The Stoics on Conceptions and Concepts

doi:10.1017/9781009369596.013

Chapter 10 - The Stoics on Conceptions and Concepts

Published online by Cambridge University Press: 25 April 2024

Katerina Ierodiakonou

Edited by

Gábor Betegh and

Voula Tsouna

Show author details

Gábor Betegh: Affiliation:
University of Cambridge
Voula Tsouna: Affiliation:
University of California, Santa Barbara

Book contents

Summary

In ‘The Stoics on Conceptions and Concepts’, Katerina Ierodiakonou offers an account of the Stoics’ distinction between ennoiai and ennoēmata (‘conceptions’ and ‘concepts’), and also of the distinctions suggested by the standard Stoic terminology of concepts also mentioned above: notably, prolēpseis (‘preconceptions’), phusikai ennoiai (‘natural conceptions’), and koinai ennoiai (‘common conceptions’). All these terms appear intended to point to general notions that play a central role in the acquisition of human knowledge, but it remains puzzling how exactly the Stoics understood them or why they introduced them into their doctrine in the first place. Ierodiakonou addresses these issues, as well as further questions debated in the secondary literature. These include whether all human beings necessarily possess concepts or just have the ability to possess them, what is the content of conceptions and how it is determined, what is the ontological status of conceptions and concepts, and what are their epistemological functions.

Keywords

concept (ennoia)conception (ennoēma)preconception (prolēpsis)articulation (diarthōsis)common conception (koinē ennoia)‘somethings’(tina)‘not-somethings’(outina)sayables (lekta)Stoics Chrysippus

Type: Chapter
Information: Conceptualising Concepts in Greek Philosophy , pp. 237 - 258

DOI: https://doi.org/10.1017/9781009369596.013 [Opens in a new window]

Publisher: Cambridge University Press

Print publication year: 2024
Creative Commons: This content is Open Access and distributed under the terms of the Creative Commons Attribution licence CC-BY-NC 4.0 https://creativecommons.org/cclicenses/

1 Introduction

The Stoics drew a fine distinction between, on the one hand, the mental state humans are in when apprehending something, that is, a conception (ennoia), and, on the other, the content of a conception, that is, a concept (ennoēma).Footnote ¹ They also claimed that conceptions and concepts are fundamental to the acquisition of human knowledge. Although we may be fairly confident in attributing this doctrine to them, it is more difficult to be certain about the arguments they put forward in order to support it, as well as about the different kinds of conceptions they postulate and the place they reserve for concepts in Stoic ontology. The textual evidence on these issues, just as on most aspects of Stoicism, is scarce and often conflicting. Moreover, the possibility of developments in the Stoic cognitive theory, from the early to the late Stoa, further complicates the situation to a significant degree.

In the existing secondary literature, there is already a long list of ingenious interpretations which have painstakingly discussed the Stoics’ account of conceptions and concepts.Footnote ² Indeed, one reasonably gets the impression that all tenable options explicating every single point connected with this topic have already been explored. My aim is not to be comprehensive in investigating all related issues or in commenting on all the diverse opinions expressed by scholars. Rather, I try to unpack the complexities in the Stoic theory of concept formation and suggest readings of the ancient sources that strike me as the most plausible concerning the classification, the ontological status and the epistemic function of the Stoic conceptions and concepts.

2 Fine Distinctions in Stoic Terminology

2.1 Conceptions and Concepts

The Greek nouns ennoia and ennoēma, translated here as ‘conception’ and ‘concept’ respectively,Footnote ³ are both cognate with the verb ennoein, which literally means ‘to have in one’s thoughts or intellect (nous)’, but they differ in their suffixes. The suffix -ia in ennoia denotes for the Stoics the mental state when having something in one’s thoughts; ennoiai are thus defined by them as impressions (phantasiai) of a certain kind.Footnote ⁴ In fact, there are various lists of the Stoics’ classification of impressions, but the one relevant to our purposes is to be found in Diogenes Laertius (7.51.8–10), who distinguishes between the non-rational (alogoi) impressions of most animals and the rational (logikai) impressions of humans, which are thoughts (noēseis). Stoic ennoiai are said, in our surviving sources, to be stored thoughts:Footnote ⁵

T1 Conception (ennoia) is a kind of impression (phantasia), and impression is a printing (tupōsis) in the soul … They [i.e. the Stoics] define conceptions as a kind of stored thoughts (enapokeimenai noēseis).Footnote ⁶

(Plut., Comm. not. 1084F–1085A; trans. LS 39F)

The suffix -ma in ennoēma, on the other hand, denotes for the Stoics the result of the act of ennoein, that is, what the conception is a conception of; ennoēmata are thus defined by them as the contents of impressions (phantasmata) that occur in the rational human souls.Footnote ⁷

The Stoics argued that, during the course of our natural development, the acquisition of conceptions and concepts results in the emergence of human reason. More specifically, they are reported to have described the emergence of human reason in an empiricist way, very similar to what Aristotle had stated about the same issue at the beginning of his Metaphysics (A.1) and at the end of his Posterior Analytics (2.19): On the basis of their sense-impressions (aisthēseis), human beings store in their souls memories (mnēmai); many similar memories result in what is called ‘experience’ (empeiria), which subsequently leads to the formation of our conceptions and, in general, to all human knowledge:

T2 When a human being is born, the Stoics say, he has the commanding-part of his soul like a sheet of paper ready for writing upon. On this he inscribes each one of his conceptions. The first method of inscription is through the senses. For by perceiving something, e.g. white, they have a memory of it when it has departed. And when many memories of a similar kind have occurred, we then say we have experience. For the plurality of similar impressions is experience. Some conceptions arise naturally in the aforesaid ways and undesignedly, others through our own instruction and attention. The latter are called ‘conceptions’ only, the former are called ‘preconceptions’ as well. Reason, for which we are called rational, is said to be completed from our preconceptions during our first seven years.Footnote ⁸

(Aëtius 4.11.1–4; trans. LS 39E, slightly modified)

So, conceptions are first formed in our rational souls on the basis of the experience we accumulate from memories of repeated sense-impressions. Some conceptions are acquired naturally (phusikōs) and undesignedly (anepitechnētōs), and these the Stoics also called ‘preconceptions’ (prolēpseis), whereas others are the result of instruction (didaskalia) and attention (epimeleia), and these were called simply ‘conceptions’ (ennoiai), in accordance with a standard pattern Stoics used in divisions and subdivisions. In other words, the Stoic term ‘ennoiai’ has two different senses, namely as the genus of which preconceptions are a species and as a distinct species of conceptions, that is, conceptions in the narrow sense. It is also worth noting that this distinction between preconceptions and conceptions in the narrow sense comes very close to the one we find in Diogenes Laertius (7.51.10–12), according to which rational impressions are divided into those of an expert (technikai) and those that are not of an expert (atechnoi). But how exactly are we to understand the distinction between preconceptions and conceptions in the narrow sense?

2.2 Preconceptions and Conceptions in the Narrow Sense

Preconceptions and conceptions in the narrow sense were distinguished by the Stoics on the basis of the way they are generated. Even the use of the term ‘pro-lēpsis’ suggests that preconceptions are conceptions that provide us with a preliminary apprehension of what we perceive through our senses. Cicero (Nat. D. 1.44) informs us that it was the Epicureans who first coined the term ‘prolēpsis’, in order to refer to the self-evident general notions humans form on the basis of repeated sense-impressions.Footnote ⁹ Under their influence, the Stoics claimed that preconceptions arise naturally and, in particular, Chrysippus is said to have defined them as ‘natural conceptions of universals’ (ennoia phusikē tōn katholou: Diog. Laert. 7.54). What does it mean, however, that preconceptions arise naturally?

