by Corry Shores
[Search Blog Here. Index tabs are found at the bottom of the left column.]
[Central Entry Directory]
[Stoicism, entry directory]
[Émile Bréhier, entry directory]
[Bréhier, La théorie des incorporels, entry directory]
Chapitre II:
Théorie des “exprimables”
2.1
[The Sayable as Predicate]
Brief summary
The primordial element in Stoic logic is the sayable (λεκτόν / exprimable), which is something along the lines of {1} the rational organization of an active, physical situation, {2} the conceptual sense of the thought of that situation, and {3} the sense of a statement about that situation. The sayable should be understood also in terms of what is being affirmed, namely {a} that which is physically affirmed through the sayable’s actualization in a corporeal situation, {b} that which is conceptually affirmed by conceiving the rational impression that is organized in accordance with the sayable, and {c} that which is logically affirmed by asserting the utterance expressing the sayable. In all three cases, the sayable is what is being expressed by the corporeals involved (note: words and thoughts are corporeals). Thus the corporeal situation, the corporeal thought, and the uttered corporeal proposition “say” or “express” the sayable. The uttered proposition, however, does not signify the sayable. It signifies the thought. But what grants the meaning relationship of utterance signifying the thought about the real world situation is that they all say the same sayable. The way our mind forms the thought of the situation is by first having fragmentary and disorganized simple impressions, and then organizing them into a rational impression in accordance with the sayable that the rational impression will come to conceptually express. Now, there are complete sayables, where there is a subject and predicate, and incomplete sayables, which lack the subject. But rational impressions are always complete sayables. So there are sayables which are not rational impressions. There is a problem with how to understand propositional judgements under the Stoic conception. For the Stoics, there are no universals. But that means the subject cannot be thought of as being included in a general class indicated by the grammatical predicate. The Stoics solve this problem by understanding the predicate as the activity of bodies, articulated as a verb, and not as the property of bodies, articulated with a copula and adjective. So they would not say “a body is warm” but instead “a body is warming up”. Thus for the Stoics, the relation between subject and predicate is not that of essential or accidental predication but is rather the relation of an event to its subject.
[Note: the idea that the utterance “says” the sayable is not something explicitly stated in Bréhier’s text, and I may be misrepresenting his thinking. It seems rather that the utterance “says” nothing and simply means the thought, which in turn “says” the sayable. In this picture, the speaking of the utterance would still affirm the sayable predicate, but only through its indicative relation with the thought. See section 2.1.4. It is still not entirely clear to me, because there Bréhier interprets Sextus Empiricus as describing the sayable as needing vocal utterance to be said. Why can it not just be thought? One answer would be that the thought depends on linguistic or syntactical structures to be formed (like subject predication), and thus in that case the utterance in a sense would still somehow be involved in the saying of the sayable. I offer the interpretation that the utterance says the sayable without indicating it, because this to me seems the most consistent with Sextus and Bréhier. Otherwise, perhaps Bréhier is arguing that Sextus Empiricus is wrong and the sayable does not need to be vocally uttered in order to be said.]
Contents
2.1
2.1.1
2.1.2
2.1.3
[The Difference between the Sayable and the Signified]
2.1.4
[Sayables are Not Signifieds or the Uttered Words Themselves.]
2.1.5
2.1.6
[The Sayable as Possibly a Type of Rational Impression]
2.1.7
[Not All Objects of Reason are Sayable. Rational Notions Are Not Sayables]
2.1.8
2.1.9
[The Coincidence of the Real Predicates of Bodies and the Logical Predicates of Propositions]
2.1.10
[Predicates as Acts Expressed by Verbs and not as Properties Expressed by Adjectives]
2.1.11
[For the Stoics, the Subject and Predicate Cannot Coincide by Means of Class Inclusion]
2.1.12
[The Stoics Use Verb Predicates (“The Tree Greens”) Instead of Copula Predicates (“The is Green”)]
2.1.13
[Judgments Only Have Active Predicates]
2.1.14
[The Only Type of Connection in Propositions is Between Subject and Event]
2.1.15
[The Coincidence of Irreal Logical Predicates and Irreal Predicates from Corporeal Activity]
2.1.16
2.1.17
[Sayables Are not Rational Representations]
Summary
2.1
[The Sayable]
2.1.1
[The Sayable of Stoic Logic]
(p.14: “La réalité logique, l'élément primordial ...”)
[In sum: The primordial element in Stoic logic is not, like in Aristotelian logic, the concept, nor is it the representation (φαντασία / représentation) or the notion (ἔννοια / notion) that come into formation through a modification of the corporeal soul by exterior objects. Rather, it is the sayable (λεκτόν / exprimable).]
The primordial element in Aristotelian logic is the concept. But in Stoic logic, the primordial element is something else entirely. It is neither the representation (φαντασία / représentation) that is a modification of the corporeal soul by an exterior body nor is it the notion (ἔννοια / notion) that is formed in the soul under the action of similar sorts of experiences. Rather, it is something entirely new, and the Stoics call this primordial element of their logic a “sayable” (λεκτόν / exprimable).
La réalité logique, l'élément primordial de la logique aristotélicienne est le concept. Cet élément est pour les Stoïciens tout autre chose ; ce n'est ni la représentation (φαντασία) qui est la modification de l'âme corporelle par un corps extérieur, ni la notion (ἔννοια), qui s'est formée dans l'âme sous l'action d'expériences semblables. C'est quelque chose de tout à fait nouveau que les Stoïciens appellent un exprimable (λεκτόν).
(14)
2.1.2
[The Sayable (λεκτόν / exprimable) as the Signified or Meaning (σημαινόμενον / objet signifié) of the Word (mot) and as the Predicate (κατηγόρημα / attribut) of the Object in the World (τύγχανον / objet) and of the Thought (νόημα / pensée)]
(p. 14-15. “Voici une difficulté que, selon Sextus ...”)
[In sum: When a word (mot) is uttered, it makes a sound (son). To those who know the language, the word refers to a thing (or situation) in the world and to a thought in the mind, both of which are corporeal, just as the sound is. But to the thought and to the thing in the world (τύγχανον / objet) belongs an incorporeal predicate (κατηγόρημα / attribut) (but not a physical property), which is the sayable (λεκτόν / exprimable). (It is like the rational sense of the situation in the world and the rational sense of the thought corresponding to that situation. And it is identically the same for the object/situation, thought, and word/sentence.)]
We have encountered a difficulty that, according to Sextus Empiricus, dissolves the theory of sayables, and it is not implausible that the theory arose from this difficulty. Suppose a Greek and a non-Greek speaker hear the same word (mot). They both have the representation (φαντασία / représentation) of the thing designated by the word (chose désignée par le mot) (that is to say, since they both see the same things, that both have formed the same rational impressions of the situations around them); yet, the Greek will understand the word and the non-Greek speaker will not. And in fact, there is no other reality [that is to say, no other corporeality] other than the corporeal sound (son) on the one hand and designated corporeal object (objet) [or situation] in the world. Now, the object in the world (τύγχανον / objet) has for both of them the same properties [or qualities]. But the object for the Greek has a predicate (κατηγόρημα / attribut) that it does not have for the non-Greek speaker. That predicate is something like, ‘being signified by the word’. It is this predicate of the object that the Stoics call the sayable (λεκτόν / exprimable). According to Sextus Empiricus, what makes that which is signified (the meaning of the word / τὸ σημαινόμενον / l’objet signifié) differ from the corporeal object in the world (τὸ τύγχανον / l’objet) is that the signified (σημαινόμενον / objet signifié) is bound up with this predicate (κατηγόρημα / attribut)whose affirmation changes nothing about the [corporeal features of the] object. The concept of the sayable (λεκτόν / exprimable) was so new and unique that interpreters of Aristotle, like Ammonius, had great difficulty in placing it within the Peripatetic classifications. According to Ammonius, for Aristotle, the thing signified by the word (chose signifiée par le mot) is the thought (νόημα / pensée). And the thing signified by the thought (chose signifiée ... par la pensée) is the object (πρᾶγμα / objet). Ammonius adds that the Stoics conceived a third thing, the sayable (λεκτόν / exprimable), which is the intermediary between the thought (νόημα / pensée) and the thing (chose). Ammonius in fact is critical of the Stoic’s addition. And really Aristotle’s theory is sufficient so long as the thought (νόημα / pensée) is itself the designated object (object désigné). But the Stoics consider thought (pensée) to be a body (corps), and as a body, it has its own unity and independent nature. [The word is a corporeal, and the thought tied to the word is corporeal. But they are both complete in themselves and thus bear no inherent relation to one another. In other words, each is what it is independently of the other. In Aristotle’s system, however, the thought is perhaps something without its own independent existence, and it only finds its expression in the words used to denote it. So what brings about the link between the thought and the word for the Stoics cannot be something corporeal. For, if it were corporeal, it again would have its own independent existence and thus no intrinsic link to the word and thought (so it would not be a third body. If we just say it is an extrinsic relation between bodies, and if that relation is not a corporeal body, then it would need to be incorporeal). Instead, it must be incorporeal. The corporeal thought has a sayable predicate (that sayable predicate being the incorporeal predicate the thought is about), and the word has a sayable predicate (that sayable predicate being the incorporeal predicate that the word expresses), and that sayable predicate is the same for both, hence the meaning relationship. It would also seem that the thing or situation in the world that the word (or sentence) refers to would also be linked to the same predicate. So the meaning relation between situation, thought, and utterance would seem to be a incorporeal linking to a common incorporeal predicate. Consider a classic example. Dion is actually walking right now. The corporeal situation here at some certain moment is fundamentally a mixture of corporeal elements with certain properties or qualities, depending on the composition of the mixture at that particular moment in time. Each moment with its properties forms what we might call a “temporal mixture” in which there is a change in composition over time. Dion is walking now. At any moment his body is in a certain posture. But if we expand the present out to include the other moments of the change, we would get a situation. At each present moment of this situation, the corporeals “express” (in a non-linguistic sense) or “present” the predicate “is walking” (with the complete predicate being, “Dion is walking”.) The Greek and the non-Greek speaker both witness the situation. They form rational impressions in the mind. These impressions are corporeal (even the mind is corporeal). And as corporeals they are their own mixtures with their own qualities and properties. But they “express” (not in a linguistic sense, but in a conceptual sense) the same sayable predicate, ‘Dion walks’. The Greek then produces a series of sound vibrations, symbolized by “Δίων περιπατεῖ”. The sounds are corporeals. As such, these sounds are mixtures with their own properties. But they too have the same sayable predicate, ‘Dion walks’. For the non-Greek speaker, however, the sounds of the utterance do not have that sayable predicate, and thus the meaning relation is not understood, despite both the Greek speaker and the non-Greek speaking having rational impressions with the predicate corresponding to the word’s meaning.(Now suppose there is another Greek who hears “Δίων περιπατεῖ” but does not see the situation. Presumably, for this second Greek, the corporeal body of the sounds has the predicate ‘Dion walks’, and then they can form the thought of Dion walking even without seeing it).] [Now, since there is no intrinsic relation between the corporeal sound and the thought or referenced object/situation in the world, that means that] the same name can designate many things. [This seems to imply that the way the sound body comes to have a predicate is not on account of its own properties but by arbitrary fiat.]
