1. Articulation of letters

Having thus described the frame of the Sophist’s discussion, I can begin the examination of the passages concerning letters, nouns and logoi. Immediately after exposing the dilemmas in his opponents’ arguments (253a), Plato speaks of a form of communion that at first glance appears to be simply an analogy for his new ontology, namely, the communion of letters (γράµµατα): “Visitor: So, if some things will mix and some won’t, it will be pretty much with them as it is with the letters of the alphabet. These too have the feature that some of them fit together in one way or another, while others don’t. Theaetetus: Of course. Visitor: But the vowels, now: they differ from the other letters, running through them all and serving like a bond between them, so that without a vowel it’s impossible for any of the others to fit together either” (Soph., 253a1-6).

At first glance, the passage may seem like an analogy, but the analogy takes a different turn when considering the dialogue’s context, which is about discovering the logos to capture the sophist. When, in a text about logos, the letters, which are parts of the logos, are used as analogies to explain logos, the analogy goes beyond merely being an analogy. In this respect, these passages are not only an analogy but also inform us about the reality of logos and letters⁸. Plato, who criticizes the paradigms in which things stand alone without mixing with each other and everything is intermingled endlessly, first resorts to the example of letters when starting to develop his own ontology as an alternative to these: some letters can be together, while others cannot. Socrates develops this in the text as follows: Letters can come together thanks to the vowel, which functions as a link by establishing a connection between them. How should this be understood? Indeed, vowels become a continuous phoneme without consonants, in which each letter loses its unity, or the multiplicity disappears into a whole, as in the Parmenidean hypothesis, or the parts of the multiplicity become inseparable in absolute mixing, as in the absolute mixing hypothesis. Moreover, consonants cannot be pronounced without vowels, as in the second hypothesis of the plurality without interaction (pseudo-Plato’s hypothesis).

In his article Letters and Syllables in Plato, Ryle⁹ comments on the passage that assumes that consonants cannot be pronounced without vowels but rather emerge together with vowels (in this sense, I need the vowel “e” to pronounce the letter “b”). At first glance, this commentary seems reasonable but, as Gallop¹⁰ states in his article which is a response to Ryle’s article, the etymology of gramma is related to the meanings of “drawing” and “that which is written”. Gallop shows that Plato uses this for the written characters in his various texts by criticizing Ryle’s interpretation. As a second counterargument, Gallop says that we can represent the letter “b” in writing without needing a vowel. In addition, we can contribute to this critical lecture by pointing to a third group of letters in Greek, namely semi-consonant letters -neither vowels nor consonants- which cause issues in this regard (Poetica, XX, 1456b). So, how should one interpret the question of letters not appearing without commonality?

I try to answer this question by suggesting that the concept which Plato is trying to point out in this passage is the articulation of letters and that letters can only appear as letters in an articulated structure. Plato expresses this in passage 253a with the sentence “But the vowels, now: they differ from the other letters, running through them all and serving like a bond between them, so that without a vowel it’s impossible for any of the others to fit together either” (Soph., 253a). Thus, according to Plato, letters in human language can be articulated to produce multiple names because they are differentiated into vowels and non-vowels. Therefore, letters can only be letters if they are articulated in a harmonic communion and can only form names with groups of letters that can signify the multiplicity within to pan. Instead, sounds that are not articulated do not consist of letters.

At this point, I would like to cite and discuss an important use of graphō, which has the same root as gramma in Aristotle’s De Interpretatione, who kept using Plato’s conceptualization of gramma and stoicheion. Aristotle states, “That which is in sounds are signs of pathēmata in the soul, and that which is written are signs of that which is in sounds”¹¹ (De Interpretatione, 1, 16a3-4). This statement basically asserts that the pathēmata are formed by the things, we signify them with what is in the sound, and we signify what is in the sound with grammata. Giorgio Agamben¹² comments on this in his book Infancy and History by consulting to the ancient commentaries: “Aristotle’s ancient commentators had asked why the philosopher had introduced the gramma as the fourth ‘hermeneut’ alongside the other three (voice, pathēmata, things) which explain the circle of linguistic signification. So they attributed the particular status of the gramma to the fact that, unlike the other three, it is not just a sign, but also an element (stoicheion) of the voice, as articulation”.

