Chapter 6 – Documenting lexical knowledge
141
(4) Tzotzil words for ‘stuck’
Kakal ‘stuck (between two surfaces)’
Ch’ikil ‘stuck (into a narrow or tight crevice)’
Katz’al ‘stuck (in a jaw-like orifice)’
Xojol
‘stuck (inside an enclosing hole)’
Tz’apal ‘stuck (a pointed thing anchored in a surface)’
As the detailed glosses show, however, each word specifies different con-
figurations, kinds of attachment, and different shapes, in both figure and
ground.
10
The exact conflation, I believe, involves such factors as the fol-
lowing, taking the root tz’ap as an illustration.
(5)
Conflation in
tz’ap
a.
the “end” of the Figure is “inside” the Ground;
b.
the Ground need not have a
y-ut ‘inside’ (or perhaps it must not be
so structured, conceived of instead as a mere surface);
c.
the Figure has a “pointed” “end” (in Tzotzil, s-ni` ‘nose’);
d.
typically the Figure is “stuck” into the Ground pointed end-first,
i.e., attached somehow, and self-supporting; and
e.
typically it is vertically oriented.
Linguists have posited various classifications of semantic types, in different
root classes, and the field lexicographer should borrow shamelessly from
such typologies: from frames, to verb types (Dixon 1972), to verb classes
based on patterns of diathesis (Levin 1993), and so on.
The multiplicity of “language games” – something that cannot long re-
main hidden from a serious field linguist – further complicates the tradi-
tional referential view of lexical meaning. We use words to refer; but also
for many other things. Here is part of Wittgenstein’s list:
Giving orders, and obeying them – Describing the appearance of an object,
or giving its measurements – Constructing an object from a description (a
drawing) – Reporting an event – Speculating about an event – Forming and
testing a hypothesis – Presenting the results of an experiment in tables and
diagrams – Making up a story; and reading it – Play-acting – Singing
catches – Guessing riddles – Making a joke; and telling it – Solving a prob-
lem in practical arithmetic – Translating from one language into another –
Asking, thanking, cursing, greeting, praying.
(Wittgenstein
1958:
sect.
23)
142
John B. Haviland
Cruse (1986: 270 ff.) reminds us of the differences between what he calls
“semantic modes,” as in the contrast between the following two utterances.
(6) “Semantic modes”
I just felt a sudden sharp pain.
Ouch!
If semantics is only about reference and predication, then it will be difficult
to capture the meaning of ‘ouch!’ semantically, because the word involves
neither reference nor predication. Instead, it will be important to understand
such things as interjections (see Kockelman 2003) in terms of very differ-
ent semiotic modes: indexing speaker stance, interlocutor’s relationship to
speaker, putative bodily and affective states, expected responses, and so on.
That words like ‘ouch’ are hard to model in terms of denotata does not re-
lieve us of the lexicographer’s responsibility of recording them and ex-
plaining how they work – a problem which I return to below.
A broader and more appropriate conception of meaning derives from
one of the well-known trichotomies of ways that signs can signify or “stand
for” other things, due to C. S. Peirce (1932). The three semiotic modes are
based on very different principles, although they generally co-mingle in
most signs, linguistic or otherwise. Peirce pointed out that some signs stand
for other things because of a resemblance between the sign vehicle and the
thing signified – thus a photograph of a person can stand for that person
(for example, in a directory or catalogue). The sign bears an “iconic” re-
semblance to what it signifies, although the nature of the “resemblance”
can vary tremendously (consider diagrams, drawings, silhouettes, graphs,
for example, or conventionalized but nonetheless onomatopoetic words
whose sounds suggest their meanings: ‘moo’ or ‘caw’ or ‘cackle’, perhaps).
There can also be an “indexical” relationship between sign and signified,
such that physical, spatial, or direct causal relationships exist between the
sign vehicle and what it signifies. A footprint, for example, may not “re-
semble” the person who made it (although it may, of course, “resemble” his
or her foot), but it stands as an ‘index’ of the person by virtue of the fact
that it took the person’s foot to make the mark (hence, indicating, for ex-
ample, that that person has been at a certain place). In language, ‘ouch!’
stands for (indeed, displays) sudden pain precisely because we imagine that
the pain itself somehow (involuntarily?) produces the utterance. In a similar
way, we know what person ‘I’ or ‘you’ refers to by observing the contex-
tual relationship between the sign – the word – and the person uttering it or
Chapter 6 – Documenting lexical knowledge
143
to whom it is uttered. Such words, then, rely on an indexical relationship
(in a context) to convey their meanings. Finally, there are signs whose sig-
nificance is essentially unmotivated by either resemblance or context: these
are Peircean ‘symbols’ which rely on a conventional relationship between
signifier and signified – Saussure’s “arbitrariness” of the linguistic sign, in
which ‘cat’ means cat only because that is what a particular linguistic tradi-
tion has legislated.
Figure 1 shows a sign which transparently combines all three Peircean
semiotic modalities: the iconic resemblance between the drawing and a
(stylized) smoking cigarette; the conventional meaning (at least in much of
the Western world) of the shaded circle with the diagonal bar as a “prohibi-
tion”; and finally, the location of the sign itself, whose physical position
signals indexically exactly where smoking is prohibited.
Figure 1. A semiotically trichotomous sign
An adequate description of the meaning of linguistic elements must capture
all three modes of signification, although the major lexicographic traditions
limit themselves largely to “conventional” or symbolic meaning, almost ex-
clusively in referential terms.