Indexes, deixis and pronouns

By Ole Togeby

How to describe personal pronouns in a semiotic framework? In Peircean semiotics personal pronouns, such as I and you, are often classified as indexes because they sort of ‘point to’ one of the interlocutors. But the concept of ‘index’ has, by Peirce and others, been defined in many ways that are not always compatible nor consistent. It has caused a lot of confusion in semiotics. Indexes are said in various definitions of the concept to involve four phenomena: pointing, causality, particularity and deixis.

Pointing is in itself not a sign, not even a natural sign (in Grice’s sense); apes don’t react to pointing. Pointing is only part of a sign; it always occurs in a situation in which some other information is accessible for the interpreters, information that make up a necessary part of the sign: (pointing at a bird): That is a bird. Pronouns can be accompanied by pointing, but are not – in any sense – an example of pointing. It is not polite to point at the adressee.

Causality is said to be the ground for dark clouds being an index of coming rain, and footprints being a sign of someone having been there. Causality is always particular, omnipresent and in it self not criterial for something being a sign. Personal pronouns have no connection whatsoever with causality. But many other communicative signs involve causality, e.g. photographs and measuring instruments such as thermometers and weathercocks.

Particularity is often mentioned as characteristic for indexes, and pronouns always refer to individuals, and not to categories the same way nouns do. That is due to the fact that pronouns are ‘shifters’, i.e. shift their reference depending on the situation in which they are uttered. But in all occurrences they also denote categories, but dynamic and functional categories like speaker, addressee, time and place of the actual utterance.

Deixis is definitely a feature of personal pronouns. They mention and identify what sin talked about (the arguments of a predication), but do not describe the state of affairs these arguments take part in (the predicate of a predication). And they can only mention and identify persons, times and places in an actual speech situation.

Personal pronouns have to be defined in other ways than as indexes, and that requires a definition of the concept of a ‘communica- tive sign’ that involve not only (according to Peirce’s definition) “somebody for whom something stands for something”, but also the one who produces and utters the form of the sign as well as the persons who interpret it, and the situation in which the sign is uttered as a means by which the interlocutors share thoughts. Personal pronouns are as a subclass of signs both conventional and situational, and will always in addition to the conventional, non-natural meaning have a unintended natural meaning as a symptom that reveals information about the interlocutors and their relations. Such non- natural and unintended meaning of signs are interpreted as symptoms, and not as intended relevant messages. A suitable definition of personal pronouns has consequently to take into account the distinction between different types of information: explicated (implied) information, focussed information, presupposed information and implicated information.