Vocabulary Size and Structure

The vocabulary (lexicon) of a human language, whether spoken or signed, is more than a list of the pronunciations (phonology) and meanings (lexical semantics) of words.  It also includes morphological and syntactic information. Morphological information includes the grammatical category of a word (noun, verb, adjective, etc.), the gender and/or declension class of nouns, the conjugation class of verbs, and any irregularities of form when a word is inflected for number, gender, person, and the like (cf. “Rules of word formation (morphology)”). Syntactic information includes the contexts in which a word typically occurs. For example, part of the lexical entry of the English verb give is that it requires three arguments - a giver, a recipient, and a thing given - and that the recipient and thing given can be realized either as a prepositional phrase following a direct object (give the book to her) or as two objects in a row (give her the book). There is evidence that the human brain also keeps track of how frequently it has encountered a particular word over the course of a lifetime of experience. The information associated with lexical entries - even for the same concept - varies a great deal cross-linguistically, i.e. within the same species. The vocabularies of human languages both signed and spoken are open-ended in that new words can always be derived via phonological recombination and by various morphological processes (e.g., suffixation, prefixation, etc.). The communicative inventories of animals in the wild are generally fixed, closed systems. The exceptions to this appear to be dolphins and parrots, both of whom are capable of lifelong vocal learning. In the laboratory, studies of the signs produced by chimpanzees exposed to American Sign Language have yielded no evidence that they have either phonological or morphological internal structure; rather, they are learned as fixed, frozen forms. The symbols used in Yerkish, an artificial system for representing words used with some language-trained apes, do have human-imposed internal structure in that graphic elements are repeated and recombined across symbol tokens. However, these graphic elements do not carry intrinsic meaning, nor are the apes trained to decompose and recombine the elements in meaningful ways, and there is no documented evidence that the apes spontaneously attempt to make use of this built-in structure on their own.