Voici une difficulté que, selon Sextus, résout la théorie des exprimables, et il n’est pas invraisemblable qu’elle soit sortie de cette difficulté 1. Un Grec et un Barbare entendent un même mot ; ils ont tous deux la représentation de la chose désignée par le mot ; pourtant le Grec comprendra et le Barbare ne comprendra pas. Quelle autre réalité y a-t-il donc que le son d’une part, l’objet de l’autre ? Aucune. L’objet comme le son reste le même. Mais l’objet a pour le Grec, je ne dis pas une propriété (car son essence reste la même dans les deux cas), mais un attribut qu’il n’a pas pour le Barbare, à savoir celui d’être signifié par le mot. C’est cet attribut de l’objet que les Stoïciens appellent un exprimable. L’objet signifié (τὸ σημαινόμενον) diffère, d’après le texte de Sextus, de l’objet (τὸ τύγχανον), précisément par cet attribut qui en est affirmé sans en changer la nature. Le λεκτόν était quelque chose de si nouveau qu’un interprète d’Aristote, comme Ammonius, a la plus grande peine à le loger dans les classifications péripatéticiennes. Pour Aristote, la chose signifiée par le | mot était, dit Ammonius, la pensée (νόημα), et, par la pensée, l’objet (πρᾶγμα). « Les Stoïciens, ajoute Ammonius, conçoivent en outre un intermédiaire entre la pensée et la chose, qu’ils nomment l’exprimable » 1. Ammonius n’approuve pas cette addition, et en effet la théorie d’Aristote se suffit à elle-même, si la pensée est en elle-même l’objet désigné. Il ne pouvait plus en être de même pour les Stoïciens. Pour eux la pensée était un corps, et le son était aussi un corps. Un corps a sa nature propre indépendante, son unité. Le fait d’être signifié par un mot doit donc lui être ajouté comme un attribut incorporel, qui ne le change en rien. Cette théorie supprimait tout rapport intrinsèque entre le mot et la chose : on peut sans doute y ramener les vues de Chrysippe sur l’amphibologie. Par celle-ci, en effet, le lien entre la parole et la pensée devient assez lâche pour qu’un même nom puisse désigner plusieurs choses 2.
(14-15)
1. Sextus, Math. VIII 11 (S.V.F. II 48, 19). Le σημαινόμενον de ce texte revient à λεκτόν ; cf. I. 23.
(14)
1. Amm. in Ar. de interpr., p. 17, 24 (S.V.F. II 48, 31).
2. Galien. de sophism. 4 (S.V.F. II 45, 35).
(15)
2.1.3
[The Difference between the Sayable (λεκτόν / exprimable) and the Signified (σημαινόμενον/ objet signifié)]
(p.15: “Si la théorie d es exprimables n’avait pas cependant une autre portée...”)
[In sum: Logical elements like the predicate (κατηγόρημα / attribut) and judgment are sayables, but they are not a signified (σημαινόμενον).]
But we need to say more about the Stoic theory of sayables in order to understand the role it plays in their logic. All the elements playing a role in logic, like predicates, judgments, judgment connectors, are all sayables. But these logical elements, even though they are sayables, are not signified by a word. For example, the predicate is what is affirmed of a being or property ((l’attribut par exemple (κατηγόρημα) indique ce qui est affirmé d’un être ou d’une propriété)). But just by being something that is affirmed does not imply that it is also what is signified by the linguistic expression. So just because all signifieds are sayables does not mean that all sayables are signifieds. Yet the Stoicorum veterum fragmenta confuses these notions in the section on logic titled Σημαινομένων ἢ Λεκτῶν [See SVF 2, pp.48ff.]
Si la théorie d es exprimables n’avait pas cependant une autre portée, on ne comprendrait pas le rôle qu’elle joue dans la logique. Tous les éléments qui servent à la logique, les attributs, les jugements, les liaisons de jugements sont aussi des exprimables. Il est visible au premier abord que ces éléments ne peuvent se réduire à des choses signifiées par un mot : l’attribut par exemple (κατηγόρημα) indique ce qui est affirmé d’un être ou d’une propriété ; nous ne trouvons nulle part cette idée, à laquelle d’ailleurs il serait bien difficile de donner un sens plausible, que le fait d’être affirmé est identique au fait d’être signifié, que le κατηγόρημα est σημαινόμενον. D’une façon générale, si le « signifié » est une « exprimable », nous ne voyons nullement que tout exprimable soit un « signifié ». Cette interprétation erronée de l’ « exprimable » est pourtant assez répandue pour qu’Arnim dans son édition des anciens Stoïciens ait pu la consacrer, en intitulant les fragments relatifs à la logique : περὶ Σημαινομένων ἢ Λεκτῶν.
(15)
2.1.4
[Sayables are not Signifieds or the Uttered Words Themselves.]
(p.16: “Cette erreur a pu venir de ce qu’il y a, en tous cas”)
[In Sum: Sayables (λεκτά / exprimables) are said, (that is, they have the predicate, “being said”), but they are not signified (that is, they do not have the predicate, “being signified”). Signified objects, however, can take the predicate “being signified”. What is being said is that which is being affirmed, and being signified is that which is meant. They are not the same thing. So we should not think that sayables must be signifieds. Nor should we think that the words themselves are sayables.]
