No label defined (Q5569200)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3282718
Language Label Description Also known as
default for all languages
No label defined
    English
    No label defined
    scientific article; zbMATH DE number 3282718

      Statements

      0 references
      1967
      0 references
      simple Lie \(p\)-algebras
      0 references
      nilpotent elements
      0 references
      analogy between degenerate Lie algebras of special class and infinite Lie algebras of zero characteristic
      0 references
      The subset \(C^*= \{x\in\mathcal L \mid x\ne 0, (\text{ad}\,x)^2 = 0\} \) of a Lie algebra is closed under the bracket operation, provided the base field has characteristic \(p\ne 2\). (The author slurs over this proviso. Counter-example for \(p=2\) can be given.) For \(p=0\) the subspace spanned by \(C^*\) is an ideal; so assume \(p>0\) henceforth. In a joint paper with \textit{I. R. Shafarevich} [Sov. Math., Dokl. 7, 715--718 (1966); translation from Dokl. Akad. Nauk SSSR 168, 740--742 (1966; Zbl 0158.03803)], the author gave comprehensive construction of four families of simple Lie \(p\)-algebras (called ``Cartan type'', incorporating the earlier constructions of Albert, Block, Frank, Jacobson, Witt and Zassenhaus) distinguished by the property \(C^*\ne\emptyset\) (empty set); and they conjectured that together with the ``classical'' ones of Chevalley and Seligman-Mills these exhaust all the simple ones (at least for \(p>5)\). It was proved by the author [Tr. Mat. Inst. Steklova 64, 79--89 (1961; Zbl 0199.07301), translation in Am. Math. Soc., Transl., II. Ser. 55, 195--206 (1966)] that a simple Lie \(p\)-algebra with \(C^*\ne\emptyset\) must be ``degenerate'' (= ``nonclassical'' = has vanishing trace form). In the present paper the converse, that degeneracy implies \(C^*\ne\emptyset\) (when \(\mathcal L\) is called ``strongly degenerate''), is proved (for \(p>5)\) under the extra hypothesis: There exists a Cartan decomposition \(\mathcal L = \mathcal H + \sum_{\alpha\ne 0}\mathcal L_\alpha\) such that \((\text{ad}\,a)^{p-1}=0\) for some nonzero \(a\in \mathcal L_\alpha\) (for some \(\alpha\), possibly \(\alpha =0)\). This greatly improves Theorem 1 of his announcement ``On the strong degeneracy of simple Lie \(p\)-algebras'' [Sov. Math., Doki. 4, 637--640 (1963); translation from Dokl. Akad. Nauk SSSR 150, 246--250 (1963; Zbl 0129.02002)]. The interest in strong degeneracy stems from the possibility of ``existence of long filtrations'', by which one means the existence of a subalgebra \(\mathcal L_0\) such that the smaller subalgebra \(\mathcal L_2 = \{x\in\mathcal L_0 \mid [x,y] \in \mathcal L_0\), and \([[x,y]z]\in\mathcal L_0\), for all \(y,z\in \mathcal L\}\) is nonzero. This property is crucial for the Cartan type Lie algebras. For \(p>7\) author proves \(C^*0\ne \emptyset\Leftrightarrow \mathcal L_2\ne 0\), and claims that this is true also for \(p =7\). The proof of \(\Leftarrow\) is trivial, whereas that of \(\Rightarrow\) involves the stronger conclusion \(\mathcal L_{(p-5)/2} \ne 0\), where one defines recursively \(\mathcal L_i = \{x\in\mathcal L_{i-1}\mid [x,\mathcal L]\subseteq \mathcal L_{i-1}\}\); for this one needs the technical fact that for suitable \(c\in C^*\), \[(\text{ad}\,c)(\text{ad}\,x_1)(\text{ad}\,x_2)\cdots(\text{ad}\,x_k)(\text{ad}\,c) = 0\] holds for all \(x_1, x_2,\ldots, x_k\), and all \(0\le k\le p-4\). (Proof of this is not given, but is said to follow by a reasoning similar to that utilized by the author in his solution of the restricted Burnside problem.) This paper also contains detailed proof of Theorem 2 of the above-quoted 1963 announcement, and also that of a weaker version of the main result, Theorem 1, of his joint announcement with Shafarevich.
      0 references

      Identifiers