are on the right track) will have a rhythmic organization based on the
emission of timed initiator powerbursts,each burst having a single peak
(Lass 1984:250).It should be noted that these two aspects of language,
the melodic and the rhythmic,are largely left by the wayside when trying
to reconstruct protoforms of language.
If we move on to morphosyntax,we see that it clearly constitutes one
of the specific characteristics of human language,which distinguishes it
from other known forms of animal communication.We can describe
morphosyntax from two points of view:formal and functional-semantic.
From the formal point of view,it appears as a capacity to link and
combine lower-level sequences.In this regard,it cannot be said that
there is anything language-specific about this function since one finds
analogous capacities in many other domains,from gesture to manual
skills to the articulation of speech.The same cannot be said for the
other aspect of morphosyntax,whose fundamental structure is the oper-
ation of predication.But here again,one should avoid getting stuck in
strictly linguistic analyses.In fact,predication,that is,the association
between a subject and a predicate,or,if you will,between a function and
its arguments—as in the sentence “Peter hit Paul”—depends on prior
capacities and operations (i.e.,naming and categorization) preceded by
pointing.
Pointing is particularly important.It leads us to distinguish two
domains in language,which Karl Bühler (1965) called Zeigfeldand
Symbolfeld:the deicticfield,in which words directly refer to the speaker
and the world and are dependent on context (e.g.,indexical expressions
such as “I,”“here,”“now,”“this”);and the symbolicfield,which proceeds
through the intermediary of general concepts.The existence of the
deictic field in language suggests a social origin for relations in the world:
it is for their “socius”that humans designate and name objects.More-
over,the act of predication must not be interpreted as a logical or
abstract operation,but instead as the representation of a scene.It is here
that one can establish a link between language and the way in which
problems of visual perception are conceptualized today:the major
concern deals with understanding how the cerebral cortex represents an
environmental scene.It is the same thing for language:it represents,that
is,it “plays out,”a scene,and we will soon see the importance of this
process for the origins of language.
The lexicosemantic dimension of language has been largely ignored
since the triumph of structural and generative linguistics and the empha-
sis that it placed on grammar.I would like to focus on one final compo-
nent of language:affective semantics,something that I would relate to
musical signification.Linguistic semantics is generally,and almost exclu-
sively,conceived of in terms of a referential semantics couched in the171 Toward an Evolutionary Theory of Music and Language