Intentionality

2. The nature of intentionality

Despite the interest in intentionality in twentieth-century philosophy, there is still controversy about how to characterize it. All writers agree that intentionality is the directedness of the mind upon something, or the aboutness of mental states, but the disagreements start when we try to explain these ideas in more detail.

To begin with, calling intentionality ’directedness’ makes it look as if it is a relation between the mind and the thing on which the mind is directed. After all, if A is directed on B, then A and B are related – if an arrow is directed on a target, the arrow and the target are related. But if the arrow is genuinely related to the target, then the arrow and the target must exist. And similarly with other relations: if Antony kisses Cleopatra, Antony and Cleopatra must exist. But this is not so with intentionality, as Brentano observed (Brentano [1874] 1973: appendix). I can desire to possess a phoenix without there being any such thing. So what am I related to when I am in an intentional state?

One reaction is to postulate that intentional relations are relations to intentional objects. A phoenix is not the material object of my desire, but it is the desire’s intentional object. However, it is not obvious that this really solves any problems. For what are intentional objects? Are they real objects? Brentano and Husserl both thought so: intentional objects are just ordinary objects. But if this is so, then what does it mean to say that intentional objects need not exist? Alexius Meinong, on the other hand, thought that intentional objects have a different kind of existence from real material objects. But this seems to misrepresent intentionality: if I want a phoenix, I want a real phoenix, with wings and feathers – not something with a different kind of existence. In any case, the idea that there are different kinds of existence is of dubious coherence. So whatever we say about intentional objects, they do not offer a satisfactory explanation of intentionality (see Scruton |1970–1).

A second important difference between intentionality and other relations is that with other relations, the way you describe the relata does not affect whether the relation holds. But with intentionality this is not so: you can believe that George Orwell wrote Animal Farm without believing that Eric Blair wrote Animal Farm, simply because you do not know that Orwell is Blair. But since Orwell is Blair, then your belief surely relates you to the same thing – so how can the obtaining of the relation (belief) depend on how the thing is described?

For these reasons, it seems impossible to regard intentionality as a relation at all. One way of avoiding these difficulties is to distinguish, as Brentano did not, between the intentional object of a state and its intentional content. Intentional content (like Husserl’s noema or Frege’s thought) is what makes it possible for a mental state to be directed on an object. Thus understood, intentional contents are not representations. Rather, they are what constitute something’s being a representation: it is in virtue of the fact that a mental state has an intentional content that it represents what it does. It is in virtue of the fact that my belief that pigs fly involves some relation between me and an intentional content – the proposition or Fregean thought that pigs fly – that it represents what it does. This is what is meant by saying that intentional states, or propositional attitudes, are relations to propositions or contents (see Propositional attitudes).

Although beliefs, desires and other intentional states are sometimes described in this way – as relations to propositions or contents – this idea should be sharply distinguished from the idea, just discussed, that intentional directedness is a relation. The intentional content expressed by the sentence ‘Pigs fly’ is not what my belief that pigs fly is directed on: the belief is directed on pigs and flying. Some have thought that all intentional directedness can ultimately be reformulated in terms of relations to intentional contents: an intentional state is directed on an object X in virtue of the fact that it is a relation to an intentional content concerning X. However, this thesis has difficulty dealing with certain intentional phenomena, most notably the phenomenon of loving: no one has given a satisfactory reformulation of the notion ’X loves Y’ in terms of X’s relations to intentional contents.

There is much controversy about exactly what intentional contents are and how we should individuate them (see Salmon and Soames 1988). Some philosophers attempt to clarify (or even sidestep) these ontological and epistemological difficulties by adopting what Quine calls ’semantic ascent’: they examine sentences which report intentionality rather than intentionality itself. A distinctive feature of many sentences reporting intentional states is that their constituent words do not play their normal referential role. Part of what this means is that the apparently uncontroversial logical principles of existential generalization (from Fa infer (∃x)Fx) and Leibniz’s Law (from Fa and a=b infer Fb) fail to apply to all sentences reporting intentionality. For example, from ‘I want a phoenix’ we cannot infer that there exists a phoenix that I want; and from ‘Vladimir believes that Orwell wrote Animal Farm’ and ‘Orwell is Blair’ we cannot infer that ‘Vladimir believes that Blair wrote Animal Farm’ (see Propositional attitude statements).

Contexts where these two principles fail to hold are known as ’non-extensional’ contexts – their semantic properties depend on more than just the extensions of the words they contain. They are also called ’intensional’ contexts, or contexts which exhibit intensionality (see Intensionality). The connection between intensionality and intentionality is not merely typographical: the failure of existential generalization in intensional contexts is the logical or linguistic analogue of the fact that intentional states can be about things which do not exist. And the failure of Leibniz’s Law is the logical or linguistic analogue of the fact that the obtaining of an intentional relation depends on the way the relata are characterized.

However, the notion of intensionality must be distinguished from the notion of intentionality, not least because there are intensional contexts which are nothing to do with the direction of the mind on an object. Prominent among these are modal contexts: for example, from ‘Necessarily, Orwell is Orwell’ and ‘Orwell is the author of Animal Farm’ we cannot infer ‘Necessarily, Orwell is the author of Animal Farm’. Other concepts which can create intensional contexts are the concepts of probability, explanation and dispositionality. But it is very controversial to hold that these concepts have anything to do with intentionality.

Another (more controversial) reason for distinguishing between intensionality and intentionality is that intentionality can be reported in sentences which are extensional. Some philosophers have argued that the context ’x sees y’ is like this. Seeing seems to be a paradigm case of the direction of the mind on an object. But if Vladimir sees Orwell, then there is someone whom he sees; moreover, if Vladimir sees Orwell, then surely he also sees Blair, and he also sees the author of Animal Farm, and so on. So although seeing is intentional, ’x sees y’ seems to be an extensional context.