We have already seen in Aëtius’ passage T2 that our preconception of something being white is formed by having many sense-impressions of something white. Indeed, most preconceptions seem to be generated, according to the Stoics’ empiricist doctrine, directly from sense-perception, memory and experience. It is puzzling, though, that the author of this passage, when presenting the developmental process that gives rise to preconceptions, talks of ‘the first method of inscription’ (Aëtius 4.11.2) and uses the plural ‘in the aforesaid ways’ (Aëtius 4.11.3). Jaap Mansfeld (Reference Mansfeld2014: 615–7) reviews the scholarly debate on this issue: Some scholars have suggested that it is memory and experience that could be thought of as adequately representing, next to sense-perception, a plurality of ways of forming preconceptions; others have argued that the list is incomplete and there must be a lacuna in the text which probably included some further natural processes like, for instance, those described by Diogenes Laertius:Footnote ¹⁰

T3 It is by confrontation (kata periptōsin) that we come to think of sense-objects. By similarity (kath’ homoiothēta), things based on thoughts of something related, like Socrates on the basis of a picture. By analogy (kat’ analogian), sometimes by magnification, as in the case of Tityos and Cyclopes, sometimes by diminution, as in the case of the Pigmy; also the idea of the centre of the earth arose by analogy on the basis of smaller spheres. By transposition (kata metathesin), things like eyes on the chest. By combination (kata sunthesin), Hippocentaur. By opposition (kat’ enantiōsin), death. Some things are also conceived by transition (kata metabasin), such as sayables and place. The idea of something just and good is acquired naturally (phusikōs). That of being without hands, for instance, by privation (kata sterēsin).Footnote ¹¹

(Diog. Laert. 7.53; trans. LS 39D)

Mansfeld himself argues that T2 is an abridgement of Aëtius and most likely something is missing from this text, although it should not be supplemented by Diogenes’ list, which is in his view ‘a mixed bag’ (2014: 617).Footnote ¹² But whether or not all or some of these processes were considered as natural and actually mentioned in Aëtius’ text, it seems that the Stoics regarded certain additional mental processes as useful for generating preconceptions needed in acquiring knowledge. And what these mental processes were supposed to have in common is their being based on sense-impressions and, most importantly, not involving any special instruction and attention.Footnote ¹³

But Stoic epistemology does not postulate that the attainment of human knowledge depends exclusively on our preconceptions. As we have seen in T2, the Stoics also talked of conceptions in the narrow sense, namely conceptions that human beings acquire through instruction and attention. Assuming, in line with the Stoic empiricist dogma, that these conceptions, too, are ultimately derived from the senses, how are we supposed to understand more precisely the distinction between preconceptions and conceptions in the narrow sense? What does it mean, according to the Stoics, that instruction and attention are involved only in the acquisition of conceptions in the narrow sense?

No doubt preconceptions manage to organise, at least at a first level, our diverse sense-experience; but they are still inchoate and inarticulate. To turn the rather rough preconceptions into refined conceptions, a certain intellectual effort seems to be required for filling them out and sharpening them. In fact, it is by carefully defining our preconceptions that, according to the Stoics, we succeed in transforming them into conceptions in the narrow sense. For there are sources attributing to the members of the Stoa a certain process, which is meant to transform our inchoate conceptions into articulated ones by supplying their definitions:Footnote ¹⁴

T4 [The Stoics say that from the senses] the mind forms conceptions (notiones) – ennoiai, as they call them – of those things, that is, which they articulate by definition. The entire method of learning and teaching, they say, stems and spreads from here.Footnote ¹⁵

(Augustine, De civ. D. 8.7; trans. LS 32F)

This process was called ‘articulation’ (diarthōsis), a rather common medical metaphor used by the Stoics to denote both the generation of speech from inarticulate sounds as well as the sharpening of undeveloped conceptions.Footnote ¹⁶ This is perhaps the process that Chrysippus discussed in his book entitled Ethics: Concerning the Articulation of Ethical Conceptions (Diog. Laert. 7.199),Footnote ¹⁷ although there are no surviving extracts from it to support such a view. Still, it may not be far-fetched to suggest that, according to the Stoics, we naturally apprehend the ethical preconception of something being just, for instance, but we need to articulate it further to have a full understanding of it. For although our preconceptions arise naturally during the first seven years [T2], the Stoics also claimed that during the next seven years these preconceptions can be transformed into conceptions in the narrow sense, due to appropriate instruction and learning.Footnote ¹⁸

Cicero gives us an insight into how the preconception of something being a human being can be transformed by articulation into a well-defined conception in the narrow sense. In his Academica, the Antiochean Lucullus says the following in defense of Stoic epistemology:Footnote ¹⁹

[T5] Such are the things we claim are apprehended by the senses. The next set are just like them, though we don’t claim that these are apprehended by the senses themselves, but by the senses in a certain respect – e.g., ‘That is white’, ‘This is sweet’, ‘That is melodious’, ‘This is fine-scented’, ‘This is rough’. Our apprehension of this set now comes from the mind rather than the senses. Next comes: ‘That is a horse’, ‘That is a dog’. Then we get the rest of the series, which connects more significant things and encapsulates what we might call a filled-out apprehension of things – e.g., ‘If something is human it is a mortal animal partaking in reason.’ It’s from this set [of impressions] that our conceptions of things are stamped on our minds, and without them there can be no understanding, investigation, or argument.Footnote ²⁰

(Cic., Acad. 2.21; trans. Brittain)

So, although we form the preconception of a human being on the basis of our senses, it is the definition of a human being that provides us with a ‘filled-out apprehension’ (expleta conprehensio) of what a human being is, that is, with the articulated conception of the human being that constitutes a conception in the narrow sense. Later, Epictetus, too, talks about the process of articulation, which he also understands as the attentive transformation of our inchoate preconceptions into well-defined conceptions:Footnote ²¹

T6 Who among us doesn’t talk about ‘good’ and ‘bad’, and about what is ‘advantageous’ or ‘disadvantageous’? For who among us doesn’t have a preconception of each of these things? Is it properly articulated (diērthōmenēn), however, and complete? Show me that it is. How am I to show that? By applying it properly to particular cases. Plato, for instance, classifies his definitions under the preconception of the ‘useful’, but you under that of the ‘useless’. Now is it possible that both of you could be right? How could it be? Or again, with regard to wealth, doesn’t one person apply the preconception of the ‘good’ to it, while another doesn’t? And likewise with regard to pleasure, and likewise with regard to health? In general, then, if all of us who utter these terms possess more than an empty knowledge of each, and we do not need to devote any attention (mēdemias epimeleias) to the articulation (diarthrōsin) of our preconceptions, why do we disagree, why do we come into conflict, why do we criticise one another?Footnote ²²

(Epictetus, Diss. 2.17.10–13; trans. Hard, slightly modified )

Finally, another relevant example, this time from mathematics, shows how the definition of a circle turns our unrefined conception of it into an articulated one; in this particular case, though, Simplicius applies the process of articulation in an Aristotelian context without making reference to its Stoic provenance:

T7 That what is indefinite and confused seems more familiar to us, as for instance the whole, he [i.e., Aristotle] confirms by taking each name as a kind of whole, and the definition of the name as providing the articulation (diarthrōsin) of the parts and elements of the name. For it is clear that the knowledge by name of the circle is readily available even to the multitude, while the definition of the circle, – that it is a plane figure encompassed by a single line such that all the lines drawn from a single point that meet it […] are equal to each other –, this definition, in contrast, is not readily available to all, since it provides the particulars of the circle and sets them out with respect both to its parts and its elements.Footnote ²³

(Simplicius, in Phys. 16.31–17.5; my translation)

To sum up, the Stoics distinguished two kinds of conceptions, namely preconceptions and conceptions in the narrow sense: Preconceptions are acquired naturally, whereas conceptions in the narrow sense are acquired through instruction and attention, that is, through the articulation of preconceptions. Admittedly, however, the distinction between preconceptions and conceptions in the narrow sense is not always to be found as clear-cut in our sources. For there are also passages referring to ‘natural preconceptions’ (phusikai prolēpseis: e.g., Epictetus, Diss. 1.22.9), ‘common preconceptions’ (koinai prolēpseis: e.g., Epictetus, Diss. 1.22.1; 4.1.41–3), ‘natural conceptions’ (phusikai ennoiai: e.g., Epictetus, Diss. 2.17.7) and ‘common conceptions’ (koinai ennoiai: e.g., Plutarch, Comm. not. 1059F; Sext. Emp. M. 11.22).Footnote ²⁴ But although the cases of natural preconceptions and common preconceptions as well as those of natural conceptions do not strike us as problematic, since after all preconceptions are acquired by all of us naturally, and conceptions in the narrow sense are ultimately derived from the naturally acquired preconceptions, to talk in the Stoic context about ‘common conceptions’ sounds baffling. For if conceptions in the narrow sense differ from preconceptions in being articulated, how could they be characterised as ‘common’? Did the Stoics think that all people carefully define their inarticulate preconceptions and manage to turn them into refined conceptions in the narrow sense?

