The Origins of Grammar

New Directions in Cognitive Linguistics, Cardiff University, 27-30 August, 2007

PhD Dissertation, University of East London, 2007

Grammar is more than just order and hierarchy; it is a way of expressing complex multidimensional schemas in one dimension. The need to communicate these schemas is the concern of language, but how they are communicated is the concern of grammar. Because grammar does not necessarily rely on the preexistence of language, it is possible for the elements of grammar to be prototyped as features of other mental systems before language appears. These elements can then be exapted as needed for language. So the genesis of language and the genesis of grammar do not necessarily need to be considered as a single process.

13th International Cognitive Linguistics Conference, Northumbria University, Newcastle, 20-25 July, 2015

Pronouns seem to express the components of communication (the sender, receiver and referent) in reduced terms, acting as placeholders for more complex noun meanings. For von Humboldt (1836), pronouns appeared early in the genesis of grammar, allowing the expression of communicator roles before personal nominalisation began. They appeared in the order of first person (the sender), second person (the receiver) and third person (the content of the utterance). For Heine and Kuteva (2007), pronouns emerged from noun usage as desemanticised and decategorised placemarkers for nouns. The three persons probably emerged together-and in quite different ways in different languages. Pronouns are devalued in current grammars. For Hurford (1994), "A pronoun is typically a little word that stands in place of a noun phrase". It is grammatically more constrained than a noun phrase, and therefore simpler; and the different natures of the three persons is trivial. Hudson (1998), using a very different approach to grammar, nonetheless largely agrees with Hurford. Evans and Green (2006), give only two pages on the role of pronouns in cognitive linguistics, and largely agree with Hurford and Hudson about their nature. Benveniste (1970) sees first and second person pronouns as different from third person: "I" and "you" reflect roles in the communicative act, the third person is just part of the signal. Benveniste therefore describes the third person as a "non-person". Van Hoek (2007) builds a similar model from a cognitive linguistics viewpoint, discussing the "on-stage" and "offstage" nature of pronouns. She sees the first and second person as representative of the actual sender and receiver, who are able to share a view of the utterance as a performance which may include themselves. The third person, in contrast, is "held at arm's length". This paper uses van Hoek (2007) and Edwardes (2014) to explore the origins of pronouns. It proposes that pronouns come from the exchange of social models, and they represent a meta-awareness of the communication act. They require hierarchical cognitive modelling: the communication model (sender-signal-receiver) is viewed by the sender and receiver as if they are outside the communication, and their communicative roles are then represented within the signal (for example, as actor-action-patient). I model myself as the sender or receiver, and then model the sender or receiver modelling themself within the signal-our models of me and you can also model me and you. Pronouns allow the different modelling levels to be collapsed into a single conceptual whole of selfhood. The paper discusses the capacity to model selves within modelled selves, and its role within the genesis of language. It shows that pronouns represent a sophisticated form of cognition which probably came relatively late in language genesis.

1997

Combining insights from work on Relevance Theory, contextualizations cues, 'emergent grammar', and grammaticalization, this paper argues that inference is the basis of all communication, and of all aspects of linguistic communication, and that grammar is a set of evolved social/cultural conventions that aid the inferential process by constraining that process. Morphological forms develop as speakers repeatedly use lexical items with conceptual information in contexts where the conceptual information is not necessary to the main information of the utterance, and the form is used simply to constrain the hearer's search for the relevance of the utterance, such as making the relationships between various parts of the total message more explicit or situating the message temporally. A fully grammaticalized form is said to contain only procedural information, but all grammaticalized forms develop from full lexical forms used for marking procedural information, and this is what allows them to become grammaticalized. Syntactic patterns become grammaticalized as implicatures and associated aspects of particular repeated patterns of use (such as same subject patterns in coordination) become fixed. These then can aid in referent identification and other aspects of interpretation. The development is not teleological, but evolutionary. Languages then come to differ in terms of what aspects of the interpretation are obligatorily constrained and what are not, and in terms of the degree to which interpretation of a particular functional domain is constrained. Languages also differ in the way they constrain interpretation of a particular functional domain.

The Role of Perception and Action in Memory, Language, and Thinking, 2005

Page 109. 4 The Emergence of Grammar from Perspective Brian MacWhinney Carnegie Mellon University Successful communication rests not just on shared knowledge and reference (Clark and Marshall, 1981), but also on a process of mutual perspective taking. ...

International Journal of American Linguistics, 1990

UNlvERslry oF CALTFoRn_rA. SANT.\ BARBARA l. Introduction. The function of pronouns might be assumed to be universal among languages. The use and interpretation of particular forms within a language, however, depend significantly on the full repertoire of referential devices with which they contrast. In English, thirdperson pronouns alternate with full lexical noun phrases and reflexives. In many other languages, free pronouns appear only for contrast or emphasis, while bound pronouns appear obligatorily in every clause, whether additional nominals are present or not. Many languages lack regular third-person pronouns altogether; reference is indicated by means of full noun phrases, by demonstratives, or not at all. The general third-person pronouns of Central Pomo contrast not only with full noun phrases, but also with a set of empathetic pronouns and with the absence of any referential device whatsoeyer. When these alternatives are compared in natural speech, it becomes clear that their selection and interpretation are highly systematic, governed by factors of cognition and interaction. Work by Wallace Chafe is useful in understanding cognitive aspects of referential choice. He relates three components of thinking: consciousness, information, and the self (Chafe 1980:12): Consciousness can be rcgarded as the activation of restricted pieces from the vast array of available information. Consciousness is highly limited ir capacity. It is also highly lidited in the time it can dwell on one piece of information. lt isjerk-r;1hat is, it rests on one thing and then another and does not scan information continuously. And it has a focus and a periphery; that is, there is a maximum degree of activalion and lesser degrees. The self deploys this mechadsm over cente$ ofintercst which are related to its needs and goals. In this deployment, it presumably follows paths which are determined by associations between one focus and the next, and by schemas-already known parhs which have been established for the same o. similar areas of informatio[ in the thinker's past experience. Referential choice in Central Pomo, as in all languages, reflects the activation state of information within the mind of the speaker as focal, peripheral, or absent. It also reflects the speaker's identification of self.

International Conference on Language, Culture and Mind, Portsmouth, July, 2004

Evolang 5, Leipzig, February, 2005

After our first study in the field of Indo-European personal pronouns , where we focused on their paradigmatic levelling, we now consider internal reconstructions of partial microsystems of first and second person pronouns in all individual branches, projecting them back to the late Indo-European protolanguage. The next step was reconstruction of the primary pronominal protosystem, allowing us both to deduce forms of pronouns in historically attested languages and to understand pronominal-verbal congruence, assumed as the axiomatic starting point.