Delahunty and Garvey
meanings are complex and can be viewed as composed of basic, indivisible
units of meaning. These units are usually called components, though some-
times you will see them referred to as features or sememes. Components
are often regarded as pure concepts, not to be equated with the words of any
language, which is why they are typically written in capital letters. From
this point of view, a word is essentially a shorthand way of grouping a set of
concepts under a single label. Some of the concepts that have been proposed
by various linguists as components are listed in Table 2.
animate (alive) become cause
curved female flat
horizontal human ingest
intention know male
married not old
place self size
vertical young
table 2. some proposed universal semantic components
The components listed in Table 2 are just a sample of those that have
been proposed, but they are adequate to illustrate the thrust of the C-model.
For instance, in this model the word alive is shorthand for the component
ANIMATE; dead is shorthand for NOT, ALIVE; die for BECOME, NOT,
ALIVE. Kill adds the component CAUSE, and suicide adds SELF. (The com-
ponents are independent of the parts of speech of the words to which they
apply.)
You might object that such definitions are grossly oversimplified. A valid
concern. At the very least, how the components are related to each other is
a very important aspect of word meaning. Simplistically adding the compo-
nents BECOME, NOT, and ALIVE together does not adequately define die.
These issues raise technicalities which need not detain us here. For ways to
deal with them you might read work on this topic, e.g., Ch. 2 of Jackendoff
(1995).
It is important to distinguish between the universality of the list of com-
ponents and their language specific uses. The features mentioned in Table 2
are quite likely to be universal, that is, having the potential to be used in the
creation of word meanings in any and all languages. While there may be com-
ponents that are specific to individual languages, there are linguists who claim
to have identified a universal set of semantic primitives. (Anna Wierzbicka
probably makes the strongest claim in that regard—see Wierzbicka 1992, for