The Indexical Point of View

On Cognitive Significance and Cognitive Dynamics

Vojislav Bozickovic

• Publisher: Routledge
• Year: 2020
• ISBN:
  9781000206944
• Edition: 1
• Topic: Philosophie
• Subtopic: Sprache in der Philosophie

This book argues that there is a common cognitive mechanism underlying all indexical thoughts, in spite of their seeming diversity.

Indexical thoughts are mental representations, such as beliefs and desires. They represent items from a thinker's point of view or her cognitive perspective. We typically express them by means of sentences containing linguistic expressions such as 'this (F)' or 'that (F)', adverbs like 'here', 'now', and 'today', and the personal pronoun 'I'. While generally agreeing that representing the world from a thinker's cognitive perspective is a key feature of indexical thoughts, philosophers disagree as to whether a thinker's cognitive perspective can be captured and rationalized by semantic content and, if so, what kind of content this is. This book surveys competing views and then advances its own positive account. Ultimately, it argues that a thinker's cognitive perspective - or her indexical point of view - is to be explained in terms of the content that is believed and asserted as the only kind of content that there is which thereby serves as the bearer of cognitive significance.

The Indexical Point of View will be of interest to philosophers of mind and language, linguists, and cognitive scientists.

1 Character, Content, and Cognitive Significance

1.1 Introduction

We have seen that Kaplan tries to account for the cognitive differences between the true a = b and a = a in terms of the difference in the characters when ‘a’ and ‘b’ are indexical expressions of different meaning types, while Perry speaks of their different roles. But Kaplan has noticed that in the case of perception-based demonstratives such as ‘this’ and ‘that’, characters are unfit for this role. Having noticed this, he makes a couple of attempts to account for the cognitive differences between an informative and uninformative use of ‘That = that’, where the two occurrences of the demonstrative ‘that’ refer to the same object, in terms of various accompanying features. In this chapter, I argue that none of these attempts are successful. I also point out that in the case of indexicals such as ‘today’, character is unfit to account for relevant cognitive differences. This should help pave the way for the subsequent discussion that is aimed at discerning the bearers of cognitive significance that capture the subject’s cognitive perspective.

1.2 Character and Cognitive Significance

As the bearer of cognitive significance, each Kaplanian character or Perry’s role is supposed to account for the common nature that different belief states might have. Grasping the character or role of, for example, the indexical ‘today’ by different people or by the same person at different times encapsulates the same way of thinking about the day(s) being referred to.

While contents are represented by functions from possible circumstances to extensions (Carnap’s intensions), characters are represented by functions from possible contexts to contents. Indexicals have context-sensitive (but stable) character: their content varies with context. The notion of cognitive value is then accounted for in terms of the notion of character. When on different days I say, ‘Today is beautiful’, I am said to be in the same cognitive state while believing different things: the two utterances have the same cognitive value or character but different contents (see Kaplan 1989a, XVII).

Like for Frege, cognitive value is of a piece with the semantic determinant of reference. For Frege, a (non-indexical) definite description designates an object in virtue of the descriptive condition that it encapsulates. In understanding the meaning of ‘The evening star’, one grasps its sense that contains a mode of presentation of Venus whereby it designates Venus via a cognitive path that leads to it. Similarly, an indexical such as ‘today’ designates its referent in virtue of the descriptive condition it encapsulates which amounts to its character the present day. In both these cases, the encapsulated descriptive condition is the bearer of cognitive value in that it presents the object in such-and-such a way. But while in the case of such a definite description the object is supposed to be picked out in a context-free way, the referent of an indexical is determined in virtue of its character given the context. Now, Frege’s sense first introduced to represent the cognitive significance of a sign is also taken by him to represent truth conditions (Kaplan 1989a, 501, n. 26). Since there is a pull between these two notions that even Frege felt, Kaplan, as noted, suggests that we should tease out of Frege’s notion of sense two different kinds of meaning: the character or linguistic meaning of an expression and the content or the proposition expressed which is (in the indexical case) Russellian in that it consists solely of objects, properties, or relations. And it is the character of an indexical expression, i.e. a rule that takes us from context to content, that accounts for cognitive as well as semantic significance (given the context). However, the character is not part of the content which represents the truth conditions of an utterance of an indexical sentence. Perry’s original view is very similar. In addition to the relational mode of presentation (or identifying condition for reference, in his later writings), he also speaks of the relative mode of presentation which in the case of the indexical ‘tomorrow’ he glosses by stating that ‘tomorrow’ denotes a day only given a day, i.e. relatively not absolutely. The relative mode of presentation corresponds to what Perry calls the ‘role’ of such a term, i.e. a rule taking us from an occasion of utterance to a certain object (day), that is akin to Kaplan’s character. It is a rule for determining reference by its relation to an element of the context of utterance. What language associates with the indexical word is such a relative mode of presentation (Perry, Postscript to 1977/1993, 27–28).

