
Theoretical stages in the prehistory of grammar

Leonard Rolfe

Abstract

This chapter conjectures that the hierarchical structure of
present-day grammars might be the result of an evolutionary
process. Grammar is taken to be a communicative device
patterned to cater for various communicative intentions such
as asking questions, making statements, and expressing
comments. These intentions are themselves elaborated in the
course of an evolving dialogic system. Each communicative
intention might be thought of as having a corresponding
pattern for its expression. If so, this would lead to a
non-integrated grammatical system. Hence the notion of
'recency dominance' is introduced here, whereby a
newly-emerged pattern becomes dominant and 'reworks' older
patterns into conformity with it. Eight stages of elaboration
are proposed. These are probably not discrete; rather,
language evolution should be viewed as proceeding in a more
mosaic pattern.
The first stage - necessarily without any prior pattern to
build on - concerns getting is motivated by the seeking of
some obligement: termed here 'solicitation'. Granting these
obligements constitutes 'compliance'. Finally, a 'close-out'
indicates the exchange is finished. This sequence constitutes
a 'frame of dialogue' that represents the source of later
grammatical functions, which become necessary to handle
communicative functions as they become more elaborated -
vocative from address; imperative and interrogative moods of
the verb from solicitation; affirmative (and negative) from
compliance (and refusal); and various markers for turn-taking.
This frame also contributes to the provision of first and
second persons of the verb when they are later
grammaticalized
A minor complexification of these abilities yields the next stage, ostension þ pointing at a visible item with one's index
finger. Ostension has three important facets: it is for
another (and is hence situated in the earlier dialogic frame);
it implies the addressee understands what is being pointed
at; and it is oriented on the speaker - that is, it is
'deictic'. Ostension primarily concerns visible items, and
distinguishes between those within and beyond reach, but can
be extended to indicate non-visible phenomena. In this
instance, however, ostension can no longer be contextually
supported, but needs a new form, 'identification', which is
secured by 'naming'. This builds on ostention, but extends it
to constitute a new stage.
Identification can be secured by gestural imitation of shape
or activity. Vocal units may have taken the first step
towards being words by reformulating some of this gestural
inventory, or by iconic representation. The notion of
referential symbolism via words may not have been fully
perceived at their creation, and they may well have arisen and
been worked into the communicative system in an ad hoc
manner. One method of securing referential symbolism is
'thematization' and this is proposed as the next stage.
'Thematic roots' are an organisational principle exploited in
some language systems whereby a set of phonological
alternances around a vowel allow the expression of particular
aspects of the notion contained in the root. Verbs and nouns
are not distinguished, but are implicit in the particular
semantic features expressed by members of a theme. Thematic
clusters constitute a semantic route towards grammar, in that
their semantic features are sorted out in modern languages
into the basic noun/verb distinction and grammaticalized into
syntax.
Topic-comment structures are proposed as the first stage in
the transition toward syntax. Initial topic-comment
structures are asyntactic, since their relations are purely
pragmatically based. These motivate the possibility of
expressing third-person action, and the realization of this
enables a speaker to depict scenes, which itself leads to a
forefronting of implicit case relations, and moves
topic/comment structures toward the realm of 'narrative' which
requires considerable grammatical support for its effective
handling. Narrativity begins to shift the relating of events
from the intersubjective realm of the dialogic participants
towards the objectification of events, yielding an 'epistemic'
patterning of discourse, and motivating syntactic devices that
handle the hitherto implicit features of narrative. This
last, epistemic, stage may be relatively recent, and
characteristic of a level of social organisation that produces
the state [Eds].
Links
Preconditions for the evolution of protolanguages by Merlin Donald.
References
Carstairs-McCarthy, A. (1999) The origins of complex language: An enquiry into the evolutionary beginnings of sentences, syllables and truth. Oxford: Oxford University Press