ARTICLE (arthron)@ IN@ ARISTOTLE@ AND@ IN THE@ STOA@ AND@ ITS@ SIGNIFICANCE@ FOR@ THE@ SYSTEMATIZATION OF THE@ GREEK@ EIGHT@ PARTS@ OF@ SPEECH@@@

@@@

@@@@@@@@@@@@@@@@@@ Koji NIHEI

@@@@@@@@@@ Department of Philosophy (Ethics)

@@@@@@@@@@ Hokkaido University of Education at Kushiro (1991-1993)

 

TABLE OF CONTENTS

 

PART I. PLATO AS PREDECESSOR - The Greek parts of speech in Plato

 

1. The Platonic system of the Greek parts of speech as the foundation of their systematization@@

2. The Platonic dialectic as the philosophical basis for a system of the parts of speech

3. The dialectic logic of one = many enabling our speech

4. Categorical application of the Platonic dialectic

5. Our precept of investigation according to the Platonic dialectic

6. General condition of our speech (logos) according to Plato

7. Words (onomata) as a kind of elements in Plato; morphological, syntactical and semantical definitions of a word@@@

8. Alphabet (stoicheion) and syllables (syllabai)

9. A virtual list of all the Greek words@@@@

10. Noun (onoma) and Verb (rhema)

11. Sentence (logos) and three kinds of words

12. From Plato to Aristotle

13. A diagram illustrating Plato's system of the Greek three parts of speech tending to Aristotle's

 

PART II. ARISTOTLE AS THEORETICAL DEVELOPER

@@@@@@@@@@@@@@

14. The tripartite system of parts of speech

15. Hebraic and Arabic tripatite system of parts of speech

16. A grammatically heterogenous system of parts of speech in Aristotle and its significance

17. A great deal of intellectual efforts by a series of classical students to solve the Aristotelian riddle: arthron

18. ARTHRON (ARTICLE) in its first definition in Aristotle is an INDICATOR of a Sentence-branch: PREPOSITION

19. Comparison of Aristotelian system of parts of speech with that of Dionysius of Thrax

20. The second definition of ARTHRON (ARTICLE) as regards Adverbs in Aristotle

21. Aristotle as theoretical developer in Syntax of sentence

 

Part III. Theodectes, Theophrastus and the Stoics

 

@@ @22. Semi-predecessor and Collaborator of Aristotle: Theodectes

@@ @23. Immediate successor to Aristotle: Theophrastus and his school showing the existence of arthron bequeathed to them by Aristotle

@@ @24. Mediate successor to Aristotle: the Stoa and the problem of@ Adverbs left

@@ @25. The Stoic stage I consisting in 4 parts of speech

@@ @26. arthron in the Stoa

@@ @27. The second and the third stage in the Stoa and the Stoic conception of Adverb: mesotes or pandectes

@@ @28. Development of the Greek word class system according to R. H. Robins (1966): Diagram A

@@ @29. Development of the Greek word class system according to our new researches and@ Robins' diagram corrected: Diagram B

 

Appendix: Aristotle's phonology reexamined

 

PART I. PLATO AS PREDECESSOR - The Greek parts of speech in Plato

1. The Platonic system of the Greek parts of speech as the foundation of their systematization

The Greek parts of speech in the sense of the Greek word classes imply in principle a classification of all the Greek@@ words. And it is evident that this classification presupposes the identification of the words of Greek as materials to be classified and at the same time some discrimination of their linguistic qualities introducing necessary classes. Now, this kind of intellectual effort about Greek we find in Plato’s philosophical works such as Cratylus [421d-426b, 431a-c], Theaetetus [201d-208b] and Sophist [251a-264b], etc. and it is probably the first authentic beginning of the systematization of the Greek parts of speech as word classes, because it is Plato himself in these works that has made clear the possibility of the existence and our knowledge of such a system in Greek (*1). Then, we will follow his fundamental doctrines concerning the Greek word classes in order to identify the post-Platonic development of the Greek parts of @speech.

 

2. The Platonic dialectic as the philosophical basis for a system of the parts of speech

A system of the parts of speech is evidently a logical system of classification, and the Platonic dialectic as a philosophical methodology gives us the ultimate principle of classification, I believe; so every system of the parts of speech is inevitably an application of this principle.

 

3. The dialectic logic of one = many enabling our speech@

Any effort of classifying the parts of our speech can be made only when the speech is in our power, and it is in our power only when it is possible “for the many to be one and for the one to be many (ta polla hen kai to hen polla einai)“ [Sophist 251b](*2) within the range of our speech, because “ we come to be constantly predicating many names of one and the same thing ”. [id.251a]

 

Theaetetus. Give an example.

Stranger. I mean that we speak of man, for example, under many names -- that we attribute to him colours and forms and magnitudes and virtues and vices, in all of which instances and in ten thousand others we not only speak of him as a man,@@@@ but also as good, and having numberless other attributes; and in the same way anything else which we originally @supposed to be one is described by us as many, and under many names.

Theaet. That is true.

Str. And thus we provide a rich feast for tyros, whether young or old; for there is nothing easier than to argue that the @one cannot be many, nor the many one; and great is their delight in forbidding us to say that a man is good; for man, they insist, is man and good is good. I dare say that you have met with persons who take an interest in such matters -- they are often elderly men, whose meagre sense is thrown into amazement by these discoveries of theirs, which they believe to be the height of wisdom.

Theaet. Certainly, I have.

Str. Then, not to exclude anyone who has ever speculated at all upon the nature of being, let us put our questions to them as well as to our former friends.

Theaet. What questions ?

Str. Shall we refuse to attribute being to motion and rest, or anything to anything, and assume that since they do not mingle, and are incapable of participating in one another, we must represent them accordingly in our discourse ? Or shall we gather all into one class of things communicable with one another ?@ Or are some things communicable and others not ? -- Which of these alternatives, Theaetetus, will they prefer ?

Theaet. I have nothing to answer on their behalf.

Str. Suppose that you take all these hypotheses in turn, and see what are the consequences which follow from each of them ?@

Theaet. Very good.

Str. First let us assume them to say that nothing is capable of participating in anything else in any respect; in that case rest and motion cannot participate in being at all.

Theaet. They cannot.

Str. But would either of them be if not participating in being @?

Theaet. No.

Str. Then by this admission everything is instantly overturned, as well the doctrine of universal motion as of universal @rest, and also the doctrine of those who distribute being into immutable and everlasting kinds; for all these add on a notion of being, some affirming that things “are” truly in motion, and others that they “are” truly at rest.

Theaet. Just so.

Str. Again, those who would at one time compound, and at another resolve all things, whether making them into one and out of one creating infinity, or dividing them into finite elements, and forming compounds out of these; whether they suppose the processes of@ creation to be successive or continuous, would be talking nonsense in all this if there were no @admixture.

Theaet. True.

Str. Most ridiculous of all will the men themselves be who want to carry out the argument and yet forbid us to call anything, because participating in some affection from another, by the name of that other.

Theaet. Why so ?

Str. Why, because they are compelled to use the words “to be”, “apart”, “from others”, “in itself”, and ten thousand more, which they cannot give up, but must make the connecting links of discourse; and therefore they do not require to be refuted by others, but their enemy, as the saying is, inhabits the same house with them; they are always carring about with them an adversary, like the wonderful ventriloquist, Eurycles, who out of their own bellies audibly contradicts them.

Theaet. Precisely so; a very true and exact illustration.

Str. And now, if we suppose that all things have the power of communion with one another -- what will follow ?

Theaet. Even I can solve that riddle.

Str. How ?

Theaet. Why, because motion itself would be at rest, and rest again in motion, if they could be attributed to one another.

Str. But this is utterly impossible.

Theaet. Of course.

Str. Then only the third hypothesis remains.

Theaet. True.

Str. For, surely, either all things have communion with all; or nothing with any other thing; or things communicate with some things and others not.

Theaet. Certainly.

Str. And two out of these three suppositions have been found to be impossible.

Theaet. Yes.

Str. Everyone then, who desires to answer truly, will adopt the third and remaining hypothesis of the communion, of some with some.

Theaet. Quite true.

Str. This communion of some with some may by illustrated by the case of letters; for some letters do not fit each other, while others do.

Theaet. Of course.

Str. And the vowels, especially, are a sort of bond which pervades all the letters, so that without a vowel one consonant cannot be joined to another.

Theaet. True.

Str. But does everyone know what letters will unite with what ? Or is art required to make a man a reliable judge of this ?

Theaet. Art is required.

Str. What art ?

Theaet. The art of grammar.

Str. And is not this also true of sounds high and low ? -- Is not he who has the art to know what sounds mingle, a musician, and he who is ignorant, not a musician ?

Theaet. Yes.

Str. And we shall find this to be generally true of art or the absence of art.

Theaet. Of course.

Str. And as classes are admitted by us in like manner to be some of them capable and others incapable of intermixture, must not he who would rightly show which kinds will unite, and which of them repel each other, proceed by science in the path of argument ? By science, too, he must know whether there are some allpervading connecting terms, which enable the other kinds to blend; and, conversely, in divisions, whether there are not others which cause whole classes to become separate ?

Theaet. To be sure he will require science, and if I am not mistaken, the very greatest of all sciences.

Str. How are we to call it ? By Zeus, have we not lighted unwittingly upon our free and noble science, and in looking for the Sophist have we not entertained the Philosopher unawares ?

Theaet. What do you mean ?

Str. Should we not say that the division according to classes, which neither makes the same other, nor makes other the same, is the business of the dialectical science ?

Theaet. That is what we should say.

Str. Then, surely, he who can divide rightly is able to see clearly one form pervading a scatterd multitude, and many different forms contained under one higher form; and again, one form knit together into a single whole and pervading many such wholes; and many forms existing only in separation and isolation. This is the knowledge of classes which @determines where they can have communion with one another and where not.

Theaet. Quite true.

Str. And the art of dialectic would be attributed by you only to the Philosopher pure and true ?

Theaet. Who but he can be worthy ?

Str. In this region, then, we shall discover the Philosopher, either now, or at any later time, if we look for him; like the Sophist, he is not easily discoverd, but for a different reason.

Theaet. It seems to be so.

Str. And the Philosopher, always holding converse through reason with the idea of being, is also dark from excess of light; for the souls of the many have no eye which can endure the vision of the divine.

Theaet. Yes; that seems to be quite as true as the other.

Str. Well, the Philosopher may hereafter be more fully considered by us, if we are disposed; but the Sophist must clearly not be allowed to escape until we have had a good look at him.

Theaet. Very good.

