In the swampman argument, Davidson imagines a character named Swampman, who is spontaneously formed from a swamp due to a freak lightning strike. This strike coincidentally arranges a swarm of molecules in such a way that they perfectly replicate the body and brain of Davidson himself.

Davidson concludes that because the mental representations of the swampman aren't causally connected to what they represent (either; directly; or indirectly, say, via a teacher), the swampman cannot mean anything when he speaks.

I understand that direct referentialist believe that that causal connection is important for words to refer. However, I would have thought that the Swampman experiment would have counted as evidence that the causal connection is an absurd requirement for reference! It seems even more of a strech to say, as Davidson does, that this being can't mean anything.

(aside: Davidson doesn't define meaning, but to claim that the Swampman doesn't mean anything is a more ambitious claim than to say that the noises Swampman makes don't refer, since it's a matter of debate that reference is necessary for meaning.)

If I examine myself as I am right now, it would be impossible, via Descartes' cogito, for me to deny that I'm having thoughts (of which Davidson also deprives the Swampman). Moreover, for all intents and purposes, the words that I'm writing now have meaning and refer to things. If I were to soon discover that I was randomly constituted a few minutes before the time of my writing of this question, it would make no difference. It seems that the requirement that my mental representations and so on be causally related to what they represent would be inconsequential as far as I, or anyone I encounter, would be concerned. In light of that, it seems that any claim that reference requires the speaker to stand in the right causal relation to the referenced thing is an arbitrary stipulation.

To use an extreme example to to make the point: in my country, holocaust denial is illegal. It seems absurd to say that Swampman could become a prolific holocaust denier and be immune to punishment.

To quote Davidson:

[Swampman] can't mean what I do by the word 'house', for example, since the sound 'house' it makes was not learned in a context that would give it the right meaning—or any meaning at all. Indeed, I don't see how my replica can be said to mean anything by the sounds it makes, nor to have any thoughts.

From that, it appears that, according to Davidson, we couldn't punish Swampman for prolifically denying the Holocaust. By the same line of reasoning, we couldn't punish Swampman for encouraging a person to commit suicide (which is also a crime in my country) and so on. On my reading of Davidson, we would be as justified in punishing this holocaust denier as we would be in punishing a parrot for making noises that sound like a sentence that denies the holocaust.

The absurdy of that conclusion would seem to demand that we conclude that Swampman can refer to things, mean things, and think about things; and therefore, that standing in the right causal relationship to a referenced thing cannot be a requirement for words and concepts to refer, signify, and so on.

  1. To me, Swampman is evidence that, contra Davidson, a causal relationship between the refering person and the world isn't necessary for reference (and especially not for meaning). Without presupposing Davidson's claim that causal relations are necessary for reference/meaning, how does Swampman help him make that claim?
  2. Are there philosophers who use Swampman to the make the opposite claim that Davidson does?
  • The Swampman was meant as an illustration of externalism about meaning, not a supporting argument for it, and makes "extremely brief appearance in Davidson’s writing" (SEP). Externalism explicitly rejects Cartesian analysis of self-knowledge and what follows from it. But Descartes's cogito is often considered invalid even by non-externalists, and "luminosity" of mental states to their possesor is also controversial.
    – Conifold
    Dec 29, 2023 at 7:03
  • @Conifold We can rephrase my question to ask how it serves as an example/evidence;it seems to do the opposite.The linked cogito answer pertains to the"I"portion of the cogito.The relevant fact here is that if I were randomly constituted and reading that I don't have thoughts(such as the ability to doubt),then I couldn't agree with what I was reading(thereby doubting my ability to doubt)or disagree with it (also doubting).We can all doubt something that we read,and there is no effective difference between us and the Swampmen,so Davidson's example seems to fail.How would he disagree with that?
    – Hal
    Dec 29, 2023 at 15:48
  • You (your Swampman) and Davidson mean the word "thoughts" differently. Davidson himself postulates that Swampman's physical state, behavior and even dispositions are identical, so, of course, there is no "effective difference". But it is not the "effective difference" that makes the difference. Davidson is after God's eye theory of meaning, not phenomenological theory of behavior or mental states. There causal histories make the difference, so Swampman's "thoughts" are not thoughts. Phenomenology, including Swampman's self-reports, is irrelevant by design, so cannot count against the theory.
    – Conifold
    Dec 30, 2023 at 0:34
  • @Confiold yeah. By effective difference, I don't mean functional. I mean irrelevant. If there's no way that we can say his theory is relevant to anything, why do we care? If it's inconsequential in every respect, how could we even say whether it's true or false.
    – Hal
    Dec 30, 2023 at 0:58
  • We certainly care about having a "proper" theory of meaning. And there is nothing unusual about theories interpreting "effectively" indistinguishable scenarios differently or contradicting habitual intuitions, especially when the scenarios generously avail us of the all-knowing perspective. Davidson's contention is that the causal theory, with his modifications, is a good candidate for a number of general reasons he lays out. One does not have to agree with him on that.
    – Conifold
    Dec 30, 2023 at 1:17

1 Answer 1


I think Davidson is talking nonsense. If Swampman is an exact replica, he would have the same memories as Davidson. When Swampman says 'house', the word will for him be just as laden with meaning as it would be for Davidson, since he will have Davidson's memories of all those disappointed nights upon which he has shredded his bingo card after that Doreen friend of his mum's- the one with the blue rinse- has taken the prize pot again.

  • Lol. I agree. But given how much traction the Swampman case received, I figure there must be good replies to that objection. Plus, it seems that Davidson is actually using that example to make a point against the intuition (that I share!) in the answer. How did he manage to do this?
    – Hal
    Dec 28, 2023 at 23:18
  • Sadly, we can't punish Davidson for inciting people using meaningless noises.
    – Scott Rowe
    Dec 29, 2023 at 0:04
  • On externalist theories of meaning, internal states, which are the only identical part, are insufficient to constitute the content of thoughts or memories. Putnam's Twin Earth thought experiment is a famous illustration why. So they will not have identical memories because they are connected to their environment in different ways. Externalism is controversial, but so is internalism, and it is a well-developed and argued position, so I would not say that Davidson is "talking nonsense".
    – Conifold
    Dec 29, 2023 at 7:11
  • 2
    @Conifold you are far more tolerant than I when it comes to nonsense spouted by philosophers. Even if I accept all the claims of externalism, I can hardly see how the newly created swampman could be so dissimilarly connected to his marshy environment that he could not equate the word house with a bingo call, a place of residence, a TV doctor, a sub-division of a school etc. Thanks for the link, btw. Dec 29, 2023 at 8:22
  • @Conifold besides, if an inhabitant of tEarth was transported to Earth and encountered water, they would assume it was twater, which suggests that they bore the concept of twater with them internally as some persisting part of their mental configuration. Dec 29, 2023 at 9:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .