9 May 2017

Luhtala ( On the Origin of Syntactical Description in Stoic Logic, “Stoic versus Peripatetic Notion of Subject”, summary


by Corry Shores


[Search Blog Here. Index tabs are found at the bottom of the left column.]


[Central Entry Directory]

[Stoicism, entry directory]

[Anneli Luhtala, entry directory]

[Luhtala, Syntactical Description in Stoic Logic, entry directory]


[The following is summary. All boldface and bracketed commentary are my own. Paragraph enumerations are also my own, but they follow the paragraph breaks in the text. Please forgive my distracting typos, as proofreading is incomplete.]




Summary of


Anneli Luhtala


On the Origin of Syntactical Description in Stoic Logic


Ch.5 The Stoics


5.5 Stoic Logic



The Component of Meaning (Σημαινόμενα)

Stoic versus Peripatetic Notion of Subject





Brief summary:

The Stoics and Peripatetics held different notions of the grammatical subject and of case. Specifically, the Peripatetics saw the nominative as a primitive or neutral sort of form of the substantive, with all other syntactically meaningful formations of that noun being case inflections of it. Grammarians used the metaphor of the other forms “falling from” the nominative form. The Stoics, however, saw all formations of the substantive as cases, including the nominative, with the metaphor now being that all these cases fall from a neutral form of the noun as it manifests abstractly in the mind (that is to say, in the mind we can conceive of the noun by itself, but as soon as we conceive its meaningful relations to other things, properties, actions, etc., it enters into syntactical relations and thereby takes on a case formation that indicates its syntactical role.) The causative case could be understood as indicating a body being acted upon by the state of affairs.




[The Stoic notion of the subject differs from the Peripatetic notion. In Stoic grammar, the subject function is related more to the direct case than to the oblique case in transitive clauses.]


The Stoic notion of subject is radically different from the Peripatetic notion. There is a canonical passage in Plutarch on the Stoic notion of subject, but this is not the source that bears out the radical distinction (103). The problem with this passage is that it “obscures the Stoic position by treating the Stoic, Peripatetic and Platonic propositions in comparable terms” (103). Luhtala then says that there are certain ways that the three traditions coincide, as in “the case when the object of description is the same, namely the ‘intransitive’ clause” (103). [I do not follow these grammatical concepts very well, so I will quote. Maybe the point is the following. For intransitive clauses, the Stoics and Peripatetics say something similar. And maybe that similarity is that they both think that the nominative case in intransitive clauses is joined to a predicate in some sense, with the Stoics saying it joins to a simple predicate and the Peripatetics to a finite verb in the predicate position. That cannot be right, so let me quote. Maybe Luhtala is not even saying this is a similarity.]

This passage rather obscures the Stoic position by treating the Stoic, Peripatetic and Platonic propositions in comparable terms. In some cases the description does indeed coincide in the various traditions. This is the case when the object of description is the same, namely the ‘intransitive’ clause. The Peripatetics would say that an ὄνομα (which is a noun in the nominative) is joined to ῥῆμα (which is a finite verb in predicate position) | while the Stoics would say that the nominative case is joined to a simple predicate.


[Luhtala’s next point might be that for transitive clauses the Stoics say something very different from the Peripatetics, but I am not sure. We begin with a notion of a ‘transitive’ state of affairs. To make a description of it, we would join the predicate to cases. I am still not sure I understand this notion of cases yet, because I would think that a case is a modification of a noun or similar sort of thing rather than that term itself. Maybe ‘case’ refers to such a term under a case modification, but I am not certain. At any rate, the Stoics distinguish two cases involved in transitive clauses: direct and oblique. From what I understand, the oblique cases would designate participants in the action, but doing so in the predicate. Luhtala says that the direct case comes closest to designating the subject function.]