Str. Since, then, we are agreed that some classes have a communion with one another, and others not, and some have communion with a few and others with many, and that there is no reason why some should not have universal communion with all, let us now pursue the inquiry, as the argument suggests, not in relation to all Ideas, lest the @multitude of them should confuse us, but let us select a few of those which are reckoned to be the principal ones, and consider their several natures and their capacity of communion with one another --" [id.251a-254c tr.by B.Jowett](*3)@

 

4. Categorical application of the Platonic dialectic

This logic of dialectic is first applied to the five most universal Categories, i.e. Being, Motion, Rest, the Same, and the Other. And that naive refutation that it is impossible for the many to be one and for the one to be many is completely rejected by answering to it that it is possible for the many to be one in a sense and the one to be many in another, because this way of thinking and speaking is just the utmost ground for the possibility of our thought and speech. For instance, ”we must admit that motion is the same and is not the same, and we must not be disturbed thereby; for when we say it is the same and not the same, we do not use the words alike. When we call it the same, we do so because it partakes of the same in relation to itself, and when we call it not the same, we do so on account of the participation in the other, by which it is separated from the same and becomes not that but other, so that it is correctly spoken in turn as not the same”.@ [Sophist 256a-b tr. by H.N.Fowler](*4)

 

Thus, in general, "the classes mingle with one another, and being and the other permeate all things, including each other, and the other, since it participates in being, is, by reason of this participation, yet is not that in which it participates, but other, and since it is other than being, must inevitably be not-being. But being, in turn, participates in the other and is therefore other than the rest of@ the classes, and since it is other than all of them, it is not each one of them or all the rest, but only itself; there is therefore no doubt that there are thousands and thousands of things which being is not, and just so all other things, both individually and collectively, in many relations are, and in many are not.” [id.259a-b](*5)

 

5. Our precept of investigation according to the Platonic dialectic

Now we have our own process of investigation where is “ something both difficult and beautiful” [Sophist 259c]. For, any system in the world is not a system of universal mixture nor a system of no mixture, but a system of particular mixtures, so, ”when a man says that the same is in a manner other, or that other is the same, we should be able to follow his arguments, criticizing one by one of them from his own point of view and in the same respect in which he modifies either of@ these proposition. But it is no real refutation to show that somehow or other the same is other or the other same, or the great small, or the like unlike; and to delight in always bringing forward such objections in our arguments, but it is clearly the new-born babe of someone who is only beginning to approach the beings“.@ [Sophist 259d-e]

 

6. General condition of our speech (logos) according to Plato

First of all, a system of no mixture is not an appropriate framework of investigation within which our speech is treated,@ for “the complete separation of each thing from all is the utterly final obliteration of all discourse, because our power of @discourse is derived from the interweaving of the classes or ideas with one another”. Hence, “the attempt to separate everything from everything else is not only out of tune but also belongs to such a person as is entirely uncultivated and unphilosophical”, [id.259d-e] because our intellectual cultivation depends mostly upon our verbal exercises. And now we will take this problem into closer consideration in this same direction.

 

7. Words (onomata) as a kind of elements in Plato; morphological, syntactical and semantical definitions of a word

How do we have the idea of a word as unit (tango in Japanese) ? It seems to be easy, but very and very difficult indeed to answer the question. And Plato gives us the first in its history and at the same time@ most@ perfect and penetrating answer to it. He writes;@@@@@@@@@@@@@@@@@@@ @@@

Socrates. But let us bear in mind that if a person asks about the words by means of which names are formed, and again about those by means of which those words were formed, and keeps on doing this indefinitely, he who answers his questions will at last give up; will he not ?

Hermogenes. Yes, I think so.

Soc. Now at what point will he be right in giving up and stopping ? Will it not be when he reaches the names which are the@ elements (stoicheia) of the other names and words ? For these, if they are the elements, can no longer rightly appear to be composed of other names.” [Cratylus 421d-422a tr. by H.N. Fowler](*6)

 

And as to a word, Plato says in another context that a name (onoma) is the smallest part of the sentence (morion smikrotaton logou). [cf. Crat.385c7-9]

 

And in addition to this process of decomposition with sufficient mutual comparison of words qualifying the so-called morphological characters of a word, we are given by Plato another definition of a word in the process of@ propositional utterance, i.e. a syntactical one; he says that if names are the smallest parts of a speech and if a speech is able to be uttered truly or falsely and if then its parts, smaller or larger, are to be uttered also truly or falsely, so the names are also able to be utterd truly or falsely [cf. Crat.385b2-d1]. (And as to being true or false Plato says that the speech which says @things as they are is true, and that which says them as they are not is false. [id.385b7-8 tr. by H.N.Fowler](*7))

 

And this subtle argument given by Plato is just nothing but a syntactical definition of a word, being supported of course by the morphological one in their reciprocal interdependence. And closely along with these morphological and syntactical definitions of a word we are also given an essential and most wonderful definition of it by Plato;

 

Soc. And can you say something of the same kind@ about a name ? The name being an instrument, what do we do with it when we name ?

Her. I cannot tell.

Soc. Do we not teach one another something, and separate things according to their natures ?

Her. Certainly.

Soc. A name is, then, an instrument of teaching and of separating reality, as a shuttle is an instrument of separating the web ?

Her. Yes.

Soc. But the shuttle is an instrument of weaving ?

Her. Of course.

Soc. The weaver, then, will use the shuttle well, and well means like a weaver; and a teacher will use a name well, and well means like a teacher.

Her. Yes.” [Crat. 388b6-c8 tr. by H.N.Fowler](*8)

 

Now, this third definition of a word is in fact the first and most principal one, because it depends upon and at the same time originates from the two most universal Categories: Being and Other. (cf. Sophist 259a-b) And this definition is to be said to be semantic, in itself sufficient and therefore supporting with its own self-sufficiency of meaning the other two @definitions, morphological and syntactical, these being said to be developping the meaningful inwardness of the first to the outer and vaster context. And this semantic definition is to be related to Plato’s onto-epistemological doctrine that: ”every existing object has three things which are the necessary means by which knowledge (episteme) of that object is acquired; and the knowledge itself is a fourth thing; and as a fifth one must postulate the object itself which is cognizable and true. First of these comes the name (onoma); secondly the definition (logos); thirdly the image (eidolon); fourthly the knowledge.” [Epistle VII 342a7-b3 tr. by Bury, R.G.] (*9)

 

Thus we obtain elementary words and complex words composed of elementary words. And these are the words in general. And as to a word, Plato says as seen above that a name (onoma) is the smallest part of the Sentence (morion @smikrotaton logou) [cf. Crat. 385c7-9]. And this definition will be valid for both the elementary and the complex words, because even a complex word will not be other than a word.

 

8. Alphabet (stoicheion) and syllables (syllabai)

But, those elements out of which even an elementary word could@ be composed are no more words but alphabetical elements or alphabet (stoicheion) and alphabetical groups or syllables (syllabai) [cf. Crat.424e-425a].

 

9. A virtual list of all the Greek words

Thus, if we use these Platonic definitions, semantical, syntactical and morphological, of a word and a technique of alphabetical permutation besides to identify the Greek words among the conglomeration of the Greek language, we will be able to obtain a complete list of the Greek words which will be a virtual lexicon of Greek in an alphabetical order. For, at first we are not able to indentify every word and such verbal continuation as "He walks." or "He lies." may be a word on our first hypothesis, and under our investigation using the Platonic definitions of a word and trying to analyse the complex into the elements, the words "he", "walks" and "lies" have a chance to be identified for the first time as such words.

 

10. Noun (onoma) and verb (rhema)

And, then, our task is raised to a higher level based on the ground of words, i.e. the level of syntax. And for the first time on this level we encounter a problem of assorting the words. And Plato's arguement of assorting the words is essentially founded on that of the shortest and most fundamental sentence composed of a noun and a verb, this composition being approached "as before we were speaking of forms and letters; for that is the direction in which the answer may be expected.

Theaet. And what is the question at issue about names ?

Str. The question at issue is whether all names may be connected with one another, or none, or only some of them.

Theaet. Clearly the last is true.

Str. I understand you to say that words which have a meaning when in sequence may be connected, but that words which@ have no meaning when in sequence cannot be connected ?

Theaet. What are you saying ?

Str. What I thought that you intended when you gave your assent; for there are two sorts of intimation of being which are given by the voice.

Theaet. What are they?

Str. One of them is called nouns, and the other verbs.

Theaet. Describe them.

Str. That which denotes action we call a verb.

Theaet. True.

Str. And the other, which is an articulate mark set on those who do the actions, we call a noun.

Theaet. Quite true.

Str. Now a succession of nouns only can never form a sentence; neither can a succession of verbs without nouns.

Theaet. I do not understand you.

Str. I see that when you gave your assent you had something else in your mind. But what I intended to say was, that a mere succession of nouns or of verbs is not discourse.

Theaet. What do you mean ?

Str. I mean that words like "walks", "runs", "sleeps" or any other words which denote action, however many of them @you string together, do not make discourse.

Theaet. How can they ?

Str. Or, again, when you say "lion", "stag", "horse", or any other words which denote agents... neither in this way of stringing words together do you attain to discourse; for the sounds convey no expression of action or inaction, or of the being of anything which is or is not, until verbs are mingled with nouns; then the words fit, and the smallest combination of them forms a sentence, and is the simplest and least form of discourse.

Theaet. Again I ask, What do you mean ?

Str. When anyone says "A man learns", would you not call this the simplest and least of sentences ?

Theaet. Yes.

Str. Yes, for he now arrives at the point of giving an intimation about something which is, or is becoming, or has become, or will be. And he not only names, but he achieves something, by connecting verbs with nouns; and therefore we say that he discourses, and to this connexion of words we give the name of discourse [or sentence].

Theaet. True.

Str. In conclusion, then, just as there appeared to be some things which fit one another, and things which do not fit, so there are some vocal signs which do, and others which do not, combine and form discourse.

Theaet. Quite true." [Sophist 261d-262 tr. by Jowett, B.]

 

11. Sentence (logos) and three kinds of words

Now, since Plato says that there are two sorts of voice [id. 261e4-6] and that one of@ them is called nouns, and the other verbs (to men onomata, to de rhemata klethen) [id. 262a1], he seems to admit of no other kind of words besides them. But this assorting is not necessarily to be viewed as comprehensive, for it is effectually proposed only from the standpoint of the first and shortest form of discourse (ton logon ho protos te kai smikrotatos) [id. 262c6-7], or a least and first sentence (logos elachistos te kai protos) [id. 262c9-10]. And, even if it is possible to admit of a third kind of words besides nouns and verbs, it is not a kind of any specific quality, but an open one, under which all the other except @the names called nouns or verbs are called. And it is indeed the third kind of words in Plato, insofar as he treats in fact some Greek words which he will recognize to be other than nouns and verbs as genuine "names" (onomata ) in describing their etymologies so as to reach their truth in Cratylus. For instance, "endon" (within [Adverb]) and "entos" (within, inside [Preposition and Adverb]) are said to be so named (onomasen) because the sound "n" is observed to be sounded from within, and therefore to have a notion of inwardness [427c1-3]. Then, Plato seems to admit of the type of words that are not nouns nor verbs, which is probable to be the third sort of words. And Plato says also in one of his epistles that "every existing object has three things which are the necessary means by which knowledge of that object is acquired; and the knowledge itself is a fourth thing; and as a fifth one must postulate the object itself which is cognizable and true. First of these comes the name; secondly the definition; thirdly the image; fourthly the knowledge. If you wish, then, to understand what I am now saying, take a single example and learn from it what applies to all. There is an object called a circle, which has for its name the word we have just mentioned; and, secondly, it has a definition, composed of names and verbs; for "that which is everywhere equidistant from the extremities to the centre "(to ek ton eschaton epi to meson ison apechon pantei) will be the definition of that object which has for its name "round" and "spherical" and "circle." (Epistle VII 342a7-c1) Thus, in the sentence composed of nouns and verbs we can recognize the words which today we call the definite article (e.g. to), or preposition (e.g. ek and epi), or adverb (e.g. ison and pantei), and moreover those which function as verbs (rhemata) (e.g. apechon) may not necessarily be verbs in the strict sense but can be nouns in the sense of adjective of today (these are allowed to be said to be of predicative use.) Thus, we have three sorts of words; the nouns, the verbs and the words other than these two according to Plato, and this last has no specific naming other than "name" (onoma). And the existence of the third sort of words will show that there are more sentences than the least and first @sentences that are composed out of these minimal sentences and of words which may be nouns, verbs, or of the third kind.

 

12. From Plato to Aristotle

Thus, we will be allowed to conclude that with Plato the Greek grammatical sciences have begun and that his main achievements are the philosophical groundwork and the first theoretical systematization of them. And it is perhaps the problem of compound sentences that Plato bequeathed to his successors in Greek grammar, and Aristotle, one and the@ most brilliant of them, is faced with it to give new prospects for the newborn science, Greek grammar.

