In this paper we try to answer the following questions: Why do frequently used words tend to polysemy and homophony? And what comes first-frequency or the higher number of meanings per word? We shall stress the key role of frequency in the emergence of polysemy and assume an interactive step-up initiated by frequency: High frequency not only favors reduction processes of words or the bleaching of meanings that can result in polysemy; it also plays a crucial role in the creation of metaphors or metonymies, ie, the main sources of polysemy. Only familiar or frequent source words/concepts tend to be used in metaphorical or metonymical expressions. Through the conventionalization of the metaphors and metonymies, the source words get additional meanings. They now can be used in a higher number of contexts what in turn favors a more frequent use.

A similar explanation might hold for the development of homophony: Shorter words are known for their tendency to homophony [Jespersen, 1933] and high token frequency. Our explanation: High frequency favors backgrounding processes, such as vowel reduction, lenition and deletion of consonants or even of syllables. This frequency-induced shortening of words often results in sound merger and in a relatively high proportion of homophonous words, ie, words encoding unrelated meanings.