This error [of thinking that all sayables are signifieds] comes from the fact that there is an intimate fusion between the sayable (λεκτόν / exprimable) and language (langage). According to Sextus Empiricus, all sayables are necessarily said, that is to say, enunciated by a word signifying the thought (un mot significatif de la pensée) [Sextus writes: “every sayable has to be said (this is how it came by this name)”; Against the Logicians p.104]. But, being said (λέγεσθαι / le fait d’être exprimé), which is a predicate of the sayable, should not be confused with being signified (τὸ σημαινόμενον / le fait d’être signifié), which is a predicate of the object [it is not here specified if the object is the corporeal object in the world that is being signified (τύγχανον / objet) or the (σημαινόμενον / objet signifié)]. But we should not thereby conclude that every sayable must be designated by words, thinking that it is of the very nature of the sayable to be designated or signified by words. [This is a bit perplexing. (Note also that Luhtala in On the Origin of Syntactical Description in Stoic Logic, section 5.5.2.1, says that the sayable and the signified are equivalent: “Λεκτόν (‘sayable’) is an equivalent of σημαινόμενον (‘that which is signified’) in technical Stoic usage (70)). For Bréhier, being said is different than being signified. Roughly I understand the distinction in the following rough way (and I will probably revise this account as I learn more). The utterance is the signifier. It is a body. It signifies the thought, which is also a body (un mot significatif de la pensée). But neither the word nor the thought signify the predicate/sayable that is shared by (or expressed by) the thought, utterance, and situation in the world. So it seems that signifying is a relation holding strictly between bodies. But being expressed or being said is a relation holding between a corporeal and an incorporeal. Suppose we utter “[‘di.oːn wɔks]”. This signifies the conception DION WALKS. It expresses the sayable, which is not the concept DION WALKS, but is rather the predicate; and the predicate is that which is affirmed with regard to a being or a property (“l’attribut par exemple (κατηγόρημα) indique ce qui est affirmé d’un être ou d’une propriété ; nous ne trouvons nulle part cette idée” p.15, section 2.1.3 above). But still, it would be odd to say that the utterance signifies the thought but not what we are affirming. When we say “Dion walks”, we are affirming that Dion is presently walking. To claim that “Dion walks” signifies DION WALKS, but expresses (and does not signify) that Dion is walking, seems to be making primarily a terminological distinction, namely, that we use the term “signify” or “mean” when the referential relation is between an utterance and a thought, and we use “say” or “express” when the referential relation is between an utterance and what is being affirmed. It seems that the relation between the utterance and the predicate/sayable is not one of meaning like it is between the utterance and the thought. Why? Signification is a matter of corporeal mixture, namely, it is a mixture of a thought a in a speaker with a sound pattern b uttered by the speaker and with another thought c in a listener, whereby thoughts a and c are identical on account of sound b having taken on the role of being crucially involved in the formation of such symmetrical mixtures, normally by means of learning and custom. But all of this can be explained physically without appeal to incorporeality. I cannot tell an accurate story about this now. I will just say for the moment that there is an issue of temporality and causality that we need to think about. Only the present exists, and corporeals are not causes of other corporeals, but rather of incorporeal predicates. So perhaps we cannot explain meaning as if there is causal corporeal chain from observed event to speaker’s mind to speaker’s utterance to listener’s mind. Rather, we might need to appeal to an incorporeal with an extra-present temporality that somehow is like the glue between object/situation in the world, concept, and utterance. We say that the utterance means the concept. It is not clear how these relate to the situation in the world. It could be that the utterance and thought mean those things, just the thought means them, or there is another relationship other than meaning, but I do not know what. But somehow or other all three are tied together. What they share is that they all express the same predicate. The predicate is what is affirmed. Dion’s actual present walking, I claim, expresses the predicate of him walking. It is expressed, because as a physical event, it “affirms” that predicate by actualizing it or embodying it. The thought of this situation is itself a body. But it does not signify the predicate of his walking; rather, it expresses that predicate, because the concept DION WALKS affirms that Dion is walking. And the utterance is a body. It signifies the thought. But it expresses the predicate, because the statement “Dion walks”, when it is said in the presence of Dion walking, affirms that Dion is walking. It is still difficult under this account to distinguish signifying from expressing without resorting to a simple terminological distinction that one is between bodies and the other between an incorporeal and a corporeal. My best guess now is that we would need to distinguish “to affirm” from “to signify”. To signify might simply be to be taken as equivalent to. To affirm might be to claim or evince as true. A predicate cannot be equivalent to an utterance or thought, because these are bodies and predicates are not. But, an utterance or thought (and even a situation itself) can claim or evince something as true.] Ammonius makes the inverse error by identifying the sayable with the uttered word itself.
Cette erreur a pu venir de ce qu’il y a, en tous cas, une fusion intime entre l’exprimable et le langage ; d’après Sextus tout exprimable doit être exprimé, c’est-à-dire énoncé par un mot significatif de la pensée3. Mais le fait d’être exprimé (λέγεσθαι) qui est un prédicat de l’exprimable ne doit pas du tout être con- | fondu avec le fait d’être signifié (τὸ σημαινόμενον) qui est lui-même un exprimable et un prédicat de l’objet. On a conclu trop vite de ce que tout exprimable devait être désigné par des mots, que toute sa nature était précisément d’être désigné ou signifié par des mots. Une erreur inverse mais de même nature a été commise par un critique ancien du stoïcisme, Ammonius, celle qui consista à identifier les exprimables avec les mots du langage 1. Cette erreur, d’après les termes, repose sur l’exposition même de Sextus ou une exposition très analogue. « Les pensées, dit Ammonius, peuvent être proférées (ἐκφορικά). Mais nous les proférons par des mots, et les mots sont les exprimables ». Ici exprimable (λεκτόν) a été confondu avec ce qui est exprimé et proféré (λεγόμενον, ἐκφερόμενον), c’est-à-dire le mot. Nous avons donc à rechercher ce qu’est réellement l’exprimable.
(15-16)
3. Sextus, Math. VIII 80 (S.V.F. II 48, 27).
(15)
1. Amm. In Arist. An. pr., p. 68, 4 (S.V.F. II 77, 7)
(16)
2.1.5
[Sayables can be Complete (Having Both a Predicate and Subject) or Incomplete (Being Just a Predicate). But in Stoic Logic and Phenomenology, We Deal Only with Complete Sayables.]
(p.16-17: “La place de l’exprimable dans le système des objets ...”)
[In sum: We want to know where to look for sayables so we can study them better. The first two possibilities we consider are problematic for the same reason, namely, they do not involve a certain class of sayables, and thus we cannot learn about sayables in general from them. A sayable can be complete or incomplete. A complete sayable is one that when articulated has both a subject and predicate, while a stated incomplete sayable would have a predicate but no subject. The first possibility is that we look to rational impressions, which always express sayables. A state of affairs in the world has a rational sense to it and thus a sayable that it expresses. The exterior event makes [ordinary] impressions (φαντασίαι / représentations ordinaires) on the leading-part of our soul (ἡγεμονικόν / partie hégémonique de l’âme). But this direct impression does not by itself express the original sayable, because it is not yet rationally ordered. So our mind then organizes those impressions rationally such that they form a different sort of impression, called the rational impression (λογικὴ φαντασία / représentation ordinaire), which expresses a sayable that is identical to the one expressed by the situation out in the world. But since the situation out in the world expressed a complete sayable, given that its activities or features involved things doing those activities or having those features, and since that sayable is identical to the one in the rational impression corresponding to that situation, that means rational impressions can only express complete sayables, leaving us to wonder still about incomplete ones. Now, since an incomplete sayable is described in propositional terms as being something that when stated gives a predicate but without a subject, we might then be tempted to look to Stoic logic to study both complete and incomplete sayables. But Stoic logic, which studies judgments or propositions, does not admit of any proposition without a subject, which again would leave us curious about incomplete sayables.]
It is quite difficult to determine what place sayables have within the system of representations in the mind. On the one hand, we have Sextus and Diocles saying that the sayable is found in the rational representation (λογικὴ φαντασία / représentation rationnelle). [Sextus writes: “They [the Stoics] say that a ‘sayable’ is what subsists in accordance with a rational impression, and a rational impression is one in which the content of the impression can be exhibited in language” (Long & Sedley I, p.196). “They say that what subsists in accordance with a rational presentation is a thing said [lekton] and that a rational presentation is one according to which the content of a presentation can be made available to reason” (Inwood & Gerson p.90). The Diocles passages are found in Diogenes Laertius’ Lives (see footnote 24 at the Perseus site, where it is explained that sections 49-83 come from Diocles). “[The Stoics] say that a sayable is what subsists in accordance with a rational impression” (Long and Sedley I p.196 translating Diog. Laert. VII, 63).] The ordinary representation is produced through contact with a body that leaves its imprint in the leading part of the soul (ἡγεμονικόν / partie hégémonique de l’âme), while the rational impression (λογικὴ φαντασία / représentation rationnelle) seems to arise more spontaneously. [Note, by “la représentation ordinaire” versus “la représentation rationnelle”, I am guessing Bréhier means the distinction between φαντασία, which he previously translated as just la representation, and λογικὴ φαντασία, or rational impression. But I am not certain. Here is from Sextus Empiricus:
(1) For they [the Stoics] say, just as the trainer or drill-sergeant sometimes takes hold of the boy’s hands to drill him and to teach him to make certain motions, but sometimes stands at a distance and moves to a certain drill, to provide himself as a model for the boy – (2) so too some impressors [φανταστῶν] touch, as it were, and make contact with the commanding-faculty to make their printing in it, as do white and black, and body in general; whereas others have a nature like that of the incorporeal sayables (lekta), and the commanding-faculty [ἡγεμονικόν] is impressed in relation to them, not by them.