According to this interpretation, things, their pathēmata in the soul, and that which is in sound as elements of the circle of signification, are completed only when they can be symbolizable in writing, i.e., have the possibility to be articulated¹³. In his commentary on De Interpretatione, Boethius remarks on this structure as follows: Thus, that which is in sounds (onomata and rhēmata) indicate thoughts that correspond to things; while not every sound (vox) can be an onoma and a rhēma, that which is in sounds only is written/represented by letters, the linguistic circle is completed, and one may assign a sense to this utterance (locutio) (42,15-30). In passage 5.1-5 of the same commentary, Boethius comments that a sense is not assignable to every sound (vox) but only to the articulated utterance (locutio). He states, “The components of utterance are letters, which when joined create one conjoined and combined spoken sound which is called utterance”¹⁴ (Boethius, On Aristotle On Interpretation, 5.9-10).

We could interpret Boethius’ idea as a retroactive movement in which only the sounds having the power to be symbolized by letters can be assigned meaning since only the parts of these sounds can be decomposed and captured as units in relation to each other. This means that the sounds must be presented in a unification model in such a way that the letters can be captured separately from each other, which is what we call articulation.

I think that we can explain Plato’s evaluation of “Letters combine because some letters function like a bridge bonding others” using a reading of De Interpretatione that is parallel to the approach in these commentaries. In the Sophist, Plato uses the concept of gramma, which is connected to the etymology of drawing and that which is written, instead of using the concept of stoicheion, which means both the elements and the letters as used in the Theaetetus. However, given the context, the etymology of stoicheion is much more suitable for discussion.

In this regard, considering 1) the ancient commentaries on Aristotelian gramma¹⁵, 2) Plato’s choice of the concept of gramma instead of stoicheion, 3) his articulative model of the communion of letters, and 4) his ontological position in which everything (to pan) is possible only in a measured way of the communion of things, we can develop the following idea about Plato’s approach to letters and therefore logos: articulated sounds emerge only if they are capable of being symbolized by singular letters in writing even though they are not practically symbolized. In this sense, we can say that Plato thinks that each of the grammata, which is part of the logos, preserves its singularity, but at the same time the grammata come together in an articulated way, and hence, each one is kept as a letter. Thus, the activity of putting them in communion while grasping them separately is the reason that we have articulated symbols, i.e., letters.

When a human cries and inanimate objects or animals make noises, they are not composed of letters even though they look like articulated sounds, since their parts cannot be grasped as singular symbols. By this, I mean that the ability to grasp symbols as entities allows us to make words to which we can assign a sense once they are assembled. For example, a wolf can howl, or a human can groan in pain. The noises that the wolf makes when howling cannot be represented by letters articulated in writing, and therefore in sounds, since the noise contains an inseparable continuity. That is, the parts of these noises cannot be grasped as a singular symbol¹⁶. From Plato’s point of view, howls and groans can be taken as an analogy of both Parmenidean ontology and of the extreme mixture hypothesis (in which the parts lose themselves in an absolute mixture) by seeing their implications as identical because they do not let the letters emerge as singular symbols in a kind of unification. In this sense, the Parmenidean position, as the holistic position that rejects the multiplicity, and the extreme mixture hypothesis, in which the mixture of the multiplicity leaves no separable singularity, can both be seen as a wolf howl as opposed to Plato’s articulated logos. Thus, letters can only be letters when captured as interconnectable singularities and meaning can only be assigned to names that emerge from their communion. Here we can refer to the distinction between abstractable and extractable developed by Ryle: An abstractable element cannot be captured from the whole but can still be considered separately, while an extractable one can be captured by separating it from the sum¹⁷. According to this distinction, we can qualify letters as abstractable but not extractable. I continue to justify this claim by referring to how it is similarly sustained in the Sophist, in which the predication of names and the accord of logoi are parallel to the letters.

