Scholar 

  • Marichal, J.-L., Mathonet, P.: A description of \(n\)-ary semigroups polynomial-derived from integral domains. Semigr. Forum 83, 241–249 (2011)

    Article  MathSciNet  Google Scholar 

  • Post, E.L.: Polyadic groups. Trans. Am. Math. Soc. 48, 208–350 (1940)

    Article  MathSciNet  Google Scholar 

  • Pouzet, M., Rosenberg, I.G., Stone, M.G.: A projection property. Algebra Universalis 36(2), 159–184 (1996)

    Article  MathSciNet  Google Scholar 

  • Rotman, J.J.: An Introduction to the Theory of Groups, 4th edn. Springer, New York (1995)

    Book  Google Scholar 

  • Szendrei, Á.: Clones in Universal Algebra. Séminaire de Mathématiques Supérieures, Vol. 99. Les Presses de l’Université de Montréal, Montréal (1986)

Download references

Acknowledgements

The authors would like to thank the anonymous reviewers for their insightful remarks that helped improving the current paper. They are especially grateful for Proposition 1.10 and Corollary 1.11. The second author is supported by the Luxembourg National Research Fund under the project PRIDE 15/10949314/GSM.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Miguel Couceiro.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix A. Alternative proof of Corollary 1.7

Appendix A. Alternative proof of Corollary 1.7

We provide an alternative proof of Corollary 1.7 that does not use Couceiro and Devillet (2019, Corollary 2.3).

To this extent, we first prove the following general result.

Proposition A.1

Let \(F \in {\mathcal {F}}^n_n\). The following assertions are equivalent.

  1. (i)

    F is reducible to an associative and idempotent operation \(G:X^2 \rightarrow X\).

  2. (ii)

    \(F((n-1)\varvec{\cdot }x,y) = F(x,(n-1)\varvec{\cdot }y)\) for any \(x,y\in X\).

Proof

The implication (i) \(\Rightarrow \) (ii) is straightforward. Now, let us show that (ii) implies (i). So, suppose that

$$\begin{aligned} F((n-1)\varvec{\cdot }x,y) ~=~ F(x,(n-1)\varvec{\cdot }y) \qquad x,y\in X, \end{aligned}$$
(7)

and consider the operation \(G:X^2 \rightarrow X\) defined by \(G(x,y) = F((n-1)\varvec{\cdot }x,y)\) for any \(x,y\in X\). It is not difficult to see that G is associative and idempotent. Now, let \(x_1,\ldots ,x_n\in X\) and let us show that \(G^{n-1}(x_1,\ldots ,x_n)=F(x_1,\ldots ,x_n)\). Using repeatedly (7) and the idempotency of F we obtain

$$\begin{aligned} G^{n-1}(x_1,\ldots ,x_n)= & {} F^{n-1}((n-1)\varvec{\cdot }x_1,(n-1)\varvec{\cdot }x_2,\ldots ,(n-1)\varvec{\cdot }x_{n-1},x_n) \\= & {} F^{n-1}((2n-3)\varvec{\cdot }x_1,x_2,(n-1)\varvec{\cdot }x_3,\ldots ,(n-1)\varvec{\cdot }x_{n-1},x_n) \\= & {} \cdots \\= & {} F^{n-1}(((n-2)(n-1)+1)\varvec{\cdot }x_1,x_2,x_3,\ldots ,x_{n-1},x_n)\\= & {} F(x_1,\ldots ,x_n), \end{aligned}$$

which shows that F is reducible to G. \(\square \)

Remark 2

Let \(\le \) be a total ordering on X. An operation \(F:X^n \rightarrow X\) is said to be \(\le \)-preserving if \(F(x_1,\ldots ,x_n) \le F(x_1',\ldots ,x_n')\) whenever \(x_i\le x_i'\) for any \(i\in \{1,\ldots ,n\}\). One of the main results of Kiss and Somlai (2019, Theorem 4.8) is that every \(\le \)-preserving operation \(F\in {\mathcal {F}}^n_n\) is reducible to an associative, idempotent, and \(\le \)-preserving binary operation. To this extent, they first show (Kiss and Somlai 2019, Lemma 4.1) that any \(\le \)-preserving operation \(F\in {\mathcal {F}}^n_n\) satisfies

$$\begin{aligned} F((n-1)\varvec{\cdot }x,y) ~=~ F(x,(n-1)\varvec{\cdot }y) \qquad x,y\in X. \end{aligned}$$

Thus, we conclude that Kiss and Somlai (2019, Theorem 4.8) is an immediate consequence of Kiss and Somlai (2019, Lemma 4.1) and Proposition A.1 above.

The following result is the key for the alternative proof of Corollary 1.7.

Proposition A.2

Let \(F \in {\mathcal {F}}^n_{n-1}\). The following assertions are equivalent.

  1. (i)

    F is reducible to an associative and quasitrivial operation \(G:X^2 \rightarrow X\).

  2. (ii)

    F is reducible to an associative and idempotent operation \(G:X^2 \rightarrow X\).

  3. (iii)

    \(F((n-1)\varvec{\cdot }x,y) = F(x,(n-1)\varvec{\cdot }y)\) for any \(x,y\in X\).

  4. (iv)

    \(|E_F|\le 1\).

Proof

The equivalence (i) \(\Leftrightarrow \) (ii) and the implication (iii) \(\Rightarrow \) (iv) are straightforward. Also, the equivalence (ii) \(\Leftrightarrow \) (iii) follows from Proposition A.1. Now, let us show that (iv) implies (iii). So, suppose that \(|E_F|\le 1\) and suppose to the contrary that there exist \(x,y\in X\) with \(x\ne y\) such that \(F((n-1)\varvec{\cdot }x,y) \ne F(x,(n-1)\varvec{\cdot }y)\). We have two cases to consider. If \(F((n-1)\varvec{\cdot }x,y)=y\) and \(F(x,(n-1)\varvec{\cdot }y)=x\), then by Lemma 2.5 we have that \(x,y\in E_F\), which contradicts our assumption on \(E_F\). Otherwise, if \(F((n-1)\varvec{\cdot }x,y)=x\) and \(F(x,(n-1)\varvec{\cdot }y)=y\), then we have

$$\begin{aligned} x ~= & {} ~ F((n-1)\varvec{\cdot }x,y) ~=~ F((n-1)\varvec{\cdot }x,F(n\varvec{\cdot }y))\\ ~= & {} ~ F(F((n-1)\varvec{\cdot }x,y),(n-1)\varvec{\cdot }y) ~=~ F(x,(n-1)\varvec{\cdot }y) ~=~y, \end{aligned}$$

which contradicts the fact that \(x\ne y\). \(\square \)

Proof of Corollary 1.7

This follows from Proposition A.2 and Dudek and Mukhin (2006, Lemma 1). \(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Couceiro, M., Devillet, J., Marichal, JL. et al. Reducibility of n-ary semigroups: from quasitriviality towards idempotency. Beitr Algebra Geom 63, 149–166 (2022). https://doi.org/10.1007/s13366-020-00551-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s13366-020-00551-2

Keywords

Mathematics Subject Classification

Navigation