In this way, what is believed (in the context) is determined by how it is believed. Thereby, in the relevant cases, the Fregean requirement that sense determines reference is met by splitting up the sense, i.e. thought, of the classical Fregean doctrine, which takes it to be the object or content of belief, into two components. Reference becomes the object or content of belief (i.e. Thought in Kaplan’s (1989a, 530) and Perry’s (1977/1993) terminology) and is determined (in the context) as well as presented in a particular way by Sense (Kaplan’s character, Perry’s role).

As noted, Perry adopts (A), i.e. that to the sign there corresponds a definite (though incomplete) sense, and abandons (B), the claim that to the sense belongs a definite referent. So does Kaplan who, to recall, claims that while in Frege’s theory, a given manner of presentation presents the same object to all mankind, character will in general present different objects of thought to different persons or even different objects of thought to the same person at different times (Kaplan 1989a, 530).

What is left of the Fregean sense does not fully determine reference by specification as does the sense of a (non-indexical) definite description. Sense amounts to the descriptive content of an indexical expression that is encapsulated by its character. It plays a part in the determination of reference, i.e. of the proposition expressed, but does not amount to a full-fledged descriptive content. This is what enables Perry to urge that references of singular terms do not depend on Fregean senses or identifying descriptions in the mind of the speaker and that expressions used do not have such senses attached to them by the conventions of language.¹ ‘The beliefs of the speaker need not supply conditions that single out a unique individual. Even if the speaker has such beliefs, the reference is not determined by those beliefs’ (Perry 1988/1993, 227).

1.3 A Problem with Characters: Demonstratives

While urging that characters are the bearers of cognitive significance, Kaplan has noticed that they are too coarse grained to capture the differences in cognitive significance (i.e. psychological states) for all indexicals (and so has Perry in his 1977/1993, 13). He wonders how the differences in cognitive significance are to be explained in the case in which different utterances of the same demonstrative expression (type) refer to the same object, but one does not believe that, as it were, ‘That₁ = that₂’.² The issue concerns perception-based (deictical) demonstratives which are in Kaplan’s view unlike pure indexicals such as ‘today’ and ‘yesterday’ or ‘here’ and ‘now’ (at least in some of their uses) in that they are in need of supplementation which both accomplishes the referring job as well as accounts for the cognitive value.³

Kaplan’s explanation of the informative case of ‘That₁ = that₂’ fails on the same grounds as does tying cognitive significance to character itself. Although the two demonstrative utterances have the same character, Kaplan’s original claim (Kaplan 1989a) was that there is a difference in the speaker’s demonstrations supplementing, respectively, the two utterances of ‘that’, where demonstrations are understood as types which can be defined as narrow psychological states. Their form is “The individual that has appearance A from here now”, where an appearance is something like a picture with a little arrow pointing to the relevant subject (Kaplan 1989a, 526), and it does not seem to be essential to a demonstration that it be mounted by any agent at all (Kaplan 1989a, 525). The notion of demonstration is a theoretical concept, meaning that there are no cases in which an utterance of a demonstrative is not accompanied by a demonstration since “a demonstration may also be opportune and require no special action on the speaker’s part, as when someone shouts ‘Stop that man’ while only one man is rushing toward the door” (Kaplan 1989a, 490). However, our problem is not solved by this move since there are still cases in which the identity of the linguistic character and that of the demonstration are kept fixed, while the cognitive value varies. An illusionist may come up with a trick creating the impression that an object sitting in one spot has been replaced with another one qualitatively identical with it, whereas this is not so. In the process, he may say, ‘That₁ is that₂’, where both utterances of ‘that’ are consecutively taken to refer to the same object, while relying on demonstrations that are of the same type. Similarly, under the same circumstances, one object may unnoticeably be replaced with a similar one in a flash such that it will be informative to the audience to be told ‘That₁ is not that₂’ and so on.⁴

