First, it is simply incorrect to assert, as Professor Tushnet does, that a ’socially constructed’ meaning of speech leads to ‘madness’ or is inconsistent with the approach taken to the First Amendment by the Courts. The US Supreme Court, in its cross burning jurisprudence, had great regard to the particular socially constructed meanings attached to cross burning in the US (see Virginia v. Black, particularly Justice Thomas’ moving dissent). Indeed, much non-representational art (such as flags, logos and images) can communicate political ideas. The art communicates such ideas because of its socially constructed meaning.

Second, contrary to Professor Tushnet’s implication that it is difficult to afford a coherent theoretical justification for the protection of non-representational art, I would suggest that such speech can, without much difficulty, be brought within the traditional rationales for free speech protection:

Political speech is often at its most powerful when it relies upon rich imagery. The protection of such political speech should not be predicated on the evaluation by a judge as to whether or not the images used are ‘representational’ or ‘non-representational’. Indeed, the very images by which a political movement defines itself may be ‘non-representational’ – should such images be left unprotected?

Moreover, if we see the First Amendment as protecting the equality or autonomy of speakers, then there is no logical ground for a distinction between representational and non-representational art. The mere fact that an visual expression does not ‘look like something’ does not alter the fact that repressing that expression limits the communication rights of the non-representational artist and sends a clear message that he or she is not a full member of the political community.