But this is only one aspect of Stoic propositional analysis. As to the description of a ‘transitive’ state of affairs, the Stoics probably said that a predicate was joined to cases (πτώσεις). The Stoics distinguished between the two cases involved in the ‘transitive’ clause in terms of the distinction between direct (ὀρθή) and oblique cases (πλάγιαι πτώσεις), of which the direct case (ὀρθὴ πτῶσις) comes closest to what we understand as the subject function. In the following discussion I will explore to what extent the distinction between direct and oblique cases is related to the ‘syntax’ of the propositions.


[The Stoics revised the notion of case to include the nominative.]


Πτῶσις (‘case’) was a linguistic term current at the time of the Stoics, and they adopted it for their own use. But they restricted its sense “to nominal inflection whereby even the nominative was regarded as a case (Schmidt 1979: 80)” (104). Luhtala views “the revision of this notion as a reflection of the development of Stoic propositional analysis” (104). Aristotle’s sense of case was different. [I do not follow these points so well, and I will not be able to summarize them. Before I offer a guess, let me quote from a prior section:

Aristotle’s effort to single out the subject function results in a distinction between ὄνομα (‘noun’ / ‘subject’) and πτῶσισ ὀνόματος (‘nominal case’), | a distinction we know as being that between the nominative case and the oblique cases. But for Aristotle ὄνομα is not a case at all. The ὄνομα is described as the only kind of noun that can appear in a proposition, which is in the form ‘man is’ or ‘man walks’, thus the nominative case. Given that the introductory chapters of the Peri hermeneias are influenced by Plato’s Sophist, in which the proposition is in the form ‘Theaetetus sits’, Aristotle probably has this kind of a proposition in mind when he defines the subject (ὄνομα). For such a proposition, this description sufficiently distinguishes it from the predicate (ῥῆμα). But this is not the case in propositions of the form ‘Socrates is white’ or ‘Man is an animal’, which are much more central to Aristotle's theory, because both parts of the proposition contain a noun in the nominative case.


Now let me next quote the current passages we are summarizing:]

In Aristotle’s propositional analysis the notion of πτῶσις (‘case’) was used in opposition to ὄνομα and ῥῆμα, the subject and the predicate, probably in the sense of a deviation from the norm, which was the minimal clause. For Aristotle, the notion of case covered all the inflectional forms of the noun and the verb, which were not relevant for the simple proposition. The nominative, i.e. the subject function, was not a case at all for Aristotle. In its Stoic usage this term was applied to nominal inflection including the nominative, obviously because all cases are relevant for the Stoic proposition. Grammatical commentators were keen to explicate the Stoic distinction between ‘direct’ and ‘oblique’ cases by means of a metaphor of a pencil falling in an upright as opposed to an oblique position. I take such justifications to be invented later to explain something that had been introduced for the purposes of propositional analysis. Observations of the facts of the Greek language are likely to lie behind the revised notion of case.


[My guess is that the idea here is the following. For Aristotle, the subject function is expressed only in the nominative case, because by means of that case the subject in question can then be predicated in the proposition. However, he perhaps did not term this situation as a ‘case’, because he defined ‘case’ as a matter of an inflected form derived from that predicatable substantive form. And maybe the Stoics thought instead that the subject function can be taken by other cases. In transitive clauses, the substantives can be of a “direct” or “oblique” case. Maybe the “direct” case is one where substantive is thought of as agent and oblique as patient, but I do not know right now.]

[The Stoics include all forms of substantives as inflected cases because they think the unmodified form is in the mind and any linguistic expression of it would be one or another modification of that conception.]


There Peripatetics challenged the Stoics reconception of case to include the nominative. [The idea as I understand was that the nominative form is not to be thought of as a case, because it is not to be understood as inflected from some other more basic form. But I am not sure. Then maybe also the Stoics considered the nominative form as the direct case, and the others as the oblique cases, where the Peripatetics saw all cases as oblique, because they fall from the non-inflected nominative form, like pencils landing obliquely. Then it seems the Stoic defense is to use this metaphor of falling obliquely, but now, rather than falling from a more basic form, to instead fall from some conception of the substantive that in its conceptual form is not yet modified in the sorts of ways that inflections do.]