(Sextus Empiricus, Against the professors 8.409; SVF 2.85; Long & Sedley I.p.163)
] Thought constructs the rational impression by assembling, increasing, and diminishing the sensible objects that are given to it. But the active cause in the construction of rational impressions is not these sensible objects but is rather reason. Diocles enumerates the different processes by which the reason acts: resemblance, analogy, displacement, composition, contradiction, transition, and privation. [Diogenes Laertius, speaking for Diocles, writes (see the last line):
According to the Stoics, ‘sense-perception’ refers to [a] the pneuma that extends from the leading part to the senses and [b] the ‘grasp’ that comes through the senses and [c] the equipment of the sense organs (which some people may be impaired in), and [d] their activation is also called ‘sense-perception’. According to them, the grasp occurs [a] through sense-perception (in the case of white objects, black objects, rough objects, smooth objects) and [b] through reason (in the case of conclusions drawn through demonstration, for example, that there are gods and that they are provident). For of conceptions, some are conceived on the basis of direct experience, some on the basis of similarity, some on the basis of analogy, <some on the basis of transposition,> some on the basis of composition, and some on the basis of opposition.
(Inwood & Gerson p.13: Diog. Laert. VII, 52)
] Sextus says that in the case of rational impressions, the soul has a representation related to the objects but not a representation directly by means of them. [See the last line of the above Sextus quote from Adv. Math. 8.409-410. The idea seems to be that we have simple impressions that are made directly though sensory affection, and we have rational impressions that are made by rationally reconfiguring the simple impressions]. [Let me quote the next line, because I am not sure about the next point: “Le λεκτόν serait donc identique, d’après ce premier témoignage aux notions dérivées de l’expérience par la raison”. I am not certain, but I gather the following. Sextus said in the above quotation that the sayable is a matter of the rational impression but not the direct sort of impression. So I am guessing here that the sayable is identical in both the rational impression and in the state of affairs that did the direct impressing on the soul. So, firstly we have the situation out in the world. It is corporeal, but it expresses an incorporeal sense or sayable. It impresses on our soul (which is a corporeal), leaving imprints. But the imprints are given in the order of perception, which might accord with the way our eyes for example scanned the scene and the way the moments unfolded in fragments each instant over the course of its unfolding. The mind then organizes those imprints in a rational way in order to create a rational impression that expresses a sayable or sense that is identical to the one expressed in the exterior state of affairs that originally made a direct impression on us. (Note, from what is said in the next section, the idea might more simply be that the sayable is identical to the rational impression. And the distinction being made might be that the sayable is identical to the rational impression but not necessarily to the disorganized simple impressions.)] [I may not get the next ideas right, but they seem to be the following. Stoic logic, with regard to sayables, is more complex than this phenomenology of sayables. In the phenomenology, the sayable out in the world and in the mind are both complete. But in the logic, there are compete sayables and incomplete sayables. (See Luhtala section 5.5.4.1.) Incomplete sayables are the predicates of judgments, articulated as verbs without subjects, like, “…writes” and “…speaks”. But in complete sayables, the verb is accompanied by its subject. (Bréhier’s idea here seems to be that we cannot understand all sayables in terms of rational impressions, because rational impressions can only express complete sayables, but in Stoic logic, there are also incomplete sayables.) So we should not look exclusively to rational impressions to understand what sayables are, as there is a class of sayables that will not be found in our rational impressions. (The next idea seems to be the following. We also cannot turn to Stoic logic to study incomplete sayables, because Stoic logic only recognizes three sorts of judgements, and in all three types, there is an individual that takes the role of the subject, although that individual might be indefinite). So the phenomenology and the logic deal only with complete predicates while there exist incomplete predicate/sayables.]
La place de l’exprimable dans le système des objets représentés à l’esprit est fort difficile à déterminer. D’une part Sextus, confirmé par Dioclès, nous dit que ce qui est dans la représentation rationnelle, c’est là l’exprimable 2. Tandis que la représentation ordinaire se produit par le contact d’un corps qui marque son empreinte dans la partie hégémonique de l’âme, au contraire, il paraît y avoir plus de spontanéité dans la représentation rationnelle. C’est la pensée qui la construit, en assemblant, en augmentant, en diminuant les objets sensibles qui lui sont donnés d’abord ; les objets ne sont pas ici cause active, mais c’est la raison. Dioclès énumère ainsi les différents procédés par lesquels elle agit la ressemblance, l’analogie, le déplacement, la composition, la contradiction, la transition, la privation 3. On peut dire avec Sextus que dans ces cas l’âme a une représentation à propos des objets et non par eux. Le λεκτόν serait donc identique, d’après ce premier témoignage aux notions dérivées de l’expérience par la raison. Mais si nous envisageons le contenu de la logique, cette idée est bien malaisée à admettre. En effet, nulle part nous n’y voyons intervenir de notions de ce genre, bien que l’exprimable soit son élément propre. De plus, la suite même du texte de Sextus et de Dioclès contredit l’interprétation qu’on pouvait, semble-t-il, légitime- | ment en tirer : « Dans les exprimables, disent-ils, les uns sont incomplets, les autres complets ». Les exprimables incomplets sont les attributs de jugements, énoncés dans des verbes sans sujet : « écrit, parle ». Les complets sont, pour ne considérer maintenant que les plus simples, le verbe accompagné de son sujet. Si ce sont là tous les exprimables (et il n’y a aucune raison de croire le contraire), nous y cherchons vainement les objets de représentation rationnelle, les notions que nous avons définies tout à l’heure. Les exprimables se bornent aux attributs tantôt sans sujet, tantôt accompagnés de leur sujet. Dira-t-on que la notion se retrouve précisément dans le sujet des jugements ? Mais nous ne voyons pas que les Stoïciens aient admis dans leur logique d’autres jugements que ceux que la logique moderne a appelé singuliers, dont le sujet est un individu. Dans la classification des jugements simples chez Sextus, parmi les trois sortes de jugements, les jugements définis ont pour sujet un individu que l’on indique (celui-ci), les indéfinis un être que l’on n’indique pas (un homme), mais que reste un individu 1.
(16-17)
2.1.6
[The Sayable as Possibly a Type of Rational Impression]
(p.17-18: “A vrai dire, dans d’autres sources, les exprimables ...”)
[In sum: In other texts, sayables are portrayed as a type of rational impression.]
We have been looking at texts about sayables from Sextus Empiricus and also from Diocles (as given in Diogenes Laertius).] In other sources, sayables are not said to be identical to the rational impressions but are rather a type of rational impression. [We saw before Diocles’ classification of ways that reason forms conceptions (section 2.1.5). One of those is “by transposition” (Long & Sedley I p.238. “on the basis of transference” Inwood & Gerson p.13; “suivant une transition” Bréhier p.17.). And place and sayables are conceived by transposition. (The idea here seems to be that place and sayables are two sorts of rational impressions, among other sorts, given that they are formed by one means and not the others.)] Diocles says that place and sayables are conceived “following a transition” (κατὰ μετάβασιν). Now, because we are using the notion of “transition”, that means that the object of the impression has a composition (of parts) and that thought goes from one part to another. [At this point we might think that the transition is from subject to predicate. But Bréhier says we do not see this movement in simple judgements. I am guessing that is because Bréhier is emphasizing this notion of “follows”. A predicate does not follow from a subject. It is a part of it already.] But we do not see this feature of “following a transition” applying to all cases of sayables. For example it does not apply to incomplete sayables or to simple judgements. However, hypothetical judgements [of the form “if…then…”] involve this feature, as do arguments where there is a movement of consequence [a transition from premises to conclusion.] [So since incomplete sayables and simple judgements are sayables, but do not involve the element of following, that means we can say that] Diocles here is not speaking of a feature of all sayables, and he is not saying that all sayables fall under this category. In another text, Sextus Empiricus opposes the sensible impressed from the rational impressed [See the Sextus quote above (section 2.1.5) Against the professors 8.409; SVF 2.85, about the drill sergeant. The first “impressors” touch the leading part and impress upon them, as in cases of sense impression. The second sort of impressor has the nature of incorporeal sayables, where the leading part of the soul is impressed in relation to the sensory impressors, but not by them.] But it is not clear from those passages whether Sextus means that all sayables are like this, or just a limited kind of them is like this.
A vrai dire, dans d’autres sources, les exprimables sont cités non pas comme identiques aux représentations rationnelles, mais comme une espèce d’entre elles. Le premier texte est la classification des notions de Dioclès que l’on a déjà citée, où l’exprimable est cité avec le lieu, comme un exemple de notions obtenues « suivant une transition » (κατὰ μετάβασιν). Celte « transition » implique que l’objet de la représentation est composé et que la pensée va d’une partie à l’autre. Si nous cherchons à quels exprimables ce caractère est applicable, nous verrons qu’il ne correspond pas à tous les cas. Ni dans les exprimables incomplets, ni dans les jugements simples, on ne saurait le trouver. Au contraire les jugements hypothétiques et les raisonnements contiennent un passage du principe à la conséquence qui, seule, peut expliquer le mot « métabase » 2. Il s’ensuit que dans l’exemple qu’il donne ici, Dioclès ne veut pas parler de tous les exprimables, ni les faire rentrer tous dans cette catégorie. Dans un autre texte de Sextus qui oppose le « représenté » (φανταστόν) sensible au « représenté » rationnel, il reste douteux si les exprimables incorporels qu’il cite dans la seconde définition sont donnés comme un simple exemple parmi d’autres, ou | comme l’ensemble de tous ces représentés 1. Mais l’opposition des corps qui sont certainement tous les représentés sensibles aux incorporels, nous ferait pencher vers la seconde alternative.
