In this dissertation, I argue that the view developed by Nelson Goodman, known as "irrealism," is a coherent position distinct from the standard conceptions of both realism and idealism. However, a careful examination of his central argument for irrealism shows that, at best, there is much work left to do in defending the position.
Special effort is made to clarify Goodman's use of such terms as 'version', 'world', and 'rightness', while also taking seriously several of Goodman's crucial claims, including the claims that there are many actual worlds and that we "make" such worlds. On the interpretation developed (Chapters 1 and 2), irrealism is distinct from realism and idealism. On the one hand, it rejects the notion of a unique independent world along with the idea that one system of representation is preeminent and all inclusive. On the other hand, it gives no neutral world a role in "underpinning" representational systems (or versions), but actual worlds are belief-constraining; since versions must meet requirements of rightness, a complete relativism to representational system is avoided.
Goodman's central argument rests on the following crucial assumptions: (1) There are alternative ostensibly conflicting versions that present good and equal claims to truth. (2) There is no unique independent world. (3) While some cases of alternative ostensibly conflicting versions may be reconciled by relegation to different parts or subclasses of the same world or by relativization to a representational system or by relativization to a frame of reference, there exist some cases (at least one such case) which cannot be so reconciled and there are no other ways to reconcile them. I discuss these assumptions in detail (Chapters 3-5). Issues arise for Goodman's support of each assumption. Since irrealism is a coherent position, but Goodman's support for it inadequate, I conclude that much work remains in defending the position.
