3
$\begingroup$

I asked this question on math stack exchange, but it didn't get any responses. So, I am asking it here. Suppose we are working in the signature of a single binary operation $*$. We are given a finite set $S$ of identities in that signature, and a single identity $E$ which is fixed in advance. Consider the decision problem of figuring out whether the set $S$ implies $E$. Is it the case that that problem is algorithmically decidable only in the trivial case of $E$ being a reflexive identity of the form $t=t$? Or, are there other identities for which the problem is algorithmically decidable? If there are, can someone give me an example of such an identity?

$\endgroup$
5
  • 1
    $\begingroup$ The order of quantifiers is a bit unclear (to me). Are you asking whether $\exists E. \mathrm{computable}(\lambda S.S\Rightarrow E)$? $\endgroup$ Commented Jan 18 at 15:33
  • 1
    $\begingroup$ See also mathworld.wolfram.com/BirkhoffsTheorem.html $\endgroup$ Commented Jan 18 at 15:51
  • 1
    $\begingroup$ I like this theorem (Jacobson): If a ring satisfies $(\forall x) x^3=x$, then it also satisfies $(\forall x , y) xy=yx$. $\endgroup$ Commented Jan 18 at 16:01
  • 1
    $\begingroup$ @DanielWagner I am asking whether there is a specific identity, like maybe the associative identity or the commutative identity, such that the decision problem for that identity is decidable, aside from identities like $x=x$ $\endgroup$ Commented Jan 18 at 23:54
  • $\begingroup$ As a test case, consider the identity $x=y$. This is implied by $S=\{x=x*y, x*y=y\}$, but it is not implied by $S_1=\{x=x*y\}$ or $S_2=\{x*y=y\}$. My guess is that for any nontrivial identity the problem is undecidable. $\endgroup$ Commented Jan 23 at 15:15

0

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.