 

13. A diagram illustrating Plato's system of the Greek three parts of speech tending to Aristotle's

 

Plato

 

Aristotle

words

sentences

words

Noun (onoma)

a noun + a verb =

a simplest sentence

Noun (onoma)

Verb (rhema)

Verb (rhema)

the third kind of words [Conj., Prep., Adv., etc.]

 

compound sentences

Conjunction (syndesmos)

Preposition

 

(arthron)

Adverb

 

@

PART II. ARISTOTLE AS THEORETICAL DEVELOPER

14. The tripartite system of parts of speech

Plato's system of Greek parts of speech is a tripartite one, as we have seen it in PART I, composed of nouns (onomata), verbs (rhemata) and words of the third kind other than these two. And from the viewpoint of Greek system of eight parts of speech described in Art of Grammar by Dionysius of Thrax (*10), such a tripartite system seems to be too defective to analyze Greek. And in fact, a great deal of intellectual efforts had to be made by Greek philosophers and grammarians after Plato, beginning with his system, in order to reach a practically appropriate level of developed systematization such as in Art of Grammar by D.T. And we too intend to follow, in outline but with rigour, the historical course of that development. But, nevertheless, such a tripartite system is in itself sufficient to describe such a language as has Subject and Predicate to construct its basic form of expressions, other components of it being@ qualifiers and modifiers in addition to either of them. Therefore, even after the Greek system of eight parts of speech was realized by Aristarchus and standardized by his pupil Dionysius of Thrax, such a learned man as Augustine could assert that of the eight parts of speech set forth by grammarians Aristotle himself had taught us to admit only of noun and verb as genuine parts of speech capable of indicating and signifying somethig vocally, others being to be correctly called auxiliary compages of speech (compagines orationis)(*11).

 

15. Hebraic and Arabic tripartite system of parts of speech@

And we have other examples of the tripartite system of parts of speech, namely in the intrinsic analyses by Hebraic and Arabic grammarians of their own mother tongue into a word system having three word classes: noun (inflectional), verb (inflectional), and particles (not inflectional). Then, according to Johann Reuchlin (1455-1522), one of great German classicists and Humanists, the Hebraic three word classes correspond to the@ Latin eight parts of speech as follows;

 

Hebraic word class

Latin part of speech

 

noun

noun@

pronoun

participle

verb

verb

 

particles

adverb

conjunction

preposition

interjection

@@@ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@(De rudimentis Hebraicis,1506)

 

And the oldest academic Arabic grammar (Kitab <The Book>) by Sibawaihi (8c.A.D., Persian) admits of as many as three parts of speech: noun (inflectional), verb (inflectional), and particles (not inflectional), independently of the influences from the Greek or the Latin grammar. (*12)@@@@

 

16. A grammatically heterogeneous system of parts of speech in @Aristotle and its significance

Now, Aristotelian system of parts of speech seems to be heterogeneous and inconsistent from the viewpoint of a usual parallelism of all the parts of speech. But, Aristotle himself intends to develop willingly such a heterogenous system according to his own principal distinction between what signifies (semainei) something by itself and what is non-significant (asemos) among@ the diction in general.(*13) And by way of further differentiation of each of the two genre, we get the Aristotelian eight parts of speech that is able in principle to cover all the diction of any kind in Greek (tes lexeos hapases ta mere) (Poetics, chap.20,1456b20-1457a30);

 

Signifing

Non-significant

Sentence (logos)@@

Noun (onoma)@@@@@@@@

Verb (rhema)@

Case (ptosis)

Conjunction (syndesmos)@@@@@

Preposition (arthron)

Adverb (arthron)@@@@@@@@@@

Syllables (syllabe)@@@@@@@@@@@@

Alphabet (stoicheion)

 

To begin with, Alphabets and Syllables are sometimes used as representative signs and symbols.(*14) Therefore, they are not @only analytic elements of a word as might be expected from our modern standpoint, but also a word or words in themselves, and in this sense they are of onomata as well as Pronouns, however, their usual representative capacity does not reach the standard level of independent meaning which is attained, as Aristotle believes, by Names (onomata) in general. According to Aristotle, indeed, a Noun and a Verb are such a qualified Name. And a Sentence is what contains one or more Names and itself signifies something as a unified whole. Then, a Case (ptosis): an oblique case of a Noun belongs to the genre : Non-significant, because when "is", "was" or "will be" is added, it does not then form a proposition, which either is true or is false, as the noun itself always does then (De Inter.16b). And both a Conjunction and a Preposition belong to the same genre as a Case on the same ground. So, if we attach only to Names defined by Aristotle and at the same time if we take Names to be words, all the words shoud be either a Noun or a Verb, otherwise there @would be the third kind of words other than Nouns and Verbs, which is the option made by Aristotle and is what is called @"Non-significant (asemos)" by him. Therefore, for instance, the Aristarchan and Dionysian system of eight parts of speech will be able to correspond to the Aristotelian system of eight parts of speech in a somehow sophisticated way, because

@ 1) each is proposed to cover all the Greek words, and

@ 2) both are not identical with each other as to their inner configuration.

 

Now, before we make a comparison between them, we must confirm our interpretation of arthron in its first definition in Aristotle as Preposition. For this grammatical term to interpret decisively has been indeed a great perplexity in the@ history of @Aristotle scholarship for more than two thousand years.

 

17.@ A great deal of intellectual efforts by a series of classical students to solve the Aristotelian riddle:arthron

In fact, arthron (article) in Aristotle has been a riddle to the classical students these two thousand and three hundred years. Soon after it was introduced in Poetics by Aristotle, the Stoic philosophers began to use it to signify the definite article & relative pronoun, besides pronouns in general afterwards.(*15) But the definition of arthron given by Aristotle himself does not seem to apply to any of those parts of speech; "Arthron deloi logou archen e telos e diorismon." (1457a6-7) (An article shows a beginning or an end or a division of sentence.)(*16)

 

Moreover, the examples of arthron given here by Aristotle are prepositions such as around (amphi), about (peri), etc. (kai ta alla), which@ seem to be in no harmony with the definition. And in addition to this, there is given the second definition of arthron, and besides there are given two definitions of syndesmos (Conjunction), the first of which seems also to be problematic, whereas the second is regular and identified in other treatises of Aristotle. But these three or four definitions seem to be confused with one another because of their mutual resemblance to shut off from our clear understanding;

 

"Conjunction (syndesmos) is, now, in one case, a non-significant sound which neither obstructs nor produces the unity of a significant sound out of several sounds and which is proper to be together-put in the middle of each of the sounds and which it is not fitting to put at the beginning of a sentence by itself. Such are men, etoi and de. Such is -men-de-,or -etoi-e-, or -e-etoi-, or -e-e-, or the like". (by the author) (DEFINITION-1)

 

"And in another case, it is a non-significant sound which is apt to produce one significant sound out of several sounds, more than one of which being significant by itself." (DEFINITION-2)

 

"And, then, Article (arthron) is a non-significant sound which shows a beginning or an end or a division of sentence; e.g. amphi, peri and so on." (DEFINITION-3a)

 

"Or it is a non-significant sound which neither obstructs nor produces the unity of a significant sound out of several sounds and which is proper to be put on both sides and in the middle of the sounds." (DEFINITION-4) (1456b38-1457a10)(*17)

 

Thence, the history of the intellectual efforts to solve the problem is filled with the grand astonishment of the human reason in the presence of a grammatical mystery, the disbelief in the extant texts, shrewd devices of correcting the texts to get to a reasonable reading, and the like. For instance, Steinthal says;(*18)

 

"The@ paragraph@ which contains the definitions of@ syndesmos and arthron is so grievously corrupted that I can admit of no plausible conjecture as to its reading, which is not necessarily an extraordinary thing, because the corruption is indeed so grievous. And I cannot realy understand how it is possible to put so hastily off the authority of Dionysius of Halicarnassus, who maintains twice (de comp. verbb. c.2 in. and de Demosth. praest. p.1101. ed. Reiske) that Aristotle proposed only three parts of speech: onomata (Nouns), rhemata (Verbs) and syndesmoi (Conjunctions). It is not only astonishing, but also unbelievable that such a man as Dionysius did not know or did neglect that place of Poetics (Classen, de primord. gr. gr. p.60). And It seems quite questionable to me that Quintilian copied off Dionysius when he referred to three parts of speech of Aristotle (I,c.4). And when the later Roman grammarians (Lersch, Sprachphilos. der Alten II, p.11) say that Aristotle set forth only two parts of speech, they explain further that these are the two essential parts of speech and that the other parts of@ speech are appendixes (appendices), or (as it is said in Augustine's catt. decem 1.) compages (compagines); then, here too, just three parts of speech are ascribed to Aristotle. Therefore, arthron is suspicious. Moreover, arthron is between rhema and ptosis when it is enumerated among mere lexeos (parts of diction) in the beginning of the chapter 20, but it is between syndesmos and onoma when defined. And further, arthron appears only in pseudo-Aristotelian Rhetoric to Alexander, c.26, in the meantime it never does in his great Rhetoric, nor in his Organ, although the use of the definite article is relevant there in his Prior Analytics, I,c.40. And how can we take an appropriate one for arthron of the given definitions ? And if we admit that the definition here given of arthron is not to be explained (just as Lersch in p.270 finds himself to do so), we have no right to ascribe arthron to Aristotle against Dionysius on the ground of Poetics. And furthermore, we can see in his Rhetoric III,5, how Pronoun, Conjunction and Article get together for Aristotle, when he gives as mutually corresponding and demanding syndesmoi: ho men-ho de (the one - the other) and ego men-ho de (me on the one hand, he on the other), whereby it is not the case that only men (one) and de (other) are meant, but just as sy-sy (you-you) is referred to in the same relation (Rhetoric to Alexander, c.26).

@And further, to get aware of the cumulative wonder of the definitions, let us see what definition is yet possible ,if all of the words that are not Noun, nor Verb, nor Adjective nor Adverb thence derived, nor Numerals shoud be syndesmoi ! Then, therefore, Pronouns, all kinds of abstract Adverbs, Prepositions and Conjunctions ought to be one thing! But, to assume with Ritter that this chapter is not at all Aristotelian and that it was interpolated by a later grammarian is impossible because even the worst grammarian could have done his work better than this. Just because of their @wonder those definitions have not been undersood at all by grammarians, I think, and so they have been inclined to be altered. He who was surprised not to find arthron in Aristotle might have interpolated it by assigning one of the definitions of syndesmos to it. Perhaps it was originally called syndesmos e arthron (Conjunction or Article), and then these were seperated.

@The second definition is yet the best and in harmony with the name: syndesmos and with the expression: ho gar syndesmos hen poiei ta polla (For a Conjunction makes many things one. Rhet., III,12). The other definition: "a non-significant sound which neither obstructs nor produces the unity of a significant sound out of several sounds" may be related to the so-called expletive particles such as ge (at any rate), and de (now); cf. Probl. XIX, p.919a22B."

 

And Vahlen (*19), after he eliminated from the text the second definition of arthron, being an incorrect repetition in vain of the preceding phrases, and only one of them being transferred to the first definition of syndesmos so as to be read: "Syndesmos is a non-significant sound which neither obstructs nor produces the unity of a significant sound properly composed of several sounds and which is proper to be put on both sides and in the middle of the sounds and which it is not fitting to put at the beginning of a sentence by itself; e.g. men, etoi and de.", proposes that this definition together with the second signifies such conjunctions as neither obstructs nor produces the unity of a sentence which is able to have its unity without a conjunction; e.g. de, nu, proteron. And that these are the so-called expletive conjunctions (parapleromatikoi syndesmoi), but as Demetrius showed in his On Style, c.55-58, they can be used efficaciously and should not be used in vain and pointlessly.