2.3 Preconceptions and Common Conceptions

Two different interpretations have been suggested concerning the relation between prolēpseis and koinai ennoiai: It has been claimed that they differ in meaning and scope, since preconceptions are inarticulate conceptions whereas common conceptions are thought-out definitions (Sandbach Reference Sandbach1930; Todd Reference Todd1973); on the other hand, it has been argued more recently that they should be treated as interchangeable terms (Dyson Reference Dyson2009: 1–22). The more recent interpretation has rightly been criticised for disregarding the textual evidence that describes common conceptions as articulated preconceptions and as functionally distinct from them (Klein Reference Klein2011: 115). So, if we are to stay with the standard view that preconceptions and common conceptions differ from each other, we need to further specify the difference between them. It is not enough to say that common conceptions are the thought-out definitions of inarticulate preconceptions, because in this case all conceptions in the narrow sense would have to be common conceptions. So, what is exactly the distinguishing characteristic of common conceptions?

The first thing to examine is, of course, the sense in which common conceptions are said to be ‘common’. Different interpretations have also been suggested concerning this issue: According to Robert Todd (Reference Todd1973: 60–63), the term ‘koinai’ is ambiguous; sometimes it has its standard meaning, namely ‘shared by all’, while at other times it means ‘basic’, especially when common conceptions are used to justify doctrines in all three parts of Stoic philosophy. Dirk Obbink (Reference Obbink1992: 225–27), on the other hand, favours the view that common conceptions were understood by the Stoics as basic underlying notions for agreement in inquiry, and are thus comparable to Aristotle’s koinai doxai and endoxa. However, there are several passages in our sources explicitly stating that common conceptions, just like preconceptions, are actually shared by all, or at least that all humans have potential access to them (e.g., Sext. Emp. M. 9.124; 138; 199). Besides, as it has been rightly pointed out, there is no real tension between the two meanings of ‘koinai’; since the common conceptions shared by all were also shared by the Stoics and the wise, they were meant to be at the basis of Stoic doctrines in the three parts of their philosophy (Brittain Reference Brittain2005: 177; Dyson Reference Dyson2009: 48–53).

Therefore, my contention is that common conceptions are conceptions in the narrow sense insofar as they are articulated preconceptions, and common insofar as they are possessed or can be possessed universally; for instance, the ethical conceptions of something being just, good or bad are common conceptions, insofar as they are articulated and all humans possess them or are able to possess them. But there are also other articulated preconceptions, that is, other conceptions in the narrow sense, which are not shared by all, since their articulation requires some technical instruction; for instance, the mathematical conception of a circle is an articulated preconception, and in this regard a conception in the narrow sense, but it is only geometers who possess it, and thus it is not a common conception. In other words, according to my understanding of the Stoics’ classification, conceptions were divided into preconceptions and conceptions in the narrow sense; conceptions in the narrow sense were further subdivided into common conceptions, which included the articulated preconceptions that are possessed universally and other conceptions in the narrow sense that were technical and accessible only to the few.

Needless to say, even if the Stoics drew these distinctions with great precision, our sources may not be so careful when presenting them; so, perhaps it should not surprise us if sometimes preconceptions and common conceptions are used interchangeably .

3 The Ontological Status of Stoic Concepts

3.1 The Supreme Genus ‘Something’

According to the Stoics, only bodies (sōmata) can be said to exist (einai), because they are capable of either affecting something or being affected by something (e.g., Sext. Emp. M. 8.263). Hence, the Stoics insisted that there are no such items in reality as Plato’s Forms (e.g., Syrianus, in Metaph. 105.21–30). But they also diverged from Epicurus’ pure materialism and listed in their ontology some additional items, which fall short of proper existence but are said to subsist (huphistanai), and even on occasion to obtain (huparchein). These are the incorporeals (asōmata), which together with the existent beings (onta) form the supreme genus of what the Stoics called ‘something’ (ti) or, according to Seneca, ‘quid’.Footnote ²⁵

Having in mind these principles of Stoic ontology, we should next investigate the ontological category in which Stoic conceptions and concepts belong. As we have already seen, conceptions (ennoiai) are rational impressions, and the Stoics defined impressions as printings (tupōseis) in the soul, which should be understood, at least according to Chrysippus, as alterations (alloiōseis) or modifications (heteroiōseis) of the soul (e.g., Diog. Laert. 7.50; Sext. Emp. M. 7.228–31). Under appropriate conditions, the external objects affect our sense-organs, and these affections are subsequently transmitted to the commanding-part of the soul. Given the Stoic view that the soul is corporeal (e.g., Nemesius, De nat. hom. 78.7–79.2; 81.6–10), the impressions we receive alter or modify the physical state of our soul and are regarded as bodies. Hence, conceptions, too, are bodies, being physical states of the commanding-part of the soul.

On the other hand, what our conceptions are conceptions of, namely our concepts (ennoēmata), are not capable of either affecting something or being affected by something, and thus are not corporeal. What then is their ontological status? This issue has recently been subject to extreme controversy among scholars, some of whom have argued that concepts, according to the Stoics, do not belong to the supreme genus of something, that is, they cannot be considered as ‘somethings’ (tina), whereas others have claimed that the earlier Stoics thought of them as non-existent somethings but Chrysippus later rejected them altogether. In what follows, I first present in a schematic way the main points of this debate; I then offer my own interpretation of the issue.

3.2 Concepts as Not-Somethings

In my view, there are three principal arguments put forward by scholars claiming that concepts cannot be somethings:Footnote ²⁶

(I) Concepts are said in our sources to be phantasmata, and this should be understood as meaning that concepts are figments of the soul, that is, mental constructs; since neither existent beings nor incorporeals are figments, and the supreme genus of something consists only of existent beings and incorporeals, concepts cannot be somethings:
- T8 (Zeno’s doctrine) They say that concepts (ennoēmata) are neither somethings nor qualified, but phantasmata of the soul which are quasi-somethings and quasi-qualified. These, they say, are what the old philosophers called Ideas. For the Ideas are of the things which are classified under the concepts, such as human beings, horses, and in general all the animals and other things of which they say that there are Ideas. The Stoic philosophers say that there are no Ideas, and that what we ‘participate in’ is the concepts, while what we ‘bear’ is those cases which they call ‘appellatives’.Footnote ²⁷ (Stobaeus 1.136.21–137.6; trans. LS 30A, slightly modified)
- T9 A concept is a phantasma of the mind, which is neither something nor qualified, but a quasi-something and quasi-qualified, in the way that the pattern of horse arises even though none is present.Footnote ²⁸ (Diog. Laert. 7.61; trans. LS 30C, slightly modified)
(II) Concepts are also said to lack substance, whereas beings exist and incorporeals are meant to subsist; since concepts can neither be said to exist nor to subsist, they cannot be somethings:
- T10 Some Stoics consider ‘something’ the first genus, and I shall add the reason why they do. In nature, they say, some things exist, some do not exist. But nature includes even those which do not exist – things which enter the mind, such as Centaurs, giants, and whatever else falsely formed by thought takes on some image despite lacking substance (substantiam).Footnote ²⁹ (Seneca, Ep. 58.15; trans. LS 27A)
(III) Concepts cannot be somethings, because the hallmark of the supreme genus of something is particularity; to be something is to be some particular thing, and concepts are universals.
Moreover, all scholars who have defended the position that Stoic concepts cannot be somethings have also maintained that they cannot be considered as nothing at all. For instead of using the common negative pronoun ‘ouden’ that has no plural, the Stoics constructed the unusal plural ‘outina’, a neologism composed from the negative particle ‘not’ (ou) and the neuter plural of ‘something’ (ti), marking thus the difference between not-somethings and nothing at all. So, concepts were rather regarded by the Stoics as ‘not-somethings’ (outina), which are connected in our sources with a Chrysippean argument known as the ‘Not-someone’ (Outis) argument:Footnote ³⁰
- T11 Indeed, Chrysippus too raises problems as to whether the Idea will be called a ‘this something’ (tode ti). One must also take into account the usage of the Stoics about generically qualified things – how according to them cases are expressed, how in their school universals (ta koina) are called ‘not-somethings’, and how their ignorance of the fact that not every substance signifies a ‘this something’ gives rise to the Not-someone sophism, which relies on the form of expression. Namely: ‘If someone is in Athens, he is not in Megara; <but man is in Athens; therefore man is not in Megara.>’ (For man is not someone, since the universal is not someone, and that is why the argument has this name, being called the ‘Not-someone’ argument.Footnote ³¹ (Simplicius, in Cat. 105.8–16; trans. LS 30E)

