Some conditions implying that an infinite group is Abelian (Q1284632): Difference between revisions

If \(\phi(x_1,\dots,x_n)\) is some word in \(n\) variables then one can define \(G\) to be a \(W^\#\)-group if every infinite set of elements in \(G\) contains a set of \(n\) elements \(\{x_1,\dots,x_n\}\) such that \(\phi(x_{1\phi},\dots,x_{n\phi})=1\) for all permutations \(\phi\) of the set \(\{1,\dots,n\}\). A group \(G\) is said to be a \(W^*\)-group if, given infinite subsets \(X_1,\dots,X_n\) of \(G\), there are elements \(x_i\in X_i\), \(i=1,\dots,n\), such that \(\phi(x_{1\phi},\dots,x_{n\phi})=1\) for all permutations \(\phi\) of the set \(\{1,\dots,n\}\). The authors consider the conditions \(W^\#\) and \(W^*\) for various words \(\phi\) which determine the variety of Abelian groups and also consider \#- and *-conditions for classes related to these words. More precisely, let \(\phi_n(x,y)=[u,v,w][y,x]\) with \(u,v,w\) being taken from the set \(\{x,y,x^{-1},y^{-1}\}\). Here appear 32 possibilities for \([u,v,w]\). The paper contains a list of all these possibilities (p.188). Let \(n\in\{1,\dots,32\}\). The group elements \(x,y\) satisfy (i) \(C_n\) if and only if \(\phi_n(x,y)=1\), (ii) \(D_n\) if and only if \(\phi_n(x^k,y^l)=\phi_n(y^l,x^k)=1\) for all \(k,l\in\{1,-1\}\). Let \(G\) be a group. The authors consider the following classes of groups. (1) \(G\in C_n^\#\) if and only if every infinite set of elements in \(G\) contains \(x,y\) such that \(\phi_n(x,y)=\phi_n(y,x)=1\). (2) \(G\in C_n^*\) if and only if, for any two infinite subsets \(X,Y\) of \(G\) there are \(x\in X\) and \(y\in Y\) such that \(\phi_n(_x,y)=\phi_n(y,x)=1\). (3) \(G\in D_n^\#\) if and only if every infinite set of elements in \(G\) contains \(x,y\) satisfying \(D_n\). (4) \(G\in D_n^*\) if and only if, for any two infinite subsets \(X,Y\) of \(G\) there are \(x\in X\) and \(y\in Y\) such that \(x,y\) satisfies \(D_n\). The main results of this paper are the following. Proposition 2.1. Let \(S=\{17,18,19,21,22,29\}\). If \(n\in S\) and the ordered pair \((x,y)\) satisfies \(C_n\), then \([x,y]=1\). Furthermore, any infinite \(C_n^\#\)-group is central-by-finite and any infinite \(C_n^*\)-group is Abelian. Proposition 2.2. Let \(S=\{23,\dots,31,32\}\). If \(n\in S\) and the elements \(x,y\) satisfy \(D_n\), then \([x,y]=1\), but \(C_n\) alone does not imply \([x,y]=1\). Furthermore, any infinite \(D_n^\#\)-group is central-by-finite, and any infinite \(D_n^*\)-group is Abelian. Proposition 2.3. Let \(S=\{1,2,3,5,6,7,10,12,13,15,16\}\). For \(n\in S\) condition \(D_n\) is equivalent to \(D_1\). Furthermore, \(D_1\) does not imply \([x,y]=1\). Proposition 2.4. Conditions \(D_8\), \(D_9\) and \(D_{11}\) are equivalent and they do not imply \([x,y]=1\). Proposition 2.5. Between the nonabelian \(D\)-conditions \(D_1\), \(D_4\), \(D_8\), \(D_{14}\), \(D_{20}\) and \(D_{30}\) we have exactly the following implications: \(D_{30}\Rightarrow D_4\Leftarrow D_1\Rightarrow D_8\Rightarrow D_{14}\Leftarrow D_{20}\). Theorem 3.6. Let \(G\) be an infinite soluble-by-finite \(C_n^*\)-group, \(n\in\{1,\dots,32\}\). Then \(G\) is Abelian. Theorem 3.7. Let \(G\) be an infinite \(C_n^*\)-group, \(n\in\{1,\dots,32\}\). If \(FC(G)\) is infinite then \(G\) is Abelian. Theorem 4.10. An infinite \(C_{14}^*\)-group is Abelian. Corollary 5.6. For any \(n\in\{1,\dots,32\}\) an infinite \(D_n^*\)-group is Abelian. Theorem 6.7. An infinite \(C_n^*\)-group, \(n=1,2,3,5,6,13\), is Abelian.