 

And Vahlen admits the second definition to be a sharper expression of the nature of syndesmos as conjunction by joining together more than one significant words, namely onomata or rhemata, into one sentence, though the concept of a sentence is not completely congruent with ours. Then, the bisection of conjunctions in Aristotle is, according to Vahlen, nearly the same as that in Problems 919a22-26; "If certain conjunctions, such as te and kai, are omitted, the language ceases to be Greek, but the omission of others gives no such offence, because there are some conjunctions which one must use often, if there is to be sense, but with others it is not so."

@

And Vahlen says about arthron that only its first definition is to be taken into account, the second being eliminated as seen above, but that in the definition no one will manage to find out any meaning which shall contain at least a reasonable, not to say a right indication of the arthron in the sense of@ the definite article.

 

In the meantime, he says that it will be well understood, if we take such conjunctions as serve to bind parts of a @periodic sentence to the greater members of it, and therewith to show, as the definition runs, a beginning, an end or a division of sentences in its largest sense: e.g. ei (if), epei (since), gar (for); hoste (that), ara (then), oun (now); hina (in order that), hos (so as to), hoti (that), ara-e (whether-or); this assumption is very near G. Hermann's.

 

Furthermore Vahlen writes (*19);

« The term arthron in the sense of the definite article which came into common use afterwards appears already in Rhetoric to Alexander, which is not by Aristotle perhaps, but originated in his times: chap.26(1435a34-36), «Pay attention also to the so-called definite articles in order that they are used in the necessary place», which is then (1435b11-15) explained more in detail as follows; «It is of importance to pay attention to the definite articles so as to be used in the necessary place; this is to be seen in these ways: The man here (houtos ho anthropos) does an injustice to the man there (touton ton anthropon). Now, the addition of the definite articles make the sentence evident, on the contrary their removal will make it obscure. And sometimes it will take place the reverse.»@@@@@

 

« But it would be doubtful to win the arthron in the sense of the definite article for Aristotle on the grounds of these phrases. For, Aristotle uses the word arthron in its proper and original sense of joint or articulation (Hist.anim. 536a4@ and b11); but it is never known to me [Vahlen] that Aristotle had used the word anywhere else also in the grammatical sense. However the definite article as such and its functions could be never hidden from him, since he shows without using a @grammatical term to indicate the definite article the difference of the meaning according to the addition or the omission of it by citing examples; cf. e.g. Soph.Elench.166a3 and Analyt. Prior. 49 b10.

 

« And where, one will ask, did he classify it ? @I [Vahlen] think that he did not dissociate it from the onoma (word), whose constant attendants the arthra are, (like the handle bound to a cup or the crest to a helmet as Plutarch says in Quaest. Platon. X 3, 1010d.e), and with which they could been accounted to be one, like the negation to a verb or to a noun, or the adverb to a verb. There is a deficiency of abstraction to be perceived therein, but it is only too natural and too reasonable in these opening times of the investigation of the philosophy of language. It is certain through trustworthy evidences that in Aristotle the onoma contains also the pronouns and above all the demonstrative ones within, and that he never did think otherwise the pronominal use of the definite article, when he, e.g. in@ Rhetoric, III,5,1407b9, cited «She (he), having come and having conversed with me, went away.» as an example of the agreement of the gender, and «They (hoi), having come, began to beat me.» as an example of the agreement of the number. Therefore it is little probable that Aristotle had dissociated the definite article from the pronouns and nouns and set it up as a specific part of speech, and it is less probable that he had had the definite article, which is so tightly bound to a noun in gender and number, etc. and sometimes takes the place even of it in the pronominal use, called non-significant sound, this terminology being proper, according to Aristotle's mode of vision, only to that class of words@ we [Vahlen]@ group under@ the@ name@ of@ particles, in opposition to onoma and rhema. (Cf. Diagram B)

 

« Thence we [Vahlen] get a further evidence that the definition of arthron in Poetics cannot imply the definite article, although it seems to do because of the term: arthron itself, yet it cannot on account of the terminology: " non-significant sound ", and if the definite article was, as we sought to make it probable, not yet dissociated from onoma as a specific part of speech in Aristotle's conception, then this will be an indirect basis of our assumption that the definitions of arthron concern the syndesmoi (Conjunctions) and protheseis (Prepositions). (Cf. Diagram B)

 

« Now, Ammonius in his commentary on De Interpretatione p.99a12 Brand. referred explicitly to the parts of diction (mere lexeos) mentioned in the chapter 20 of Poetics, and in a41 just there he takes the arthron for the definite article and then defines it as that which is tightly woven into the nouns and a relative to them (kai ten anaphoran pros ekeina echon).

 

« But this is scarcely a well-founded objection against our [Vahlen's] interpretation of arthron, because what he says there about the parts of speech is less Aristotelian than congruent with the later views.

 

« And, finally, it is to be observed that in the Peripatetic school the arthra were known as a specific part of speech besides syndesmoi, as seen in Simplicius' note about Theophrastus on Aristot.Categ. f.3b. Bas.: «As far as the dictions are concerned, there are other problems, which Theophrastus has stirred up in his Matters about the parts of speech and his@ followers have studied, e.g. whether only the Noun and the Verb are the parts of speech (tou logou stoicheia) or also the arthra and the syndesmoi are, and how about the others, these being indeed the parts of diction (mere lexeos), which are not the parts of speech that the Noun and the Verb are, and so on.»

@

« Although it is not known to us [Vahlen] in what sense Theophrastus understood the arthron, and any certain conclusion is not to be made from the fact that he ranked it as level with syndesmos and opposed both against onoma and rhema, the fact itself that Theophrastus knew arthra and syndesmoi as distinct parts of speech will be of a little use to the confirmation, not to say the accounting for, of the definition of arthron in Poetics. » (*19)

 

And LUCAS, D.W. (Aristotle POETICS, 1968, OUP, Oxford, com. on 1456b38-57a1) says, "Editions are agreed in despairing of this passage. The text with its repetitions and alternative definitions is suspect, the illustrations are inadequate, and the meaning of the terms, especially of arthron, is not the same as in later writers. Arthron: lit. joint, is the term used by later writers for the article, first perhaps in Rhet.ad.Alex.1435b13. This is not the meaning here, and the prepositions which are offered as an example do not conform to the definition. It is impossible to say what kinds of non-significant word Aristotle here intends. Rostagni following Susemihl and some earlier editions rejects the definition of arthron here and removes the word at 56b21 because, according to Dion.Hal.Comp.48, Aristotle and Theodectes recognized only three parts of speech onoma, rhema, syndesmos, and the distinction between syndesmos and arthron in the sense of article was due to the Stoics. In any case Dionysius seems to be inaccurate in that this sense of arthron is found in the Rhet.ad.Alex.1435b13 and perhaps in Theophrastus (Simplicius on Aristotle's Categories,p.10,24@ Kalbfleisch) which are earlier than Stoic grammar. If the passage were a later interpolation one would expect the account of arthron to be that current in the interpolator's own time." (pp.201-202)

 

18. ARTHRON (ARTICLE) in its first definition in Aristotle is an INDICATOR of a Sentence-branch: PREPOSITION

Now, Aristotelian definition of arthron: "Arthron deloi logou archen e telos e diorismon." (1457a6-7) should not be translated into: "An article shows a beginning or an end or a division of sentence." (DEFINITION-3a), but, as we have shown it in another of our studies concerning the problems, into: "An article indicates a branch or a fraction or a portion of sentence." (DEFINITION-3b) And a branch or a fraction or a portion of sentence is what we call now "prepositional phrase", so article (arthron) is its first component, namely a preposition, not the definite article.(*20)

 

But why? Because arche (beginning) may mean "beginning or origin of a river, that is, branch of a river" (cf. LXX Gen.2,10), and telos (end) may mean "troop (of children) or column (of chariots) or territorial division (e.g. to Lokrikon telos, the Locrian territory)"(*21), and diorismos (division) means "division" of any kind according to its cotext; and the conjunction e (or) may be used not only alternatively but also as a connective implying "that is to say", or "in other words"; then, we get: "a branch or a portion or a fraction of sentence: that is, in a word, a sentence-branch (a germ of sentence)."

 

And "a sentence-branch" is not the same as "a branch of@ a sentence", because the latter presupposes an already constructed sentence, but the former is only a syntactical fraction or a portion in regard to the construction of sentences (this@ is meant just by "of sentence", not "of a sentence"). And "a sentence-branch" is not but "a prepositional phrase" in our sense. For, a prepositional phrase is not yet "a sentence" in Aristotle, but "a sentence-branch," "a germ of sentence".

@@@@

Indeed, the Aristotelian definition of logos (sentence): "A sentence is, now, a significant compound sound, some parts of which signify something by themselves." (Poet.1457a23-24) excludes the prepositional phrases from the category of logos, because a prepositional phrase does not contain names (nouns substantive or adjective in the nominative, or verbs in the non-finite form) nor verbs (in the finite form) that are said by Aristotle to be able to signify something by themselves in the sentences (cf. Poet.1457a10-14; On Inter.16a20,16b26-28,17a17-19).

 

For, a prepositional phrase, i.e. a sentence-branch (logou arche) is made of a preposition (arthron) and a case (ptosis, i.e. an oblique case of nouns) sometimes with a definite article in the same case with itself as its predecessor, e.g. en archei (in the beginning), epi ton akron (at the extremities), none of these being, by definition, able to signify something by itself.

 

Therefore a prepositional phrase is not yet sufficient to be "a sentence", but only to be called "a branch or a portion or a fraction of sentence (logou arche e telos e diorismos)".

 

Then, the first component of a prepositional phrase is a preposition, which is allowed to be said "the Introducer or the Indicator of a sentence-branch"; so, Aristotle defined it as "deloi logou archen e telos e diorismon (It shows a branch or a portion or a fraction of sentence).

 

And this "Indicator of a sentence-branch" was given the name "ARTHRON (ARTICLE)" by Aristotle.

 

And the post-Aristotelian generations including the Stoic philosophers soon after Aristotle could not see but some inflectional particles in the term "arthron" with a nuance of joint or articulation, but this was far off its genuine sense. Yet, in Aristotle indeed, "arthron" can signify some non-inflectional particles such as prepositions, because, for him, "arthron" or "kampe" (joint) implies two things: rest and motion. For instance, Aristotle writes; "Now that the origin of all the other movements is that which moves itself, and that the origin of this is the immovable, and that the prime mover must necessarily be immovable, has already been determined when we were investigating whether or not eternal movement exists. And this we must apprehend not merely in theory as a general principle but also in its individual manifestations and in the objects of sense-perception, on the basis of which we search for general theories and with which we hold that these theories ought to agree. For it is clear also in the objects of sense-perception that movement is impossible if there is nothing in a state of rest, and above all in the animals themselves. For if any one of their parts moves, another part must necessarily be at rest; and it is on this account that animals have joints. For they use their joints as a centre, and the whole part in which the joint is situated is both one and two, both straight and bent, changing potentially and actually because of the joint. And when the part is being bent and moved, one of the points in the joint moves and one remains at rest." (*22)

 

19. Comparison of Aristotelian system of parts of speech with that of Dionysius of Thrax

Now that the meaning of arthron in Aristotle is determined, we try to make Aristotelian system of parts of speech correspondith that of Dionysius of Thrax in order to make it clearer as to the points not yet defined.

 

Dionysius of Thrax

Aristotle

onoma (Noun)

antonymia (Pronoun)

metoche (Participle)

arthron (Definite Article & Relative Pronoun)

 

onoma (Noun)

rhema (Verb)

rhema (Verb)

prothesis (Preposition)

arthron (Preposition)

syndesmos (Conjunction)

syndesmos (Conjunction)

epirrhema (Adverb)

???

stoicheion (Alphabet)

stoicheion (Alphabet)

syllabai (Syllables)

syllabai (Syllables)

logos (Sentence)

logos (Sentence)

 

Now, if, as supposed before, two systems correspond with each other as a whole, what (???) is then that which does in the Aristotelian with the epirrhema (Adverb) of Dionysius of Thrax ? And where are Adverbs in Aristotle ?

