Summary of some responses to Donald Davidson’s omniscient interpreter argument against skepticism


Author: Doctor Terence Rajivan Edward (or 0161__Rajivan, if that helps) Date: 13 Jan 2026


The anti-skeptical argument. Davidson thinks that the words of a person do not mean whatever the person intends them to mean, rather meaning is public and social in this way: whatever meaning a radical interpreter would assign to the words is their meaning. A radical interpreter knows one language already and learns the person’s language from scratch. To make progress with learning, they must interpret charitably Davidson thinks: taking the person to express true beliefs, by the interpreter’s lights, wherever plausible. There is room for disagreement but only against a background of shared belief. But could not this shared background be false? To counter this skepticism, Davidson invites us to imagine an omniscient interpreter. They are possible, he says, and as radical interpreter would interpret the person as having largely true beliefs, by their lights, but since they are omniscient the beliefs are really true.


Foley and Fumerton. In “Davidson’s Theism,” Foley and Fumerton argue that in order for Davidson’s argument to succeed in combatting skepticism, there actually has to be an omniscient interpreter. If such an interpreter exists and interprets you, then your beliefs must be true, given Davidson’s constraints on interpretation. But what if they don’t exist?


Brueckner. Brueckner objects to Foley and Fumerton by arguing that Davidson does not require a world in which both you and the omniscient interpreter exist, to combat skepticism about your beliefs. He argues this assumption will do: “Some possible world W* contains an omniscient interpreter who has perfect knowledge about all possible worlds, including the actual world; thus he believes, among other things, all and only true propositions about the actual world.”


Reynolds. Reynolds objects to Foley and Fumerton roughly by appealing to the idea that for any possible world in which you have certain beliefs but there is no omniscient interpreter, there could be an almost identical possible world except there is an omniscient interpreter. In that second world, this interpreter interprets you as largely sharing their beliefs, as they must according to Davidson, so your beliefs must be largely true. Since your beliefs are the same in the world without an omniscient interpreter and since this first world is almost the same, your beliefs are largely true in this world too.


Vermazen. Davidson is committed to facts about what the omniscient interpreter’s beliefs are, it seems, but Davidson does not believe there are determinate facts about beliefs. (I took this objection from Ludwig’s presentation of it.)


Ludwig. Ludwig objects to Davidson’s assertion that there is nothing absurd about an omniscient interpreter. If a skeptic accepts Davidson’s claims that interpreting another always involves attributing to them a number of beliefs one takes to be true and that a person whose supposed speech cannot be interpreted (or translated) is not actually speaking, then it is open to the skeptic to deny that there could be an omniscient interpreter.


Crane and Svoboda. They argue that, taken literally, an omniscient interpreter seems incoherent because they learn your beliefs through interpretation, but then they are not omniscient. So they use a less literal interpretation according to which an omniscient interpreter knows all hard facts: facts about the physical environment and physical properties of people. Soft facts are facts about meaning, beliefs, and other attitudes. But Crane and Svoboda ask us to consider an omniscient interpreter’s own beliefs in something. Davidson cannot say this is a hard fact, because this would contradict his view that beliefs do not go beyond what is revealed by radical interpretation. But they argue that he cannot say that it is a soft fact either. Take the example of an omniscient interpreter’s belief that a particular cat is on a particular mat. This is caused by the cat’s being on the mat and there is a causal law ensuring this, because the interpreter is omniscient. But the items in a relation of causal law are themselves hard facts. So the omniscient interpreter’s belief cannot be a soft fact about them. (If I did not know Crane, I am not sure I would motivate myself to include this intricate argument, being lazy sometimes with intricacies.)


References
Brueckner, Anthony. 1991. The Omniscient Interpreter Rides Again. Analysis 51(4): 199-205.
Crane, Tim and Svoboda, Vladimir. 2004. Causation, Interpretation and Omniscience: A Note on Davidson’s Epistemology. Organon F 11(2): 117-127.
Davidson, Donald. 1977. The method of truth in metaphysics. Midwest Studies in Philosophy 2 (1):244-254.
Foley, Richard & Fumerton, Richard (1985). Davidson's theism? Philosophical Studies 48 (1): 83 - 89.
Ludwig, Kirk. 1992. Skepticism and interpretation. Philosophy and Phenomenological Research 52 (2):317-339.
Reynolds, Steven L. 1993. Skeptical hypotheses and ‘omniscient’ interpreters. Australasian Journal of Philosophy 71 (2): 184-195.


Also try:
Ward, Andrew. 1989. Skepticism and Davidson's Omniscient Interpreter Argument. Crítica: Revista Hispanoamericana de Filosofía Vol. 21, No. 61: 127-143.


Apologies to anyone omitted. (“Hey, I’m not in this immortal handout.”)