(17-18)
2. Cf. la μεταβατικὴ διανοία (Sextus S.V.F. II 43, 22) se rapportant à l’ἀκολουθία.
(p. 17. Note for footnote 2. This is Adv. Math. VIII, 275. I do not see διανοία in the Sextus part, but it appears in the Galen part above it, in the same section 153 of SVF II.)
1. Sextus, Math. VIII 409 (S.V.F. II 29, 2).
(18)
2.1.7
[Not All Objects of Reason are Sayable. Rational Notions Are Not Sayables]
(p.18: “Malgré ces difficultés, il y a des raisons sérieuses ...”)
[In sum: Not all objects of reason are sayables, because rational notions are corporeal and are thus not sayables, which are incorporeal.]
But we still should not think that any object of reason is a sayable. [From Diogenes Laertius, speaking for Diocles: “They [the Stoics] divide impressions into those which are sensory and those which are not. Sensory impressions are ones obtained through one or more sense-organs, non-sensory are ones obtained through thought such as those of the incorporeals and of the other things acquired by reason” (Long and Sedley pp.236-237: Diog. Laert. VII, 49-51; SVF 2.52, 55, 61). “According to them, some presentations are sensible, some are non-sensible. Those received through one or more sense organs are sensible; non-sensible are those which come through the intellect, for example, presentations of incorporeals and the other things grasped by reason” (Inwood & Gerson p.13).] Diocles classifies impressions as being either sensory or non-sensory. Non-sensory impressions are received through thought, and they include incorporeals and other things acquired by reason. Now, among these incorporeals are sayables. So since there are other things involved in the rational impressions, there must be other objects of reason that are not incorporeals [and thus not sayables], and in fact rational notions are not incorporeal. They are composed of real corporeal traces that the sensing body leaves in the leading part of the soul [which is also a body]. As we can see, for the Stoics there is physiology of the rational notion that is not distinct from the psychology of the rational notion. When Zeno says that notions are neither substances nor qualities, that would seem to suggest that they are not corporeal, since bodies are only found under these two categories. However, when we look at what Zeno goes on to say, we see that what he has in mind is not so much the substance of a rational notion but more about the rational notion’s relation to the object that it represents. The rational notions resemble bodies that left their imprint, and in that way are like the substances and qualities [of those external things. I do not grasp this part, but I will guess the idea is the following. When Zeno says that rational notions are neither substances nor qualities, he means they are not the same substances and qualities of the bodies that originally left their mark on the soul.] But that does not mean that rational notions cannot themselves be corporeals.
Malgré ces difficultés, il y a des raisons sérieuses de ne pas confondre l’exprimable avec n’importe quel objet de la raison. Dioclès classant les représentations en sensibles et non sensibles distingue dans les secondes qui arrivent « par la pensée » celle des « incorporels et des autres choses perçues par la raison » 2. Comme les exprimables doivent sûrement être rangés dans les incorporels, il y a donc encore d’autres objets de la raison qui ne sont pas les incorporels et en effet les notions rationnelles ne sont nullement des incorporels. Elles sont issues et composées de traces réelles que les corps sensibles laissent dans la partie hégémonique de l’âme. Il y a là une physiologie de la notion que les Stoïciens ne distinguent pas du tout de sa psychologie 3. Lorsque Zénon dit que les notions ne sont ni des substances ni des qualités4, il semble bien par là leur refuser un corps, puisque les corps se trouvent seulement dans ces deux catégories ; mais la suite du texte fait voir qu’il songe moins à la substance même de la notion qu’à son rapport avec l’objet qu’il représente : c’est en ce sens qu’elles sont comme des substances et comme des qualités, c’est-à-dire semblables aux corps qui en ont laissé l’empreinte ; mais ceci n’empêche pas qu’elles ne soient en elles-mêmes de nature corporelle : comment, sans cela, pourrait-on dire que la science qui contient de tels objets de représentation est un corps? 5 L’art et la science reposent toujours sur la conservation des empreintes par la mémoire.
(18)
2. Diog. La. VIII 51 (S.V.F. II 24,18).
3. Nous entendons ici par notions (ἐννοίαι) non pas tous les objets pensés (νοούμενα) qui contiendraient aussi les sensibles et les exprimables, mais seulement les notions générales comme celles du bien.
4. μήτε τινὰ μήτε ποία Stob. Ecl. I. p. 136 (S.V.F. I 20).
(18)
2.1.8
[The Corporeals of Reason are Rational Notions and the Incorporeals are Sayables and Rational Impressions]
(p.18-19: “On comprend par là la distinction qu’il y a à établir ...”)
[In sum: Some objects of reason are corporeal (like rational notions) and others are incorporeal (like the sayables of the rational impressions).]
So there are objects of reason, some of which are corporeal, including the rational notions, and others are incorporeal, including the sayables. [Recall from section 2.1.5 these quotes from Sextus Empiricus: “They [the Stoics] say that a ‘sayable’ is what subsists in accordance with a rational impression, and a rational impression is one in which the content of the impression can be exhibited in language” (Long & Sedley I p.196: Adv. Math. III, 70; SVF 2.187); “… others [other impressions, namely, rational impressions] have a nature like that of the incorporeal sayables (lekta), and the commanding-faculty [ἡγεμονικόν] is impressed in relation to them, not by them” (Long & Sedley I. p.163: Adv. Math. VIII, 409; SVF 2.85;)] Sextus Empiricus, by describing the object of a rational representation as a sayable, has confused the species for the genus. [I am not sure, but the idea here might be that the sayable is a species of the genus rational impression, so by saying that all rational impressions are sayables, he has confused the species with the genus. Or perhaps a distinction should be made between rational notions and rational impressions, with one being the species of the other, and with one being equated with or including sayables.] This confusion is understandable given that he is limiting his topic to the objects of logic, all of which are reducible to sayables.
On comprend par là la distinction qu’il y a à établir entre l’exprimable qui est incorporel et les autres objets de la raison qui sont corporels. On voit aussi que Sextus, en qualifiant d’exprimable l’objet de la représentation rationnelle en général, a pris l’espèce pour le genre. Ceci est d’ailleurs fort explicable, puisque, dans le passage en question, il a l’intention de parler seu- | lement des objets de la logique, et que ces objets se réduisent aux exprimables.
(18-19)
2.1.9
[The Coincidence of the Real Predicates of Bodies and the Logical Predicates of Propositions]
(p.19: “Il nous fallait d’abord indiquer ces fausses conceptions de l’exprimable ...”)
[In sum: The incomplete sayable, when articulated verbally, forms an incomplete predicate. Physical things have predicates, but we should not think of these predicates as being corporeal but rather as being incorporeal sayables. Also, propositions have predicates, but we should not think of the incorporeal predicate that the propositions express as being logical entities built into the proposition but rather as senses that are expressed by the proposition. Under these conceptions we can view the real predicate (of corporeal bodies) and the logical predicate (of propositions about corporeal bodies) as coinciding entirely.]
We needed first to consider these incorrect conceptions of the sayable before moving on to the correct ones. Besides substances and properties, which are both bodies, there exists nothing else in nature. [I am not sure if this is relevant, but considering seeing the four Stoic categories or genera: {1} ὑποχείμενον (substance / substrate), {2} ποιόν (quality / qualified), {3} πως ἔχοv (disposition / disposed), and {4} πρός τί πως ἔχον (relative disposition / relatively disposed). See Luhtala On the Origin of Syntactical Description in Stoic Logic, section 5.5.3.7 and the chart in 5.5.3.13 of that entry, and Sambursky Physics of the Stoics, section 1.4.] [Recall the following from section 1.2.6: There are two planes of being. {1} There is the plane of real and profound being, in which there are corporeals (physical bodies) that act and are acted upon by other corporeals on account of physical forces driving their interaction. {2} There is the plane of surface (and unreal) being. It lies at the metaphysical limit of the corporeal plane, remaining directly related to the interactions of corporeals, but not existing as something physically real among those interacting bodies. Here facts (predicates) play at the surface of physical being. They are the effects which are caused by the corporeal interactions. As such, the internal forces driving interaction on the deeper real plane are not exhausted by having produced effects on the incorporeal plane, since these effects are not matters of physical force (“De la même façon l’action d’un corps, sa force interne ne s’épuisent pas dans les effets qu’il produit” pp.12-13).] But, although the only real things existing in nature are bodies, these bodies have internal forces whose creative effects do not arise in the corporeal realm but rather produce incorporeal predicates on the [metaphysical] “surface” of bodies. The first type of sayables that we come across are incomplete sayables, which are identical to the predicates of bodies [caused by the internal forces at work in their non-causal interactivity.] To understand these incorporeal predicates properly, we need to do away with two incorrect notions. We must do away with the idea {1} that the predicate of a thing is something that exists physically, and {2} the predicate, understood as a logical entity that is a component of a proposition, is something that exists in thought. If we remove those notions from our conception of sayables, then we can instead conceive of the logical predicate and the real predicate as both being incorporeal and inexistent and as being wholly coincident.