 

In fact, the grammatical position of Adverbs in Aristotle is not yet clear. Although some people include it within onomata (names), saying according to their own arbitrary criterion that an adverb signifies something by itself, but an adverb does not reach the Aristotelian criterion of onoma (a Name) that when "is", "was" or "will be" is added, it does then form a proposition, which either is true or is false, as a Noun itself always does then (De Inter.16b), with the exception that it is called "a name or word (onoma)" in an application of one of the rhetorical rules of paromoiosis (the repetition of the same word at the end of the clauses of a periodic sentence) (cf. Rhet.III,9, 1410a24-37).

 

Because, e.g. when "is", "was" or "will be" is added to an adverb "koinos (in common)", it does not then form a proposition, which either is true or is false. Is an adverb, then, a Case (ptosis) ? Yes, it is, in@ general (cf. Top.I.15,106b29-39). But, "ptosis is used of any modification of a word, such as cases and genders of nouns and adjectives, adjectives derived from nouns, adverbs formed from adjectives (as in the examples which Aristotle gives here), and the tenses of verbs" (Forster's note on Top.106b29,tr.p.314(*23)) So, if we are to find a further and proper definition of the Adverb in Aristotle, it will be elsewhere possible.

 

20. The second definition of ARTHRON (ARTICLE) as regards Adverbs in Aristotle@

Now, although we eliminated the second definition of arthron (DEFINITION-4) of the text as a confusion before, it will be worth while seeking to consider its possibility of implying the adverbs in Aristotle, because the definition: "Or it is a non-significant sound which neither obstructs nor produces the unity of a significant sound out of several sounds and which is proper to be put on both sides and in the middle of the sounds." may suggest that this peculiarity of the arthron shall conform to, as it were, the third horizen of the syntactical relations of words, i.e. the adverbial relations in syntax, the 1st horizen being the predicative ones, and the 2nd being the attributive ones, these two together with the conjunctive ones consisting in principle in compound sentences by conjunctions (ho logos ek pleionon heis syndesmoi: a compound sentence unified into one by conjunction as the Iliad. Poet.1457a28-30)

 

In fact, the logos (sentence) in Aristotle implies only the 1st and the 2nd of these, because they have a subjective substantive and a predicative verb (e.g. Socrates sleeps.) or a noun adjective and a noun substantive (e.g. beautiful horse) or at least one Name capable of utterance (phasis) necessary to be significant by itself (e.g. his eyes) (Cat. chap.2).

 

But, if we hold to the definition of logos (sentence) by Aristotle, an adverbial relation too is to be regarded as one of sentences, because it has at least one Name capable of utterance (phasis) necessary for it to be significant by itself (e.g. to work hard, or very beautiful) in the same way as the example: his eyes.

 

Therefore, the second definition of arthron: "Or it is a non-significant sound which neither obstructs nor produces the unity of a significant sound out of several sounds and which is proper to be put on both sides and in the middle of the sounds." may imform us of the adverbial relation set up by the addition of an adverbial modifier to an already standing sentence, e.g. "well" added to "Socrates sleeps." or "very" to "beautiful horse". And in such cases, in general, modifiers may be posted on both sides and in the middle of the sounds according to the context (in this case, the subject is singular, i.e one adverb, then, it is put (tithesthai: Poet.1457a9), not together-put (syn-tithesthai: Poet.1457a2) like the conjunctions used in couple or in ensemble; therefore, «on both sides and in the middle (kai epi ton akron kai epi tou mesou: Poet.1457a9-10)» signify disjunctively one of three positions implied there, on the contrary the same phrase «on both sides and in the middle (kai epi ton akron kai epi tou mesou: Poet.1457a2-3) signify conjunctively as regards the conjunctions used in couple or in ensemble.) and they neither obstructs nor produces the unity of a significant sound out of several sounds, because the unified structure of the sentece is there presupposed and they function as its modifiers.

 

Although, in principle and by definition in Aristotle, an adverbial relation for itself is capable of forming an original sentence, this possibility is not yet fully perceived by him, because his analysis and theory of sentence (logos) is centered on his epistemological and ontological ground of the subject that cannot be itself predicate of anything. And to this ultimate subject, a predicate or an attribute is, as it were, the first and the nearest modifier, an adverbial being the next because of its mediate relation to the ultimate subject via a predicate or an attribute of which it is a modifier. In a word, a predicate or an attribute is the immediate and direct modifier of the ultimate subject and an adverbial is the mediate and indirect modifier of it.

 

Besides, in Greek, "adverbs, like prepositions and conjunctions, were originally case forms, made from the stems of nouns and pronouns. Some of these nominal and pronominal stems have gone out of common use, so that only petrified forms are left in the adverbs. Some of these words were still felt to be live cases; in others no consciousness of their origin survived. Many adverbs show old suffixes joined to the stem or to a case form. It is sometimes uncertain whether we should speak of@ adverbs or of nouns with local endings." (Smyth, H.W., Greek Grammar, Harvard U.P., Cambridge, 1920, p.99)

 

Thus, we are allowed to see Adverbs in the second definition of arthron (DEFINITION-4), and to understand why Aristotle classified them under arthron. For, Prepositions and Adverbs are non-inflectional particles, then they are both called arthron, and nevertheless their syntactical functions are different from each other, then each is defined in a different manner. In conclusion, Aristotelian system has the arthron in its second sense that corresponds to the epirrhema in the system of the Greek eight parts of speech of @Dionysius of Thrax; that is, ADVERB.

@@@@@@@@@@@

21. Aristotle as theoretical developer in Syntax of sentence@

The standpoint from which Aristotle analyzes Greek grammatical phenomena is that of Syntax of sentence, his theory of parts of speech defined thereby, as we see in the following TABLE;

 

Sentence (logos)

Parts of speech (lexeos mere)

Simple Sentence (hen semainei)

(1) Predicative Relation

onoma (Noun Substantive) +

rhema (Verb)@

(2) Attributive Relation

onoma (Noun Adjective) +

onoma (Noun Substantive) @

(3) Adverbial Relation (*24)

(1) or (2) + arthron (Adverb)@@@@@@@@@

@@@@@@ (DEFINITION-4)@

@@@ (2', 3') Sentence-branch

@@@@@ @@@@@(logou arche)

@arthron (Preposition)

@@@@@@ (DEFINITION-3b)@

 

Compound or Complex Sentence@ (syndesmoi heis)@

(4) Conjunctive Relation

syndesmos (Conjunction)

(4a) conjunctive relation

syndesmos (DEFINITION-2)

(4b) disjunctive relation

syndesmos (DEFINITION-1)

 

Now that Greek parts of speech in Aristotle is clear to the last detail, the definitions and their corollaries of logos (sentence) in Aristotle are also clearer and vice versa. In fact, the definition (Y): "A sentence is, now, a significant sound organized into one, some parts of which signify something by themselves." (Poet.1457a23-24) is essentially equivalent with the definition (Z): "A sentence is, now, a significant sound, some of whose parts are significant for themselves as an utterance, but not as a positive or a negative statement." (On Inter.16b26-28) and an utterance can be made by a name (onoma; i.e. a Noun or a Verb. Cf. On Inter.17a17-19), therefore, that which has names as its parts can be called a sentence, as contrasted with a name, no part of which is significant by itself (cf. Poet.1457a11-12; On Inter.16a20.).

 

On the other hand, a sentence is one in two ways, that is, by signifying one thing, or by being unified with conjunctions (cf. Poet.1457a28-30.) So, the latter type of sentence is capable of having a positive or a negative statement as its parts despite the definition (Z), and of being called a complex or a compound sentence, as contrasted with that whose parts never surpass an utterance, this type of sentence allowed to be called simple.

 

Thus, we have four types of sentence according to Aristotle;

@(1) a name (Noun) + a name (Verb)@

@(2) a name (Noun Adjective) + @a name (Noun Substantive)

@(3) a name (Noun or Verb) + non-significant (ptosis or arthron)

@(4) a sentence + a conjunction + a@ sentence

@(2') a name (noun) + a sentence-branch (attributive)

@(3') a name (verb) + a sentence-branch (adverbial)

 

For instance;

@(1) anthropos manthanei. (Man learns.)

@(2) kalos hippos (a beautiful horse)

@(3) basilei philoi (friendly to the king)

@@@@ hen semainein (to signify one thing) (*25)

@@@@ dichos legesthai (to be said in two ways)

@@@@ mala polla (very many)

@(4) hippos trechei kai anthropos manthanei.

@@@@@@@@@@@@ (A horse runs and man learns.)

@(2') phone aneu chronou (a sound without tense)

@(3') en archei tithenai (to put in the beginning)

 

And we have, in the concrete, any mixed sentence of;@

@(1) and (2): kalos hippos trechei. (A beautiful horse runs.)

@(1) and (3): hippos trechei tacheos. (A horse runs quickly.)

@(2) and (3): zoion epistemes dektikon

@@@@@@@@@@@@@ (an animal receptive of knowledge)@@@@@@@@@@

@(1), (2), (2') and (3'): onoma esti phone synthete semantike aneu chronou hes meros ouden esti kath' hauto semantikon.

@@@@@@@ (A noun is a compound significant sound without tense, no part of which is significant by itself.)

and so on.

@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @@@@@@@@@@@@@@@

Part III. Theodectes, Theophrastus and the Stoics

22. Semi-predecessor and Collaborator of Aristotle: Theodectes

Dionysius of Halicarnassus tells us very important facts about the early history of the development of the Greek parts of speech in his On Literary Composition, 2; "Composition is, as the name itself indicates, a certain process of arranging the parts of speech, or the elements of diction, as some call them. These were restricted to three only in number by Theodectes and Aristotle and the philosophers of their day, who made nouns, verbs and conjunctions the primary parts of speech. Their successors, and in particular the leaders of the Stoic school, raised the number to four, separating the articles from the conjunctions. Subsequent grammarians distinguished appellatives from the other substantives, and represented the primary parts as five. Others detached the pronouns from the nouns, and thus introduced a sixth element. Yet others divided the adverbs from the verbs, the prepositions from the conjunctions and the paticiples from the appellatives."(*26)

 

And in his On the style of Demosthenes, 47-48, too, he writes;

"The primary parts of speech, which some call the elements, whether they be three, as Theodectes and Aristotle believe -- nouns, verbs and conjunctions -- or four, as Zeno and the Stoic school say, or more, are always accompanied by two phenomena of equal importance, tone and time."(*27)

 

Now, his attribution of three parts of speech (onomata, rhemata and syndesmoi) to Aristotle is to be in doubt, because Dionysius of Halicarnassus is the first person ever in history that has been identified as reader of the Corpus Aristotelicum through Andronicus' edition(*28), which contains our Poetics, therefore he must have read the 20 chapter of it and been acquainted with the arthron there introduced by Aristotle in addition to onoma, rhema and syndesmos. And an indirect, but most uniquely useful evidence for it is just his remark: "Their successors, and in particular the leaders of the Stoic school, raised the number to four, separating the articles from the conjunctions." For, how else could he know that the Stoic arthra had been @separated from the Peripatetic syndesmoi ? (He might have meant also Theophrastus and his followers by "their successors" in accordance with his premises.)

 

As a matter of fact, he never seems to have get to the Aristotelian sense of arthron, which he identifies with the definite article and perhaps also with the relative pronoun, the pronoun in general being, according to his remarks, "detached from the nouns". Thus he shows implicitly the terminological origin of the Stoic arthrton in Aristotle and at the same time he neglects the Peripatetic problem about the Aristotelian definition of the term in Poetics and among the Peripatetics such as Theophrastus and his followers.

