Navigation – Plan du site
Dire et vouloir dire dans les arts du langage anciens et tardo-antiques

Imposition of Words in Stoicism and Late Ancient Grammar and Philosophy

Sten Ebbesen

Résumés

Si les penseurs anciens s’accordaient pour dire que les mots reçoivent leur signification d’une institution (θέσις), leurs vues divergeaient, en revanche, quant à la question de savoir jusqu’à quel point leurs instituteurs étaient libres de choisir tels ou tels sons pour désigner telle ou telle chose. Il ne fait pas de doute que, pour les Stoïciens, ceux qui ont imposés les noms visaient d’ordinaire à créer des expressions linguistiques capables d’évoquer d’une certaine manière les choses signifiées. L’article rejette cependant l’idée que les Stoïciens présupposaient une situation initiale où l’ensemble des termes d’une langue donnée étaient parfaitement adaptés à leur fonction, les aberrations ou anomalies du langage ordinaire étant liées à une dégradation progressive de conditions de départ idéales. Ils pensaient, au contraire, que l’imposition des mots est un processus continu, que les locuteurs ordinaires entretiennent pour des raisons essentiellement utilitaires ; processus qui n’est que partiellement réglé par leur rationalité commune. Ce qui explique pourquoi le Grec qu’ils se trouvaient parler présentait à leurs yeux un défaut de systématicité que les Stoïciens se sont dès lors appliqués à corriger par le biais d’une nomenclature qui se voulait rigoureuse et cohérente.

Haut de page

Notes de l’auteur

A first version of this article was delivered orally in December 2016 at a colloquium in Lille organized by the project Sêmainô. A Danish version with reduced scholarly apparatus is due to appear in 2019 in the electronic journal Aigis at http://www.igl.ku.dk/aigis/.

Texte intégral

Ι. Θέσις

  • 1 Ancient literature is cited without indication of edition used, as long as the authors involved are (...)

1In this essay1 I intend to shed some light on ancient ideas about the origin of human language, and in particular about the origin of the vocabulary. I will pay particular attention to Stoic theory and argue that Chrysippus et al. were much less ‟naturalists” than often assumed.

2Θέσις is not a particularly fancy word, but it represents a notion that was central in European philosophy of language for a couple of thousand years. No ancient or medieval philosopher or grammarian seems to have doubted that words start their career as carriers of signification by an act of institution, a θέσις: somebody sets up a certain complex of sounds to function as a sign of something. The paradigmatic situation was that of putting a name on some thing, labelling it, as it were, and the Greek for doing so is to τίθεσθαι “to put”, ‟set” or ‟lay down” a name, with the verbal noun θέσις indicating the act of so doing. In Latin people were to use the terms pōnere and positiō or impōnere and impositiō.

  • 2 Plato, Cratylus 401b.
  • 3 Used by Alcinous, Ammonius, Philoponus and Simplicius, among others. See the electronic Thesaurus L (...)

3Plato in his Cratylus famously speaks about a θέσις τῶν ὀνομάτων2, a ‟setting” or ‟institution of the names”, and later authors created the compound ὀνοματοθέτης ‟name-setter”3.

  • 4 Aristotle’s in De interpretatione 2.16a19-29 stresses that names owe their signification to συνθἠκη(...)

4While the ancients agreed that one must operate with a θέσις τῶν ὀνομάτων, they disagreed about whether that institution was in principle arbitrary, any pronounceable combination of sounds being as good a name of some thing X as any other combination (as Aristotle probably thought4), or whether the nature of X put some restrictions on which combination or combinations would be suitable to name it, so that the natures of things provide a framework within which any imposition of names –whether the original or a later one– must operate if the result is to be words that can properly fulfil their function as bearers of information.

  • 5 For the latter, see Plato, Cratylus 384d.
  • 6 Bonitz, Hermann (1870), Index Aristotelicus (= Aristotelis Opera V), Berlin, Reimer, p. 327b, s.v. (...)
  • 7 Thus already in Epicurus, Epistula ad Herodotum, apud Diogenes Laertius, Vitae philosophorum X.75. (...)

5To express the ‟by nature / by convention” alternative Plato does not say ‟φύσει / θέσει”, neither in regard to words nor in other contexts; his pair is φύσει / νόμωι, both when the question is about moral values and when it is about words.5 The same holds for Aristotle,6 but soon afterwards the φύσει / θέσει pair of opposites became widely used, especially when the question was about the naturalness or conventionality of names7. In Latin the pair became nātūrā / positiōne, which survives in the expressions natural and positive law.

  • 8 Simplicius, in Cat., CAG VIII: 40.5-13: ‟Διὰ τί δὲ Ἀρχύτας παραλέλοιπεν ταύτην τὴν περὶ τῶν ὀνομάτω (...)

6The change from νόμωι to θέσει may look innocent, but has sometimes created confusion. Everybody agreed that new words are introduced by a ‟setting”, while only some held that such a ‟setting” can be independent of the nature of the thing to be named. However, with the possible exception of some late-ancient Pythagoreanizing Platonists8, no ancient ‟naturalist” seems to have thought that nature dictates exactly one sequence of sounds for each sort of thing; rather, the idea was that an ‟onomatothete” could select one from a certain repertoire of suitable sequences, some of the possible choices probably being better than others to transmit information about the thing to be designated.

  • 9 Philo Iudaeus, De cherubim 56: “ὁ μὲν ἄλλος ἅπας ἀνθρώπων ὅμιλος ὀνόματα τίθεται πράγμασι διαφέροντ (...)

7As an oddity, it may be mentioned that Philo of Alexandria explains away certain eccentricities in the book of Genesis by claiming that9

“Everybody else imposes names on things that differ from the things, thus producing a gap between the bearers of appellations and their appellations. With Moses, by contrast, the institutions (θέσεις) of names with all clarity express the things, so that the name necessarily and straightaway is the thing and there is no difference between the name and that on which it is put.”

  • 10 Diodorus Siculus, Bibliotheca historica V.67.3: ‟Τῶν δὲ Τιτανίδων φασὶ Μνημοσύνην λογισμοὺς εὑρεῖν (...)

8So, everybody else has failed to match names and things named properly, only Moses got it right! But then he was divinely inspired, I suppose, and divine intervention in the creation of language was also occasionally invoked by others. Thus the Muse Mnēmosynē could be cast in the role of the original name-giver, since she is a hypostatization of memory (mnēmē), with which the need and ability to communicate is intimately connected10.

  • 11 Aristotle, De interpretatione 3.16b19-20: ‟αὐτὰ μὲν οὖν καθ' αὑτὰ λεγόμενα τὰ ῥήματα ὀνόματά
    ἐστι κ (...)
  • 12 About function words, see S. Ebbesen, Commentators and Commentaries on Aristotle’s Sophistici Elenc (...)

9All the old philosophers thought less about the grammar of a language than about its vocabulary and they all took as their model for language creation the situation in which something hitherto nameless is to be assigned a name –a dangerous simplification that neglects both the grammatical system and the important part of the vocabulary that is constituted by function words, pronouns and prepositions in particular, the sort of words the medievals called syncategoremata with a term of Stoic origin. As Aristotle remarked11, a verb like to run may well be considered a name of something, namely a certain type of action, but –and this he forgot to remark– a pronoun like every or a preposition like except cannot be assigned a corresponding thinglike semantic content. Later in Antiquity people were aware of this, and said that while function words do not signify anything they are consignificative (συσσημαντικά), i.e. they modify the meaning that the genuinely significative words contribute to the sentence12. On the question how such function words entered our language the sources are silent, as far as I can tell, but the fate of function words in ancient linguistics is an understudied topic.

  • 13 See, e.g., Flavius Sosipater Charisius, Ars grammatica, Karl Barwick (ed.), Leipzig, Teubner, 1925 (...)
  • 14 See Dante’s De vulgari eloquentia.

10Another simplification in both ancient and medieval thinkers’ treatment of the rise of language consists in pretending that there is only one human language –Greek in the Greek culture and Latin in the Latin culture. Admittedly, it is understood, and occasionally made explicit, that other languages have the same sort of origin as Greek or Latin, so that common mechanisms underly the creation of them all, but considerations about a genetic community between different languages are rare and superficial –usually limited to suggestions that maybe Latin has developed from Greek13. To the best of my knowledge, Dante Alighieri is the first to have presented a sensible theory about how one language can develop into several14.

  • 15 See, e.g., Schmidt, Rudolf (1839), Stoicorum grammatica, Halle (rp. Amsterdam, Hakkert, 1967), 21ff (...)
  • 16 Pohlenz, Max (1959), Die Stoa, Göttingen, Vandenhoeck & Ruprecht (here quoted from 4. Auflage, 1970 (...)

11The ancient Stoics are known to have spent much thought on linguistic matters, including the origin of language. They have often been seen as rather hard-core naturalists15. Thus in his –for a long time influential– Die Stoa Max Pohlenz said that according to the Stoics16,

  • 17 I suspect a printing error, an indefinite article having fallen out, so that the text should read ‟ (...)

“Die Namen sind vom menschlichen Logos durch17 bestimmten Willensakt gegeben (θέσει), aber sie sind zugleich natürlichen Urspungs (φύσει), weil die beigelegten Bezeichnungen der Physis der benannten Dinge entsprachen. Daß die ersten Menschen tatsächlich so vorgegangen seien, glaubten die Stoiker noch am empirischen Sprachmaterial zeigen zu können.”

12In English:

“Names have been given by the human logos through a certain act of will (θέσει), but they are at the same time of natural origin (φύσει) because the accorded designations matched the physis of the things named. The Stoics believed that on the basis of the empirical linguistic material they could still show that the first humans had actually proceded in that way.” [My translation.]