Il nous fallait d’abord indiquer ces fausses conceptions de l’exprimable, possibles grâce à la pénurie et à l’obscurité des textes, pour établir la véritable. En dehors des substances et des propriétés qui, toutes deux, sont des corps, il n’existe rien dans la nature. Mais, nous l’avons vu, leur force interne se manifeste à la surface, et ces aspects extérieurs ne sont ni corps, ni parties de corps, mais des attributs (κατηγορήματα) incorporels. Le premier genre d’exprimables que nous rencontrons, l’exprimable incomplet, est identique à cet attribut des corps. Il faut, pour bien le comprendre, se défaire de cette idée que l’attribut d’une chose est quelque chose existant physiquement (ce qui existe c’est la chose même), et de cette autre idée que l’attribut, sous son aspect logique, comme membre d’une proposition est quelque chose existant dans la pensée. A cette condition on pourra concevoir que attribut logique et attribut réel qui, à la vérité, sont tous deux incorporels et inexistants coïncident entièrement.
(19)
2.1.10
[Predicates as Acts Expressed by Verbs and not as Properties Expressed by Adjectives]
(p.19: “Les attributs des êtres sont exprimés non pas par des épithètes ...”)
[In sum: The predicates of beings are not expressed by adjectives indicating their properties but rather by verbs indicating their actions.]
The predicates of beings are not expressed by adjectives that indicate their properties but rather by verbs that indicate their acts. [See épithète as attributive adjective.]
Les attributs des êtres sont exprimés non pas par des épithètes qui indiquent des propriétés, mais par des verbes qui indiquent des actes (ἐνέργήματα).
(19)
2.1.11
[For the Stoics, the Subject and Predicate Cannot Coincide by Means of Class Inclusion]
(p.19-20: “Si nous considérons maintenant la nature de la proposition ...”)
[In sum: We need to understand how the logical and physical predicates can coincide. With regard to the subject and predicate coinciding, this can happen as participation in Plato’s philosophy or inclusion in Aristotle’s philosophy. However, for the Stoics, there are no universals. ]
[Recall from Luhtala On the Origins section 5.5.0.1 that the Stoics regarded logic as being comprised of rhetoric and dialectic. Rhetoric deals with continuous speech, while dialectic deals with argument by means of question and answer. And recall from section 5.5.2 that Stoic dialectic has two parts. Generally speaking, there is a part that lies on the level of language itself and a part that deals more with the conceptual meanings the language expresses. Specifically, the two parts are: {1} “expression,” equivalent to σημαίνοντα (that which signifies), to φωναι (vocal sounds), and to λέξις (word), and {2} “meaning” or “sayables,” equivalent to σημαινόμενα (that which is signified, meaning), and to λεκτά (sayables). And recall from section 5.5.4.7.1 that “The Stoic proposition [ἀξίωμα] is an item of meaning (σημαινόμενον), i.e. a complete sayable (λεκτόν)” (Luhtala 109, bracketed insertion mine). So the proposition (ἀξίωμα) is an item of meaning (σημαινόμενον), and an item of meaning falls under Stoic dialectic.] If we consider the nature of the proposition in dialectic, we find a solution to the problem of predication where the logical predicate of a proposition coincides entirely with the predicate as we have defined it [as the predicate of a corporeal.] This problem was one of the greatest concerns of the schools that followed Socrates, and we can say that it is for the sake of resolving these difficulties that Plato devised his theory of Ideas. The Stoics offer a profound an ingenious solution that requires no appeal to a theory of Ideas. If, in a proposition, the subject and predicate are considered to be concepts of the same nature, namely, that both are concepts indicating classes of objects, we will encounter great difficulty in understanding the nature of the connection indicated by the copula. But, if they are different classes, then each exists apart from and outside of the other, and thus they cannot be said to be linked. [I am not sure why, however. It seems a link between classes, at least a syntactical sort, requires that the classes somehow overlap. The idea might be that by being of different classes, subject and predicate are too different to enter into a syntactical sort of relationship. For example, if subjects were corporeal and predicates were incorporeal, then their metaphysical incompatibility would prevent them from being linked by a copula. That is my guess so far.] If the two classes of objects (of subjects and predicates) are identical, then we can only have judgments that state identity. The relations of participation in Plato or inclusion in Aristotle present solutions to this problem. [I am not sure what is meant here. Perhaps the idea is of falling under a class. The subject falls under the more broader class indicated by the predicate. Balls are round. The subject and predicate are both classes of things (the class of balls and the class of round things) but the class of balls is included in the class of round things. Or we might say that balls participate in the more general idea of roundness, maybe.] [The next idea might be the following, but I am not certain. Modern thinkers would understand these classes of things involved in this sort of class inclusion to be thoughts or concepts. But the ancients saw these classes or Ideas as real beings and not just thoughts or concepts. Yet, for the Stoics, reality is not found in any generalities; it is only found in individuals. (See Luhtala section 5.5.0.3; and section 5.5.3.9 might also be relevant.) And since there are no universals, this link between subject and predicate cannot be understood as a matter of participation or inclusion. And in fact, each individual is a particular idea irreducible to any other (as with class inclusion). In order for these realities (of the individual) to participate in or be included in another, the two realities would need to fuse indiscernibly with one another (or, that one same individual can have in itself more of its own quality), which is absurd. (I probably have that second possibility wrong. My guess is the following. Suppose we have “balls are round”. This absurd conception would say that balls, which have a certain degree of roundness, given their physical imperfections, include also within themselves the more perfect roundness which allows us to say that they have a degree of roundness in the first place. But if the ball already contains within it perfect roundness, then it cannot also have imperfect roundness.)] So it seems then that two different realities cannot coincide.
Si nous considérons maintenant la nature de la proposition (ἀξίωμα) dans la dialectique, nous trouvons une solution du problème de l’attribution qui fait coïncider entièrement l’attribut logique de la proposition avec l’attribut tel que nous venons de le définir. Ce problème avait été une des plus grandes préoccupations des écoles qui avaient suivi Socrate, et on a pu dire que c’est pour en résoudre les difficultés que Platon avait construit sa théorie des idées. Les Stoïciens, précédés sur ce point, semble-t-il, par quelques philosophes de l’école mégarique, en ont donné une solution profonde et ingénieuse, qui n’exige aucun appel à une théorie des idées. Si, dans une proposition, le sujet et le prédicat sont considérés comme des concepts de même nature, et particulièrement des concepts indiquant des classes d’objets, on aura grand’peine à comprendre la nature de la liaison indiquée par la copule. Si ce sont des classes différentes, chacune existe à part, en dehors de l’autre, et elles ne peuvent se lier. Si elles sont identiques, nous sommes réduits à des jugements d’identité. La liaison de participation que Platon avait trouvée, et celle d’inclusion qu’Aristote utilisait de préférence, étaient | une solution possible à ces difficultés. Mais de telles solutions, qui, pour les modernes, ne concernent que les pensées, avaient, pour les anciens, une portée métaphysique, que l’on n’en pouvait disjoindre. Les termes du jugement désignent en effet non seulement des pensées, mais des êtres réels. Or si la réalité se concentre, comme chez les Stoïciens dans l’individu, une pareille théorie est inadmissible. En effet chaque individu non seulement possède, mais est une idée particulière (ἰδίως ποιόν) irréductible à tout autre. Pour que ces réalités participent l’une à l’autre ou soient incluses l’une dans l’autre, il faudrait que deux individus fussent indiscernables l’un de l’autre, ou qu’un même individu pût avoir en lui plus d’une qualité propre, ce qui est absurde 1. Deux réalités ne peuvent coïncider.
(19-20)
1. Philon de incorruptib. M. 236, 6 B (S.V.F. II 131, 6).
2.1.12
[The Stoics Use Verb Predicates (“The Tree Greens”) Instead of Copula Predicates (“The is Green”)]
(p.20: “Il restait une solution, c’était d’envisager...”)