 

On@ the@ other@ hand, although it is hardly credible besides astonishing, if Dionysius of Halicarnassus "did not know or did neglect the relevant text of Poetics" (Classen, de primord. gr. gr. p.60)(*29), his remarks in On literary Composition would be literally taken. Then, he must have thought that Aristotle had counted what the Stoics meant by their original term "arthron" as belonging to conjunctions, possibly on the ground that Rhetoric, perhaps one of his main sources, might indicate it (e.g. III,5,1407a22-23), though this indication is false, because the case is that the pronominal use of the definite article here (ho) is not treated as conjunction by Aristotle, but that it is one of the proper applications of the coupled conjunctions (men----de) to any coupled words or phrases; i.e., here, to the coupled personal pronouns: I (ego) and he (ho).

 

Or rather, we estimate that his phrase "separating the@ articles from the conjunctions" is his rhetorical way of implicit referring to the arthron which he found in Aristotle's Poetics, but whose real definition could not be conciliated to him with the Stoic and current use of it: the definite article and the relative pronoun.

 

Moreover, an ordinary reader of Aristotle would find that the definite article as well as the pronoun including the relative pronoun in Aristotle had been treated as onoma, i.e. its substitute, not as syndesmos (cf. Rhet.III,5,1407b8-11;III,6,1407b35-37; Anal.Pr. I,40,49b10-13; Poet.1457a18-23; Sophist.Ref. 32,182a8-182b6).

 

Thus, we are allowed to say that Dionysius of Halicarnassus has conserved and protected the Aristotelian origin of the Stoic arthron by means of his highly rhetorical expressions of the matter.

 

As to Theodectes, however, his remarks indicate the fact itself, or his is the only information about Theodectes' view of the parts of speech. And we are here to ask why Theodectes precedes Aristotle in both of his explanations; "These were restricted to three only in number by Theodectes and Aristotle and the philosophers of their day", and "they be three, as Theodectes and Aristotle believe -- nouns, verbs and conjunctions".

 

Now, according to Capps(*30), Theodectes must have been the friend rather than the disciple of Aristotle, who was several years junior, and not some ten years his senior, as one has supposed hitherto.

 

This almost revolutionary view of Capps depends on his severe analyses of the catalogues of victorious tragic and comic poets found in the Greek inscriptions. Though the results of his studies are not yet fully taken into consideration concerning the problems like ours, we ought to reconsider their academic and objective value for a new prospect of the matter, getting rid of our confirmed and subjective romanticism like Ernst Diehls's in his contributions to Pauly=Wissowa's Realencyclopedie (s.v.Theodektes).

 

Capps writes;

"The order in which the names occur in these catalogues was determined by the date of the first victory of each@ poet. If, then, we can fix the date of any given name in the lists, we shall know within very narrow limits the dates of the first victories of the poets immediately succeeding and following, and if we can fix the date of any two names in a given list, the limits are known within which the intervening names must fall. With the information thus gained we may hope in some cases to be able to correct or correctly interpret the often vague or corrupt chronological notices found in Suidas, the hypotheses prefixed to the extant dramas, Eusebius and the other chronographers, Anonymous peri komoidias II (Kaibel), the Parian Chronicle, and the statements scattered throughout Greek literature. This has not yet been attempted except in a desultory way and where the conclusions are most obvious. I propose to apply the new information thus derived mainly to some of the better known of the minor poets. The results which we shall reach may not always seem conclusive; it is hoped that they may at least be of value in suggesting a new line of inquiry or in giving a new point of view.

 

Theodectas/es - Suidas furnishes almost all of the data which we possess concerning this poet: "Theodectes, son of Aristandros from Phaselis of Lycia. Rhetor, becoming a tragic poet. Disciple of Plato and Isocrates and Aristotle. He spoke in the funeral competition for the sake of the late Mausolus in the 107th (MSS103rd;corr.Clinton) Olympiad. Wrote 50 dramas. Departed in Athens at the age of 41, his father being still alive."

 

Welcker (Die griech.Trag. p.1070) finds a terminus ante quem for his death in the story of Alexander's homage to the poet's statue at Phaselis (Plut.,Alex.17). This was in 334/3. Since Theodectas was 41 years old at the time of his death, he must have been born as early at least as 375, probably a few years earlier. This result has been universally accepted, being consistent with the statement of Suidas that Theodectas was a pupil of Aristotle, who came to Athens in 368, and accounting for the marked respect shown by Alexander, who became the pupil of Aristotle in 343. The young prince may even have known the poet personally.@@

 

But the victors' catalogue upsets this most reasonable combination. In frag.b we find [Carci]nus XI, [Ast]ydamas V[II]I, [Th-eo]dectas VII, [Apha]reus II. According to Vit.X Orat.839d, Aphareus began to exhibit in the archonship of Lysistratus, 368/7, and appeared last in the archonship of Sosigenes, 342/1, winning two victories at the City Dionysia in this period. The acme of Carcinus is placed by Suidas in Ol.100 (380-77). We learn from Diod.Sic.5,5 that he was often in Syracuse during the reign of the younger Dionysius (368 to 356). He must have attained a high position as a tragic poet before he was invited to Syracuse, and probably had won the larger number of his eleven victories before the accession of Dionysius II. The date of the first victory of Astydamas is fixed by the Parian Chronicle in the year 372, as we shall see later. The order of the names Carcinus, Astydamas, and@ Aphareus is therefore entirely in harmony with the chronological data. If we should assume an interval of three years between each of these four names- and certainly this is a liberal estimate - we should have as approximate dates of the first victories: Carcinus, ca.376; Astydamas, 372; Theodectas, ca.368; Aphareus, ca.362. Since the acme of Carcinus is given as 380-77, it is more probable that his first victory was won before 376 than that Theodectas won later than 368. However, in order to keep as near to Suidas as possible, let us set the first victory of Theodectas forward to 365, though so long an interval is intrinsically improbable.

 

Theodectas produced 50 tragedies - that is, took part in more than 16 contests. That he devoted himself more especially to the City Dionysia is a fair inference from the fact that seven of his eight victories (Epigram apud Steph.Byz., s.v. Phaselis) were won at this festival. By all accounts he had gained an enviable reputation as a rhetor before he turned his attention to tragedy. His talents must have been recognized at an early age. And yet he could hardly have entered upon his career as a poet before the age of 25. To assume a later date would make it necessary to crowd more than three tragedies into each year. Accepting this age for his first appearance at the Dionysia, and assuming that he was victorious in his first competition, his death would fall ca.350. If he was not successful at once, his death must be placed still earlier - a supposition that is excluded by the fact of his participation in the Mausolus competition in 351. On the other hand, even if nine years elapsed between the first victory of Astydamas and that of Theodectas, and even if the latter took up tragic poetry before the age of 25, his death could not be placed more than a few years after 350. At the closest possible estimate he died from 10 to 15 years earlier than was assumed in Welcker's combination. The year of Theodectas' birth was accordingly not far from 390. He may well have been a pupil of Plato and Isocrates, but he must have been the friend rather than the disciple of Aristotle, who was several years his junior, and not some ten years his senior, as one has supposed hitherto. In this connection it is significant that the Vit.X Orat.(837c) and Dionysius of Halicarnassus (Isaeus,sub fin.) both report that the poet was a pupil of Isocrates, but say nothing of Aristotle. It is evident that Suidas or his source was tempted to bring together the great trio. Theodectas was not a youth of 24 when invited to do honor to the memory of Mausolus in 351, but a mature and accomplished man of 40, whose reputation was firmly established. Alexander could not have known him personally, but learned to esteem the man and his works through the poet's friend and collaborator, Aristotle. This is the meaning of association (homiliai) in Plutarch's reference to Alexander's act of homage: thus in pleasantry returning no ungraceful honour for his association with the man that took place through Aristotle and the philosophy (ouk acharin en paidiai timen apodidous tei genomenei di' Aristotelen kai philosophian homiliai pros ton andra).

 

Astydamas, father and son. - Since the date which we have been able to reach for the first victory of Theodectas depends somewhat upon our interpretation of the notice in the Parian Marble for the year 372, it may be well to state here the reasons which oblige us to assume that this chronicle records only first victories. It contains six notices of dramatic victories in a form sufficiently complete to be of service. In three of these the phrase is: won the first prize (proton enikesen) - Aeschylus in 484, Euripides in 441, and Menander in 315 (new.frag., Athen.Mitth.1897,p.187). The victory of Sophocles in 468 we know from Plutarch, Cimon 8, to have been his first victory, won at the City Dionysia. Philemon is set down as victorious in 327; he could scarcely have won before this date, and we know that his first Lenaean victory was not gained for some years afterward. The omission of the first prize (proton) in the case of Astydamas consequently signifies nothing. When, now, in the catalogue of victors at the City Dionysia we find that a poet Astydamas won his first victory between 376 and 362 - and both of these dates, though approximate, are derived from evidence independent of the inscription - the conclusion is irresistible that the victory of Astydamas in 372 was his first victory - indeed, his first City victory - determining the position of his name in the victors' list. Thus what was only a shrewd conjecture of Clinton must now be recognized as a demonstrated fact. (And in the succeeding@ arguments Capps checks the ancient informations about the elder and the younger Astydamas, pointing out the contradictions among them, and concludes that) The facts that refer to the son have been transferred, by a simple error of transmission, such as abound in Suidas, to the father."

 

Now, the studies by Capps on the order of t he names of the tragic victors may influence by the formalism itself upon our study of the order of the names: Theodectes and Aristotle in the descriptions of Dionysius of Halicarnassus, apart from the contents gained from them. That is, Theodectes as the primer may have had the principal merits of proposing the three parts of speech, adding the Conjunction (syndesmos)(*31) to the traditional Noun (onoma) and Verb (rhema) set up by Plato.

 

This fact must have been recognized by Dionysius of Halicarnassus or any other who could have read Theodectes' rhetoric or its compendium or the compendia of all the preceding rhetorics by Aristotle (*32). If so, then Theodectes is the founder of the explicit theory of the three parts of speech, whom Dionysius of Halicarnassus puts in the first place in giving supporters of the theory, Aristotle being for him a suspect of having the theory of four parts of speech including arthron in Poetics, which he will never acknowledge explicitly. Therefore, Aristotle seems to be a follower of Theodectes as to the Conjunction (syndesmos), and the original founder as to the arthron.

 

Concerning the former, Lersch seems to be right in saying (*33); "Perhaps Aristotle, who himself had already observed many times the phenomena of syndesmos, had mentioned willingly or at least with no reluctance that Theodectes gave it a very high rank, and that Theodectes counted it as indispensable part of speech as well as onoma and rhema" in his rhetorical works.