  • 18 Origenes, Contra Celsum 1.24 (= FDS F 643): ‟Λεκτέον δὲ καὶ πρὸς τοῦτο ὅτι ἐμπίπτει εἰς τὸ προκείμε (...)

13The foundation of this view of the Stoic theory is in particular a passage in Origines, where he speaks about18

“the deep and abstruse discussion about the nature of names, whether names are conventional (as Aristotle thinks) or natural, as the Stoics hold, the original speech-sounds imitating the things of which they were to be names, for which reason the Stoics also introduce certain etymological elements – or whether, as Epicurus teaches, names are natural in another sense than the Stoics think, namely because the first humans spontaneously broke into certain speech-sounds about the things <they encountered>”.

  • 19 See, e.g. Cicero, De natura deorum III.xxiv.63 (= FDS F 606) and the examples of Stoic etymologizin (...)
  • 20 Plato, Cratylus 396a-b: ‟ἔστι δὲ οὐ ῥᾴδιον κατανοῆσαι. ἀτεχνῶς γάρ ἐστιν οἷον λόγος τὸ τοῦ Διὸς ὄνο (...)

14That the founders of Stoicism were fond of etymologies is well-attested19, and their etymologies are of the same type as many of Plato’s in Cratylus: the word under consideration is reminiscent of some other word or a phrase that reveals something characteristic of the thing signified. Chrysippus is even reported to have repeated Plato’s etymologizing of the two accusative forms of Ζεύς, Δία and Ζῆνα: they point at the important fact that Zeus is the God through (διά) whom all living beings live (ζῆν)20.

15I wish to argue in this article that the Stoic’s naturalism was a very moderate one. To get at the aim, I shall make a detour to the late-ancient commentators on Aristotle, who are often described as Neo-Platonists, but ought rather to be called Aristoteli-Platonists. They composed their works at a time in the early 6th century when Stoicism was dead, but they drew upon a commentary tradition that reached back to the first centuries A.D. when the school was still alive and in many ways dominated the discussion about matters linguistic. Via the learned 3rd-century commentator Porphyry, in whose youth there were still some Stoics around, much material of Stoic origin reached the late Aristoteli-Platonists.

  • 21 There is a long, somewhat rambling, discussion supporting the compromise view in Ammonius, in Int., (...)
  • 22 For words in the category of when, see Simplicius, in Cat., CAG VIII: 350.4-5: ‟διὰ τοῦτο δὲ οἱ ὀνο (...)

16The Aristotelian commentators typically go for some sort of compromise between φύσει and θέσει: yes, words are instituted, and in that sense θέσει, but at least the fundamental human vocabulary was not instituted at random, the givers of names let their choice of vocal expressions be determined by considerations about the natures of the things to be named21. The commentators call the creation of the fundamental vocabulary ‟the first imposition” or ‟institution”: πρώτη θέσις. It established a first-order vocabulary for speaking about the physical world, or, in Aristotelian terms, the world of the ten categories, and included such words as ‘man’ ‘woman’ ‘white’ ‘black’ ‘yesterday’ ‘today’ ‘tomorrow’ etc.22. Later, a second-order vocabulary was created, to which I shall return.

  • 23 Simplicius, in Cat., CAG VIII: 15.8-12: ‟ πρώτη θέσις τῶν φωνῶν ὡς ὀνομάτων ἐστίν· δεηθέντες γὰρ ο (...)
  • 24 Ammonius, in Cat., CAG IV.4: 11.8 ff. Ammonios offers a hopelessly naive picture of the first impos (...)

17What occasioned the rise of language? Simplicius23 offers a sort of Original Fall as the explanation: having seceded from the common understanding (ἡ κοινὴ νόησις), men had no choice but to institute names if they were to signify things to one another. Other authors just make the original institution of names a consequence of man’s being by nature a social, communicative animal. This also seems implied by Simplicius’ account, but unlike him they mention no Original Fall24. It is remarkable how little highfaluting Platonism one finds in the commentators’ remarks about the imposition of names. To me this suggests that it was not a matter in which they were terribly interested, and that consequently most of what they say about it is not something they had thought out themselves but heirlooms from an earlier period.

  • 25 Dexippus, in Cat., CAG IV.2: 11.14-15: ‟προηγουμένως ἡ σημαντικὴ λέξις τῆς πρώτης ἔχεται χρείας τοῦ (...)
  • 26 Thus in Philoponus, in Cat., CAG XIII.1: 85.23.

18The commentators agree that the first things to be named were the things ordinary people would notice and needed to be able to talk about in order to make human life possible, so the first vocabulary could also be called “words of the first need” or ‟use” (τῆς πρώτης χρείας)25, and the commentators repeatedly speak of συνήθεια, that is common usage, normal behaviour, as what determined which words were coined. Sometimes this common usage becomes personified, as it were, and itself speaks and institutes words26.

  • 27 There is a beautiful book in Danish about Quintilian’s use of ancient theories of language: Pinborg (...)
  • 28 Quintilian, Institutio oratoria I.6.43-45. Notice his concluding remark: ‟Ergo consuetudinem sermon (...)

19There can be little doubt that the commentators understood their συνήθεια in much the same way that the first-century Roman rhetorician Quintilian, whose views on language show many Stoic traits27, understood consuētūdō. For a word to belong to consuetudo it is not sufficient, he says, that it is being used by many people: it is also required that it is used by decent, educated speakers of the language28. Such an élitist attitude to language also makes it easier to maintain that –generally speaking, at least– the original name-instituters were clever people who made good choices when laying the foundations for a language.

  • 29 Simplicius, in Cat., CAG VIII: 19.26-20.5. The context is a discussion of why Aristotle put the ant (...)

20While the commentators clearly favour the view that the first namegivers somehow attempted to make their coinages fit the things signified, possibly even successfully, there is some wobbling in the ranks, and they are very vague about what exactly is required for a name to make a natural fit with its significate. Simplicius, citing Porphyry, claims that we use certain words handed down to us by common usage without possessing an exact knowledge of their natures, i.e. the features of their significates on which the name-institutors were focusing when instituting those words.29 So, the properties of a word that make it a good fit for its meaning may no longer be recognizable. Or perhaps it was not a good fit in the first place, except in the mind of its creator.

21In another context, Simplicius, talking about what makes a good imposition, says:

  • 30 Simplicius, in Cat., CAG VIII: 187.8-9: “εἰ μὲν θέσει τὰ ὀνόματα, στοχαστέον τοῦ σκοποῦ τῆς θέσεως· (...)

“If names <become names> by institution, one must aim to hit the goal of the institution. If they are by nature, one must pay attention to the nature of the things, and thus combine the proper sounds.”30

  • 31 See note 19, above.

22This suggests that words may have other virtues than sounding in a way suggestive of the nature of their significates, whether directly, or indirectly via some other word or phrase, the way Δία and Ζῆνα point to Zeus’s role as the god through (διά) whom all living beings live (ζῆν)31 What other virtues could Simplicius be thinking of? Brevity, perhaps, and certainly fitting into an existing linguistic pattern.

  • 32 Philoponus, in Cat., CAG XIII.1: 157: 7-13: ‟oὔτε γὰρ ὁ δρομικὸς οὔτε ὁ πυκτικός, οἱ φυσικήν τινα ἐ (...)

23The scope of the first wave of naming was limited by the limited horizon of ordinary people. They concentrate on what is concrete and actual. This explains why ordinary Greek has no name for a still latent talent for running or for boxing. Common usage, Philοponus says32, has concentrated on assigning names to things in which a potency has been fulfilled and actualized, so that is why there are words to describe good running and boxing skills as well as good runners or boxers, but, as Aristotle had noticed (Cat. 8.10a32-b5), none to describe somebody with a talent for becoming a good sportsman or his talent.

  • 33 Philoponus, in Cat., CAG XIII.1: 113.19 ff.: ‟λέγει οὖν <sc. ο Ἀριστοτέλης πρὸς τοῦτο ὅτι οὐδὲν χαλ (...)
  • 34 See Dexippus, in Cat., CAG IV.2: 26.29-27.2. In 15.23-16.13 Dexippus further includes ‘whole’ and ‘ (...)

24The discovery of crafts and scientific disciplines required an extension of the vocabulary, a second wave of institution, a δευτέρα θέσις. Geometry and music are cited as examples of disciplines that are beyond the horizon of common usage, and thus beyond the ken of the original givers of names, but which in time came to need a technical vocabulary33. Most often when the ancients speak of a second institution they are thinking of meta-linguistic terms like ‘noun’, ‘verb’ and ‘trisyllabic’, but some, at least, also counted the names of second-order concepts, such as ‘genus’ and ‘species’ as words of the second institution, and Dexippus notes that words of the second institution transcend the Aristotelian categories34.

  • 35 Simplicius, in Cat., CAG VIII: 187.5-6: ‟ἢ τοῖς κειμένοις ἤδη αὐτοὶ χρώμενοι ἐπ’ ἄλλοις, ὡς τῷ χρώμ (...)

25How, then, to proceed about imposing names? There are two main ways to create a new term, the commentators hold: either you create a new word or you assign a new meaning to an existing one. Examples of the latter are easily found if one peeps into Aristotle’s tool-box: in ordinary usage ὕλη must have been a piece of wood functioning as a blank for a carpenter, and εἶδος was someone’s looks. Simplicius mentions the technical senses of χρῶμα “colour” and δίεσις “release” in music and of κέντρον “goad” in geometry35.

26Using a term in a new sense is what the ancient grammarians called ἀκυρολογία –a broad term comprising metaphor, metonomy and several other types. Its opposite is κυριολογία, which one grammarian defines it as follows:

  • 36 Trypho, De tropis, in Rhetores Graeci, Leonhard Spengel (ed.), Leipzig, Teubner, 1856, III, p. 191: (...)