According to this interpretation, therefore, Stoic concepts are not-somethings, meaning that neither do they belong to the supreme genus of something nor are they nothing at all. They are utterly mind-dependent items, which do not even subsist, as incorporeals do, but result from the formation of conceptions in the commanding-part of the rational soul .

3.3 Concepts as Non-Existent Somethings

Victor Caston (Reference Caston1999: 158–71) objects to the view that concepts should be treated as not-somethings. I understand the main points of his reasoning briefly as follows:

(I) The distinction between not-somethings and nothing at all is rather vague, if not unintelligible. Besides, if the Stoics drew such a distinction, there should have been a supreme genus beyond the genus of something, which would encompass both somethings and not-somethings; but the genus of something is meant to include, according to our sources, every item in Stoic ontology.
(II) Stobaeus’ and Diogenes’ passages [T8 and T9], which describe concepts as neither somethings nor qualified, do not have to be read as banishing concepts from the supreme genus of something. They simply deny that concepts are existent beings and, more precisely, that they belong either in the first Stoic category of substrates (hupokeimena) or in the second category of qualified things (poia), since both of these imply existence. But the fact that concepts are not among the existents is still perfectly compatible with the claim that they are somethings.
(III) Concepts cannot be not-somethings, since concepts are the contents or the intentional objects of thought, whereas not-somethings are said to have no subsistence with respect to thought (anupostata tē dianoiai), that is, we cannot even think of them:
- T12 If something is taught, it will be taught either through not-somethings, or through somethings. But it cannot be taught through not-somethings, for these have no subsistence with respect to thought, according to the Stoics.Footnote ³² (Sext. Emp. M. 1.17; trans. LS 27C, slightly modified)
(IV) Concepts cannot be not-somethings, since they play an important role in Stoic epistemology; for Stoic divisions involve genera and species, which are defined in terms of concepts and are clearly said to have a subsistence of their own (idian hupostasin: Sext. Emp. PH. 2.219; Diog. Laert. 7.60).

But if concepts cannot be said to be not-somethings, what is their ontological status? According to Caston (Reference Caston1999: 204–13), they belong in the supreme genus of something and, since its division into existent beings and incorporeals seems exhaustive, they should be considered as incorporeals, that is, non-existent somethings. More specifically, Caston claims that there was a significant divergence of opinion between Zeno and Cleanthes, on the one hand, and Chrysippus, on the other with regard to the way they treated concepts: The first two scholarchs of the early Stoa retained concepts in their ontology and thought of them as incorporeals, whereas Chrysippus did not even use the term ‘ennoēma’ but only spoke of sayables .Footnote ³³

3.4 Concepts as Sayables

I find convincing the arguments defending the view that concepts cannot be, according to the Stoics, not-somethings, but belong in the supreme genus of something and, more specifically, among incorporeals, that is, non-existent somethings. In particular, I agree with Caston’s view that Stobaeus’ and Diogenes’ passages [T8 and T9] should not be read as banishing concepts from the supreme genus of something. For the use of the noun ‘phantasma’, in these passages, should not be taken to refer to a figment of the mind; it rather has the Aristotelian sense of the content of an impression, that is, it is equivalent to the Stoic noun ‘phantasia’. This is what Jaap Mansfeld (Reference Mansfeld2014: 625) convincingly shows, I think, in the case of the relevant passage from Aëtius:

T13 A concept is a phantasma in the thinking faculty of a rational animal; for a phantasma is only then called a concept (ennoēma) when it occurs in a rational soul, deriving its name from the intellect (nous). Accordingly, all phantasmata that occur in non-rational animals are mere phantasmata. But those that occur in the gods and to us are phantasmata as to genus and concepts as to species. Just as denarii <and> staters, if you consider them in themselves, are simply denarii and staters. But if you use them to pay for a naval voyage these are not only denarii, but are called ‘ship fare’ as well.Footnote ³⁴

(Aëtius 4.11.4–5; trans. Mansfeld, slightly modified)

Most importantly, Mansfeld concludes that we should not consider it particularly odd that in the doxographical tradition, more generally, the Stoic doctrine of concept formation ‘is represented in more or less Aristotelian terminology; that is to say, in a terminology that is more Aristotelian than a Stoic, presumably, would have used himself’. Therefore, Stobaeus’ and Diogenes’ passages [T8 and T9] can be read accordingly.

Furthermore, there are no uncontroversial and reliable ancient sources that characterise Stoic concepts as mental constructs lacking subsistence. It is worth pointing out that, in Seneca’s letter 58 T10, we are presented with cases that are ‘falsely formed by thought’, that is, Centaurs and giants, and not with standard examples of concepts like human being or horse; besides, these are cases that are said to lack substance and no indication is given whether or not they lack subsistence.Footnote ³⁵ Also, a passage from Alexander of Aphrodisias’ commentary on the Topics, which allegedly claims that concepts neither exist nor subsist, has been rightly criticised as ‘probably too polemical to carry much weight’ (LS Reference Long and Sedley1987: 165):

T14 In this way it will be shown that not even ‘something’ is the genus of everything. For there will also be a genus of ‘one’, which is either equal to it or broader than it – if, at any rate, ‘one’ is predicated of the concept, whereas ‘something’ is said only of bodies and incorporeals, and the concept is neither of these according to those who speak of these things.Footnote ³⁶

(Alexander, in Top. 359.12–16; trans. LS 30D)

So, there is no clear evidence that concepts, according to the Stoics, were mind-dependent items.

But if concepts are indeed non-existent somethings, does this mean that we need to assume a fifth category to the standard four incorporeals listed by our sources – namely, void, place, time and sayables (lekta) (e.g., Sext. Emp. M. 10.218; 234; 237; 11.230)?Footnote ³⁷ I agree with Caston that there must have been a development in the Stoics’ attitude to concepts, but I am not convinced that the change was, as he suggests, drastic in that Zeno and Cleanthes treated concepts as somethings, whereas Chrysippus decided to abandon them altogether and expel them from his ontology. The fact that no surviving text attributes to Chysippus the use of ennoēmata cannot be a decisive argument, for there is no doubt that the textual evidence on Stoic ontology is deplorably meagre. Briefly stated, my suggestion is the following: All Stoics regarded concepts as non-existent somethings, that is, as incorporeals, although they seem to have dealt with them differently. For they initially considered them as predicates (katēgorēmata), but from Chrysippus’ time onwards they also thought of them in terms of their corresponding definitions. Still, Stoic concepts were always treated as sayables, that is, as the fourth category of incorporeals and, hence, as belonging to the supreme genus of something.

