This is from an interview with Susan Haack (Haack.interview–warning pdf). Aside from her Philosophy of Logics textbook, I’m completely ignorant of her work, for the most part.
CB: Could you tell us more about Innocent Realism?
SH: It is, I hope, a metaphysical position that can accommodate the most robust realist intuitions to the most sophisticated anti-realist objections. The main ideas are something like this. The world — the one, real world — is independent of how we believe it to be. In saying this, obviously, the Innocent Realist repudiates both the irrealist thesis that there is no real world, and the pluralist thesis that there are many. However, she of course allows that human beings intervene in the world, and that we, and our physical and mental activities, are part of the world. The one, real world, in other words, is heterogeneous: there are, besides natural things and events, human artifacts of every kind, social institutions, and the theories, depictions, and imaginative constructions of scientists, artists, poets, novelists. etc..Adapting an idea from Peirce (who was in turn adapting an idea from Duns Scotus), the Innocent Realist construes “real” as meaning “independent of how you, or I, or anyone believes it to be”; and as contrasting with “fictional, a figment, imaginary.” Scientific theories are real; and so are works of fiction. But the explanations scientists imagine, when they are successful, are true, and the laws they imagine real; while fictional characters and events are precisely not real, but imaginary. Though very fallibly and imperfectly, we humans are able to know something of how the world is. This is possible only because we have sense organs able to detect information about particular things around us, and the intellectual capacity to make generalizations about them; and because the things around us are of kinds and subject to laws.
We describe the world, sometimes truly, sometimes falsely. Whether a synthetic description is true or is false depends on what it says (which is a matter of human convention) and on how the things in the world it describes are. There are many different true descriptions of the world, in different vocabularies. All these many different truths must somehow fit together: there can’t be rival, incompatible truths or “knowledges.” But this doesn’t mean that all the truths about the world must fit together by being reducible to a privileged class of truths in a privileged vocabulary; I see the truths of the social sciences as “fitting together” with the truths of the natural sciences more in the way a road map can be superimposed on a contour map of the same territory.
Here’s a pdf of an interesting article by Haack, “Six signs of scientism.”
Perverse Egalitarianism 
Without a reference to Cthulhu she’ll never catch on. But seriously, the line that interests me is “though very fallibly and imperfectly, we humans are able to know something of how the world is.” Echoes of the neo-Platonic ‘through a glass darkly’ idea of course. Which is at least an advance on the thesis that we are constitutionally unable to know the world as it is in itself, in its silliest form the contention that since subject can never actually become object we can never know object fully, as if that is what is meant by knowledge or truth. This thesis and Haack’s apparently tempered version of it run into problems when faced with the possibility of truth, a possibility which can no more be ontologically ruled out than falsehood. Pete Wolfendale rightly talked about the problem of falsehood for realism last year but truth is also a problem for it, as it cannot really cope with the occurance of a truthful insight or statement. Which is why I think the idea that we are constitutionally unable to know reality as it is in itself reverses from realism into the most subjective-idealist apriorism. As Hegel would say, there’s no labour of the concept there, no experience, but rather an abdication of it.
Well, my realism makes use of Cthulhu, but I spell it with a “K.”
I’m somewhat intrigued by Haack’s innocent realism, though as of yet, I can’t really tell what makes it innocent, or, different from well, realism. Regardless, I’m intrigued. I don’t know enough about Haack’s position to address your concern, which I think is completely legitimate, but I’m not sure it’s such a problem for her. Consider this passage from Defending Science (which gels with what she says in the interview):
Pragmatism always tended to leave me cold, but this watered down (?) middle of the road version is interesting. I’m hoping to be able to carve out some time to read some more of this stuff.
“I can’t really tell what makes it innocent, or, different from well, realism. “
I am interested in her choice of adjectives. “Innocent” as opposed to, say, disenchanted? I’m guessing that this is the realism formerly known as “naive realism,” and this because, as you point out, it is hard to tell just how robust this version is; from her description here, it reads very much like a kind of “common-sensism,” a la the later G.E. Moore, maybe. Still, I’m all for drawing a line in the sand against scientism, and I did like some of her Logic book– her resistance to the trendiness of (what she calls) “deviant” logics (I’m more open to these than she, but I still like her feistiness).
Her remarks on psychology & logic in the interview are good.
About “common-sensism” ( turning everything into isms is really ugly but it seems one has to get over it … ).
I wonder if anyone has made the connection to a wisdom-of-the-crowd style argument, which is basically about finding an expectation value among a broad range of opinions which are considered independent of each other. Common-sensism may be reflective in the twofold sense that 1) I don’t expect my view deviating strongly from that of all others and 2) there are no two large distinct clusters of opinions: dividing the statistical data into two equally sized sets I get the same result [1], no matter how the elements are distributed among the subsets of the partition.
Of course Platon would argue that the herd is still sitting in caves and sharing the experience of a fake world and he would be beaten up by the crowd for sure. How can unnecessary cruelty towards him being avoided?
[1] In order to avoid problems with arbitrary subdivisions one may allow error bounds at each division s.t. after a certain number of divisions the deviation relative to the original distribution can become arbitrary large and clusters may in fact emerge.
To add something, prompted by the Prof who has just contended that Kant’s Ding-an-sich is ‘uncontroversial’. No it precisely is not uncontroversial. What does he think philosophy was doing between 1790 and 1910, that glaring lacuna in his reading of its history?
The Innocent or Naive Realist cannot repudiate either the irrealist thesis that there is no real world, nor the pluralist thesis that there are many real worlds.
That’s because naive realism is a special case of metaphysical pluralism, where the world is held fixed in time in flat, Euclidean space. This follows from the metaphysical restrictions of Plato’s empirical derivation of Platonic Realism of particulars at the Republic 4:436b-437a.
Aristotelian Realism (Met. G) starts with the analytical version of Plato’s result (the Principle of Non-contradiction). Anything and everything with that most simple logic, and discrete objective identity drops into this metaphysical market basket.
Modern Realism simply tries to organize the disparate ontological existents in that basket.