κυριολογία is the way of expression (φράσις) that signifies the things through the first institution of the names”36

27or, in a freer translation,

“using words in their proper sense means using them to signify what they were originally imposed to signify”

28This implies that words carry with them the meaning that they were assigned at the original imposition, though they may be assigned further, improper ones. Anyone familiar with medieval semantics will know that this idea of words never being able to get rid of the meaning originally assigned to them was to have a great, but troubled, future.

  • 37 Simplicius, in Cat., CAG VIII: 186.34-187.1: ‟καὶ γὰρ ἀσφαλέστατος τρόπος ὀνοματοποιίας ὁ ἀπὸ τῶν ἤ (...)

29As for creating genuinely new words, Simplicius holds that the best procedure is to start from existing ones that belong to prevalent usage, and just modify them slightly, as Aristotle did when coining πηδαλιωτόν “ruddered” from πηδάλιον “rudder”. This, he adds, also seems to have been the strategy followed by the first instituters of names, for etymology shows that they started with simple and commonly used (συνήθη) names and then created more complex ones ‟by composition” – by which he seems to mean derivation by means of suffixes37.

II. Θέμα

  • 38 Apollonius, Pron. 26.6-8: ‟οὐκ ἄλογον δὲ τὸ τὴν τῶν ὀνομάτων ἐκφορὰν ἐν τρίτοις νοεῖσθαι, ἐπεὶ αἱ ἴ (...)
  • 39 Priscian, Partitiones 68, GL III: 474-5: ‟quaedam enim adverbiorum sunt πρωτότυπα, id est primitiva (...)
  • 40 Priscian, Inst. VIII.xii.63, GL II: 421-2: ‟quia prima positio verbi, quae videtur ab ipsa natura e (...)

30Now it is time to visit the second-century A.D. grammarian Apollonius Dyscolus and his sixth-century disciple Priscian, who also operate with a θέσις or positio of words. Thus Apollonius explains that nouns are inherently third-person items because the institution of a noun has the form This is to be called a stone rather than You are to be called ‘stone’38. Similarly, Priscian says that some adverbs are πρωτότυπα, id est primitiva vel primae positionis, while others are derivative39. He also holds that the first institution of the verb, ‟which seems to be the work of nature itself” is in the indicative, just as that of the nouns is in the nominative40.

31It is well known that much of Apollonius’ grammatical theory and terminology bears a Stoic stamp, although it is not clear how aware he himself was of the fact. Priscian certainly was unconscious of the Stoic heritage latent in Apollonius. I believe the reason why the first person singular in the present indicative active became the citation form for Greek and Latin verbs was the Stoic notion that our awareness of the world starts with a self-awareness I, here, now, so the original θέσις of the verb ‘to see’ would be in the first person present form ὁρῶ ‟I see”. From a lexicographical point of view, the Stoic obsession with I, here and now was a catastrophe. The imperfective stems of Greek verbs are unstable over time, and even synchronically, several variants being in use simultaneously, whereas an aorist active stem tends to survive without changes for millennia.

  • 41 Stobaeus, Eclogae I.13.1c, p. 138 (= SVF I: 25, frgm. 89; FDS F 762): ‟αἴτιον δ’ ὁ Ζήνων φησὶν εἶνα (...)
  • 42 Diogenes Laertius, Vitae philosophorum VII.58 (= SVF III.213, frgm. 22; FDS F 536).
  • 43 E.g., Plutarch, De communibus notitiis 30, 1074D; Sextus Empiricus, Adversus Mathematicos IX.211.

32The Stoics were famous for their peculiar terminology, and it does not take any special keenness of sight to see that they tried to create a systematic terminology. However, remarkably little has been done in modern times to exploit this realization in attempts to reconstruct pieces of their philosophy. According to Zeno, the founder of Stoicism, causes are bodies and their effects are predicates41. The Zeno fragment does not explicitly say that Stoic predicates, κατηγορήματα –the sort of quasi-entities that verbs signify–42, are incorporeal, but implicitly it contrasts the corporeal causes with their incorporeal effects, and the incorporeal nature of κατηγορήματα is explicitly attested in other sources43. Their incorporeality is shown by their having a name with the suffix -mat like ἀξίω-μα, σύμβα-μα etc., the stems of which end in -mat, though, in accordance with the phonological rules of Greek, the final t has been shed in the citation form.

  • 44 For the difference between λέξις and λόγος see Diogenes Laertius, Vitae Philosophorum VII.56-57 (= (...)

33Now, supposing someone institutes a name, say ‘stone’, that person is a body effecting something. In Stoic terminology, there would be a θέσις, which would also be a body, namely the person, or more precisely his centre of control, his ἡγεμονικόν, qua laying down a name. The effect would be an incorporeal θέμα “the fact that an institution has taken place”. The institution would also imply a further bodily entity, namely the sound constituting the words produced by the onomatothete when saying ‟Let this be called a stone”. Qua physical entity, the utterance would be a λέξις in Stoic terminology, and qua carrier of information a λόγος, while the incorporeal signified content would be a complete λεκτόν44 with the content ‟that ‘stone’ is to signify stone”. Presumably, the utterance and its content could also be called θέσις and θέμα, respectively

  • 45 For these θέματα, see Frede, Michael (1974), Die stoische Logik, Göttingen, Vandenhoeck & Ruprecht, (...)

34As we shall see, θέμα is a technical term in Apollonius’ grammar and in a sense having to do with the imposition of words, but it also occurs in other contexts, and in ways that support my above proposal about its use in connection with the creation of words. The first thing to notice is that the noun θέμα is not attested before the founding fathers of Stoicism, but it is attested as a Stoic word. In Stoic logic a θέμα was a rule that allowed reduction of irregular arguments to one of the five indemonstrables45. The exact reason for the choice of the word in this context is unknown, but it should probably be understood as a primitive rule introduced by a ‟positing” (θέσις), a rule that cannot be proved, yet must be assumed prior to all reasoning, so that it makes no sense to offer it for acceptance in a disputation so that it can become a λῆμμα.

  • 46 Sextus Empiricus, Adversus Mathematicos VIII.301: ‟ἀπόδειξις κατὰ μὲν τὸ γένος ἐστὶ λόγος· οὐ γὰρ δ (...)

35Λῆμμα and θέμα make a joint appearance in book VIII of Sextus Empiricus’ Adversus Mathematicos, when he tries to explain what proof, an ἀπόδειξις, is. It is a sort of λόγος, he says, though not in the sense of something one can hear, but qua a movement and consent of the mind46. There may be an echo of a Stoic description here, though probably somewhat distorted. But then he goes on with an explanation that seems to be directly lifted from some Stoic source:

  • 47 Sextus Empiricus, Adversus Mathematicos VIII.301-302: λόγος δέ ἐστιν, ὡς ἁπλούστερον εἰπεῖν, τὸ συ (...)

“An argument is, to put it simply, a complex of lemmas and a conclusion. By λήμματα (‟takings”) we do not understand θέματα (‟positings”), which we just grab, but such <claims> as our collocutor gives and concedes because they are obvious.”47

36In this context, the θέμα must be the content of what we posit, and the effect produced by our positing. Like other sentential contents, it must be a complete λεκτόν.

  • 48 One non-logical instance in Poetics 1457b7, and two in De spiritu (482b10, 483a14).
  • 49 The description of an argument in first period of the passage from Sextus Empiricus quoted two note (...)

37Someone might object that λῆμμα in the relevant sense of a premiss which the questioner in a dialectical debate “takes” is found already in Aristotle’s Topics, but the word used for a conclusion, ἐπιφορά, is not a technical term of his48; the combination of λῆμμα and ἐπιφορά is decidedly Stoic terminology49.

  • 50 Sextus Empiricus, Adversus Mathematicos I.148-153: ‟ τὰ δὲ αὐτὰ ὀνόματα οὐ πᾶσίν ἐστι τὰ αὐτά, ἀλλὰ (...)

38Notice again the Stoic choice of a word with the -mat suffix: what we take and the answerer in the debate assents to is of course not the physical string of words we hear but their semantic content, the speaker’s claim (ἀξίωμα), and by accepting it the answerer makes it legitimate for the questioner to take it, whereby it becomes a λῆμμα; without the answerer’s assent, it would remain a ‟positing” in the sense of a mere postulate with nothing but the speaker’s authority behind it. A similar connotation of one-sided decision making clings to the rare verb θεματίζειν and its corresponding noun θεματισμός, which Sextus Empiricus uses four times to characterize as arbitrary decisions about what gender nouns should have, some people have ‟thematized” one way and others differently; on one further occasion he uses it about an arbitrary decision about how a sign is to signify50.

39It is hardly surprising that it turns out that the Stoics employed θέμα in more contexts than one, with the meaning of the word varying somewhat according to what one can set/lay down or posit in the given context. After all, τίθημι is a verb with very little meaning of its own when not used about literally putting something in some place. And now to Apollonius Dyscolus.

  • 51 GG III.3: 211a.

40In 1910 Richard Schneider (1835-1917), the greatest Apollonius scholar there ever was, in the index of terms that concludes his and Uhlig’s monumental edition51, translated θέμα as forma non derivata, stirps, Grundform, and, indeed, the main characteristic of lexemes or word-forms that Apollonius characterizes by means of the word θέμα or the derivative adjective θεματικός is that they are primitive in the sense of not being derived from other lexemes or word-forms.

  • 52 Apollonius, De pronomine, GG II.1: 60.7-8: ‟ἄμεινον οὖν ἦν θέμα καταλιπεῖν, ἢ ἀποκοπὴν τοῦ ὁδεῖνα, (...)