Let me try to defend my interpretation: It has been said above that the Stoics defined conceptions as rational impressions or thoughts (noēseis). The Stoics also defined sayables as what subsists in accordance with a rational impression (kata phantasian logikēn huphistamenon: Diog. Laert. 7.63; Sext. Emp. M. 8.70; PH. 2.104), or what subsists upon or along with our thought (parhuphistamenou dianoia: Sext. Emp. M. 8.12).Footnote ³⁸ It is thus reasonable to credit the Stoics with the view that, as in the case of all thoughts, the contents of conceptions, namely what is thought, are sayables. For instance, when we perceive a human being, we have the thought of something being human, that is, we conceive of the predicate ‘being human’; and Cleanthes is said in our sources to have regarded predicates as sayables (Clement of Alexandria, Strom. 8.9.26.3–4), in particular, as incomplete sayables. But it has also been argued that the content of inchoate preconceptions is articulated by thought-out definitions, which are of course complete sayables. More precisely, Chrysippus formulated definitions as conditionals ranging over particulars (e.g., Sext. Emp. M. 11.8–11).Footnote ³⁹ For instance, the standard definition of a human being ‘Human beings are mortal, rational animals.’ is properly expressed by the Stoics as follows: ‘If something is a human being, then that something is a mortal, rational animal.’

Therefore, the Stoics seem to have thought of concepts as incorporeal sayables, whether incomplete or complete; incomplete in the case of the predicates, complete in the case of the definitions corresponding to articulated conceptions. Most importantly, being sayables, concepts were regarded by the Stoics as ontologically mind-independent items that subsist whether we think of them or not, and hence as belonging to the supreme genus of something. In this way, they had no reason to assign to concepts the ontological status that Plato had previously assigned to the Forms, and developed a theory that explains our capacity to form concepts merely on the basis of our repeated sense-experience of particular existent bodies .

4 The Epistemic Function of Stoic Conceptions and Concepts

4.1 Building Blocks of Rationality

The Stoics’ empiricist predilections are clearly attested in Aëtius’ passage [T2]: The process of acquiring knowledge starts with sense-perception, which provides us with impressions of particular observable facts. But we cannot rely on these if we want to attain the absolute knowledge that constitutes wisdom. To this purpose, the Stoics introduced in their epistemology the general conceptions and their corresponding concepts that are generated from experience based on repeated memories. As we have said, they followed the Epicureans in talking about those conceptions and concepts that are directly derived from our senses but, in addition, they talked about those that require some technical instruction and attention. So, according to the Stoics, although the human soul has no content at birth, it has the capacity to acquire a sufficiently rich set of conceptions and concepts in terms of which it starts to think rationally, so that human beings come to be distinguished from non-rational animals:

[T15] Owing to it he [i.e. Zeno] also rated the senses as trustworthy, since, as I said before, he thought that an apprehension caused by the senses was true and reliable – not because it apprehended all the features of its object, but on the ground that it omitted nothing detectable by it. Another reason was that nature had given apprehension as a standard and starting point for scientific knowledge of the world: it was the source from which our conceptions of things were later stamped on our minds, which in turn give rise not just to the starting points but to certain broader paths for discovering reason.Footnote ⁴⁰

(Cic., Acad. 1.42; trans. Brittain)

Conceptions and concepts, therefore, owe their special epistemic function as building blocks of rationality to the way they are formed. In fact, it is our conceptions and concepts that, according to the Stoics, ensure the possibility of knowledge, by offering an alternative solution to Meno’s paradox and by rendering Plato’s theory of recollection redundant. That is to say, having even a rough preconception of something enables one to recognise it when encountering it, or to continue inquiring about it in order to form an articulated conception and the corresponding concept, so that the desired knowledge is finally reached.Footnote ⁴¹

Nevertheless, scholars have expressed doubts as to whether it is really the case that the Stoics managed to avoid Plato’s innatism. For there are some passages in our Stoic sources that refer to preconceptions as innate or implanted (emphutoi):

T16 He [Chrysippus] says that the theory of good and bad things introduced and approved by himself is most in harmony with life and connects best with the innate preconceptions.Footnote ⁴²

(Plut., Stoic. rep. 1041E; trans. LS 60B)

But since such passages are mainly in later sources,Footnote ⁴³ it has been suggested that it is only among the late Stoics that we find an innatist distortion of the empiricist early Stoic theory, either because of a Platonic influence or because of an interest in countering sceptical challenges to Stoicism (Sandbach Reference Sandbach1930; Long Reference Long2002: 80–83). On the other hand, it has also been argued that the late Stoics’ innatism does not amount to radical heterodoxy but is merely a robust affirmation of the official Stoic line on the subject. The early Stoics seem to have been dispositional innatists as much as the late Stoics, that is, they also claimed that even if human beings are not born with conceptions, they are nonetheless predisposed to the formation of them. For instance, the conception of something being good derives ultimately from the inclination, innate in all animals, to distinguish what is beneficial for themselves from what is harmful (Scott Reference Scott1988; Jacson-McCabe Reference Jackson-McCabe2004).

It is true, however, that the textual evidence presenting the early Stoics as dispositional innatists is extremely limited. Besides, I find it particularly intriguing that there seems to be an emergence of innatist tendencies in the late Epicurean texts, too, although this view has also been much debated among scholars.Footnote ⁴⁴ Still, I suggest that it would be worth studying the relevant evidence from the Epicureans and the Stoics together, in order to assess whether such tendencies can be seen as reflections of significant modifications in the late doctrines of the Hellenistic schools. It is, after all, such modifications that may be said to have signalled the advent of even more changes in the theories of knowledge introduced after the end of the Hellenistic period .Footnote ⁴⁵

4.2 Criteria of Truth

But if conceptions and concepts are considered by the Stoics as the building blocks of human rationality, do they also serve as the foundations of knowledge? In other words, are they used as criteria of truth, too? Diogenes Laertius (7.54) accuses Chrysippus of inconsistency, because sometimes he claimed, in line with the standard Stoic dogma, that only cognitive impressions (katalēptikai phantasiai) are criteria of truth, whereas other times he suggested that sense-perception and preconceptions are also criteria of truth. It has been argued, though, that this conflict is merely apparent; we should rather assume a shift in the use of the term ‘kritērion’ and take into consideration the context in which it is mentioned. For when the Stoics discussed the attainability of knowledge, they talked about cognitive impressions as criteria of truth, whereas when they were interested in the details of their epistemology, they also invoked sense-perception and preconceptions (Frede Reference Frede and Algra1999: 316–18).

A further argument in favour of the view that Stoic preconceptions and, in general, conceptions should also be thought of as criteria of truth is the fact that they are characterised in our sources as ‘evident’ (enargeis; e.g., Epictetus, Diss. 1.27.6; 2.12.6). I have argued elsewhere (Ierodiakonou Reference Ierodiakonou, Morison and Ierodiakonou2011), the notion of ‘evidence’ or ‘self-evidence’ (enargeia) characterises both the Epicurean and the Stoic criteria of truth; just as in the case of cognitive impressions, it suffices to consider conceptions as evident in order to guarantee their criterial role and render them foundations of knowledge.

Assuming, therefore, that conceptions are actually criteria of truth, we next need to specify how they function as such. Alexander of Aphrodisias provides us with an illustration, according to which Chrysippus corroborated his doctrine of three different kinds of mixtures on the basis of three different conceptions:

T17 He [i.e., Chrysippus] tries to support the existence of these different mixtures through the common conceptions, and says that we take these from nature as excellent criteria of truth: we certainly have one impression for the bodies composed by joining, and a different one for those that are fused and destroyed together, and another for those that are blended and mutually coextended through and through so that they each preserve their own nature; we would not have these different impressions if all things, however they were mixed, lay side by side one another by joining.Footnote ⁴⁶

(Alexander, Mixt. 217.2–9; trans. LS 48C5–6)

Scholars have offered different models of the exact way conceptions are supposed to be used, according to the Stoics, in order to access the truth or falsity of impressions and beliefs.Footnote ⁴⁷ I do not want to enter into the details of this discussion but, briefly stated, I find the following plausible: To claim that conceptions are criterial implies that their corresponding definitions may play the epistemological role that cognitive impressions normally play, namely they can serve to judge further impressions and beliefs. The Stoics thus seem to have thought that, by using the definitions of articulated conceptions, humans are able to arrive at whole systematic bodies of beliefs and, ultimately, at wisdom.