 

In short, Theodectes found the facts of conjunctions in Greek, and Aristotle, having accepted the data from him, gave the name syndesmos to the particles and defined it grammatically, we think. And in this context, too, Theodectes' advent to Academy@ may have@ been one of the chief motives for the positive change of the attitude toward rhetoric in Aristotle and in Academy.(*34)

 

23. Immediate successor to Aristotle: Theophrastus and his school showing the existence of arthron bequeathed to them by Aristotle

Now, the Aristotelian system of parts of speech consisting in four word classes; first onoma (Noun & Pronoun & Participle & The Definite Article & The Relative Pronoun), secondly rhema (Verb), thirdly syndesmos (Conjunction), and fourthly arthron (Preposition & Adverb) was bequeathed to the Peripatetic school beginning@ with Theophrastus, the public and at the same time private (he is said to be bequeathed his master's whole library including his own original manuscripts)(*35) successor to Aristotle, the real founder of the school. And Theophrastus' Matters about the parts of speech shows his loyal heritage of the theory of the four word classes of Aristotle: Onoma, Rhema, Syndesmos and Arthron. In fact, as Vahlen pointed it out (cf. 17.);@

 

« It is to be observed that in the Peripatetic school the arthra were known as a specific part of speech besides syndesmoi, as seen in Simplicius' note about Theophrastus on Aristot.Categ.: «As far as the dictions are concerned, there are other problems, which Theophrastus has stirred up in his Matters about the parts of speech and his@ followers have studied, e.g. whether only the Noun and the Verb are the parts of speech (tou logou stoicheia) or also the arthra and the syndesmoi are, and how about the others, these being indeed the parts of diction (mere lexeos), which are not the parts of speech that the Noun and the Verb are, and so on. »

 

This Peripatetic argument about the problem of the parts of speech (tou logou stoicheia) and the parts of diction (mere lexeos) is not entirely of Aristotle, nevertheless it concerns with no doubt the four word classes (Noun, Verb, Syndesmos and Arthron), because « the others» than these four he mentioned are such as « Syllables (syllabe), Alphabet (stoicheion), Case (ptosis) » (cf. 16.).@

 

24. Mediate successor to Aristotle: the Stoa and the problem of Adverbs left

And there was another successor who made his own way to the flourishing development of the Greek parts of speech; that is, the Stoic school founded by Zeno of Citium, from Cyprus. Concerning the Stoic system of the parts of speech, we are informed that;

 

a): "The Stoic philosophers leaning upon their own particular arguments separate from the Name (Proper Noun, onoma) the Common Noun (prosegoria), which, they say, is an independent part of speech. And they catalogue the parts of speech as follows: first the Name, secondly the Common Noun, thirdly the Verb (rhema) and Participle (metoche) under the same rubric, saying that the Verb is Predicate and the Participle is an inflexion of a Verb, i.e. a derivation of a Verb, fourthly the Article (Joint, arthron) and Pronoun (antonymia) under the same rubric, asserting that the former is the Indeterminate Joint (aoriston arthron) and the latter is the Determinate Joint (horismenon arthron), and fifthly the Preposition (prothesis) and Conjunction (syndesmos) under the same rubric, calling the former Prepositive Conjunction (protheticos syndesmos) and the latter Subordinate Conjunction (hypotacticos syndesmos). But, then, they did not think the Adverbs (epirremata) worthy of belongings to the sentence or to the number, leaving them aside as if they were offshoots or small grapes for gleaners." (Scholia in Dionys.Thr.) (*36)

 

b): "There are, as stated by Diogenes the Babylonian in his treatise On Language and by Chrysippus, five parts of speech: Proper Name, Common Noun, Verb, Conjunction, Article. To these Antipater in his work On Words and their Meaning adds another part, the Middle (mesotes, i.e. Adverb).

 

A Common Noun is defined by Diogenes as part of a sentence signifying a common quality, e.g. man, horse; whereas a Name is a part of speech showing a peculiar quality, e.g. Diogenes, Socrates. A Verb is, according to Diogenes, a part of speech signifying an individual predicate, or, as others define it, an indeclinable element of a sentence, signifying something attachable to one or more subjects, e.g. write, speak. A Conjunction is an indeclinable part of speech, binding varios parts of speech; and an Article is a declinable element of a sentence, delimiting the genders and numbers of Nouns, e.g. ho, he, to, hoi, hai, ta (= the; masc., fem. and neut., singular and plural)." (Diocles ap.D.L.) [Hülser, Nr.536]

@

Therefore, from the terminological point of view, with the witness of Dionysius of Halicarnassus counted, the Stoic system of the parts of speech has developed as follows;

 

Stage I (4): Noun, Verb, Conjunction, Article.

Stage II (5): Name, Common Noun, Verb, Conjunction, Article.

Stage III (6): Name, Common Noun, Verb, Conjunction, Article, Adverb.

 

And, before the Adverb (mesotes) was introduced by Antipater at the stage III, it was lacking in the space for adverbs. Indeed, the Stoics, then, did not think them worthy of belongings to the sentence or to the number, leaving them aside as if they were offshoots or small grapes for gleaners. (Scholia in Dion.Thr.) @But Why ? We are to answer now the question.

 

25. the Stoic stage I consisting in 4 parts of speech

About the Stoic arthron, we are informed that;

 

c): "the Stoics call the Pronouns also Joints (Articles), their Joints (Articles) being different from our definite articles in that the Pronouns are the Determinate Joints and the definite articles are the Indeterminate Joints, as, they say, follows; by a joint we mean two things: one is a join of two limbs in the same sense as when we say about joints' dislocation, and the other is the joined limbs themselves in the same sense as when we say that one is with large limbs. This is just the way in which a Joint functions in the sentence. And Apollodorus of Athens and Dionysius of Thrace call the Pronouns also Demonstrative Joints." (Apol.Dysc., De Pronom.) [Hülser, Nr.550]@

 

d): "The Stoics take the definite article and the pronoun as one, calling the pronoun Determinate Joint and the definite article Indeterminate Joint. The reason why they take them as one is that the definite articles are used in place of the pronouns. For, that which is used in place of something is the same with it. And further they say about their naming that all the pronouns are entirely determinate either through demonstration (deixis) or through reference (anaphora) because a pronoun may have a use of deixis, as in the case of "I", "you" and "this", or a use of anaphora, as in the case of "he". And every deixis is the original acknowledgement of the present person here. Because of this, they call the pronoun Determinate Joint. On the contrary, Indeterminate Joints are so called, because we find them indeterminate in their context; e.g., in a sentence: "I listened to the speaker not to see him in listening.", the is used in place of "a"." (Scholia in Dionys.Thr.) [Hülser, Nr.551]@

 

e): "The reason will plead for them (the Stoics) insofar as the pronouns are used anaphorically and the joints have a relation of anaphora. In fact, the joints are substituted for pronouns as "before-placed joints" (arthra protactica; i.e. the definite articles) (e.g.Il.I,12; Od.XIV,36), or as "after-placed joints" (arthra hypotactica; i.e. the relative pronouns) (e.g. Il.XXI,198;XXIII,9; Od.II,207). (Apol.Dysc., De Pronom.) [Hülser, Nr.550]@

 

f): "The Stoics took the joint and the pronoun as the identical part of speech, calling what the grammarians call the definite article "Indeterminate Joint" (infinitus articulus), and adding there also the indefinite nouns as well as the relative nouns, which Didymus does, too, in his treatise On the syntactical correctness of Latin (De Latinitate). (On the other hand, the@ Stoics called "the Determinate Joint" what we now call determinate@ pronoun@ but is to be called pronoun simply with sound reason.) (Priscianus, Inst.gramm. XI,1) [Hülser, Nr.558]@

 

g): "the Stoics used to count indefinite nouns in general (generaliter infinita nomina), that is, relative nouns (relativa-nomina) and interrogative nouns (interrogativa nomina) as Joints because of their anaphoric character." (Prisc. ib.XVII,52) [Hülser, Nr.559]

 

And as to the Stoic syndsmos, we are informed that;@

 

h): "As is generally accepted and I, too, have demonstrated in@ the preceding work, the before-placed parts of speech owe the name (prothesis, Preposition) to their peculiar structure of syntax; i.e. to being before-placed. Because of this also the Stoics called the prepositions "Prepositional Conjunctions (protheticoi syndesmoi)"." (Apollon.Dysc., De synt. IV, 5) [Hülser, Nr.590]@

 

i): "The Stoics take the preposition and conjunction as one, calling preposi tions "Prepositional Conjunctions (protheticoi syndesmoi" and conjunctions simply "Conjunction (syndesmoi)", for, because prepositions and conjunctions are indeclinable, they take them as one." (Scholia in D.T.) [Hülser, Nr.592]

 

Thus, the first stage of the Stoic system consisting in four word classes; onoma (Noun), rhema (Verb & Participle), syndesmos (Conjunction & Preposition) and arthron (The Definite Article & The Relative Pronoun & The Pronoun) is lacking in the category: Adverb, because, when they accepted Aristotle's system or at least Aristotle's grammatical terminology via, probably, Theophrastus, they transformed some of the categories of Aristotle, syndesmos being transformed into Conjunction & Preposition and arthron into The Definite Article & The Relative Pronoun & Pronoun, so that the Adverb had to be excluded@ unconsciously out of the arthron so as to be left untouched outside the Stoic system.(*37)

 

And this is an indirect evidence that the Stoic work of@ the systematization of the parts of speech was, in the beginning, a succession in their own manner to that of Aristotle, not yet their own entirely original one, because if the latter had been the case, their systematization would have included any grammatical identification of the adverbs, which they ought to have left untouched outside their system as Aristotle, they thought as a result of their transformation of his system, did.

 

26. arthron in the Stoa

And the chief motive for this transformation was that they could not find the Aristotelian senses of arthron (Preposition & Adverb), so that they read its definition as implying The Definite Article & The Relative Pronoun joined with The Pronoun in general afterwards. For the definition: "arthron deloi logou archen e telos e diorismon." was read by them that "An article shows a beginning or an end or a division of a sentence."(DEFINITION-3a) because of their having been off its real sense defined by Aristotle, and they thought that these functions were explicit in the following uses of the Greek definite article and the relative pronoun;

 

@1) to kalon rhodon

@@@ (the beautiful rose)

@2) to rhodon to kalon

@@@ (the rose the beautiful, i.e. the beautiful rose)

@3) theaomai to rhodon, ho moi pepompen ho philos.

@@@ (I look at the rose, which the friend has sent to me.)

 

Namely, the to in 1) is a definite article showing the biginning of the sentence as well as the first to in 2). And the second to in 2) is in a sense that which shows the division or the end (as to the modified word rhodon) in the sentence. And the relative pronoun ho in 3) shows the beginning (of the subordinate clause) or the end (as to the principal clause) or the division (between the two) in the sentence. Thus we get an amalgam of the Greek definite article & the relative pronoun as a grammatical approximation of arthron in the sense of DEFINITION-3a.

 

And according to the notices c) to g) above, the Stoic arthron as Indeterminate Joint includes as one the so-called definite articles (arthra protactica) and the so-called relative pronouns (arthra hypotactica), which is no other thing than what we call the Stoic amalgam of the Greek definite article & the relative pronoun as a grammatical approximation of arthron in the sense of DEFINITION-3a, and which will include also the pronouns in general through the syntactical character of anaphora, as they say. And this unfortunate misunderstanding of the Aristotelian arthron by the Stoics has, in fact very fortunately, made its own way to the full development of the Greek eight parts of speech, otherwise the possibility of distinct identification of the Definite Article together with the Relative Pronoun would have been veiled in the commonplace of nouns or adjectives.

 

27. The second and the third stage in the Stoa and the Stoic conception of Adverb: mesotes or pandectes

We pass over, here, the Stoic second stage (having prosegoria: common noun apart from onoma: proper name) in order to insist upon the third stage where the Stoic conception of adverb; mesotes or pandectes(*38) was introduced.

 

Mesotes or pandectes was probably the Stoic answer to the question: why a certain kind of words (i.e. adverbs) were to be left outside the four word class system ? At first, mesotes, namely the Middle or the Mean, was adopted, we think, to suffice the adverbial character of modifying both nouns and verbs. For, in Greek, adjectives as well as substantives are nouns, therefore, the combination of a so-called adverb and a so-called adjective (e.g. very happy, highly admirable, etc.) is, in@ Greek in fact, a combination of an adverb with a noun, the combination of an adverb with a verb being extremely familiar. In this sense the Greek adverb is in its syntactical character a Mean between the verb and the noun, which is extraordinarily analogous to the naming of metoche (Participle) because of its partaking of both the noun and the verb. By further analysis by the Stoics, however, the adverb is discovered to modify not only nouns and verbs but also any sentence, as is the case with arthron in the second sense in Aristotle: i.e. adverb as sentence-modifier, and via a sentence also any kind of words, so that the adverb is able to be characterised as pandectes: that is, All- Receiver, or All-Receiving Particle. And the more traditional term epirrhema for adverb signifies, probably in its original use, not only the modifier of verb (rhema), but also of any kind of words, for the Greek epirrhema signifies originally "additional saying in general"(*39), just as the term rhema does "what is said in general". Therefore, the definition of epirrhema as modifier of verb in Dionysius of Thrace (§19) is an expression of the minimum condition of adverbs, because this condition alone can discriminate the adverbs from all the other parts of speech, being insufficient for referring to the other adverbial uses. Anyhow, the term epirrhema in its whole sense and also in its popular use could cover and take the place of all the extension of the Stoic pandectes, rich in its philosophical contents but a mere academic coinage. This is merely our theoretical hypothesis, which is ready to account for the mysterious history of the Stoic treatment of the Greek adverbs and for their terminological vicissitudes through the Stoa and afterwards.