[In sum: The way that the Stoics solve the problem of the subject coinciding with the predicate while being metaphysically uncoincidable is to reconceive the nature of the predicate. Normally it involves the copula, “to be,” and that copula serves to make the subject and predicate coincide as one class or object within another class. Consider “the tree is green”. Here the tree is included in the class of green things. But while conceptually this makes sense, metaphysically a corporeal thing cannot be inside a conceptual thing like a class. The Stoics, following the Megarians, solve the problem by stating judgments not with a copula but rather with a verb indicating an activity. So we would instead say, “the tree greens.” Here the predicate is no longer an object but is rather a fact or event, and the subject is not said to be included in that fact or event. Rather, the activity of the subject (done or received) has a causal result among its incorporeal predicates.]
The solution to this problem [of how to reconcile the subject and predicate] is to reconceive the nature of the predicate. The Megarians did not state judgments under the usual form, where a copula is used, as in “The tree is green”. Rather, they would say, “the tree greens”. The Stoics show us how this presents a solution to the problem of predication. When we put aside the copula and express the subject using a verb in which the predicate adjective is not directly manifest, then the predicate, considered as the whole verb, no longer appears as expressing a concept (an object or class of objects), but rather it only expresses a fact or event. [When we say, “the tree greens”, we are not saying that the tree belongs to a class of green things. Rather, we are saying that the tree is involved in an event of greening. Here we no longer have the problem of explaining how subject and predicate can coincide of they are metaphysically different types of realities in the world.] Thus the proposition no longer needs the reciprocal penetration of two objects, impenetrable by nature; rather, this construction simply expresses a certain aspect of an object, insofar as it accomplishes or undergoes an action. This aspect of the object is not a real nature, that is to say, it is not a being that penetrates the object. Rather, this aspect [expressed by the verb] is the act that is the very result of its [the object’s] activity or of the activity of another object acting upon it. The content of the proposition, which is what it signifies, is thus neither an object nor a relation of objects.
Il restait une solution, c’était d’envisager tout autrement la nature du prédicat. L’on sait que certains mégariques refusaient d’énoncer les jugements sous leur forme habituelle, à l’aide de la copule est. On ne doit pas dire, pensaient-ils « L’arbre est vert » mais « L’arbre verdoie ». Comment c’était là une solution du problème de la prédication, c’est ce que les Stoïciens nous font voir. Lorsqu’on néglige la copule est et que l’on exprime le sujet par un verbe où l’épithète attribut n’est pas mise en évidence, l’attribut, considéré comme le verbe tout entier, apparaît alors non plus comme exprimant un concept (objet ou classe d’objets), mais seulement un fait ou un événement. Dès lors la proposition n’exige plus la pénétration réciproque de deux objets, impénétrables par nature, elle ne fait qu’exprimer un certain aspect d’un objet, en tant qu’il accomplit ou subit une action ; cet aspect n’est pas une nature réelle, un être qui pénètre l’objet, mais l’acte qui est le résultat même de son activité ou de l’activité d’un autre objet sur lui. Le contenu de la proposition, ce qui est signifié par elle n’est donc jamais un objet, ni une relation d’objets.
(20)
2.1.13
[Judgments Only Have Active Predicates]
(p.20: “Il suit de là que les Stoïciens n’accepteront que ...”)
[In sum: Since the Stoics only allow propositions using a verb rather than a copula, that means they exclude all judgments where the predicate indicates a real property of the subject or indicates a relation of the subject with other concepts. So they would not say “a body is warm” but instead “a body is warming up”. The distinction in predicates then are matters of how the event is expressed, and these metaphysical or physical matters are understood linguistically as well. Thus the Stoic classification of predicates follows their grammatical classification of verbs. Under one class of distinction, we have predicates articulated using συμβάματα, which are personal verbs that indicate the action of a subject, as in “Socrates walks”, and παρασυμβάματα, which are impersonal verbs. Another distinction we make is between {1} the direct predicates, composed of a verb with a complement that undergoes action, {2} passive predicates, which are passive verbs, {3} reflexive predicates, which are reflexive verbs, and finally {4} predicates [and verbs] that are neither direct nor indirect.]
So the Stoics only allow propositions that contain a verb, which blurs the copula and predicate together. This means that they exclude all judgements where the predicate indicates a real property of the subject and that indicate a relation with other concepts. So that means that what is expressed in the judgment is not a property, as in “a body is warm” (un corps est chaud); rather, what is expressed in the judgment is an event, as in “a body is warming” (un corps s’échauffe). [I would like to make a comment here. We might wonder about bodies that have been about the same temperature for a while, or a tree that has been green all summer. I think the idea here might be the following, and I am drawing inspiration from Deleuze’s Logic of Sense. When we say that a tree greens, we mean that it is currently involved in an activity that goes from not green (being bereft of leaves), through various shades of green (as the leaves develop and change), then back to not green (as the leaves turn color in autumn). So we might be stopping it in some present instant and saying that it greens, but what we means is that we caught it in an activity with a very broad temporal distribution.] In the classification of predicates, the Stoics do not distinguish them like Aristotle does by their mode of relation to the subject, being more or less essential or accidental. Rather, the Stoics only distinguish the different ways that the event is expressed. Also, their classification of predicates follows closely and is even identical with their grammatical classification of verbs. We first distinguish the συμβάματα, which are personal verbs that indicate the action of a subject, as in “Socrates walks”, from the παρασυμβάματα, which are impersonal verbs. Another distinction we make is between {1} the direct predicates, composed of a verb with a complement that undergoes action, {2} passive predicates, which are passive verbs, {3} reflexive predicates, which are reflexive verbs, and finally {4} predicates [and verbs] that are neither direct nor indirect. [See Luhtala On the Origins section 5.5.4.3.2.]
Il suit de là que les Stoïciens n’accepteront que les propositions contenant un verbe : dans le verbe se confondent pour eux prédicat et copule. On voit par là tous les jugements qu’ils excluent, tous ceux dont l’attribut indique une propriété réelle du sujet, et qui indiquent un rapport entre concepts. Ce qui s’exprime dans le jugement, ce n’est pas une propriété comme : un corps est chaud, mais un événement comme : un corps | s’échauffe. Dans la classification des attributs, ils ne les distingueront pas comme Aristote par le mode de leur liaison au sujet, plus ou moins essentielle ou accidentelle : ils ne veulent y distinguer que les diverses façons dont l’événement peut s’exprimer. Aussi leur classification suit de près, est même identique à la classification grammaticale des verbes. On distingue d’abord les συμβάματα, verbes personnels indiquant l’action d’un sujet (Socrate se promène), et παρασυμβάματα, verbes impersonnels (Σωκράτει μεταμέλει). D’autre part on distingue les prédicats directs, composés d’un verbe avec un complément qui subit l’action ; les prédicats passifs, qui sont les verbes passifs ; et en eux les prédicats réfléchis (verbes réfléchis) ; enfin ceux qui ne sont ni directs, ni passifs (comme φρονει) 1.
(20-21)
1. Porphyr. Ap. Ammon. In Ar. De interpret., p.44, 19 (S.V.F. II 59, 25).
(21)
2.1.14
[The Only Type of Connection in Propositions is Between Subject and Event]
(p.21: “Il ne faut pas voir dans la substitution de cette forme verbale...”)
[In sum: So the Stoics only accept propositions of fact (which express events) as propositions. But the relation between subject and predicate is not that of essential or accidental predication but is rather the relation of an event to its subject.]
There is an important philosophical move here in substituting this verbal form for the copula. The Stoics only accept propositions of fact [which express events rather than a substance’s possession of properties.] But, this fact can still have certain conventional logical properties, like being necessary or contingent, true or false, possible or impossible, and so the different modalities can still apply. Nonetheless, these modalities are not entirely like those applying to formations where there are essential or accidental connections between the subject and predicate. Under the Stoic conception, there is only one kind of connection, namely, the connection between an event and a subject.
Il ne faut pas voir dans la substitution de cette forme verbale à la copule distincte une simple subtilité. Les Stoïciens veulent indiquer par là qu’ils n’acceptent d’autres propositions que les propositions de fait. Sans doute le fait peut être lui-même nécessaire ou contingent, vrai ou faux, possible ou impossible, et en ce sens, les différentes modalités sont encore admises. Mais c’est, on le voit, en un sens bien différent de celui d’une logique des concepts où ces modalités reposaient sur la liaison essentielle ou accidentelle du sujet avec l’attribut. Ici nous n’avons plus qu’un genre de liaison, une liaison qui, au sens de la logique d’Aristote, était accidentelle (et que les Stoïciens continuent d’ailleurs à désigner par le mot συμβάματα), à savoir celle de l’événement à son sujet.
(21)
2.1.15
[The Coincidence of Irreal Logical Predicates and Irreal Predicates from Corporeal Activity]
(p.21-22: “Le problème de l’attribution est donc résolu en enlevant ...”)
[In sum: Now, the fact that the predicate effected by corporeal action and the logical predicate are both irreal is the reason why they can both coincide.]
