Hinton (now of Google Brain) and DS Touretzsky (yeah, the anti-scientology guy) published a thing in 1985 claiming inferential symbol production a la lisp from connectionist (neural) models (<a href="http://www.cs.toronto.edu/~hinton/absps/symbolsIJCAI.pdf" rel="nofollow">http://www.cs.toronto.edu/~hinton/absps/symbolsIJCAI.pdf</a>). It only had the disadvantage of being a researchy piece of crap.