To: Walter F.
From: Geoffrey Klempner
Subject: Significance of Quine's ontological relativity
Date: 9 February 2005 14:24

Dear Walter,

Thank you for your email of 30 January, with your third essay for the Philosophy of Language program, in response to the question, 'What is the significance of Quine's doctrine of 'Ontological Relativity'? How does it contrast with 'linguistic relativism'?'

You should have received your Certificate for the Moral Philosophy program by now. Please let me know if it hasn't yet arrived.

This is an interesting and perceptive essay. I didn't recall your telling me that you worked for ten years with the US Public Health Service on Indian Reservations. I agree with your point that Whorf's conjectures concerning the concepts possessed, or not possessed by the Hopi, 'would probably be difficult to distinguish from native scepticism about the motives of anthropologists'

Quine's famous thesis of the indeterminacy of translation - the claim that there is no such thing as a uniquely correct translation manual - can be interpreted, as Quine himself admits, in two distinct ways. The underdetermination of meaning, which Donald Davidson has emphasised in his writings, ultimately rests on impossibility, in principle, of factoring out the contribution made by beliefs and desires in explaining an agent's actions. The very same bodily movements can, in principle, be produced by different pairs of beliefs and desires. This inability to identify desire independently of belief or vice versa infects the very process of writing a translation manual, because our only handle on what the 'natives' mean by a certain utterance is our observation the circumstances in which we observe them using that utterance, or their response when we use it: in both cases, prior assumptions are necessary concerning their beliefs and desires.

Davidson calls this process 'radical translation'. Every radical translation is a 'theory' in which multiple simultaneous assignments of beliefs, desires and meanings are made. But, then, as a last finesse, Davidson recognizes that it is not only the anthropologist who finds themself in this predicament. Whenever two human beings communicate with one another, an implicit process of radical translation is necessarily involved. We are not normally aware of this, only because we already 'speak the same language', i.e. utter the same noises. This 'homophonic' translation (the same word for the same word) can break down, however. An example might be the conversation between an analyst and his patient, when the analyst begins to suspect that the patient is saying something very different from what he thought he was saying.

Quine's ontological relativity also involves translation manuals, but at a deeper level. In the 'gavagai' case, for example, where the native utters 'gavagai' and you are unsure whether to translate this as 'rabbit' or 'rabbit part' etc. we are envisaging a permutation of the logical apparatus for individuation and identity. So, for example, the sign 'equals' or the utterance 'is the same' which we use to express the observation, e.g. 'That's the same rabbit I saw a minute ago', might, in a different translation manual, mean, 'Similar', as in, 'That's a similar momentary rabbit stage to the momentary rabbit stage I saw a minute ago.'

Whenever we identify the meaning of a sign, e.g. "'hubshcub' means 'the same" we are doing this relative to our own language. There is no way of comparing a language 'to the world', we can only map one language onto another. That's Quine's fundamental point.

Quine's reliance 'on the old standbys Behaviourism and Pragmatism', as you put it, may look suspect because it seems to imply that there is something concrete which we can identify below the level of meaning - bodily movements, sounds, scribbles - which constitutes the real 'hard stuff' or bedrock of semantics. Yet workers in the field will testify to the fact that it is notoriously difficult to identify a purely physical 'bodily movement' or 'guttural sound' independently of hypotheses concerning the agent's beliefs, desires and intentions. Every observation, at whatever level, is infected by theory. In his metaphor of 'the net', Quine seems to recognize this.

The idea of linguistic relativism is also tackled by Donald Davidson, in his well known paper, 'On the Very Idea of a Conceptual Scheme'. In the program, I try to go one step further than Davidson - who argues in his paper that the 'very idea' of the existence of a different, incompatible conceptual scheme ultimately has no real content - when I suggest that there is something which the worry about linguistic relativism *correctly* latches onto: this is the possibility that our own concepts might themselves be 'wrong' in some way, i.e. embody some kind of internal incoherence. This involves a particularly unsettling kind of scepticism, where the problem is not just that our most deeply held beliefs might be wrong, but rather the recognition of the possibility that we may think our words make sense when in reality they are nonsense.

It is difficult, as I explain, to give examples. Perhaps philosophy itself and in particular the language of philosophy is the area where there is the greatest danger of coining meaningless concepts, as many would agree!

All the best,

Geoffrey