The Stoics had to justify their novel use of case (πτῶσις), which was criticized by the Peripatetics, who explained the origin of this term so that all the oblique cases ‘fell’ from the basic form, the nominative. They thus raised the question as to the origin of the Stoic interpretation of case; which includes even the nominative. The Stoics answered that all cases, including the nominative, fall from the mind (ἀπὸ τοῦ νοήματος τοῦ ἐν τῇ ψυχῇ, ἀπὸ τῇς ἐννοίας).170 “When we have in our mind the thought of | Socrates,” as Ammonius explains, “we utter ‘Socrates’”. The Peripatetics are said to have then argued that even the other parts of speech should be called cases because they also fall from the mind; we are not informed of the Stoics’ answer to this argument (In de int. GAG IV.5: 42,30-43,20). As to the justification of the ‘direct’ case, the nominative is called ‘direct’ by the Stoics because, according to Ammonius, it is the archetype of linguistic expression (διὰ τὸ ἀρχέτυπον τῆς κατα τῆν ἐκφώνησιν προφορᾶς εἶναι).


170 Ammonius (In de int. GAG IV. 5: 42,30-43,24 = Hülser 1987 3: 918, fr. 776); | Stephanus (In de int. 10,20-11,5 = Hülser 1987 3: 920, fr. 777); Leo Magentinus (In de int. 104a n.2 = Hülser 1987 3: 924, fr. 779).


[Stoics would probably regard complete transitive or direct propositions as prototypical cases of linguistic expressions.]


Luhtala wonders whether “Ammonius is correct in thinking that the Stoics regarded ‘Socrates’ as an archetype of linguistic expression” (105). [See the text, as I am not sure I follow all the details of the next point. One idea seems to be that the Stoics would have placed the substantive into a proposition, and the prototypical sorts of propositions or predicates would be “direct” or transitive sorts. Again, please consult the text.]

[There is a passage from Apollonius and Herodian that indicates the syntactical basis for how the Stoics use the notion of case.]


[Luhtala will first quote a passage that probably comes from Apollonius and Herodian. It indicates how the Stoics understood the notion of case, including the nominative, as falling away from something broader. In this description, the case forms fall away from the generic form of the noun. I am not sure, but I assume the generic form is an uninflected form somehow. Her comments come in the next paragraph, so we will discuss it more then.]

The syntactical basis of the Stoic usage of the notion of case (πτῶσις) would seem to be revealed most clearly in a passage attributed by a scholiast to Apollonius and Herodian:

The generic noun comprises everything covered by the expression of noun, just as the genus man does; the nominative case falls from the | generic onto me, onto you, and onto every state of affairs (πρᾶγμα).  It is a case because it falls from the generic noun (γενικὸν ὄνομα) and it is called direct because it shows directly the substance of states of affairs (ὀυσία)173 and is construed with direct verbs (ὀυσία τοῦ πράγματος), that is active (ἐνεργητικοῖς).174


173 [See p.106 for Greek text in the original, citing (GG I.3: 548, 27-30).]

174 [See p.106 for Greek text in the original, citing (GG I.3: 546, 5-14).]


[For the Stoics, the direct case is a modification of a more abstract sort of noun without case. The direct case falls from this neutral noun, and the direct case shows the substance (substantives as agents or patients) involved in the state of affairs, and it takes a direct verb, as for example an active verb.]