5 Conclusion

The discussions in our ancient sources about the Stoics’ views on conceptions and concepts have been characterised as ‘seemingly byzantine’ (Brunschwig Reference Brunschwig and Inwood2003: 224); no doubt the same description applies to the scholarly interpretations of these views in recent books and articles. I resent the pejorative connotation of this description, but it is true that the Stoics’ theory of concept formation brings together some of the most bewildering issues of their ontology and epistemology. Still, there is certainly nothing wrong with highly complex and subtle theories so long as they are not unnecessarily intricate or deliberately evasive, and there are also compelling philosophical reasons why they cannot be straightforwardly articulated and easily comprehended. In this sense the Stoic account of conceptions and concepts fits perfectly Brunschwig’s characterisation.

Footnotes

* I would like to thank Gábor Betegh, Paolo Crivelli, Marion Durand, Voula Tsouna, David Wolfsdorf and the anonymous reviewers for their careful reading of an earlier version of this chapter and their helpful suggestions. Ada Bronowski’s book The Stoics on Lekta: All There Is to Say (Oxford 2019) came out after this chapter was written.

¹ On the distinction between ennoia and ennoēmata, see esp. Section 3 below.

² For references to the most influential interpretations, see notes below.

³ I translate ennoia as ‘conception’ and not as ‘concept’, because I think that ‘concept’ better translates ennoēma, given the Stoic distinction between these terms, that is, between the mental state and the content of the mental state respectively. My translation is in conformity with, for instance, LS Reference Long and Sedley1987; Brunschwig Reference Annas and Barnes1994; Crivelli Reference Crivelli2007; Bronowski Reference Bronowski2013; for a different choice in translation, see Warren (p. 58) and Irwin (p. 44) in this volume.

⁴ See also the similar use of the noun epinoia: e.g., Epictetus, Diss. 1.14.8; 3.13.7; Sext. Emp. M. 7.56; 8.453; 9.49; 10.188.

⁵ See also: Plut., De soll. an. 961C-D; [Galen], Def. Med. 19.381.12 K.; Porph., Abst. 3.22.17–18.

⁶ φαντασία γάρ τις ἡ ἔννοιά ἐστι, φαντασία δὲ τύπωσις ἐν ψυχῇ… ὥστε τὰς ἐννοίας <ἐν>αποκειμένας τινὰς ὁριζόμενοι νοήσεις.

⁷ Aëtius 4.11.4–5; Stobaeus 1.136.21–137.6; Diog. Laert. 7.61. For close discussion of these texts, see Sections 3.2–4.

⁸ Οἱ Στωϊκοί φασιν· ὅταν γεννηθῇ ὁ ἄνθρωπος, ἔχει τὸ ἡγεμονικὸν μέρος τῆς ψυχῆς ὥσπερ χάρτην εὔεργον εἰς ἀπογραφήν· εἰς τοῦτο μίαν ἑκάστην τῶν ἐννοιῶν ἐναπογράφεται. πρῶτος δὲ ὁ τῆς ἀναγραφῆς τρόπος ὁ διὰ τῶν αἰσθήσεων. αἰσθανόμενοι γάρ τινος οἷον λευκοῦ, ἀπελθόντος αὐτοῦ μνήμην ἔχουσιν· ὅταν δὲ ὁμοειδεῖς πολλαὶ μνῆμαι γένωνται, τότε φαμὲν ἔχειν ἐμπειρίαν· ἐμπειρία γάρ ἐστι τὸ τῶν ὁμοειδῶν φαντασιῶν πλῆθος. τῶν δὲ ἐννοιῶν αἱ μὲν φυσικῶς γίνονται κατὰ τοὺς εἰρημένους τρόπους καὶ ἀνεπιτεχνήτως, αἱ δὲ ἤδη δι’ ἡμετέρας διδασκαλίας καὶ ἐπιμελείας· αὗται μὲν οὖν ἔννοιαι καλοῦνται μόνον, ἐκεῖναι δὲ καὶ προλήψεις. ὁ δὲ λόγος, καθ’ ὃν προσαγορευόμεθα λογικοὶ ἐκ τῶν προλήψεων συμπληροῦσθαι λέγεται κατὰ τὴν πρώτην ἑβδομάδα.

⁹ On recent interpretations of the Epicureans’ account of preconceptions, see Tsouna Reference Tsouna2016; Németh Reference Németh2017: 27–47. See also Betegh and Tsouna in this volume.

¹⁰ Sextus Empiricus (M. 8.58–61; 9.393–5), too, describes some of these natural processes, i.e., similarity, analogy and combination.

¹¹ Κατὰ περίπτωσιν μὲν οὖν ἐνοήθη τὰ αἰσθητά· καθ’ ὁμοιότητα δὲ τὰ ἀπό τινος παρακειμένου, ὡς Σωκράτης ἀπὸ τῆς εἰκόνος· κατ’ ἀναλογίαν δὲ αὐξητικῶς μέν, <ὡς> ὁ Τιτυὸς καὶ Κύκλωψ· μειωτικῶς δέ, ὡς ὁ Πυγμαῖος. καὶ τὸ κέντρον δὲ τῆς γῆς κατ’ ἀναλογίαν ἐνοήθη ἀπὸ τῶν μικροτέρων σφαιρῶν. κατὰ μετάθεσιν δέ, οἷον ὀφθαλμοὶ ἐπὶ τοῦ στήθους· κατὰ σύνθεσιν δὲ ἐνοήθη Ἱπποκένταυρος· καὶ κατ’ ἐναντίωσιν θάνατος. νοεῖται δὲ καὶ κατὰ μετάβασίν τινα, ὡς τὰ λεκτὰ καὶ ὁ τόπος. φυσικῶς δὲ νοεῖται δίκαιόν τι καὶ ἀγαθόν· καὶ κατὰ στέρησιν, οἷον ἄχειρ.

¹² It should be noted that Diogenes’ statement that, in particular, the ethical conceptions of something being just and good arise phusikōs does not imply that the other conceptions mentioned here are not acquired naturally.

¹³ There is no reason to suppose that Diogenes’ list is exhaustive. On the contrary, it is tempting to add yet another mental process that seems to have also been considered by the Stoics among those by which we come to form preconceptions; namely, by way of co-recollection or simultaneous recollection (kata summnēmoneusin: Sext. Emp. M. 1.129; 7.276–80; 9.353–5; 10.64; 10.176; PH3.108). Sextus informs us how, according to some unnamed dogmatists, the process of simultaneous recollection functions: It is triggered initially by perceiving something through our senses; on the basis of this sense-impression, we recollect some previous sense-impression; by simultaneously having both the current sense-impression and the memory of the previous one, we conceive of something different from both, and we thus form a preconception of it. For instance, in order to form the preconception of something being in motion, what is required is to perceive something in one place and simultaneously remember it being in a different one. In such cases of simultaneous recollection, sense-perception is not adequate, since it can register only instantaneous events and not events over time, so some thought is also required. Nevertheless, the conception of something being in motion and, in general, all conceptions generated by simultaneous recollection were probably regarded by the Stoics as preconceptions and not as conceptions in the narrow sense, since they are formed without any special instruction and attention. For a more detailed account of summnēmoneusis and its attribution to the Stoics, see Ierodiakonou Reference Ierodiakonou2015: 116–29.

¹⁴ See also: [Galen], Def. Med. 19.348.18–349.4 K.

¹⁵ animum concipere notiones, quas appellant ἐννοίας, earum rerum scilicet quas definiendo explicant; hinc propagari atque conecti totam discendi docendique rationem.

¹⁶ E.g., Galen, Hipp. Epid. 6, 223.14–15; Dig. Puls. 8.880.16–881.4 K.; Simplicius, in Cat. 154.3–6; 379.12–20. On the Stoics’ use of διάρθρωσις, see Gourinat Reference Gourinat2000: 46–51; Brittain Reference Brittain2005: 179–81; Crivelli Reference Crivelli2010: 383–87.

