Diogenes Laertius (7.56-7) preserves for us the linguistic theory of the Stoic philosopher Diogenes of Babylon. Part of this theory included a distinction between voice (φωνή), which is both articulate and inarticulate, and utterance (λέξις), which is only articulate. There is also a further distinction between utterance, which contains both significant and non-significant utterances, and language (λόγος), which is a significant utterance. This classificatory scheme was adopted by grammarians and later philosophers working within the Platonist-Aristotelian tradition (i.e., the commentators). Each part of the classification was connected to set examples across all post-Stoic authors. The examples of non-significant utterance (i.e. λέξις which is not also λόγος) are two supposedly nonsensical words: σκινδαψός and βλίτυρι. Yet, there is reason to believe that the examples date back to the Stoics as well. Diogenes Laertius himself gives the example of βλίτυρι in his précis of Diogenes of Babylon’s classification, where we would expect extraneous material not to be added, but removed. Furthermore, a fragment from the satirical poet Timon connects Zeno, the founder of the Stoa, with the word κινδαψός (with the initial sigma removed metri gratia).
But the word σκινδαψός was also of interest to ancient lexicographers, precisely because it did signify something. Various authorities gives its definition as a musical instrument, a tribe in India, a plant, and the name of a slave. βλίτυρι, too, had another meaning: that of the sound made by a string. These meanings are attested in sources either contemporaneous with or earlier than the Stoics. This, of course, creates a discrepancy, as noticed by Galen (De Diff. Puls. 8.662). If the words had meanings, why would the Stoics use them as examples of words without them? The answer, I will argue, is that the Stoics originally used them as examples of certain types of λέξις, which had meanings, but for certain reasons failed to signify them properly. As suggested by Ax 1986, βλίτυρι is an onomatopoeia of the sound of a musical instrument (very possibly that of the σκινδαψός itself), and so, on the Stoic view, signifies nothing but itself, a particular sound. A further look at the lexicographical evidence confirms Ax’s view. The path of σκινδαψός is harder to trace, but a careful look at the evidence shows that above all what ties the meanings together is the idea that the word was thought of as foreign (which is, of course, a matter separate from its actual origins). And, as it happens, Diogenes of Babylon also classified foreign language as λέξις, not true λόγος. These words were technically examples of certain types of λέξις, but were misunderstood by later authors and explained as words that referred to nothing at all. Thus, we may surmise that the grammatical theory of the Stoics originally had more to say on the classification of language and language-like sound than the doxography communicates.
Linguistic Strategies and the Hermeneutics of Reading