41A rather typical example from Apollonius’ work On Pronouns (GG II.1: 59-60): Some people had proposed to derive ὁδεῖν ‟the so-and-so” (which he probably pronounced [ɔðîn]) from ὁδί “this here”, but Apollonius rejects the proposal because the supposed derivative disagrees with its supposed origin in accentuation, in signification and in the way it is declined. He then concludes52 :

“Consequently, it is better to leave it <i.e. ὁδεῖν> as a θέμα or as an apocope of ὁδεῖνα, so as to make both the accentuation and the signification agree.”

  • 53 Apollonius, De pronomine 90.12: ‟θέμα γὰρ ἴδιόν ἐστιν ὀξύτονον, οὐχὶ ἔγκλιμα τῆς σφῶιν.”

42Elsewhere he refuses to derive the oxytone Homeric pronoun σφωίν from σφῶιν with a circumflex on the first syllable, saying that the former is “an oxytone θέμα of its own, not an inflexion of σφῶιν.”53. The word used for “inflexion” deserves notice: it is ἔγκλιμα, which only occurs once or twice more in the preserved oeuvre of Apollonius, whereas there are 130 cases of ἔγκλισις. I believe that, wittingly or not, Apollonius is here following Stoic rules, which would prescribe -mat words to indicate both of the alternative classifications of the pronoun.

43So, can a word have been a θέμα in Stoic linguistic theory? Definitely not as a bodily, acustic entity. But ‘θέμα’ may well have designated a status of a word: the status of being a primitive and a foundation for the derivation of other lexemes or word-forms –a status that accrued to the word by a θέσις laying down what particular semantic content should forever after be associated with that particular complex of vocal sounds.

  • 54 Seneca, Epistulae ad Lucilium V.42.1.
  • 55 Diogenes Laertius, Vitae philosophorum VII.192: ‟Περὶ τῆς κατὰ τὰς λέξεις ἀνωμαλίας πρὸς Δίωνα α΄β΄ (...)
  • 56 Varro, De lingua latina IX.1 (= SVF II.45, frgm. 151; FDS F 640; I follow the text in FDS): ‟Crates (...)

44Origines’ description of Stoic naturalism about imposition is clearly too lacking in nuance. But the way the Stoics resorted to etymologies presupposes that they thought the original givers of names did not assign sounds to semantic content on a lottery basis. Were the onomatothetes then wise men? Hardly. Such ones are as rare as the bird phoenix54. However, being rational, the onomatothetes will have striven to create semantically transparent words entering into a systematic pattern. Alas, the result was less than perfect, as Chrysippus was so painfully aware of that he wrote four volumes about linguistic ἀνωμαλία55, that is “unevenness”, an unevenness consisting in the fact that similar things are called by dissimilar names and dissimilar things by similar names, as Varro informs us in De lingua latina56.

  • 57 Simplicius, in Cat. 102.22-24: ‟ καλῶς δὲ εἶπεν ὅτι πᾶσα οὐσία δοκεῖ τόδε τι σημαίνειν, διότι τὴν ἑ (...)

45Now, Varro’s Latin formulation has a Greek counterpart in a passage in Simplicius’ commentary on Aristotle’s Categories57. As opposed to Varro, Simplicius does not on this occasion mention Chrysippus, but the similarity between the two texts is striking:

Varro

Simplicius

Chrysippus de inaequabilitate cum scribit sermonis, propositum habet ostendere similes res dissimilibus verbis et dissimiles similibus esse vocabulis notatas

ἀπαγγέλλει <sc. Aristoteles> καὶ τὴν ἀνωμαλίαν τῆς συνηθείας, ἥτις τὰ ἀνόμοια πράγματα ὁμοίοις σχήμασι λέξεων εἴωθεν πολλάκις σημαίνειν

  • 58 Cf. ἡ τῆς συνηθείας ἀνωμαλία at Simplicius, in Cat. 160.25 and ἡ τῆς χρήσεως ἀνωμαλία at 264.22.
  • 59 Simplicius, in Cat., CAG VIII: 396.19-21: ‟πολλῆς δὲ οὔσης τῆς ἀνωμαλίας Χρύσιππος μὲν ἐν τοῖς Περὶ (...)

46The Greek text’s only serious deviation from the Latin one is that instead of saying that the bearer of the unevenness is sermo, it attributes it to our old acquaintance συνήθεια. Now, Varro’s de inaequabilitate ... sermonis may have been meant as a translation of the title of Chrysippus’ work, which Diogenes Laertius gives as Περὶ τῆς κατὰ τὰς λέξεις ἀνωμαλίας, and the introduction of συνήθεια could be Simplicius’ or his source’s (perhaps Porphyry or Iamblichus) way of twisting the title58, but another passage in Simplicius’ commentary that deals with the unevenness of usage refers to Chrysippus’ Περὶ στερητικῶν, and there are five more references to that work59.

  • 60 Cf. the title Λόγοι παρὰ τὰς συνηθείας in Diogenes Laertius, Vitae philosophorum VII.192.

47I invite you to entertain the idea that the formulation is Chrysippus’ own60. If we then supplement what is known about the Stoics with some key elements of what the commentators say about the institution of names, we reach the result that the reason why ordinary language is not semantically as transparent as we might wish for, is that it is the result of institutions performed by ordinary people, who, even though some of them were bright, failed to co-ordinate their efforts.

  • 61 Thus also κρίσις/κρῖμα, κατόρθωσις/κατόρθωμα, φρόνησις/φρόνημα, φρονίμευσις/φρονίμευμα. See the ind (...)

48Hence, there is a job left for us, Stoic philosophers could think: we must iron out the unevenness and make sure anyone can see when we are talking of a corporeal entity, which may have a role as a cause, and when we are talking about the incorporeal effects of such a cause. One way to make this clear is to reserve the suffix -sis for names of bodies and their actions, while using the suffix -mat as a marker of the incorporeal effects of such actions by bodily actors61. Thus we end up with the couple θέσις/θέμα as well as a host of other wonderfully transparent technical terms that will help mankind make some progress (προκοπή) on its never-ending journey towards wisdom.

49We just have to hope that the fools in the other philosophical schools do not manage to mess up the vocabulary again after we have tidied it up so nicely!

Haut de page

Bibliographie

Allen, James (2005), “The Stoics on the Origin of Language and the Foundations of Etymology”, in Dorothea Frede & Brad Inwood (eds.), Language and Learning. Philosophy of Language in the Hellenistic Age, Cambridge, Cambridge University Press.

Bonitz, Hermann (1870), Index Aristotelicus (= Aristotelis Opera V), Berlin, Reimer.

Ebbesen, Sten, Commentators and Commentaries on Aristotle’s Sophistici Elenchi. A Study of Post-Aristotelian Ancient and Medieval Writings on Fallacies, 3 vols., = Corpus Latinum Commentariorum in Aristotelem Graecorum VII.1-3, Leiden, Brill, 1981.

Lo Piparo, Franco (2003), Aristotele e il linguaggio, Roma-Bari, Laterza.

Long, Anthony A. (2005), “Stoic Linguistics, Plato’s Cratylus and Augustine’s De dialectica”, in Dorothea Frede & Brad Inwood (eds.), Language and Learning. Philosophy of Language in the Hellenistic Age, Cambridge, Cambridge University Press.

Pohlenz, Max (1959), Die Stoa, Göttingen, Vandenhoeck & Ruprecht (here quoted from 4. Auflage, 1970).

Schmidt, Rudolf (1839), Stoicorum grammatica, Halle (rp. Amsterdam, Hakkert, 1967).

Steinthal, Heymann (18902), Geschichte der Sprachwissenschaft bei den Griechen und Römern, Berlin, Dümmler (rp. Hildesheim-New York, Georg Olms, 1971).

Haut de page

Notes

1 Ancient literature is cited without indication of edition used, as long as the authors involved are well-known persons like Plato and Diogenes Laertius and there are no particular textual problems. When quoting Latin texts I do not necessarily follow the orthography of the edition used. Translations from the Greek are my own. Abbreviations of titles of Greek works generally follow the practice of Liddell-Scott-Jones, A Greek-English Lexicon, Oxford, Clarendon Press, 1940. The following abbreviations are used for major series and collections of fragments: CAG = Commentaria in Aristotelem Graeca, Berlin, Reimer, 1882-1907; GG = Grammatici Graeci, Leipzig, Teubner, 1867-1889; GL = Grammatici Latini, Leipzig, Teubner, 1855-1880; FDS = Karlheinz Hülser, Die Fragmente zur Dialektik der Stoiker, 4 vols., Stuttgart-Bad Cannstatt, Frommann-Holzboog, 1987-1988; SVF = Hans von Arnim, Stoicorum Veterum Fragmenta, 4 vols., Leipzig, Teubner, 1905-1924.

2 Plato, Cratylus 401b.

3 Used by Alcinous, Ammonius, Philoponus and Simplicius, among others. See the electronic Thesaurus Linguae Graecae.

4 Aristotle’s in De interpretatione 2.16a19-29 stresses that names owe their signification to συνθἠκη and not to φύσις, and only puts one restriction on the choice of phonetic material: noises that cannot be represented in letters (ἀγράμματοι ψὀφοι) are not allowed. I am unconvinced by arguments that κατὰ συνθήκην in the context equals κατὰ σύνθεσιν, as has repeatedly been suggested (for a spirited defence of the view, see Lo Piparo, Franco (2003), Aristotele e il linguaggio, Roma-Bari, Laterza).

