ABSTRACT
The reader might suspect that part of what’s behind the just-rehearsed failure to underwrite reference by causal relations (plus other naturalistically acceptable resources) is something similar to the elusivity of truth (Part II § 6). The suspicion is correct: Reference cannot be defined in terms of underlying causal relations, nor can causal procedures of any sort be used to augment or restrict our notion of reference, just as truth cannot be characterized in terms of whatever truth-gathering methods are available at a given time. There are versions of Putnam modals about reference that we intuitively find true: “There might be instances of gold we could never causally interact with no matter how much we perfect our procedures” or (for any p) “what causal process p picks out might not be (just) gold,”are examples. These can be used to show points about reference analogous to points I’ve made earlier (using Putnam modals) about truth.