Luhtala then discusses this passage (above). She first says that the abstract noun is an abstract entity that does not bear any case to it. [I am not entirely sure how to conceptualize it clearly. Perhaps the idea here is like the following. Something can be a noun, which is a grammatical sort of determination, and that thing can be conceived its pure nounness, that is, simply by itself. But whenever that thing is conceptually or linguistically placed into a meaningful relation with other things (or properties or events), then it takes on a syntactical sort of status and would thus need to have a case formation indicating that syntactical relation.] This means that “The noun (ὄνομα) is thus a non-syntactic unit of speech” (106). [Next we turn to the notion of the direct case. It is not here characterized as the nominative. I found this definition for the direct case, in Barry Blake’s Case:

direct case. In a number of Indo-Aryan languages the unmarked case covers S, A and P and is often referred to as the direct case. (Blake 199)

S. 1. The single argument of a one-place predicate. 2. In the context of word order it refers to subject as in the abbreviation SOV referring to subject–object–verb order. (Blake 204)

A. The agent argument of a transitive verb plus any other role that is treated in the same way grammatically. (Blake 195)

P. The patient argument of a transitive verb plus any other role that is treated in the same way grammatically. (Blake 203)

It seems more or less that the direct case refers either to the agent or patient of the indicated activity, and maybe it also refers to the subject that is predicated by some property, like snow is white; but I am not sure.] So from the generic noun “falls” the direct case [which could include the nominative form of the noun] “onto a state of affairs” (106). Luhtala explains that the state of affairs “in its Stoic use, is a syntactic notion” (106). [[I do not clearly understand how the state of affairs is a syntactical notion. I am also not sure how to conceive it as a physical notion, but it seemed from before to be that about the physical situation that corresponds to the predicate. So perhaps for ‘snow is white’, the state of affairs in the world is being white (or the snow’s being white, if we include the subject) and for Dion walks, the state of affairs is walking (or Dion’s walking). But  I am not sure how it is a syntactical notion. The state of affairs is syntactical perhaps because it enters into a relation with the nominative form of the noun. However, I am not sure if that means the state of affairs is always a linguistic entity or if instead a linguistic entity can represent it. I wonder if we can offer the following conception: the basic physical situation is an affair or affairs. The state of affairs is something sayable with regard to those affairs, namely, the element of the activity which could be expressed as a predicate. Consider the example of snow is white. There is the affairs of the snow physically bearing certain features that make it appear white. Suppose it were to melt. It would then become clear and not white. So its being white might somehow be understood as a persisting activity, or maybe that there are influences acting on the water particles, as cold for example, sustaining that affairs of it being white. The state of affairs is thus “being white” or “staying white” or “whiting” or something like that. Now consider Dion walks. Here the affairs are all the physical components involved in Dion’s walking, like his body’s parts and the other physical factors like gravity and so on. The state of affairs is the fact that all these factors together constitute the walking of Dion’s parts. It is the “walking” or “is walking” of the physical body/bodies. I am still not sure if the state of affairs includes the subject, but Luhtala’s account often seems to exclude it. See for example section section where we learn that for the Stoics, the predicate is defined as {1} a state of affairs construed around one or more subjects, or {2} a defective sayable which has to be joined to a nominative case in order to yield a judgement. And in section we learn that a predicate is a state of affairs that is said of something, with that something taking a nominative case (this comes directly from a quote by Diogenes Laertius; see specifically section So what do we call the situation in the world including both the bodies (perhaps as subjects or maybe more) and their states of affairs? It is no more than the complete sayable or the complete state of affairs (see section I wonder if it can also be called a fact.]] [The next idea is that this falling from the generic noun involves falling to the individual involved in the action. Her last point seems to be that Apollonius claims to reject the Stoic distinction between noun and case, but here we see that it may in fact lie at the basis of his system. Let me quote so you can interpret.]

