What Comes First: Meaning or the Word?

The most fundamental question is whether the child learns a word to describe a category or class he has already created mentally as a result of his manipulations of the world around him, or whether the existence of a word forces the child to create new cognitive categories. This may seem like a highly abstract argument, but it touches on the fundamental issue of the relationship between language and thought. Does the child learn to represent objects to himself because he now has language, or does language simply come along at about this point and make the representations easier?

...we now know that quite young babies have at least some kind of ability to represent things to them¬ selves, since they are able to remember and imitate objects and actions over periods of time—long before they have language to assist them. The develop¬ ment of a kind of gestural language before spoken language suggests the same thing. We also know that in the early stages of language development, children seem to apply words (or even to create words) to describe categories or classes they have already created in actions or images. For example, Brenda's word nene, which seemed to mean liquid food and the pleasure that goes With it, was probably a word to describe an already existing mental category or scheme.

As further support for the cognitive basis of early language, we know that some grammatical categories, such as prepositions, are not added all at once, but appear in the child's language over several years, added only when the child appears to understand the underlying relationship. So, for example, the word in is used before the word between, and both appear beifore in front of (Johnston, 1985). All of these bits of evidence point to the Ukelihood that in the early years concepts precede language in many instances.

But it seems equally clear that children's concepts and classification systerns are affected by the labels attached to objects, too. As the noted Russian psychologist Lev Vygotsky pointed out years ago (1962), there is a point some¬ where in the child's second year when she "discovers" that objects have names. In part this discovery itself seems to rest on a new cognitive ability, the ability to categorize things. In one longitudinal study, for example, Alison Gopnik and Andrew Meltzoff (1987) found that the "naming explosion" typi¬ cally occurs just after, or at the same time as, children first show spontaneous categorization of mixed sets of objects. Having discovered "categories," the child may now rapidly learn the names for already existing categories. But as Katherine Nelson (1988) argues, if the child assumes that there is a one-to-one correspondence between words and concepts, then not only does there have to be a word for every concept, there also has to be a concept for every word. So the existence of new words helps to create new concepts and new schemes.