Categoriae 1. 1a 1-15
Ὁμώνυμα λέγεται ὧν ὄνομα μονον κοινόν, ὁ δὲ κατὰ τοὔνομα λόγος τῆς οὐσίας ἕτερος, οἷον ζῷον ὅ τε ἄνθρωπος καὶ τὸ γεγραμμένον. Τουτων γὰρ ὄνομα μόνον κοινόν, ὁ δὲ κατὰ τοὔνομα λόγος τῆς οὐσίας ἕτερος· ἄν γάρ τις ἀποδιδῷ τί ἐστιν αὐτῶν ἑκατέρῳ τὸ ζῴῳ εἶναι, ἴδιον ἑκατέρου λόγον ἀποδωσει.
Συνώνυμα δὲ λεγεται ὧ τό τε ὄνομα κοινὸν καὶ ὁ κατὰ τοὔνομα λόγος τῆς οὐσίας ὁ αὐτός, οἷον ζῷον ὁ τε ἄνθρωπος καὶ ὁ βοῦς. ὁ γὰρ ἄνθρωπος καὶ ὁ βοῦς κοινῷ ὀνόματι προσαγορεύεται ζῷον, καὶ ὁ λόγος δὲ τῆς οὐσίας ὁ αὐτός· ἐὰν γὰρ ἀποδιδῷ τις τὸν ἑκατέρου λόγον, τί ἐστιν αὐτῶν ἑκατέρῳ τὸ ζῴῳ εἶναι, τὸν αὐτὸν λόγον ἀποδώσει.
Homonymous are called those things which have only the same name, but which have a different account of their essence [λόγος τῆς οὐσίας] in relation to the name; in such a way ‘animal’ [ζῷον] is used for both a human being and a drawing. For only the name is common to both, while the account of the essence for this name is different. For if someone were to explain what it is for each of them to be an animal, he will give a particular account of each.
Synonymous are called those things which have both the same name and the same account of their essence in relation to the name; in such a way ‘animal is used for both a human and an ox. For man and ox are both called animal with the same name, and the account of their essence is the same. For if someone were to give the account of what it is for each of them to be an animal, he will give the same account.
λόγος τῆς οὐσίας: translated by Ackrill as “definition”; the emphasis, however, is slightly different than that of ὁρισμὸς in the Physica above.
ζῷον: originally meant ‘animal’ or ‘living thing’, but also came to refer to paintings or other artistic representations.
Aristotle’s use of “homonym” and “synonym” is different than the modern one. Ackrill (71) points out that they apply to things, not types of words. I have followed him in translating ὁμώνυμα and συνώνυμα in the passage as though they were adjectives in order to make this relationship clearer.
The distinction Aristotle makes between homonyms and synonyms seems to involve what we would call different word senses. That is, he distinguishes between two instances of a word where it is being used in the same sense (synonymous) and two instances where the sense is different (homonymous). Note that in his description of synonyms he is not saying that ‘human’ and ‘ox’ mean the same thing, but rather that, since both are hyponyms of ‘animal’, the name applies to each of them in the same way (in Aristotle’s terms, by virtue of their being possessed of both a nutritive and a motive/perceptive soul), whereas a drawing is an animal only insofar as it represents, or has the form necessary for, an animal.
For Aristotle, as will be seen in the section from De partis animalibus below, this homonymous (we might say, ‘metaphorical,’ although where non-metaphorical extension of meaning ends and metaphor begins is difficult to say) use of language seems to be inferior; each word has a central meaning which is connected to the physical thing which it describes.
παρώνυμα δὲ λέγεται ὅσα ἀπό τινος διαφέροντα τῇ πτώσει τὴν κατὰ τοὔνομα προσηγορίαν ἔχει, οἷον ἀπὸ τῆς γραμματικῆς ὁ γραμματικὸς καὶ ἀπὸ τῆς ἀνδρείας ὁ ανδρεῖος.
Paronymous are called those things of a sort which have the name they are called by from something else, while differing in their ending; thus, the grammarian from grammar, and the brave man from bravery.
πτωσις: “falling” or “fall”. Usually used in grammar for the cases of a word, but here Aristotle is using it to refer to derivational, not inflectional morphemes.
Unlike the previous terms, “paronym” corresponds fairly closely with modern usage, although once again it probably refers to things rather than words. Ackrill also suggests that “the derivativeness in question is not etymological. Aristotle is not claiming that the word ‘brave’ was invented after the word ‘bravery’. He is claiming rather that ‘brave’ means ‘having bravery’; the brave is so called because of (‘from’) the bravery he has” (72). This interpretation would be consistent with Aristotle’s thought in general and his concern with qualia in particular, as it does not simply register a grammatical fact (we can derive words from other words), but a certain type of relationship between the name and the characteristics of the thing named. The Greek tends to support this: the use of ἀπό τινος “from something” instead of ἀπὸ ἄλλου ὀνόματος “from another word,” and the examples Aristotle chooses, which both start with the abstract quality from which the noun (for a person) is derived.
Ackrill, J.L. Trans & Commentary. 1963. Aristotle. Categories and De Interpretatione. Oxford: Clarendon Press.