5 For the latter, see Plato, Cratylus 384d.

6 Bonitz, Hermann (1870), Index Aristotelicus (= Aristotelis Opera V), Berlin, Reimer, p. 327b, s.v. θέσις mentions a possible φύσει / θέσει opposition in Aristotle, Physics III.5.205b34, but the θέσις mentioned there is probably a location. Alexander of Aphrodisias, in Top., CAG II.2: 82.13-16 wrongly attributes the dichotomy to Aristotle. Strictly speaking, Aristotle never says νόμωι about the value of linguistic items, but in Nicomachean Ethics V.8.1133a28-31 he says that money is what it is κατὰ συνθήκην ‟by convention”, and this is why it is called νόμισμα, because it is money νόμωι and not φύσει. This is closely parallelled in De interpretatione 2.16a27-28, where it is said that κατὰ συνθήκην is included in the definition of a name (noun) because no name is φύσει.

7 Thus already in Epicurus, Epistula ad Herodotum, apud Diogenes Laertius, Vitae philosophorum X.75. Laws, ethical norms: Philo, De ebrietate 34; Legum allegoriae III.126 (and elsewhere). Syllables long φύσει or θέσει: Dionysius Thrax, Ars grammatica, GG I.1.17-18.

8 Simplicius, in Cat., CAG VIII: 40.5-13: ‟Διὰ τί δὲ Ἀρχύτας παραλέλοιπεν ταύτην τὴν περὶ τῶν ὀνομάτων διδασκαλίαν ἐν τῷ Περὶ τῶν καθόλου λόγων; ἢ ἐπειδὴ φύσει τὰ ὀνόματα καὶ οὐ θέσει λέγουσιν οἱ Πυθαγόρειοι, καὶ τὰ ὁμώνυμα καὶ τὰ πολυώνυμα παραιτοῦνται, ὡς ἑνὸς ὀνόματος πρὸς ἓν πρᾶγμα κατὰ φύσιν λεγομένου. διακρίνειν οὖν ἐκείνους εἰκὸς τὰ μὲν ὁμώνυμα τῇ προγονικῇ ἢ κατ’ ἐλπίδα ἀναφορᾷ, τὰ δὲ πολυώνυμα δείξουσιν οὐ καθ’ ἓν λεγόμενα, ἀλλὰ κατὰ διαφόρους ἐτυμολογίας ἀποδιδόμενα, ὅταν ὄντως ὀνόματα ᾖ· τῶν δὲ παρωνύμων τὸν παρασχηματισμὸν οἰκείως ἐκεῖνοι ταῖς κατὰ τὰ πράγματα συζυγίαις ἐξηγήσονται.” Simplicius attributes the naturalist view to the original Pythagoreans, but his picture of the sect was one painted for him by late ancient Platonists who considered themselves Pythagoreans. He explains how they might defend their denial of genuine polyonymy by assuming a different etymology underlying each name, i.e. that each name highlights a different feature of the thing signified. The same strategy is used by Anonymus Domus Petri 205, Commentarium in librum Divisionum Boethii (mid 13th century), ms Cambridge, Peterhouse 205: 54rA: ‟licet una et eadem vox sit communis multis significationibus, non tamen est possibile recipere unam et eandem significationem communem pluribus vocibus, una tamen <res> significari potest per plures voces, non tamen una et eadem est significatio; secundum enim diversas rationes imponebantur illae voces ad significandum illam rem, et ita sunt diversae significationes. Unde sicut <trium> filiorum respectu eiusdem patris non est eadem relatio sed diversa, eodem modo trium vocum respecu unius rei non est eadem significatio sed diversa.”

9 Philo Iudaeus, De cherubim 56: “ὁ μὲν ἄλλος ἅπας ἀνθρώπων ὅμιλος ὀνόματα τίθεται πράγμασι διαφέροντα τῶν πραγμάτων, ὥσθ' ἕτερα μὲν εἶναι τὰ τυγχάνοντα, ἑτέρας δὲ κλήσεις τὰς ἐπ' αὐτοῖς· παρὰ Μωυσεῖ δὲ αἱ τῶν ὀνομάτων θέσεις ἐνάργειαι πραγμάτων εἰσὶν ἐμφαντικώταται, ὡς αὐτὸ τὸ πρᾶγμα ἐξ ἀνάγκης εὐθὺς εἶναι τοὔνομα καὶ <τοὔνομα καὶ> καθ' οὗ τίθεται διαφέρειν μηδέν.”

10 Diodorus Siculus, Bibliotheca historica V.67.3: ‟Τῶν δὲ Τιτανίδων φασὶ Μνημοσύνην λογισμοὺς εὑρεῖν καὶ τὰς τῶν ὀνομάτων θέσεις ἑκάστῳ τῶν ὄντων τάξαι, δι’ ὧν καὶ δηλοῦμεν ἕκαστα καὶ πρὸς ἀλλήλους ὁμιλοῦμεν· ἅ τινες τὸν Ἑρμῆν φασιν εἰσηγήσασθαι. Προσάπτουσι δὲ τῇ θεῷ ταύτῃ καὶ τὰ πρὸς ἀνανέωσιν καὶ μνήμην γινόμενα παρὰ τοῖς ἀνθρώποις, ἀφ’ ὧν δὴ καὶ τῆς προσηγορίας τυχεῖν αὐτὴν ταύτης.”

11 Aristotle, De interpretatione 3.16b19-20: ‟αὐτὰ μὲν οὖν καθ' αὑτὰ λεγόμενα τὰ ῥήματα ὀνόματά
ἐστι καὶ σημαίνει τι”.

12 About function words, see S. Ebbesen, Commentators and Commentaries on Aristotle’s Sophistici Elenchi. A Study of Post-Aristo­tel­ian Ancient and Medieval Writings on Fallacies, 3 vols., = Corpus Latinum Commen­tario­rum in Aristotelem Graecorum VII.1-3, Leiden, Brill, 1981, I: 151 with source references in note 29 on p. 158. About Apollonius Dyscolus’ ideas about the relation between the imposition of nouns and verbs (the primary, reality-oriented parts of speech) and that of the remaining parts of speech, see Luhtala, Anneli (2011), “Imposition of Names in Ancient Grammar and Philosophy” in Stephanos Matthaios, Franco Montanari & Antonios Rengakos (eds.), Ancient Scholarship and Grammar. Archetypes, Concepts and Contexts, Berlin, W. De Gruyter, p. 479-498.

13 See, e.g., Flavius Sosipater Charisius, Ars grammatica, Karl Barwick (ed.), Leipzig, Teubner, 1925 (rp. 1964), V, p. 380: ‟cum ab omni sermone Graeco Latina lingua pendere videatur, quaedam inveniuntur vel licentia ab antiquis vel proprietate linguae Latinae dicta praeter consuetudinem Graecorum.”

14 See Dante’s De vulgari eloquentia.

15 See, e.g., Schmidt, Rudolf (1839), Stoicorum grammatica, Halle (rp. Amsterdam, Hakkert, 1967), 21ff. Somewhat more nuanced, Steinthal, Heymann (18902), Geschichte der Sprachwissenschaft bei den Griechen und Römern, Berlin, Dümmler (rp. Hildesheim-New York, Georg Olms, 1971), I: 328ff. Also more nuanced, but still appealing to Origenes, Allen, James (2005), ‘The Stoics on the Origin of Language and the Foundations of Etymology’ in Dorothea Frede & Brad Inwood (ed.), Language and Learning. Philosophy of Language in the Hellenistic Age, Cambridge, Cambridge University Press, p. 14-35, and Long, Anthony A. (2005), ‘Stoic Linguistics, Plato’s Cratylus and Augustine’s De dialectica’ in the same book, p. 36- 55.

16 Pohlenz, Max (1959), Die Stoa, Göttingen, Vandenhoeck & Ruprecht (here quoted from 4. Auflage, 1970), I: 41. Cf. Pohlenz’ notes in II.23-24.

17 I suspect a printing error, an indefinite article having fallen out, so that the text should read ‟durch einen bestimmten Willensakt”.

18 Origenes, Contra Celsum 1.24 (= FDS F 643): ‟Λεκτέον δὲ καὶ πρὸς τοῦτο ὅτι ἐμπίπτει εἰς τὸ προκείμενον λόγος βαθὺς καὶ ἀπόρρητος, περὶ φύσεως ὀνομάτων· πότερον, ὡς οἴεται Ἀριστοτέλης, θέσει εἰσὶ τὰ ὀνόματα , ὡς νομίζουσιν οἱ ἀπὸ τῆς Στοᾶς, φύσει, μιμουμένων τῶν πρώτων φωνῶν τὰ πράγματα, καθὧν τὰ ὀνόματα, καθὸ καὶ στοιχεῖά τινα τῆς ἐτυμολογίας εἰσάγουσιν, , ὡς διδάσκει Ἐπίκουρος, ἑτέρως ὡς οἴονται οἱ ἀπὸ τῆς Στοᾶς, φύσει ἐστὶ τὰ ὀνόματα, ἀπορρηξάντων τῶν πρώτων ἀνθρώπων τινὰς φωνὰς κατὰ τῶν πραγμάτων.” The main part of the text is also reproduced in SVF II: 44, frgm. 146. Another important source usually adduced to support the notion of Stoic naturalism concerning the imposition of names is Augustine’s De dialectica, which (no doubt correctly) is assumed to contain Stoic material transmitted via Varro’s Disciplinarum libri.

19 See, e.g. Cicero, De natura deorum III.xxiv.63 (= FDS F 606) and the examples of Stoic etymologizing assembled in FDS F 650-680. Not explicitly ascribed to the Stoics, but in my opinion indisputably of Stoic ancestry, is Apollonius Dyscolus’ definition of a noun (ὄνομα) that includes a derivation from νέμει. See Scholia in Dionysii Thracis Artem grammaticam, GG I.3: 524: ‟Ἰστέον δὲ ὅτι οἱ περὶ Ἀπολλώνιον καὶ Ἡρωδιανὸν οὕτως ὁρίζονται τὸ ὄνομα· ὄνομά ἐστι μέρος λόγου πτωτικόν, ἑκάστῳ τῶν ὑποκειμένων σωμάτων ἢ πραγμάτων κοινὴν ἢ ἰδίαν ποιότητα ἀπονέμον.”