Later on, Kaplan (1989b) takes the demonstration associated with an utterance of a demonstrative as a mere externalization of the speaker’s directing or perceptual intention – the intention aimed at a perceived object which may or may not be the object that the speaker has in mind (Kaplan 1989b, 583; see also Perry 2012, 70f, for the role that this kind of intention plays, as well as Perry 2009, for a discussion of directing intentions in relation to Kaplan; see also Korta and Perry 2011, 41f). It is this intention rather than the demonstration that Kaplan now takes to be the semantically significant supplement to an utterance of a demonstrative. And, in keeping with the view that the semantic path to reference is supplied by the subject’s way of thinking of the reference, this intention is meant to also account for cognitive value.

Should directing intentions be taken as types defined as narrow psychological states, it follows that a directing intention may stay fixed while the cognitive value varies and the foregoing argument concerning the speaker’s demonstration can be re-adjusted to show that the directing intention cannot account for cognitive significance. But Kaplan is somewhat reluctant to see directing intentions as types. He says that ‘[t]he same demonstrative can be repeated, with a distinct directing intention for each repetition of the demonstrative’ (Kaplan 1989b, 586). This is to say that directing intentions are not separable from particular contexts in which they arise. In view of this, Kaplan admits that the ‘cognitive uncertainties of “that₁ is that₂” may no longer be an aspect of meaning’ (Kaplan 1989b, 588; see also Lalor 1997).⁵

While agreeing with Kaplan that cognitive significance is not an aspect of meaning as a property of linguistic types, I disagree with his claim that the speaker’s directing intentions account for the cognitive uncertainties of ‘that₁ is that₂’. In disbelieving that ‘that₁ is that₂’, the subject will be in two different cognitive (psychological) states, as evidenced by the fact that she may at the same time rationally assent to ‘That₁ is F’ and dissent from ‘That₂ is F’, although they both designate the same object. Since each of these cognitive states is coupled with a distinct directing intention, there is no mismatch between these states and the directing intentions as there is between these states and the linguistic meaning or character of the demonstrative ‘that’ in the foregoing utterances which is the same in both cases (as well as between the cognitive states and demonstrations in the sense described). But, in spite of the fact that each of these utterances is accompanied by a distinct directing intention, these intentions do not account for the fact that the subject is in two different cognitive states. As a result of her false belief that there are two different objects in play, and disbelieving that ‘that₁ is that₂’, the subject may aim these intentions at what she falsely believes to be two different objects. But, being just the intentions that are aimed at the perceived object, these intentions do not account for the difference in the subject’s cognitive states. A difference in her cognitive states is rather the result of her false belief that there are two different objects in play.⁶

1.4 A Problem with Characters: Pure Indexicals

Unlike demonstratives, the reference of pure indexicals such as ‘today’ and ‘yesterday’ is fixed without the aid of supplementary features such as demonstrations or directing intentions. The context of utterance takes care of this. If I utter ‘Today is beautiful’ on d, I refer to d. As a function from context to content, the character of ‘today’ ensures this. And unlike demonstratives such as ‘this’ and ‘that’, pure indexicals such as ‘today’ and ‘yesterday’ have fixed characters; they are the present day and the previous day. So, when I assent to ‘Today is a beautiful day’ on d, I am thinking of d as the present day and when I assent to ‘Yesterday was a beautiful day’ on d + 1, I am thinking of d as the previous day and my ways of thinking of d are adjusted with the change of context. But I may get the days wrong in the sense pointed out by Perry (1980/1993, 80). Suppose that Smith, whose watch is an hour fast, accepts ‘Today is my husband’s birthday’. But just before 11, she realizes she got it wrong. It is March 1 and...

Table of contents

• Cover
• Half Title
• Series Page
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Introduction
• 1 Character, Content, and Cognitive Significance
• 2 Other Kinds of Content and Cognitive Significance
• 3 Anti-individualism and Cognitive Perspective
• 4 Cognitive Dynamics, Belief Retention, and Cognitive Significance
• 5 Beliefs and Characters
• 6 Slicing Thoughts
• 7 How Many Modes of Presentation Do We Need?
• 8 Tracking and Reporting
• 9 Conclusion
• Notes
• References
• Index