2. Predication of names

After explaining his ontology, which is based on the letters model and established with the communion of eidos or genos¹⁸ while still preserving their identity, Plato begins to explain how the logos mechanism work: “Visitor: So come on, let’s use the same approach again as we did in relation to forms and letters, this time asking about names… Visitor: Well now, no speech is ever formed from names alone being uttered one after another, and neither is it if verbs have been strung together without names… Visitor: For example, ‘walks runs sleeps’, and the other verbs that signal actions - even if someone says all of them in succession, it won’t make them the slightest bit more into speech (logos)… Visitor: Then again, if one says ‘lion stag horse’, and all the names there are of things that do the actions, that collection of words doesn’t yet constitute speech (logos) either…” (Soph., 261d-262c1).

Plato explains the structure of logos by taking two types of communions as a model: eidē and letters. As we said at the beginning, logos, which is presented as a genos that connects ta genē, is surprisingly the same kind as the things that it unifies, that is, genos or eidos. Therefore, the communion of eidē is beyond that of an example because it is logos itself. Although Plato uses other analogies (e.g., the harmony of notes in music) to ground his ontology, other than eidē he only refers to letters when constructing logos. These choices reveal that Plato sees continuity between the two structures: letters and logos.

When examining the passages, it is clear that the logos consists of an onoma and a rhēma¹⁹. However, when a rhēma is predicated on an onoma, alogos emerges, in which we find the second linguistic meaning of koinōnia: the predication of names. In this sense, logos would not emerge if onomata and rhēmata were not combined in the form of predication. Alongside this idea, Plato hides a more interesting idea in these passages: Onoma and rhēma acquire meaning only within the logos. If we look closely at the text, we see that Plato states that although we may put rhēmata in succession/ephexēs (walksrunssleeps), they do not necessarily form a logos. Likewise, three onomata do not form a logos when we put them together continuously/synexēs (lionstaghorse). Thus, he states that if onoma and rhēma are not posited together in an expression, they cannot signify neither “praxis” nor “a-praxis”, neither being of “what is” nor “what is not”.

If we pay attention to these passages, we can see that Plato distinguishes the group of rhēmata from the group of onomata by persistently using the expression ephexēs for the group of rhēmata (262b) while using synexēs for the group of onomata (262a, 262b, 262c). To make sense of this, we can make use of Aristotle’s distinction between these two concepts in Physica and Metaphysica. According to Aristotle, continuity indicates inseparable unity, while succession points to the repetition of the same kind of thing, that is, to separable objects (Physica, VI 1, 231a20-5). In this context, we can interpret Plato’s characterization of names as continuous, as they do not denote a judgment by falling into an inseparable unity, since they do not form logos without a verb, and thus do not indicate a praxis or being.

We can think of this in parallel to the example of a wolf howling since, in both situations, the parts lose their identity in the unification. On the contrary, the verbs in succession can very well have meaning since every verb refers to praxis and since Greek, unlike modern English, allows for a logos only with a rhēma and without any onoma. Then, where there are three rhēmata, there must also be three logos. So why does Plato, who is aware that a logos can be formed with only one verb in Greek, say otherwise? The answer is hidden in the phrases just after the examples: “…since no more in this case than in the other does what has been voiced indicate action, or lack of action, or the being of something that is, or of something that is not, nor will it until someone starts blending the verbs with the names. Then they fit together, and their first interweaving is at once speech…” (262c1-5).

These groups of words do not indicate praxis and being or their absence unless they are a mixture of nouns and verbs. So, only in the case of unity do these parts gain meaning. This does not mean that we cannot make a phrase with only a verb. So, what can it mean? To fully understand this statement, we must accept the condition that every logos must be about something, which Plato expresses in lines 262e6-7. Plato states two conditions of logos, which can be seen as an activity of judgment that predicates a rhēma to an onoma: 1) There must be a subject to be predicated; 2) there must be a praxis or being to predicate to the subject, either positively or negatively. So, we can say that onomata and rhēmata have a meaning only in as much as they are a subject and predicate in a communion of predication²⁰. This is what we called Plato’s transitive structure of language before. If we follow Ryle’s conceptualization of letters here, we can say that onomata and rhēmata can be abstracted but not extracted as they originally belong to judgments²¹.