20 Plato, Cratylus 396a-b: ‟ἔστι δὲ οὐ ῥᾴδιον κατανοῆσαι. ἀτεχνῶς γάρ ἐστιν οἷον λόγος τὸ τοῦ Διὸς ὄνομα, διελόντες δὲ αὐτὸ διχῇ οἱ μὲν τῷ ἑτέρῳ μέρει, οἱ δὲ τῷ ἑτέρῳ χρώμεθα –οἱ μὲν γὰρ “Ζῆνα,” οἱ δὲ “Δία” καλοῦσιν– συντιθέμενα δ’ εἰς ἓν δηλοῖ τὴν φύσιν τοῦ θεοῦ, ὃ δὴ προσήκειν φαμὲν ὀνόματι οἵῳ τε εἶναι ἀπεργάζεσθαι. οὐ γὰρ ἔστιν ἡμῖν καὶ τοῖς ἄλλοις πᾶσιν ὅστις ἐστὶν αἴτιος μᾶλλον τοῦ ζῆν ἢ ὁ ἄρχων τε καὶ βασιλεὺς τῶν πάντων. συμβαίνει οὖν ὀρθῶς ὀνομάζεσθαι οὗτος ὁ θεὸς εἶναι, δι’ ὃν ζῆν ἀεὶ πᾶσι τοῖς ζῶσιν ὑπάρχει”. Chrysippus is reported to have adopted Plato’s etymology: see Stobaeus, Eclogae I.1.26 (= FDS 563):  ‟Χρυσίππου· Ζεὺς μὲν οὖν φαίνεται ὠνομάσθαι ἀπὸ τοῦ πᾶσι δεδωκέναι τὸ ζῆν· Δία δὲ αὐτὸν λέγουσιν, ὅτι πάντων ἐστὶν αἴτιος καὶ δι’ αὐτὸν πάντα.”

21 There is a long, somewhat rambling, discussion supporting the compromise view in Ammonius, in Int., CAG IV.5: 34-37.

22 For words in the category of when, see Simplicius, in Cat., CAG VIII: 350.4-5: ‟διὰ τοῦτο δὲ οἱ ὀνοματοθέται ἐπὶ τοῦ ποτὲ ὀνόματα διάφορα τεθείκασι, τὸ χθὲς καὶ αὔριον καὶ πέρυσιν καὶ τῆτες καὶ εἰς νέωτα”.

23 Simplicius, in Cat., CAG VIII: 15.8-12: ‟ πρώτη θέσις τῶν φωνῶν ὡς ὀνομάτων ἐστίν· δεηθέντες γὰρ οἱ ἄνθρωποι σημαίνειν ἀλλήλοις τὰ πράγματα διὰ τὸ τῆς κοινῆς ἀποστῆναι νοήσεως ἔθεντο τῷδε μὲν ὄνομα σῶμα, τῷδε δὲ δεξιόν, καὶ ἄλλῳ ποιεῖν πάσχειν, οὔπω διαφορᾶς οὔσης, καθ' ἣν τὰ μὲν ὑπόκειται, τὰ δὲ κατηγορεῖται καὶ τὰ μὲν προσσημαίνει χρόνον, τὰ δὲ οὔ.”

24 Ammonius, in Cat., CAG IV.4: 11.8 ff. Ammonios offers a hopelessly naive picture of the first imposition of names as an occasion when people gathered and decided what things were to be called. The same ridiculous scenario reappears in Olympiodorus, Prolegomena, CAG XII.1: 21.26-38: ‟ἀλλ' ἵνα μάθωμεν τίς ἡ πρώτη θέσις καὶ τίς ἡ δευτέρα θέσις, ἔνθεν ἀρξώμεθα· ἰστέον τοίνυν ὅτι ἡ φύσις κοινωνικὸν τὸν ἄνθρωπον ἐβούλετο προαγαγεῖν· οὐκοῦν τοῦτον ἔχουσα τὸν σκοπὸν καὶ τὸ διαλέγεσθαι αὐτῷ δεδώρηται, ἵνα σημᾶναι δύνηται τοῖς πέλας τὰ πράγματα. οὐκ ἂν δὲ κοινωνεῖν οὕτως ἠδύνατο, εἰ μὴ καὶ κατὰ τὰς φωνὰς ἡ κοινωνία ἦν, ὃν τρόπον οὐδὲ οἱ διαφόρους ἔχοντες νόμους τῆς αὐτῆς εἶναι δύνανται πολιτείας. κοινωνία δὲ τῶν φωνῶν ἔνθεν ἐγένετο· χορὸς σοφῶν ἀνδρῶν συνελθὼν ἑκάστῳ πράγματι ἴδιον ἀφώρισεν ὄνομα, καὶ τὸ μὲν ἐκάλεσαν οὐσίαν τὸ δὲ συμβεβηκός, καὶ ἁπλοῖς πράγμασιν ἁπλᾶ ἔθεντο τὰ ὀνόματα, καὶ οὕτω παρὰ πᾶσιν χαρακτηρίζεται. αὕτη λέγεται πρώτη θέσις. εἶτα ἐπειδὴ ἑώρων τὰ μὲν τῶν ὑπ' αὐτῶν τεθέντων ὀνομάτων ἐπιδεχόμενα ἄρθρα, τὰ δὲ μὴ ἐπιδεχόμενα ἀλλὰ καὶ χρόνον σημαίνοντα, ἐκάλουν τὰ μὲν ὀνόματα τὰ δὲ ῥήματα. καὶ ἐκάλουν ταύτην δευτέραν θέσιν.” Olympiodorus has misunderstood his source, which must have said that the first imposition created words for both substances and accidents. In his account this become that it introduced the technical terms ‘substance’ and ‘accident’, which, of course, are second imposition words.

25 Dexippus, in Cat., CAG IV.2: 11.14-15: ‟προηγουμένως ἡ σημαντικὴ λέξις τῆς πρώτης ἔχεται χρείας τοῦ λόγου, καθ’ ἣν τὰ πράγματα δηλοῦν ἀλλήλοις ἐφιέμεθα”; 15.24: “Τῆς δευτέρας τοῦ λόγου χρείας ἔχεται καὶ ταῦτα.‟

26 Thus in Philoponus, in Cat., CAG XIII.1: 85.23.

27 There is a beautiful book in Danish about Quintilian’s use of ancient theories of language: Pinborg, Jan (1963), Quintilian og den antikke sprogteori, Studier fra Sprog- og Oldtidsforskning 253, København, Gads Forlag.

28 Quintilian, Institutio oratoria I.6.43-45. Notice his concluding remark: ‟Ergo consuetudinem sermonis vocabo consensum eruditorum, sicut vivendi consensum bonorum.”

29 Simplicius, in Cat., CAG VIII: 19.26-20.5. The context is a discussion of why Aristotle put the antepraedicamenta in front of the main body of his text and the postpraedicamenta after it, although the purpose of both passages is to clarify the meaning of some words that he needed to use in the main text: “τί οὖν μὴ καὶ ταῦτα [the postpraedicamenta] προὔταξεν τοῦ περὶ τῶν κατηγοριῶν λόγου ἢ διὰ τί μὴ κἀκεῖνα [the antepraedicamenta] μετὰ τὰς κατηγορίας ἐπήγαγεν ἀπορήσας ὁ Πορφύριος λύει καλῶς τὴν ἀπορίαν· ὅτι ἐκεῖνα μὲν ἄγνωστα παντελῶς ἦν ἀπὸ τῆς συνηθείας, τούτων δὲ ἐννοίας μὲν ἔχομέν τινας, οὐ μὴν ἀκριβῆ διάκρισιν οὐδὲ αὐτῶν · μὴ βουλόμενος οὖν ἐπὶ πολὺ τὸν περὶ αὐτῶν τῶν κατηγοριῶν ἀναβάλλεσθαι λόγον ὑπερέθετο ταῦτα εἰς ὕστερον, οὔτε προδιδάξαι αὐτὰ ἀναγκασθεὶς διὰ τὴν ὁποιανοῦν προϋπάρχουσαν αὐτῶν ἔννοιαν οὔτε ἀτεχνῆ τὴν περὶ αὐτῶν γνῶσιν καταλείπεσθαι τῷ ἀκροατῇ συγχωρῶν· εἰ γὰρ καὶ χρώμεθά τισιν ἀπὸ τῆς συνηθείας ὀνόμασιν, ἀλλ’ οὐ γινώσκοντες
αὐτῶν τὰς φύσεις ἀκριβῶς, εἰς ἃς οἱ ὀνοματοθέται βλέποντες τεθείκασιν
.”

30 Simplicius, in Cat., CAG VIII: 187.8-9: “εἰ μὲν θέσει τὰ ὀνόματα, στοχαστέον τοῦ σκοποῦ τῆς θέσεως· εἰ δὲ φύσει, προσέχειν τῇ φύσει τοῖς οἰκείοις πράγμασιν τοὺς οἰκείους φθόγγους συντάττοντας.”

31 See note 19, above.

32 Philoponus, in Cat., CAG XIII.1: 157: 7-13: ‟oὔτε γὰρ ὁ δρομικὸς οὔτε ὁ πυκτικός, οἱ φυσικήν τινα ἐπιτηδειότητα ἐν τούτοις ἔχοντες, παρωνύμως ἀπὸ τούτων τῶν δυνάμεων λέγονται· οὐδὲ γὰρ κεῖται ὀνόματα ταῖς δυνάμεσι ταύταις διὰ τὸ τὴν συνήθειαν τοῖς τελείοις τῶν πραγμάτων καὶ κατ’ ἐνέργειαν οὖσιν ἐσπουδακέναι θεῖναι τὰ ὀνόματα· τὸ γὰρ πυκτικὸς ἢ δρομικὸς ἀπὸ τῶν κατ’ ἐνέργειαν ἐπιστημῶν, τῆς τε πυκτικῆς λέγω καὶ δρομικῆς, ὀνομάζονται.”