If the direct case falls from the generic noun, the generic noun is supposed to be an abstract entity of some sort, which does not bear the feature ‘case’ at all. The noun (ὄνομα) is thus a non-syntactic unit of speech. The direct case, according to Apollonius, falls from the generic noun, onto a state of affairs (πρᾶγμα), which, in its Stoic use , is a syntactic notion. This is obviously the case here, too. When involved in a state of affairs, the direct case shows directly the substance of the state of affairs (probably the definite person involved in action) and is construed with direct verbs, that is with active verbs. Active verbs are transitive verbs for Apollonius. When receiving case (πτῶσις), the noun (ὄνομα) falls from the generic to the individual. I interpret this difficult passage as if it reflected the Stoic division between noun (ὄνομα) and case (πτῶσις), even if this distinction has been abolished in Apollonius’ framework as a whole. In justifying such a fundamental point of doctrine as the notion of case, Apollonius might have resorted to genuinely Stoic viewpoints on which his own framework ultimately depended.175 According to Priscian, too, case falls from the generic noun to the special.176


175 [Luhtala notes other scholars holding similar views. See p.106.]

176 [Luhtala here offers quotation for support. See pp.106-107.]


[The Peripatetics did not think the nominative has case. The Stoics think that any substantive playing a syntactical role has case. The nominative indicates the person involved directly in the verb. Thus the nominative for the Stoics has case.]


The Peripatetics claimed that the nominative is not really a case [it is something like a form from which the other cases are derived]. Rather, for them it is at best a case metaphorically speaking (107. Here Luhtala writes in footnote: Cf. Charisius: (κατηχρηστικῶς) casus nominativus (GL I: 154,10).] Appollonius and Priscian, however, “defend the genuinely Stoic position according to which the notion of case covers even the nominative, and not only metaphorically” (107). [I am not sure I follow the next point. Recall our discussion in section (where we referenced a prior page, p.26). Here we gathered that syntax is the meaningful combination of parts of speech (but that could be inaccurate). Here her point seems to be that case for Stoics is always syntactically-based. The idea might be that so long as a substantive is related syntactically to other parts of the sentence, it has case. And since the nominative bears such syntactically relations, it must have case. Specifically, the nominative plays the syntactical role of indicating the person referred to by the verb. Let me quote so you can see:]

The Stoic usage is not simply an irrelevant question of terminology, but one which seems to reflect the basics of syntactical doctrine. I have claimed that the feature of case is syntactically-based with the Stoics, who seem to have regarded all nominal inflection as syntactically relevant. The nominative case is understood as indicating the definite reference of the verb, the ‘person’. It is labeled the direct case, being associated with direct predicates.


[The nominative is thus not a matter of naming for the Stoics. Their causative case could be understood as indicating a body being acted upon by the state of affairs.]


[The term ‘nominative’ suggests nominating or naming. But as we can see from the Stoic view, a substantive is nominative when it indicates the person directly referred to by a verb in a direct predicate. So it is not simply a matter of naming this or that object (which could be involved in any way in any sort of predicate). Luhtala then goes on it seems to explain how grammarians came to use the term ‘nominative’ for the direct case, but I might be misreading. Dionysius Thrax considered the nominative case as both the direct and the nominative case. Priscian adopts the Latin pedagogical tradition that uses the term ‘casus nominativus’ (even though Priscian could have used direct. I did not grasp the idea, but it might be that Priscian would normally call it direct but chose to substitute nominative, thereby contributing to this terminological shift.) The philological tradition also changed ‘direct’ to ‘nominative’.  Luhtala then discusses αἰτιατκὴ πτῶσις. But I do not follow these ideas, as I am unfamiliar with these grammatical notions. It seems that it means “causative case” and  it indicates “the body acted upon in a ‘transitive’ or ‘causative’ state of affairs. But it is not the accusative case for some reason. Let me quote this section:]