At this point, we can refer to the idea of ​​the concept (begrifflichkeit) developed by Immanuel Kant in his Critique of Pure Reason: “We can, however, trace all actions of the understanding back to judgments, so that the understanding in general can be represented as a faculty for judging. For according to what has been said above it is a faculty for thinking. Thinking is cognition through concepts”²² (KrV, A68/B93). Kant, who defines concepts as functions of judgments, constructs them in such a way that we can only understand them in the correlations of judgment that have a synthetic structure. Thus, it is not possible to grasp concepts without functions of judgments (KrV, A68/B93; B128). Parallel to Kant, Plato also puts forth an idea of “name” that functions as a logos and is always found in logoi. Plato’s tendency to only position names in judgments is also clearly revealed in the search of the sophist which is the main issue of the dialogue: “…we seek what the sophist is and make him visible in a logos. In fact, for now, our communion is only in his name. But the function (ergon), that we call by it, is perhaps different for each of us. However, for everything, it is always necessary to agree on (synomologein) thepragma (thing, that which has been done, act) itself through logos rather than a name alone without a logos”²³ (218c-d).

At the beginning of the Sophist, Stranger, who puts forward the problem, aims to find the name of the sophist wrapped with a logos but not “without logos”. In doing this, he desires to arrive at the activity (ergon) that the name sophist carries. An ergon can only be captured in a logos since the communion of a subject and praxis only occurs in a logos. At the end of the text, Stranger achieves this goal by providing a logos that captures the activity of the sophist (Soph., 268c5-d4).

Thus, onomata and rhēmata are seen as such only when they function as a correlation and are surrounded by a logos; they are located in a web of correlation²⁴. In this regard, our subsequent separates use of them separately should be seen as an abstraction, not an extraction. This approach is opposed to classical Aristotelian logic and substance theory. According to Aristotle, there is a drastic distinction between what is said separately and what is said together, the latter being composed of the former (Categoriae, 1a15-19).

What is said separately emerged initially from the first universals (proto katholou) that are produced from the pathēmata left in the soul by the separated substances (ousiai chōristes), to which everything is predicated in nature (Metaphysica, Δ 8, 1017b25-26; APr., I 27, 43a25-43a36). Nous (reason) draws them through the process of epagōgē (induction). Then, these universals are used as elements/principles of the logos and later of the syllogisms as they are part of the epistēmē. The synthetic epistēmē level, which works with the predicate system, is based on these isolated universals, which precede the predicate system. So, these universals owe their independent individuality and origin to the entities that exist on their own, that is, to substances (APo., II 19, 100a-100b15).

In this sense, we can position Aristotelian thought in the second ontological hypothesis (pseudo-Plato): The entities as individuals that do not mix with each other. On the contrary, according to Plato, everything (composed of all kinds) is caught in communion by logos. Parallel to this idea, onomata and rhēmata are only found in logos. Instead, Aristotle defines the logos not as a predication of names, but as the sound that is meaningful on its own when divided into parts (De Interpretatione 4, Poetica 20). With this definition, which is strange at first glance, Aristotle allows the logos to be groups of names without copula, apart from the judgments bearing copula. As the Poetics XX shows, Aristotle did not limit the definition of logos only to the communion of onoma and rhēma as predication but expanded it to onoma groups, since the definitions (horos) that do not carry predicate or copula are also logoi. The definitions are then logoi that the names point to, and they are not judgments. This Aristotelian idea is the opposite of Plato’s ontology of communion of genē, in which the logoi conjoin genē. The idea assumes that the names that are not caught in the judgments still refer to objects captured as fixed extractable entities in the outside world, which has been the dominant philosophical view for centuries and the basic assumption of substance metaphysics. However, according to Plato, things gain their singularity and identity in relation to each other. Plato clearly states this structure of the logos in the following passage of the Sophist: “Visitor: Yes, my friend, for certainly trying to separate off everything from everything not only strikes the wrong note in other respects, but above all is the mark of a completely uncultivated (amousikos) and unphilosophical (aphilosophos) person. Theaetetus: Why so? Visitor: If one separates each thing off from everything, that completely and utterly obliterates any discourse (logos), since it is the interweaving of forms that gives us the possibility of talking to each other in the first place” (259d-e).