33 Philoponus, in Cat., CAG XIII.1: 113.19 ff.: ‟λέγει οὖν <sc. ο Ἀριστοτέλης> πρὸς τοῦτο ὅτι οὐδὲν χαλεπὸν οἷς μὴ ἐφρόντισε θεῖναι ὀνόματα ἡ συνήθεια, ἐκείνοις ἡμᾶς αὐτοὺς ἐπινοεῖν· ἡ γὰρ συνήθεια ἐκείνοις τίθησιν ὀνόματα οἷς γινώσκει, αἱ δὲ τέχναι ἅτε δὴ καινοτέρων πραγμάτων οὖσαι εὑρέτιδες ὀφείλουσι τοῖς πράγμασι τοῖς ὑπ’ αὐτῶν εὑρισκομένοις τιθέναι ὀνόματα πρὸς τὴν τούτων σημασίαν· οἷον ὁ γεωμέτρης [...] καὶ ὁ μουσικὸς ὁμοίως [...]”. Simplicius, Cat. 187.1-4: ‟ἔτι δὲ πάσης τέχνης οἰκεῖον τὸ ὀνοματοποιεῖν· καὶ γὰρ ὁ γεωμέτρης καὶ ὁ μουσικὸς περὶ πράγματα τοῖς πολλοῖς ἄγνωστα διατρίβοντες ἴδια τιθέναι τοῖς πράγμασιν ἐκείνοις ὀνόματα καὶ ἀσυνήθη τοῖς πολλοῖς ἀναγκάζονται.”

34 See Dexippus, in Cat., CAG IV.2: 26.29-27.2. In 15.23-16.13 Dexippus further includes ‘whole’ and ‘part’ and several other words among those resulting from a second imposition: ‟Τὸ ὅλον καὶ τὸ μέρος εἰς ποίαν κατηγορίαν τάξομεν; –Τῆς δευτέρας τοῦ λόγου χρείας ἔχεται καὶ ταῦτα. χεὶρ μὲν γὰρ καὶ κεφαλὴ πρῶτα ὀνόματα, ὅτι δὲ ἡ χεὶρ καὶ ἡ κεφαλὴ μέρη ἐστὶν ὅλου καὶ ὡς μέρη ὀνομάζονται, δευτέρα κίνησις λεκτικῆς μηνύσεως· Σωκράτης μὲν γὰρ τῆς οὐσίας προσηγορία, ὅτι δὲ τὸ Σωκράτης ὄνομα, δευτέρα θέσις ὀνόματος ἐπὶ τῇ πρώτῃ. οὕτως οὖν ἔχει καὶ τὸ ὅλον καὶ τὸ μέρος, καὶ διὰ τοῦτο παραιτητέον εἰς τὰς κατηγορίας αὐτὸ ἀνάγειν· εἰ δ’ ἄρα, πρός τι τοῦτο τάξομεν· ὅλου γὰρ μέρος τὸ μέρος καὶ τὸ ὅλον ἐκ τῶν μερῶν ὅλον. –Τὸ καθόλου καὶ τὸ καθάπαξ καὶ τὸ ἄρδην εἰς ποίαν κατηγορίαν ἀνάξομεν;  –Ταῦτα τῶν κατηγοριῶν ἐξῄρηται· κατὰ μετοχὴν γὰρ πρώτων τινῶν καὶ ἀρχετύπων κατηγοριῶν ἀναφέρεται. –Τὸ ταὐτὸν καὶ τὸ ἕτερον καὶ τὸ ἴδιον καὶ τὸ ἀλλότριον καὶ τὸ οινὸν καὶ τὴν ἀδιαφορίαν τήν τε ἀμφιβολίαν καὶ τὴν συνωνυμίαν ποῦ τάξομεν; –Εἴρηται μὲν ὅτι κατὰ πάσας ἕκαστον τούτων λέγεται· ἐπεὶ δὲ ὀνομάτων ἐστὶ καὶ τῆς δευτέρας ὄντα θέσεως τῶν ὀνομάτων φαίνεται, τῶν κατηγοριῶν ἐκπίπτει. ἤδη γὰρ πολλάκις εἰρήκαμεν, ὅτι αἱ κατηγορίαι κατὰ τὸ σημαντικὸν τῶν ἐν τοῖς οὖσι διαφορῶν ὑφίστανται· διὸ καὶ μᾶλλον ἀξιοῦμεν τὰ ἄτομα καὶ ἔχοντα δεῖξιν θεωρεῖν ὡς κυριώτερα ἐν ταῖς κατηγορίαις καὶ μᾶλλον ἀποπληροῦν αὐτῶν τὴν ἐπαγγελίαν δυνάμενα. τούτων τοίνυν διωρισμένων ἄξιον ἐπιστῆσαι πρὸς τὴν τελεωτάτην ἐπιστήμην ἀναφέροντας αὐτῶν τὴν εἴδησιν.”

35 Simplicius, in Cat., CAG VIII: 187.5-6: ‟ἢ τοῖς κειμένοις ἤδη αὐτοὶ χρώμενοι ἐπ’ ἄλλοις, ὡς τῷ χρώματι καὶ τῇ διέσει ὁ μουσικὸς καὶ τῷ κέντρῳ ὁ γεωμέτρης.”

36 Trypho, De tropis, in Rhetores Graeci, Leonhard Spengel (ed.), Leipzig, Teubner, 1856, III, p. 191: “κυριολογία μὲν οὖν ἐστιν ἡ διὰ τῆς πρώτης θέσεως τῶν ὀνομάτων τὰ πράγματα σημαίνουσα <sc. φράσις>”.

37 Simplicius, in Cat., CAG VIII: 186.34-187.1: ‟καὶ γὰρ ἀσφαλέστατος τρόπος ὀνοματοποιίας ὁ ἀπὸ τῶν ἤδη κειμένων καὶ τῆς κρατούσης συνηθείας· καὶ γὰρ οἱ πρῶτοι ὀνοματοθέται φαίνονται ἀπὸ τῶν συνήθων καὶ ἁπλουστέρων τὰ ἄλλα συντιθέντες, ὡς αἱ ἐτυμολογίαι δηλοῦσι, καὶ μέντοι ὁ οὕτως ποιῶν περὶ τὴν λέξιν ὀλίγον καινίζει, τὴν συνήθη περὶ τῶν πραγμάτων ἔννοιαν φυλάττων.” The remark falls in a longish note on Cat. 7.7a5-7, where Aristotle excuses his neologisms πτερωτόν and πηδαλιωτόν with the remark that sometimes it may be necessary to make up a name (ὀνοματοποιεῖν) when there is no established one —ἐὰν μὴ κείμενον ᾖ ὄνομα: notice that κείμενον functions as a perfect passive participle of τίθεσθαι.

38 Apollonius, Pron. 26.6-8: ‟οὐκ ἄλογον δὲ τὸ τὴν τῶν ὀνομάτων ἐκφορὰν ἐν τρίτοις νοεῖσθαι, ἐπεὶ αἱ ἴδιαι θέσεις τῶν ὀνομάτων οὐ τὸν πρός τινα λόγον ἔχουσιν, ἀλλὰ τὸν περί τινος.” The same point is made in Synt. II.43, p. 156.

39 Priscian, Partitiones 68, GL III: 474-5: ‟quaedam enim adverbiorum sunt πρωτότυπα, id est primitiva vel primae positionis, quaedam derivativa; primitiva, ut ‘saepe’ ‘satis’, derivativa ut ‘saepius’ ‘satius’.” Cf. Inst. IV.iii.18, GL II: 127. Quintilian (Inst. I.v.60, I.vi.10-11) and several Latin grammarians use prīma posítiō about the basic form of a lexeme, which in practice means the citation form. In, Inst. I.v.65 Quintilian contrasts simple words, which ‟keep to their first imposition, i.e. their nature” (‟prima positione, id est natura sua, constant”) with compound ones.

40 Priscian, Inst. VIII.xii.63, GL II: 421-2: ‟quia prima positio verbi, quae videtur ab ipsa natura esse prolata, in hoc est modo <sc. indicativo>, quemadmodum in nominibus est casus nominativus”.

41 Stobaeus, Eclogae I.13.1c, p. 138 (= SVF I: 25, frgm. 89; FDS F 762): ‟αἴτιον δ’ ὁ Ζήνων φησὶν εἶναι δι’ ὅ· οὗ δὲ αἴτιον συμβεβηκός· καὶ τὸ μὲν αἴτιον σῶμα, οὗ δὲ αἴτιον κατηγόρημα.”

42 Diogenes Laertius, Vitae philosophorum VII.58 (= SVF III.213, frgm. 22; FDS F 536).

43 E.g., Plutarch, De communibus notitiis 30, 1074D; Sextus Empiricus, Adversus Mathematicos IX.211.

44 For the difference between λέξις and λόγος see Diogenes Laertius, Vitae Philosophorum VII.56-57 (= SVF III.213, frgm. 20; FDS F 476).

45 For these θέματα, see Frede, Michael (1974), Die stoische Logik, Göttingen, Vandenhoeck & Ruprecht, p. 172 ff.

46 Sextus Empiricus, Adversus Mathematicos VIII.301: ‟ἀπόδειξις κατὰ μὲν τὸ γένος ἐστὶ λόγος· οὐ γὰρ δή γε αἰσθητὸν ἦν πρᾶγμα, ἀλλὰ διανοίας τις κίνησις καὶ συγκατάθεσις, ἅπερ ἦν λογικά.”

