Here is a stab at summarising a thought I seem to return to often.
At the MMU workshop on placebo, Phil Hutchinson sketched an account of a form of intentionality (in the philosophical sense of aboutness) which could be shared by linguistic humans and non-linguistic animals. Subjects have an intentional directedness towards the world if they take in features of the world in experience under an aspect and conceptually, but not necessarily propositionally, structured, he claimed. In response to my question about what sort of entities such concepts were, Michael Loughlin - spotting an implication - spoke forcefully in defence of (Phil and) animals and against my apparent slight. It is merely a philosophical absurdity, he argued, to deny that a dog knows where its bone is eg., left behind the sofa, or is afraid of thunder. Hence I re-summarised what seems to be a problem for everyone in the debate: whatever one can plausibly smooth over always leaves a lump somewhere under the philosophical rug.
If one adopts a full blown Davidsonian views that thought and language are co-dependent, then Lois’ cat Snufkin is a thoughtless brute. But this leaves the intellectual ‘lump’ that Michael was stressing: it certainly seems that he knows some things and not others. He knows where I sleep and hence where is warm once the stove has gone cold at 3am, for example. He knows that he needs to get to the kitchen to be fed.
But if one takes these examples as the starting place and grants them the obvious interpretation, smoothing over the worry of denying animal thought, then a lump springs up elsewhere. If Snufkin can think then he can think thoughts. But if so, what thoughts does he think? He knows hence thinks that he is fed in a particular room: the kitchen, I might say. But does he know he is fed in the kitchen, ie. the room in which we, his owners, cook? Or does he think of it as, say, the most northerly room in the house? Or (demonstratively) that! room? But if so, what is his understanding of ‘room’? And if not room then place? If so what is the geometry of places he grasps? In general, under what aspects or concepts or Fregean senses does he think any of this? There seems to be no principled way to ascribe intensional distinctions to Snufkin (hence distinguishing between thinking of the room as the kitchen and as the most northerly one) because he cannot help out linguistically. But then the transition from his thinking to his thinking specific thoughts breaks down. He thinks, but we cannot begin to say what he thinks.
A third possibility is to try to outline a primitive form of telic directedness, short of full blown thought (eg the aim of full blown teleosemantics) but richer than mere cause and effect (Dan Hutto pressed this idea on me once.) Such a position might be an explanatory stepping stone to full blown conceptually structured thoughts. One might argue that there must have been some such stage in evolutionary history. But that doesn’t solve the problem of now describing the states of telic directedness. Use of ‘concept’ for example is illicit given its original home in articulating propositional form.
Hence whatever justification there is for irritation towards philosophers denying that animals have thoughts, that doesn’t solve the problem of saying what they have if they do have thoughts.
(My own preference is to think that the ascription of thoughts has its primary application with linguistic creatures whose thoughts have propositional structure for the reason hinted at in Philosophical Investigations §445 and Davidson’s roughly parallel account in his paper ‘What is
present to the mind?’and hence are, of necessity, conceptually articulated. It is a contingent fact that we can self ascribe such states but not that they have the satisfaction conditions they do. This is our vocabulary for discussing intentional states. It is also of some use with some animals. To deny that animals have thoughts would thus be to deny that it had the use it does have. But the usual distinctions that comprise the whole framework do not apply well to animals. We should not, however, boggle over what they are really thinking.)