The nominative case has thus nothing to do with naming in the Stoic analysis. The fact that this case was also known as the ‘nominative’ (ὀνομαστικὴ πτῶσις) by the grammarians seems to reflect a different tradition, probably that of the Alexandrian grammarians. Dionysius Thrax calls the nominative case both ὀρθή (‘direct’) and ὀνομαστικὴ πτῶσις (‘nominative case’). The Latin pedagogical tradition uses the term casus nominativus which was adopted by Priscian, who also knew the term ‘direct case’ (casus rectus).178 The philological tradition, even if it adopted many Stoic features of the notion of case, obviously preferred to change the obscure term ‘direct’ into ‘nominative’. As to the αἰτιατικὴ πτῶσις, whose misrepresentation as the accusative case in the Latin tradition is well-known, Ι relate it to the notion of cause, as is done by Dionysius Thrax (§12 GG I.1: 31, 5).179 It seems natural to associate it with the Stoic causal analysis, which involves two bodies, and which seems to have been based on the model of the proposition. The causative case would thus represent the body acted upon in a ‘transitive’ or ‘causative’ state of affairs.


178 When Priscian occasionally uses the term ‘direct’, it concerns a point of doctrine whose Stoic origin is obvious: “nec eget recto casu quem verbum habet in se” (“this construction does not lack the direct case which the verb contains in itself” GL III: 8,5,10 ff.

(107, end parenthesis missing from my copy)

179 ἡ αἰτιατικὴ κατ’ αἰτίαν (“the causal case relates to a cause” tr. Kemp 1987:178). Its origin has been sought in the Aristotelian term at1:ta1:6v ('effect ' ) by, for instance, Pohlenz (1939 : 171-172), who nevertheless associates it with the Stoic notion of cause and effect: “Die Verbindung dieses Kasus mit dem Verbum drückt also nach der Auffassung der Stoa aus, dass ein Verhältnis von Ursache und Wirkung vorliegt. Gerade dieses ist aber für die Stoa so wichtig, dass sie es in weitestem Umfange annimmt”. See also Pinborg (1975: 86).










Luhtala, Anneli. 2000. On the Origin of Syntactical Description in Stoic Logic. Münster: Nodus.



Other texts, cited by Luhtala:


Ammonius: In Aristotelis De Interpretatione commentarius. CAG IV.5. Ed. by Adolf Busse. Berlin: Reimer 1897.


GAG: Commentaria in Aristotelem Graeca. Berlin: Reimer 1881-1907.


[GG] GG: Grammatici Graeci. Ed. by Heinrich Schneider and Gustav Uhlig. Leipzig: Teubner (repr 1965 Hildesheim: Olms).


[GL] GL: Grammatici Latini. Ed by Heinrich Keil. Leipzig: Teubner 1855-80.


Hülser, Karlheinz. 1987. Die Pragmente zur Dialektik der Stoiker. Neue Sammlung der Texte mit deutscher Übersetzung und Kommentaren. 3 vols. Stuttgart-Bad Cannstatt: Frommann-Holzboog.


[For the reference, tr. Kemp 1987, I have not found it yet in the Bibliography, but it might be:

Kemp, Alan [translating and commenting Dionysius Thrax]. 1987. “The Tekhné Grammatiké of Dionysius Thrax. Translated into English,” in The History of Linguistics in the Classical Period (ed. Daniel J. Taylor. Amsterdam / Philadelphia: John Benjamins.]


Pinborg, Jan. 1975. “Classical Antiquity: Greece”. Sebeok (1975: 69-126).


Pohlenz, Max. 1939. Die Begründung der abendländischen Sprachlehre durch die Stoa. Göttingen: Vandenhoeck & Ruprecht. (Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen. Philologisch-historische Klasse. Neue Folge. 1.3, Heft 6.).


Schmidt, Rudolf T. 1979. Stoicorum grammatica. Halle 1839. (Reprinted Amsterdam: Hakkert 1967 = Die Grammatik der Stoiker. Einführung, Übersetzung und Bearbeitung von Karlheinz Hülser, Urs Egli, Braunschweig, Wiesbaden 1979.].





Texts I cite:


Blake, Barry. 2004. Case. Cambridge: Cambridge University.





No comments:

Post a Comment