@ John (post 7)
Well, you can 'prove' anything with contextualism so, yes, my point is conceptual in that regard but I think that it makes sense.
To put it a little more precisely: it seems to me that there are two variants of 'realism' to be recognised and they often become conflated. This point can be demonstrated by quoting Levi from above:
"Nothing is completely present, there is no transcendental signified."
One variant of realism would agree with this statement, one wouldn't. The former, I would argue, has taken on board, knowingly or otherwise, the kind of arguments that Derrida and his associates have been making for the past forty years -- taken on board, understood and moved on. The latter still believe in the 'really real' beneath all the mess, superstition and unreality. This is quite a big difference.
Many of the critics of the sort of realism that Levi and Graham are proposing immediately assume that they are attempting to insert some sort of transcendental signified back into the discussion -- to get 'back' to the 'really real' beneath all the sludge and detritus of sense perception. (I must admit, this was the conclusion I initially and ignorantly jumped to.) This is rather far from the truth!
Perhaps my view is a little skewed coming much more from political theory than philosophy -- in political theory 'realism' has for a very long time been the exclusive province of rock-kickers and table-thumpers, desperate to beat their opponents into submission. Levi, Graham et al. are, thankfully, a bit above that and the difference between this old, rather vulgar, foundationalist realism and what is going on now deserves recognition.
Many 'realists' in political theory talk about going 'back' to realism, the implication being that 'yes, yes, all that stuff about signifiers and whatnot is very interesting and all but lets get back to what we were doing before -- talking about reality'. The new realism isn't going 'back' it is going in its own direction, which I, for one, am pleased about.