Plato states in the passage that we would be both amousikos and aphilosophos, and would have worked against harmony, if we had not had the connection of logos, which connects things in a synthetic relationship and captures them as one in this relationship. He states that he will lose the title of mousikos, which signifies the activities that make humans human, and that without logos, man will leave his humanity and will not be able to catch the harmony of to pan. Thus, he will not be able to capture individual things and the relations between them, which basically means that there will be no knowledge. Kant expresses this point in the Critique of Pure Reason as follows: “If every individual representation were entirely foreign to the other, as it were isolated and separated from it, then there would never arise anything like cognition, which is a whole of compared and connected representations” (KrV A97). By comparing this with the aforementioned passage of the Sophist, we can see that Kant admirably translated in modern language what Plato meant by amousikos and aphilosophos. In other words, both philosophers believe that things cannot be known in isolation and without the relation of logos, as in this case there would be no knowledge. This idea is entirely parallel to the communion of articulation that we defined in the letters. Just as letters can only come together in names through articulation and can be thought of as one by abstraction, names too can only emerge in judgments but can be grasped as isolated ones through abstraction.

3. Accord (homologia) of logoi

The research on the communion of letters as articulation in names, and the communion of names as predicates in logoi, can only be completed by considering how the communion of logoi takes place. To examine this communion, we will consult Plato’s very interesting concept of homologia. The concept of homologia is so important to the dialogue that the Sophist’s opening sentence begins with it (216a), and it is repeated frequently throughout the text (241e, 249e, 260a). It has several aspects.

1.- First aspect: While the logoi separate and combine ideas, they do it in accord/agreement with their nature (Crat., 387c-390e)²⁵, for which Plato often uses the concept of harmony. This also necessitates the successive logoi to be homologous/harmonic with each other. Since kinds (ta genē) preserve their identity only in harmonic communion with each other, logoi can remain harmonic to kinds that they connect only in such an agreement²⁶. In this sense, each logos finds its meaningful place only when the whole of the language is connected to in a harmonic and homologous way. Plato explains this side of logos in passage 253: “So then given that we’ve agreed that kinds too mix in such ways as these must a person not have some sort of expertise to progress in his arguments (logos) if he is going to show correctly which sorts of kinds are in harmony with which and which are not receptive to each other, and further, whether there are some that hold them together, running through them in such a way as to make them capable of mixing; and again, in cases where they divide off, whether there are others similarly running through them all that cause the division?” (253b9-c3).

The function of the logos here can be seen as closer to consciousness or mind -which makes conceptual distinctions and separates objects as we position them today- but, in a way, without subjectivity²⁷. The logos/onoma system, which enables us to capture a human as an individual, gives, on the one hand, the possibility of separating/associating the concept of human from/with other concepts, and, on the other hand, of separating this singular human from those objects and keeping it in relation to them.

To further explain the first sense of homologia, we can refer to Plato’s conceptualization of the dialectics or epistēmē of free people, that is, philosophers: “Visitor: Are we not going to claim that dividing according to kinds, and not thinking either that the same form is different or, when it is different, that it is the same, belongs to expertise in dialectic?... Visitor: The person who can (dynatos) do this is then surely well enough equipped to see when one form is spread all through many, each of them standing separately, or when many forms that are different from one another are embraced from the outside by one; or again when one is connected as one through many forms, themselves wholes, or when many forms are completely divided off and separate. This is all a matter of knowing how to determine, kind by kind, how things can or cannot combine… Visitor: And this matter of dialectic you’ll not, I think, attribute to anyone but the philosopher, with his pure and justified love of wisdom (253c5-e5)”.

In these passages, a dialectician is defined as a person having the power (dynatos) to separate those things that are separate, keep those same things together and create relations between these entities. Therefore, dialectics is defined as this power (dynamis). To shed light on this definition, we can ask the following question: How is dynamis defined in the Sophist, and in what way is it used?

In the dialogue, Plato defines “being” as having dynamis and “having dynamis” as affecting (poiein) or being affected (pathein; 247d-e). In parallel, in passage 248e, Plato defines knowing as affecting. Based on the definition of epistēmē as power and power as affecting and being affected, we will try to understand epistēmē through the concepts of affecting and being affected. If we accept that epistēmē has two aspects, one active and other passive, we must keep in mind that, contrary to this, Plato generally states in his other texts that the activity of knowing is a passive process in which the soul is affected by the eidē (Phaid., 79d; Polit., 511d, Phil., 39a). Instead, we can see that in the Sophist Plato has added an active meaning to the concept of knowing and has not limited it to a passivity. The activities of separating the existents and gathering them together in their relations in terms of logos are seen as an active activity in this text.