¹⁷ Ἠθικοῦ λόγου τοῦ περὶ τὴν διάρθρωσιν τῶν ἠθικῶν ἐννοιῶν. On the authenticity of this title, see Brittain Reference Brittain2005: 182, n. 68.

¹⁸ Aëtius 5.23.1: περὶ δὲ τὴν δευτέραν ἑβδομάδα ἔννοια γίνεται καλοῦ τε καὶ κακοῦ καὶ τῆς διδασκαλίας αὐτῶν. For a systematic reading of this passage, see Mansfeld Reference Mansfeld2014: 623–24.

¹⁹ See also: Cic., Top. 31; Tusc. 4.53; Plut., Comm. not. 1059B-C.

²⁰ atqui qualia sunt haec quae sensibus percipi dicimus talia secuntur ea quae non sensibus ipsis percipi dicuntur sed quodam modo sensibus, ut haec: ‘illud est album, hoc dulce, canorum illud, hoc bene olens, hoc asperum’: animo iam haec tenemus conprehensa non sensibus. ‘ille’ deinceps ‘equus est, ille canis.’ cetera series deinde sequitur maiora nectens, ut haec quae quasi expletam rerum conprehensionem amplectuntur: ‘si homo est, animal est mortale rationis particeps.’ quo e genere nobis notitiae rerum inprimuntur sinc quibus nec intellegi quicquam nec quaeri <nec> disputari potest.

²¹ See also, Epictetus, Diss. 2.11.18; 2.12.9.

²² ἀγαθὸν καὶ κακὸν καὶ συμφέρον καὶ ἀσύμφορον τίς ἡμῶν οὐ λαλεῖ; τίς γὰρ ἡμῶν οὐκ ἔχει τούτων ἑκάστου πρόληψιν; ἆρ’ οὖν διηρθρωμένην καὶ τελείαν; τοῦτο δεῖξον. ‘πῶς δείξω;’ ἐφάρμοσον αὐτὴν καλῶς ταῖς ἐπὶ μέρους οὐσίαις. εὐθὺς τοὺς ὅρους Πλάτων μὲν ὑποτάσσει τῇ τοῦ χρης<ί>μου προλήψει, σὺ δὲ τῇ τοῦ ἀχρήστου. δυνατὸν οὖν ἐστιν ἀμφοτέρους ὑμᾶς ἐπιτυγχάνειν; πῶς οἷόν τε; τῇ δὲ τοῦ πλούτου οὐσίᾳ οὐχ ὁ μέν τις ἐφαρμόζει τὴν τοῦ ἀγαθοῦ πρόληψιν, ὁ δ’ οὔ; τῇ δὲ τῆς ἡδονῆς, τῇ δὲ τῆς ὑγείας; καθόλου γὰρ εἰ πάντες οἱ τὰ ὀνόματα λαλοῦντες μὴ κενῶς ἴσμεν ἕκαστα τούτων καὶ μηδεμιᾶς ἐπιμελείας περὶ τὴν διάρθρωσιν τῶν προλήψεων δεόμεθα, τί διαφερόμεθα, τί πολεμοῦμεν, τί ψέγομεν ἀλλήλους;

²³ Ὅτι δὲ γνωριμώτερον ἡμῖν τὸ ἀδιόριστον καὶ συγκεχυμένον οἷον τὸ ὅλον δοκεῖ, πιστοῦται τὸ μὲν ὄνομα ἕκαστον ὡς ὅλον τι λαβών, τὸν δὲ τοῦ ὀνόματος ὁρισμὸν ὡς τὴν διάρθρωσιν τῶν τοῦ ὀνόματος μερῶν καὶ στοιχείων παραδιδόντα. δῆλον γὰρ ὅτι ἡ μὲν κατὰ τὸ ὄνομα γνῶσις τοῦ κύκλου πρόχειρος καὶ τοῖς πολλοῖς ἐστιν, ὁ δὲ τοῦ κύκλου ὁρισμός, ὅτι ἔστι σχῆμα ἐπίπεδον ὑπὸ μιᾶς γραμμῆς περιεχόμενον, πρὸς ἣν ἀπὸ ἑνὸς σημείου πᾶσαι αἱ προσπίπτουσαι [πρὸς τὴν τοῦ κύκλου περιφέρειαν] ἴσαι ἀλλήλαις εἰσίν, οὗτος δὴ ὁ ὁρισμὸς οὐκέτι πρόχειρος πᾶσι τὰ καθέκαστα τοῦ κύκλου παραδιδοὺς καὶ τοῖς μέρεσιν αὐτοῦ καὶ στοιχείοις ἐπεξιών.

²⁴ For a systematic list of all occurrences of these terms in our sources, see Dyson Reference Dyson2009: 155–62.

²⁵ E.g., Seneca, Ep. 65.11; Plut., Adv. Col. 1116B–C; Alexander, in Top. 301.19–25. On the Stoics’ ontological principles as well as on their stance towards Plato’s theory of Forms, see Goldschmidt Reference Goldschmidt1972; Pasquino Reference Pasquino and Brunschwig1978; Brunschwig Reference Brunschwig, Barnes and Mignucci1988 and Reference Brunschwig and Inwood2003; Caston Reference Caston1999: 148–50 and 176–82; Boeri Reference Boeri2001; Sedley Reference Sedley and Salles2005: 119–21; Vogt Reference Vogt2009; de Harven Reference De Harven2015.

²⁶ Pasquino Reference Pasquino and Brunschwig1978: 378; Sedley Reference Sedley1985: 87 and Reference Sedley and Salles2005, 120-21; Brunschwig Reference Brunschwig, Barnes and Mignucci1988 and Reference Brunschwig and Inwood2003; Crivelli Reference Crivelli2007: 113–18; Bronowski Reference Bronowski2013; Bailey Reference Bailey2014.

²⁷ Ζήνωνος <καὶ τῶν ἀπ’ αὐτοῦ>. τὰ ἐννοήματά φασι μήτε τινὰ εἶναι μήτε ποιά, ὡσανεὶ δέ τινα καὶ ὡσανεὶ ποιὰ φαντάσματα ψυχῆς· ταῦτα δὲ ὑπὸ τῶν ἀρχαίων ἰδέας προσαγορεύεσθαι. τῶν γὰρ κατὰ τὰ ἐννοήματα ὑποπιπτόντων εἶναι τὰς ἰδέας, οἷον ἀνθρώπων, ἵππων, κοινότερον εἰπεῖν πάντων τῶν ζῴων καὶ τῶν ἄλλων ὁπόσων λέγουσιν ἰδέας εἶναι. ταύτας δὲ οἱ Στωικοὶ φιλόσοφοί φασιν ἀνυπάρκτους εἶναι, καὶ τῶν μὲν ἐννοημάτων μετέχειν ἡμᾶς, τῶν δὲ πτώσεων, ἃς δὴ προσηγορίας καλοῦσι, τυγχάνειν.

²⁸ Ἐννόημα δέ ἐστι φάντασμα διανοίας, οὔτε τὶ ὂν οὔτε ποιόν, ὡσανεὶ δέ τι ὂν καὶ ὡσανεὶ ποιόν, οἷον γίνεται ἀνατύπωμα ἵππου καὶ μὴ παρόντος.

²⁹ primum genus Stoicis quibusdam videtur ‘quid’; quare videatur subiiciam. in rerum, inquiunt, natura quaedam sunt, quaedam non sunt. et haec autem quae non sunt rerum natura complectitur, quae animo succurrunt, tamquam Centauri, Gigantes et quicquid aliud falso cogitatione formatum habere aliquam imaginem coepit, quamvis non habeat substantiam.

³⁰ For a logical analysis of this argument, see Caston Reference Caston1999: 200–4; Crivelli Reference Crivelli2007: 98–113.

