The Nature of Grammar, its Role in Language and its Evolutionary Origins

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.

In this dissertation, the continuity of language with other forms of signalling is reviewed. Language as a communicative act has the same structure as nonhuman signalling: the components in both cases are sender, receiver, message and referent. The sender is always me, the first person, and the receiver is always you, the second person. These roles are invariant in all signalling, including language, and there is no need for them to be explicit in the signal. The nonhuman signal has to express only the referent or context - a single unsegmented call will suffice. The specific action to be undertaken by the receiver in the presence of this signal is also implicit, but where the referent is significant only to the sender, the action is significant only to the receiver. A single unsegmented call does both jobs simultaneously.

This view of signalling, however, relies on a disinterested viewpoint. It incorporates the individual views of the sender (the context in which the call is made), the receiver (the reaction the call produces) and the
third party (the effect the call has on the third party). These three approaches to the signal are all available to the disinterested observer, here referred to as the fourth person. Being able to adopt this fourth-person viewpoint is, however, something very unusual in nature, and it may be that only humans can do it.

This dissertation also looks at both the structure and process of grammar. In linguistics these two aspects of grammar are often seen as difficult to reconcile: the structural approach of Formalist linguistics is contrasted, rather than combined, with the process descriptions of Functionalist linguistics, producing a separation of methodologies and even philosophies. The two approaches are complementary, however, and they need to be combined if the origins of grammar are to be fully understood.

Language, unlike most nonhuman signalling, is segmented. Formalist linguistics shows us that there are distinct forms involved in this segmentation (Noun Phrase, Verb Phrase, Prepositional Phrase, etc); but the fact that some roles can contain others creates a recursive hierarchy in language that is missing from nonhuman signalling. The segmented nature of language is dictated by the various forms, and by the recursive capacity that this hierarchy of forms needs. However, language also differs from nonhuman
signalling in that it is multistratal: what is passed in a language message is not a single unambiguous value but a set of interrelated meanings. The meanings involve the relation between sender and receiver, the relation between the message and the coding structure, and the relation between the message and the conceptualised world. This transfer of meaning on three levels is the Functionalist view of language.

An important cognitive difference between humans and nonhumans is the ability to make models of the self. It is shown that this ability is problematic in evolutionary terms. Self-modelling requires the capacity to be dispassionate about the self and see the other party's point of view; but how can accommodating the needs of reproductive rivals become a successful strategy? In this dissertation it is shown that the ability to model others is probably quite ancient, while the ability to model the self can only come about in a cooperative linguistic environment. Yet both self-modelling and other-modelling are deeply implicated in the grammar of language: modelling is the mechanism that powers social calculus, and social calculus is behind the two-argument instigator-action-recipient form which has clear relationships with the three argument instigator-action-recipient-context form of language grammar.

The dissertation proposes that the development of grammar structures is explicable in terms of social calculus, but the transition from internal social calculus to external language is only explicable in terms of a cultural revolution. Enhanced social modelling creates the conditions for advanced social calculus, and the syntax of social calculus corresponds to the form of simple language grammar. However, social calculus alone did not create the environment in which syntactic language appeared; it required a cultural revolution to create the necessary conditions for co-operation and sharing of the social calculus.

This thesis supplied via ROAR to UEL-registered users is protected by copyright and other intellectual property rights, and duplication of any part of the material is not permitted, except for your personal use for the purposes of non-commercial research and private study in electronic or print form. You must obtain permission from the copyright-holder for any other use. Electronic or print copies may not be offered, for sale or otherwise, to anyone. No quotation from the thesis may be published without proper acknowledgement.