This is where the meaning of the logos comes into play: presenting the singularities of the whole as harmonically in relation to each other, separating the different and combining the identical. In this sense, dialectics is not a special, local or technical knowledge: it is the knowledge of the things that are the basis of all knowledge as it relates to itself. We can see it as equivalent to the epistēmē that Aristotle establishes in Metaphysics’ Gamma, which knows being as being, since this epistēmē examines not being as a specific aspect of being but as being itself. Both epistēmai aim to function as the most general and non-localized knowing, providing us with the basis of knowing. Consequently, dialectics, namely the epistēmē of the free people, makes logoi and names homologous to the harmony and segmentation of pragmata. Therefore, the meaning of the first aspect of homologia is having made logos harmonic to to pan, which is composed of ta pragmata through the power of epistēmē.

2.- The second aspect of homologia is advancing both sides in a conversation and reaching a mutual agreement (synomologia), since the basic function of the logos, the language, is the agreement of reciprocal sides. Agreement of the sides can mean two things. On the one hand, it can occur as a person talking to himself, as Plato defines thought (which is intertwined with the first aspect of the concept; Theaetetus, 189e-190a). On the other hand, it can occur as sides that are in agreement/homologous at the moment of dialogue, which also allows dialectics to be realized.

As we said, the Sophist starts from homologia and ultimately aims to agree on the logos of “sophist”. In this sense, the condition and telos of dialectical dialogue is homologia²⁸. At the end of the text, the concept of sophist wrapped in logos is placed harmonically with its ergon, and the parties agree on this. In this sense, the dialogue takes the homologia itself (216a) as the telos (241e, 249e, 260a).

This homologia is provided not by the personal positions of the parties but by the activity of the name that is surrounded by logos. To explain the second aspect of homologia, we need to refer to the Gorgias 487d-e: “And that you are indeed able to be outspoken and not to feel shame, you yourself assert, and the speech (logos) that you were making a little while ago agrees with (homologein) you. So this is how it stands now about these things: when you agree with (homologein) me on something in the speeches (logos), this will at last have been sufficiently tested by you and me, and there will be no further need to carry it back to another touchstone. For you would never have conceded it either through a lack of wisdom or through an excess of sense of shame, nor again would you concede it to deceive me; for you are a friend to me, as you yourself say. Your and my agreement, therefore, will really at last attain the goal (telos) of truth”²⁹ (487d-e).

Dialectically written dialogues work on the principle of mutual homologia/agreement since language is what provides mutual understanding. In this regard, we can talk about the political role of the logos as it is employed by the citizens of the polis. The logos attains genuine homologia in people’s external and objective agreement in mutual dialogue, rather than in one person thinking alone³⁰. In this regard, the first sense of homologia constitutes the basis of the second and final sense of agreement/homologia. This is because, in the Platonic position, homologia cannot be reached by departing from subjective positions but can only be reached by departing from an objective position that corresponds to the harmony of to pan. Therefore, attaining the homologia through logos will permit us to reach the homologia through dialogue³¹.

As a result, we can say that the third aspect of linguistic koinōnia is logos gaining its singularity and meaning in homologia/accord/consensus. This initially occurs when the logoi attain harmony with one another and with everything/to pan and ultimately occurs as the real meaning of language: mutual agreement in dialogue as synomologia.

4. Conclusion

To conclude, I would say that, according to Plato’s Sophist, the three aspects of linguistic koinōnia -the articulation of letters, the predication of names, and the homologia (accord) of logoi- share similar whole-part dynamics which have the same synthetic structure. Unlike formal languages, the Platonic logos cannot be established from its parts or from its totality but can only be discovered as a system already working and including all its whole-part relations. The three different levels of koinōnia in language can only happen thanks to each other, and none of them can precede the other, neither logically nor ontologically. Thus, unlike modern formal logic, all these levels of linguistic koinōnia necessarily require each other³².