³¹ καὶ γὰρ καὶ Χρύσιππος ἀπορεῖ περὶ τῆς ἰδέας, εἰ τόδε τι ῥηθήσεται. συμπαραληπτέον δὲ καὶ τὴν συνήθειαν τῶν Στωικῶν περὶ τῶν γενικῶν ποιῶν, πῶς αἱ πτώσεις κατ’ αὐτοὺς προφέρονται, καὶ πῶς οὔτινα τὰ κοινὰ παρ᾽ αὐτοῖς λέγεται, καὶ ὅπως παρὰ τὴν ἄγνοιαν τοῦ μὴ πᾶσαν οὐσίαν τόδε τι σημαίνειν καὶ τὸ παρὰ τὸν Οὖτιν σόφισμα γίνεται παρὰ τὸ σχῆμα τῆς λέξεως, οἷον ‘εἴ τίς ἐστιν ἐν Ἀθήναις, οὐκ ἔστιν ἐν Μεγάροις * * *’· ὁ γὰρ ἄνθρωπος οὐ τίς ἐστιν· οὐ γάρ ἐστιν τὶς ὁ κοινός· ὡς τινὰ δὲ αὐτὸν ἐλάβομεν ἐν τῷ λόγῳ, καὶ παρὰ τοῦτο καὶ τὸ ὄνομα τοῦτο ἔσχεν ὁ λόγος Οὖτις κληθείας.

* * * <ἄνθρωπος δε ἐστιν ἐν Ἀθήναις· ἄνθρωπος ἄρα οὐκ ἔστιν ἐν Μεγάροις> suppl. Kalbfleisch

³² καὶ μὴν εἰ διδάσκεται τί, ἤτοι διὰ τῶν οὐτινῶν διδαχθήσεται ἢ διὰ τῶν τινῶν. ἀλλὰ διὰ μὲν τῶν οὐτινῶν οὐχ οἷόν τε διδαχθῆναι· ἀνυπόστατα γάρ ἐστι τῇ διανοία ταῦτα κατὰ τοὺς ἀπό τῆς Στοᾶς.

³³ Vanessa de Harven (Reference De Harvenforthcoming) endorses Caston’s arguments against the view that concepts are not-somethings but departs from his developmental account and argues that all Stoics regarded concepts as nothing at all, since they are merely private intentional objects.

³⁴ ἔστι δ’ ἐννόημα φάντασμα διανοίας λογικοῦ ζῴου· τὸ γὰρ φάντασμα ἐπειδὰν λογικῇ προσπίπτῃ ψυχῇ, τότε ἐννόημα καλεῖται, εἰληφὸς τοὔνομα παρὰ τοῦ νοῦ. – Διόπερ τοῖς ἀλόγοις ζῴοις ὅσα προσπίπτει, φαντάσματα μόνον ἐστίν· ὅσα δὲ ἡμῖν καὶ τοῖς θεοῖς, ταῦτα καὶ φαντάσματα κατὰ γένος καὶ ἐννοήματα κατ’ εἶδος· ὥσπερ τὰ δηνάρια καὶ οἱ στατῆρες αὐτὰ μὲν καθ’ αὑτὰ ὑπάρχει δηνάρια <καὶ> στατῆρες· ἐὰν δὲ εἰς πλοίων δοθῇ μίσθωσιν, τηνικαῦτα πρὸς τῷ δηνάρια εἶναι καὶ ναῦλα λέγεται.

³⁵ On the scholarly controversy around Seneca’s letter, see Brunschwig Reference Brunschwig and Inwood2003: 220–22.

³⁶ οὕτω δειχθήσεται μηδὲ τὸ τὶ γένος ὂν τῶν πάντων· ἔσται γὰρ καὶ τοῦ ἑνὸς γένος ἢ ἐπ’ ἴσης ὄντος αὐτῷ ἢ καὶ ἐπὶ πλέον, εἴ γε τὸ μὲν ἓν καὶ κατὰ τοῦ ἐννοήματος, τὸ δὲ τὶ κατὰ μόνων σωμάτων καὶ ἀσωμάτων, τὸ δὲ ἐννόημα μηδέτερον τούτων κατὰ τοὺς ταῦτα λέγοντας.

³⁷ On the related debate concerning the Stoics’ account of the ontological status of mathematical concepts, see Ierodiakonou Reference Ierodiakonou2018: 127–32. Since very few texts provide us with information on this issue (Proclus, in Eucl. 89.15–8; 395.13–18; Plut., Comm. not. 1078E–1080E; Diog. Laert. 7.135), conflicting interpretations have been suggested: Some scholars have claimed that, according to the Stoics, mathematical concepts are mental constructs and thus nothing at all, whereas others have argued that the Stoics thought of mathematical concepts as incorporeals. Moreover, some alternative positions have also been put forward; for instance, that the objects of geometry were regarded as incorporeals while the objects of arithmetic as mental constructs; that the Stoics distinguished between the geometrical limits of real bodies that are incorporeals and the mentally constructed limits that are reached through infinite division. Finally, it has been argued that the inconsistency of Stoic opinions in our ancient sources may reflect an extensive debate between the late and the early Stoics, but also some degree of disagreement among the early Stoics themselves.

³⁸ On the Stoic notion of sayables, see Frede Reference Frede and Everson1994a; Gourinat Reference Gourinat2019: 142–51.

³⁹ On Stoic definitions, see Caston Reference Caston1999: 195–99; Crivelli Reference Crivelli2007: 118–22; Reference Crivelli2010.

⁴⁰ e quo sensibus etiam fidem tribuebat, quod ut supra dixi comprehensio facta sensibus et vera esse illi et fidelis videbatur, non quod omnia quae essent in re comprehenderet, sed quia nihil quod cadere in eam posset relinqueret, quodque natura quasi normam scientiae et principium sui dedisset unde postea notiones rerum in animis imprimerentur; e quibus non principia solum sed latiores quaedam ad rationem inveniendam viae reperiuntur.

⁴¹ On the Stoics’ reply to Meno’s paradox, see Frede Reference Frede and Boudouris1994b: 54–55; Brittain Reference Brittain2005: 179–80; Fine Reference Fine2014: 257–95.

⁴² Τὸν περὶ ἀγαθῶν καὶ κακῶν λόγον, ὃν αὐτὸς εἰσάγει καὶ δοκιμάζει, συμφωνότατον εἶναί φησι τῷ βίῳ καὶ μάλιστα τῶν ἐμφύτων ἅπτεσθαι προλήψεων.

⁴³ E.g., Cic., Nat. D. 2.12–15; Seneca, Ep. 117.6; Plut., Stoic. rep. 1070C–D; Epictetus, Diss. 2.11.1–7.

⁴⁴ For a detailed discussion of the scholarly debate concerning Epicurean innatism, see Tsouna Reference Tsouna2016: 174–85. See also Betegh and Tsouna in this volume.

⁴⁵ For Epictetus’ influence on Galen’s doctrine of a kind of innate tacit knowledge in ethics, see Chiaradonna Reference Chiaradonna2018: 339–44.

⁴⁶ τὸ δὲ ταύτας τὰς διαφορὰς εἶναι τῆς μίξεως πειρᾶται πιστοῦσθαι διὰ τῶν κοινῶν ἐννοιῶν, μάλιστα δὲ κριτήρια τῆς ἀληθείας φησὶν ἡμᾶς παρὰ τῆς φύσεως λαβόντας. ἄλλην γοῦν φαντασίαν ἔχειν ἡμᾶς τῶν καθ’ ἁρμὴν συγκειμένων, καὶ ἄλλην τῶν συγκεχυμένων τε καὶ συνεφθαρμένων, καὶ ἄλλην τῶν κεκραμένων τε καὶ ἀλλήλοις δι’ ὅλων ἀντιπαρεκτεινομένων οὕτως, ὡς σώζειν ἕκαστον αὐτῶν τὴν οἰκείαν φύσιν. ἣν διαφορὰν φαντασιῶν οὐκ ἂν εἴχομεν, εἰ πάντα τὰ ὁπωσοῦν μιγνύμενα παρέκειτο ἀλλήλοις καθ’ ἁρμήν.

⁴⁷ Gould Reference Gould1970: 60–64; Todd Reference Todd1973: 55–60; Schofield Reference Schofield, Schofield, Burnyeat and Barnes1980: 291–305; Frede Reference Frede and Boudouris1994b: 57; Dyson Reference Dyson2009: 23–47.