47 Sextus Empiricus, Adversus Mathematicos VIII.301-302: λόγος δέ ἐστιν, ὡς ἁπλούστερον εἰπεῖν, τὸ συνεστηκὸς ἐκ λημμάτων καὶ ἐπιφορᾶς. λήμματα δὲ καλοῦμεν οὐ θέματά τινα, συναρπάζομεν, ἀλλἅπερ προσδιαλεγόμενος τῷ ἐμφανῆ εἶναι δίδωσι καὶ παραχωρεῖ.”

48 One non-logical instance in Poetics 1457b7, and two in De spiritu (482b10, 483a14).

49 The description of an argument in first period of the passage from Sextus Empiricus quoted two notes back is virtually identical to the one found in Diogenes Laertius VII.45, which guarantees that it is Stoic. All manuscripts of Diogenes offer εἶναι δὲ λόγον αὐτὸν συστημάτων καὶ ἐπιφορᾶς, but already in 1815 Beier corrected συστημάτων into σύστη<μα ἐκ λημ>μάτων καὶ ἐπιφορᾶς on the basis of a comparison with Sextus Empiricus, Pyrrhoneae hypotyposes II.135, where the definition is cited without attribution to any particular school. There thus is a certain circularity in the proof of the definition’s Stoic credentials, but not one that can justify serious doubt.

50 Sextus Empiricus, Adversus Mathematicos I.148-153: ‟ τὰ δὲ αὐτὰ ὀνόματα οὐ πᾶσίν ἐστι τὰ αὐτά, ἀλλὰ τοῖς μὲν ἀρρενικὰ τοῖς δὲ θηλυκὰ τοῖς δὲ οὐδέτερα· οἷον Ἀθηναῖοι μὲν τὴν στάμνον λέγουσι θηλυκῶς, Πελοποννήσιοι δὲ τὸν στάμνον ἀρρενικῶς, καὶ οἱ  μὲν τὴν θόλον οἱ δὲ τὸν θόλον, καὶ οἱ μὲν τὴν βῶλον οἱ δὲ τὸν βῶλον, καὶ οὐ διὰ τοῦτο οὗτοι ἢ ἐκεῖνοι λέγονται ἁμαρτάνειν· ἕκαστος γάρ, ὡς τεθεμάτικεν, οὕτω χρῆται. καὶ οἱ αὐτοὶ δὲ διαφόρως ταὐτὰ ὁτὲ μὲν ἀρρενικῶς ἐκφέρουσιν ὁτὲ δὲ θηλυκῶς, λέγοντες τὸν λιμόν καὶ τὴν λιμόν. οὐκ ἄρα φύσει τῶν ὀνομάτων τὰ μὲν ἀρρενικὰ τὰ δὲ θηλυκά, ἀλλὰ κατὰ θεματισμὸν τὰ μὲν τοιαῦτα γίνεται τὰ δὲ τοιαῦτα. [...] τοίνυν εἰ φύσει οὐδέν ἐστιν ἀρρενικὸν ἢ θηλυκὸν ὄνομα, ζητῶ πῶς ὁ γραμματικὸς ἐπιλήψεται τοῦ διαστρόφως λέγοντος ὁ χελιδών καὶ ἡ ἀετός. ἤτοι γὰρ ὡς φύσει τοῦ ὀνόματος τῆς χελιδόνος θηλυκοῦ ὄντος, ἐκείνου δὲ ἀρρενικὸν αὐτὸ τῷ ἄρθρῳ βιαζομένου γενέσθαι, ἢ ὡς τῆς κοινῆς συνηθείας θηλυκὸν αὐτὸ θεματισάσης ἀλλ’ οὐκ ἀρρενικόν. ἀλλ' εἰ μὲν ὡς φύσει θηλυκοῦ καθεστῶτος, ἐπεὶ οὐδὲν φύσει θηλυκόν ἐστι καθὼς παρεστήσαμεν, ἀδιάφορον τὸ ὅσον ἐπὶ τούτῳ ἐάν τε οὕτως ἐάν τε ἐκείνως ἐκφέρηται· εἰ δ' ὡς ὑπὸ τῆς κοινῆς συνηθείας ἀντὶ θηλυκοῦ θεματισθέν, γενήσεται τοῦ τε εὖ λεγομένου καὶ μὴ κριτήριον οὐχὶ τεχνικός τις καὶ γραμματικὸς λόγος ἀλλ’ ἡ ἄτεχνος καὶ ἀφελὴς τῆς συνηθείας παρατήρησις.” Id., op. cit., VIII.202: ‟διαφέρει τοίνυν τοῦ ὑπομνηστικοῦ σημείου τὸ ἐνδεικτικόν, καὶ οὐ μεταβατέον ἐστὶν ἀπ' ἐκείνου ἐπὶ τοῦτο, παρόσον τὸ μὲν ἑνὸς μόνου δεῖ μηνυτικὸν ὑπάρχειν, τὸ δὲ πλειόνων δύναται εἶναι παραστατικὸν καὶ ὡς ἂν ἡμεῖς θεματίσωμεν σημαίνειν.”

51 GG III.3: 211a.

52 Apollonius, De pronomine, GG II.1: 60.7-8: ‟ἄμεινον οὖν ἦν θέμα καταλιπεῖν, ἢ ἀποκοπὴν τοῦ ὁδεῖνα, ἵνα καὶ τὰ τοῦ τόνου συμφωνῇ καὶ τὰ τῆς σημασίας.” The manuscripts have an addition, which Schneider excises because it does not fit into the syntax, but which might actually be a parenthetical afterthought due to Apollonius himself: ‟βέλτιον θέμα, ἐπεὶ καὶ αἱ ἀποκοπαὶ ἄκλιτοι”.

53 Apollonius, De pronomine 90.12: ‟θέμα γὰρ ἴδιόν ἐστιν ὀξύτονον, οὐχὶ ἔγκλιμα τῆς σφῶιν.”

54 Seneca, Epistulae ad Lucilium V.42.1.

55 Diogenes Laertius, Vitae philosophorum VII.192: ‟Περὶ τῆς κατὰ τὰς λέξεις ἀνωμαλίας πρὸς Δίωνα α΄β΄γ΄δ΄.”

56 Varro, De lingua latina IX.1 (= SVF II.45, frgm. 151; FDS F 640; I follow the text in FDS): ‟Crates, nobilis grammaticus, qui fretus Chrysippo, homine acutissimo qui reliquit περὶ ἀνωμαλίας †lei libri† contra analogian atque Aristarchum est nixus, sed ita, ut scripta indicant eius, ut neutrius videatur pervidisse voluntatem, quod et Chrysippus de inaequabilitate cum scribit sermonis, propositum habet ostendere similes res dissimilibus verbis et dissimiles similibus esse vocabulis notatas.”

57 Simplicius, in Cat. 102.22-24: ‟ καλῶς δὲ εἶπεν ὅτι πᾶσα οὐσία δοκεῖ τόδε τι σημαίνειν, διότι τὴν ἑτέρων δόξαν ἀπαγγέλλει καὶ τὴν ἀνωμαλίαν τῆς συνηθείας, ἥτις τὰ ἀνόμοια πράγματα ὁμοίοις σχήμασι λέξεων εἴωθεν πολλάκις σημαίνειν”. I first connected the Simplicius passage with the one in Varro in Ebbesen, Sten (1981), Commentators and Commentaries on Aristotle’s Sophistici Elenchi, Leiden, Brill, I, p. 211-212, and suggested that Simplicius is here reproducing Porphyry, to whom he and the other late commentators on Categories and Peri hermeneias seem to owe just about everything of Stoic origin occurring in their works, as they no longer had access to genuine Stoic works, whereas Porphyry had (though he probably also pilfered a lot of his information about Stoics from Alexander of Aphrodisias).

58 Cf. ἡ τῆς συνηθείας ἀνωμαλία at Simplicius, in Cat. 160.25 and ἡ τῆς χρήσεως ἀνωμαλία at 264.22.

59 Simplicius, in Cat., CAG VIII: 396.19-21: ‟πολλῆς δὲ οὔσης τῆς ἀνωμαλίας Χρύσιππος μὲν ἐν τοῖς Περὶ τῶν στερητικῶν λεγομένοις ἐπεξῆλθεν αὐτῇ”; further references to the same work at 389.22, 394.31, 395.10, 401.7, 403.6. Simplicius’ τὰ περὶ τῶν στερητικῶν λεγόμενα seems to be a distorted version of the title Περὶ τῶν κατὰ στέρησιν λεγομένων reported in Diogenes Laertius’ list of Chrysippus’ writings (D.L. VII.190).

60 Cf. the title Λόγοι παρὰ τὰς συνηθείας in Diogenes Laertius, Vitae philosophorum VII.192.

61 Thus also κρίσις/κρῖμα, κατόρθωσις/κατόρθωμα, φρόνησις/φρόνημα, φρονίμευσις/φρονίμευμα. See the index in SVF IV.

Haut de page

Pour citer cet article

Référence électronique

Sten Ebbesen, « Imposition of Words in Stoicism and Late Ancient Grammar and Philosophy », Methodos [En ligne], 19 | 2019, mis en ligne le 27 février 2019, consulté le 23 octobre 2019. URL : http://journals.openedition.org/methodos/5641 ; DOI : 10.4000/methodos.5641

Haut de page

Auteur

Sten Ebbesen

Saxo Institute, University of Copenhagen

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus de la revue Methodos sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo DOAJ - Directory of Open Access Journals
  • Logo UMR Savoirs, Textes, Langage
  • Logo CNRS - INSHS
  • Logo Université de Lille
  • OpenEdition Journals