So the problem of predication is resolved by stripping predicates of any true reality. The predicate is neither an individual nor a concept. Rather, the predicate is an incorporeal, and it only exists in thought. Now, the results of the actions of corporeals are these incorporeal predicates, and there is nothing that differentiates the logical predicates [expressed in propositions] from the predicates of things, which are the results of the corporeal actions. [See Luhtala On the Origins section 5.5.4.0.1.] Both senses of predicate are designated with the same word, κατηγόρημα, and both find their expression in verbs. And of course, both are incorporeal and not real. [The next part I may missummarize, so please consult the quotation below. The basic ideas might be the following. The predicate as event is like the action producing it, except the predicate is not real while the action is. And the logical predicate in Stoicism is not like the concepts of thought that we predicate to subjects using copula formulations, because these concepts can represent certain non-temporal and essential things, while the Stoic logical predicate can only represent transitory and accidental facts. But it is because of them both lacking reality in these ways that they can coincide.]
Le problème de l’attribution est donc résolu en enlevant aux prédicats toute réalité véritable. Le prédicat n’est ni un individu, ni un concept ; il est incorporel et n’existe que dans la simple pensée. On chercherait vainement en quoi le prédicat logique de la proposition pourrait différer des attributs des choses, considérés comme résultats de leur action. Tous deux sont désignés par le même mot κατηγόρημα, et trouvent leur expression dans des verbes ; tous deux sont incorporels et irréels. Du côté du réel, la réalité de l’acte a pour ainsi dire été atténuée au profit de celle de l’être permanent qui le produit : du côté de la logique, l’attribut a été privé de sa dignité de concept objet de la pensée, pour ne plus contenir qu’un fait transitoire et acciden- | tel. Dans leur irréalité et par elle, l’attribut logique et l’attribut des choses peuvent donc coïncider 1.
(21-22)
1. Cf. Clem. Alex. Strom. VIII (S. V. F. I 263, 1).
(22)
2.1.16
[Facts Are in the Mind]
(p.22: “Les sciences expérimentales et les philosophies sceptiques ...”)
[For the Stoics, facts exist in thought.]
Experimental science and skeptical and critical philosophies have led us to understand facts or events as being matters of true objective reality and to conceive of objects as resulting from the synthesis of a great number of facts and not as the subject of the predication of these facts, like in the Stoic conception. And under this newer framework, facts are the only object of experience, and our thoughts about facts stand at a distance to them. [This is unlike the Stoic understanding of facts, which locates them in the mind.] Given this context, we struggle to conceptualize the Stoic doctrine. But the Stoics say that facts are incorporeal and can only exist in mind. Thus for the Stoics, facts become the matter of their dialectic [see the comment on dialectic above (section 2.1.11), as explained by Luhtala.] The common trait of all ancient logics is their realism. The ancients held that one can never have the thought of something that does not exist. [I will missummarize the next part, so you need to consult the quotation below. For now I am guessing the main idea is the following. The Stoics, despite appearances to the contrary, remained faithful to these ancient realist tendencies. I cannot understand why, but I am guessing that the incorporeal predicates inhere in thoughts, and those incorporeal predicates are caused by real situations. So it is not a matter of thinking things that do not exist, as if we have a thought on one side and the non-existing predicate as the object of thought on the other. Rather, we have the non-existing predicate on the side of thought and on the other side are the existing things that share that predicate. But please read the text to see what it really means.]
Les sciences expérimentales et les philosophies sceptiques ou critiques en accord avec elles nous ont accoutumés à voir dans le fait ou l'événement la véritable réalité objective et à considérer un objet comme un résultat et une synthèse d'un grand nombre de faits, plutôt que comme le sujet d'attribution de ces faits. Le centre du réel s'est pour ainsi dire déplacé. C'est cette circonstance qui peut rendre cette doctrine stoïcienne assez pénible à concevoir. Les faits sont le seul objet d'expérience et la pensée qui cherche à les observer et à découvrir leurs liaisons reste étrangère à eux. Au contraire les Stoïciens, en admettant que les faits étaient incorporels et n'existaient que dans la pensée, pouvaient en faire, nous ne dirons pas même l'objet, mais la matière de leur dialectique. Au fond le caractère commun à toutes les logiques anciennes est d'être réalistes : jamais les anciens n'ont cru que l'on pouvait avoir la pensée de quelque chose qui n'existe pas. Les Stoïciens, malgré les apparences, sont restés fidèles à ces tendances : si la pensée dialectique n'enserre plus, dans la proposition, des réalités, l'attribut pensé n'en est pas moins identique à l'attribut objectif. En refusant à la pensée la réalité telle qu'ils la conçoivent, ils ne peuvent ainsi que la refuser à son objet.
(22)
2.1.17
[Sayables Are not Rational Representations]
(p.22-23: “Les attributs ne sont qu’une certaine espèce d’exprimables”)
[In sum: Sayables can also be complete, and complete sayables can be conjoined in acts of reasoning. Rational representations are judgements and thus require a subject and predicate. But as we can see in such common Stoic examples of propositions as “it is day”, there is no subject. Thus sayables are not rational representations.]
Predicates are only a certain type of sayable. They are incomplete sayables. They become complete sayables when we answer the question, who is the subject of the action? [See Luhtala On the Origin section 5.5.4.1.1.] As such, these are simple propositions. The other complete sayables are compound propositions that are obtained through a combination of simple propositions. An example of this is what we today call the hypothetical proposition. [The whole act of reasoning that thinks on these combined propositions is not itself a sayable. Rather, it is a sequence of sayables.] The essence of the λεκτόν is thus a predicate or event, whether it be with or without a subject. It is interesting that in Porphyry’s account, the proposition itself is called a predicate, as it is a complete predicate. All the attention of the dialecticians was on the predicate sayable. In the most common examples of Stoic dialectics, like “it is day”, “ρῶς ἐστι”, etc., the propositions express facts without any subject. Thus the sayable is not at all a type of rational representation. It is instead only a kind of fact or event. [I do not follow the reasoning there, but I will guess it is the following. Rational reasoning always requires a subject, because rational reasoning involves judgments, which always predicate a subject. But in such cases as “it is day”, there is no subject, even though there is a sayable. So sayables are not kinds of rational representation.] As such, it forms the matter of all logic. We will now follow the effects of this conception in the theory of judgement and reasoning.
Les attributs ne sont qu’une certaine espèce d’exprimables 2. Ce sont les exprimables incomplets, que l’on transformera en propositions ou en exprimables complets en répondant à la question : « Qui est le sujet de l’action? » 3. Ce sont là des propositions simples : les autres exprimables complets seront des propositions composées que l’on obtient par une combinaison de propositions simples, dont un exemple est ce que nous appelons aujourd’hui la proposition hypothétique (le συνημμένον des Stoïciens). Enfin ces propositions se combinent en raisonnements qui ne sont jamais appelés des exprimables 4, mais qui sont plutôt une suite d’exprimables. L’essentiel du λεκτόν, c’est donc l’attribut ou l’événement, soit avec le sujet, soit sans le sujet. Il est intéressant de voir que, dans l’exposition de Por- | phyre, la proposition elle-même est appelée un attribut (κατηγορούμενον) ; c’est seulement un attribut complet (τέλειον) 1. Toute l’attention du dialecticien porte sur l’exprimable attribut. Dans les exemples les plus familiers à la dialectique Stoïcienne, comme : il fait jour, ρῶς ἐστι, etc., les propositions expriment des faits sans aucun sujet d’inhérence. L’exprimable n’est donc pas toute espèce de représentation rationnelle, mais uniquement celle du fait et de l’événement. Il forme, comme tel, la matière de toute la logique ; nous allons maintenant essayer de suivre les effets de cette conception dans la théorie du jugement et du raisonnement.
(22-23)
2. Clem. Alex. Strom. VIII 9, 26 (S.V.F. I 109, 24). Diocles Magnes Diog. La. VII 63 (II 59, 11).
3. Ib. (II 58, 30).
4. Cf. la classification de Philon : de agric. 139 (S.V.F. II 58, 38).
(22)
1. S.V.F. II 59, 30.
(23)
Bréhier, Émile. 1962. La théorie des incorporels dans l'ancien stoïcisme. 3rd Edn. Paris: Vrin.
PDF of a microfilm version available at:
Also cited:
Inwood, Brad, and Gerson, Loyd P. 2008. The Stoics Reader. Selected Writings and Testimonia, edited and translated by Brad Inwood and Loyd P. Gerson. Indianapolis and Cambridge: Hackett.
Long, Anthony A. and David N. Sedley. 1987. The Hellenistic Philosophers, vol.1: Translations of the Principle Sources, with Philosophical Commentary. Cambridge: Cambridge University Press.
Long, Anthony A. and David N. Sedley. 1987. The Hellenistic Philosophers, vol.2: Greek and Latin Texts with Notes and Bibliography. Cambridge: Cambridge University Press.
Sextus Empiricus. 2005. Against the Logicians, edited and translated by Richard Bett. Cambridge: Cambridge University.
.
No comments:
Post a Comment