arXiv++ Combinatorics

It is not hard to write a first order formula which is true for a given graph G but is false for any graph not isomorphic to G. The smallest number $(G) of nested quantifiers in a such formula can serve as a measure for the ``first order complexity'' of G. Here, this parameter is studied for random graphs. We determine it asymptotically when the edge probability p is constant; in fact, D(G) is of order log n then. For very sparse graphs its magnitude is Θ(n). On the other hand, for certain (carefully chosen) values of p the parameter D(G) can drop down to the very slow growing function log^* n, the inverse of the tower-function. The general picture, however, is still a mystery.

We say that a first order formula A distinguishes a graph G from another graph G' if A is true on G and false on G'. Provided G and G' are non-isomorphic, let D(G,G') denote the minimal quantifier rank of a such formula. We prove that, if G and G' have the same order n, then D(G,G')\le(n+3)/2, which is tight up to an additive constant of 1. The analogous questions are considered for directed graphs (more generally, for arbitrary structures with maximum relation arity 2) and for k-uniform hypergraphs. Also, we study defining formulas, where we require that A distinguishes G from any other non-isomorphic G'.

Given a relational structure M on n elements, let D(M) be the minimum quantifier rank of a first order formula identifying M up to isomorphism in the class of n-element structures. The obvious upper bound is D(M)\le n. We show that if the relations in M have arity at most k, then D(M)<(1-\frac{1}{2k})n+k^2-k+2. The coefficient at n, which equals 1-\frac{1}{2k}, is probably not best possible but this is the first known bound having it strictly below 1 (for fixed k). If one is content to have the worse coefficient 1-\frac{1}{2k^2+2}, then one can choose an identifying formula of a very special form: a prenex formula with at most one quantifier alternation. A few other results in this vein are presented.