 

28. Development of the Greek word class system according to R. H. Robins (1966)

The development of the word class system of the European grammatical@tradition by R. H. Robins (Foundations of Language 2, 1966, pp.3-19) was, thus far, very instructive and comprehensive in@this field of study, but it is now to be fundamentally corrected@because of its negligence(*40) of arthron in Aristotle. (The results of his studies are synopsized by his diagram A.)

 

Diagram A

 

 

Plato

Aristotle

Stoics 1

Stoics 2

Stoics 3

Dionysius

(Aristarchus)

Priscian

modern tradition

 

 

 

 

logos

 

onoma

 

onoma

 

onoma

onoma

onoma

onoma

nomen

Adjective

 

prosegoria

prosegoria

Noun

mesotes

(pandektes)

epirrhema

interiectio

Interjection

adverbium

Adverb

rhema

rhema

rhema

rhema

rhema

rhema

verbum

Verb

metoche

participium

 

 

syndesmos

syndesmos

syndesmos

syndesmos

prothesis

prepositio

Preposition

syndesmos

coniunctio

Conjunction

arthron

arthron

arthron

antonymia

pronomen

Pronoun

arthron

 

Article

 

 

29. Development of the Greek word class system according to our new researches and Robins' diagram corrected:

Our diagram B, correcting Robins', shows the development of the Greek word class system according to our new researches.

 

Diagram B

 

 

Plato

Theodectes

Aristotle & Theophrastus

Stoics 1

Stoics 2

Stoics 3

Dionysius

(Aristarchus)

Priscian

modern tradition

 

 

 

 

logos

onoma

onoma

 

 

 

Significant

 

onoma

onoma

onoma

onoma

onoma

nomen

Noun

prosegoria

prosegoria

Adjective

arthron

arthron

arthron

antonymia

pronomen

Pronoun

arthron

 

Article

rhema

rhema

rhema

rhema

rhema

rhema

rhema

verbum

Verb

metoche

participium

the third

kind of

words

syndesmos

 

Non-significant

syndesmos

syndesmos

syndesmos

syndesmos

syndesmos

coniunctio

Conjunction

 

 

arthron

prothesis

prepositio

Preposition

 

 

 

mesotes

(pandektes)

epirrhema

adverbium

Adverb

 

interiectio

Interjection

 

 

Appendix: Aristotle's phonology reexamined:

As one of our conclusions, we'll give a most reasonable analysis of Aristotle's phonology in Poetics, chap.20, which has been one of the causes for us to disbelieve in@ the@Aristotelian origin@of@the chapter including his theory of the@ parts@of speech.

@

In fact, the tripartite system of the Greek alphabet; the vowel (phoneen), the consonant (aphonon) and the semivowel (hemiphonon) has@inherited that of the precedig times (e.g. Plato; Theaet.203A-B, Cratyl.424B-C, Phileb.18B-C), and refined it (*41), in particular by defining the semivowels and with them the syllables containing them. For example, GR is said to be a syllable as@well as GRA, because the R is a semivowel and voiced, though not perfectly, like vowels. And@ S too is a semivowel. Now, the@reason why@ R and S, etc. are said to be voiced is that the liquids (R,L), the@sibilant (S) and the nasals (M, N) are such phones as are able@to@be@pronounced continuously and without intermission just like the vowels, all the other consonants being not. Whence the liquids (R, L), the sibilant (S) and the nasals (M, N) are able to@be called semivowels (*42), which can give the voiceless consonants a partner with a certain degree of voice by their competence@for continuous utterance to make a syllable together. Therefore they are called also sonant or syllabic liquids, etc.(*43).@

 

@@ NOTES:@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@@Part I.@ @@

*1) Apart from a systematization of words there can be no problem of word classes, because where is no systematization there is no classification, and vice versa. And we find many struggles for catching the essence of the human language or capacity of naming and with it the essence of names in early Greek philosophers before Plato, e.g. in Phytagoras, his deciple Archytas, Democritus, Antisthenes, and Stilpo and other Megarians, but there is no attempt of classifying words. And@ at the same time in them we can recognize the beginning of the process of classifying words before Plato, because they are in their questioning about the essence of words identifying above all and at most "names as names" or "words as words", which is to ready for someone to become aware of the different functions of names including the difference of the parts of speech. For instance, Democritus is probably the first philosopher that takes all kinds of words on equal terms with each other, just like "atoms" in his physics, numberless indivisible particles, each of them unequal with other both in form and size like broken pieces of glass, in one of his works, entitled "On@ words" (peri rhematon) [D.L. IX 48], decentrarizing the priority of names, because his use of rhema in this case is not that of verb, but that of saying or word in general, on the other hand his another work titled "On terms"@ (onomastikon) [id.] is a vocabulary arranged according to the subjects. Cf. Lersch, L., Die Sprachphilosophie der Alten, dargestelt an der historischen Entwickelung der Sprachkategorieen, zweiter Theil Geschichte der Entwickelung der Sprachkategorieen, I. Die Redetheile A. Die Griechen, die altesten Schriftsteller. Das Hauptwort, F.Baaden, Bonn, 1840, S.3-7.

*2) Burnet,J., Platonis Opera T.I - V, Oxford Univ. Press, 1901-1907, Oxford.

*3) Jowett, B, The Dialogues of Plato translated into English with analyses and introductions, 4th ed., vol III, Oxford Univ. Press, 1953, London, pp.404-408.

*4) Fowler, H.N., Plato with an Englih translation Theaetetus Sophist, William Heinemann LTD, 1961, London, p.411.@@@@@

*5) id. p.423.

*6) Fowler, H.N., Plato with an Englih translation Cratylus Parmenides Greater Hippias Lesser Hippias, W.H.LTD, 1963, London, pp,129-131.

*7) id. p.11.

*8) id. p.23.

*9) Burry, R.G., Plato with an Englih translation VII Timaeus Critias Cleitophon Menexenus Epstles, W.H.LTD, 1961, London, p.533. @

 

@@Part II

*10) @DIONYSII THRACIS ARS GRAMMATICA, edited by UHLIG,G., Teubner Verlag, 1883, Leibzig.

*11) @AUGUSTINUS, CATEGORIAE DECEM EX ARISTOTELE DECERPTAE, c.1. PL, p.1419.

*12) SUGIURA, S., History and Principle of Word Classification (in Japanese), Kobian Shobo Publ.,Tokyo,1976, pp.37-38.

*13) Cf. VAHLEN, (below), p.287.

*14) Cf. ARISTOTLE, On the Movement of Animals, tr. by FORSTER,E.S., Loeb C.L., Harvard U.P., Cambridge, 1961, 698a23-24.

*15) Cf. HÜLSER, K. Die Fragmente zur Dialektik der Stoiker, fromman-holzboog, Stuttgart-Bad Cannstatt, 1987, Nr. 542, 536, 550, 551, 558,559; BEKKER, I., Anecdota Graeca, bei G.Reimer, Berlin, 1816, p.873.

*16) Cf. BYWATER, I., Aristotle On the Art of Poetry, OUP, 1920, p.69.

*17) Text by KASSEL, R., Aristotelis de Arte Poetica Liber, OUP,Oxford, 1965.

*18) STEINTHAL, H., Geschichte der Sprachwissenschaft bei den Griechen und Romern mit besonderer Rucksicht auf die Logik, Ferd. Dummlers Verlag, Bonn,1961, (Unveranderter Nachdruck der Zweiten Auflage von 1890-1891, Berlin).

*19) VAHLEN, J., Beitrage zu Aristoteles' POETIK, B.G.Teubner Verl., Stuttgart,1914, pp.109-117, 284-290.

*20) Vahlen too denied the definite article to arthron.

*21) Cf. LIDDELL & SCOTT, A Greek-English Lexicon, OUP,1968.

*22) On the Movement of Animals, pp.441-443.

*23) Loeb C.L., Harvard U.P., Cambridge,1960.@@@@@

*24) Cf. KELLNER, L., Historical Outlines of English Syntax, ed. with notes by MIYABE, K., Kenkyusha LTD., Tokyo,1892, pp.21-28.

*25) "The relation of the accusative to its governing verb is analogous to that of the genitive to its governing substantive." KELLNER, id., p.118.@@@@@@@@@@@@@@@

@@@@@@@@@@@@@@@@@@@@@@@@

@ Part III;

*26) Dionysius of Halicarnassus The Critical Essays II, Engl. tr. by Usher, S., Loeb C.L. 466, Harvard U.P., Cambridge,1985.@@@@@

*27) Dionysius@ of Halicarnassus The Critical Essays I, Engl. tr. by Usher, S., Loeb C.L. 465, Harvard U.P., Cambridge,1974. N.B. Usher's tr. "Zeno and the Stoic School"; lit: the followers of@ Zeno the Stoic (hoi peri Zenona ton Stoicoi).@@

*28) "The earliest quotation I have foun d,which is undoubtedly taken from Andronichus' edition, is Dionysius de compos. c.25,198 and ep.ad.Amm. 8 en tei trite bybloi ton technon." Düring, Notes on the history of the transmission of Aristotle's writings, in Kurfess, H. and Düring, I., Aristotle And His Influence: Two Studies, Garland P.I., New York,1987 (reprint.) p.69.

*29) Steinthal, id., Erster Teil, S.263-4.

*30) Capps, E., Chronological Studies in the Greek Tragic and Comic Poets, Amer. Journ. of Philology, XXI,1900, pp.38-61. This is preceded by the same author, The Catalogues of Victors at the Dionysia and Lenaea, CIA. II 977, Amer. Journ. of Philology, XX, 1899, pp.388-405.

@ And as a matter of wonder, the literary sources of these inscriptional buildings of the Hellenistic age were probably nicai tragicai kai komicai (The Tragic and Comic Victories) [Capps,1899,p.398] or didaskaliai (The Catalogues of the Dramas) [D.L.V,26], compiled by Aristotle and his assistants.

*31) Theodectes, having been a pupil of Isocrates, might have taken his master's rhetorical teachings about the Greek conjunctive functions into account to identify the conjunctions. Cf. "In the lessons of rhetoric, one paid attention especially to the linking particles (syndesmoi, Conjunctions), because upon their correct setting did the clearness (sapheneia) and survey of our discourse depend." Pohlenz, M., Die Stoa, Geschichte Einer Geistigen Bewegung, Bd.I,

Vand. & Rup., Göttingen,1949, 6.Aufl.1984, S.43; as to Isocrates,cf. Bd.II, Erläuterungen, Vand. & Rup., Göttingen,1955,5.Aufl.1980, S.25. About the extraordinary importance of the conjunctive structure, a Greek grammarian in Japan says; Greek has a large number of means to coordinate sentences, and it is nearly one of@ the grammatical rules to use them to express the relation of reference to a preceeding sentence. Then, it is an exceptional case not to use them in juxtaposing sentences, only for the sake of expressing one's strong emotions. (Kozu,H., Elementary Greek Grammar, Hokuseido publ., Tokyo,1951, p.163). And only against this universal explicit connectivity of Greek the peculiar form of asyndeton in Greek could shape up; @@"ASYNDETON: Two or more sentences (or words) independent in form and thought, but juxtaposed, i.e. coördinated without any connective, are asyndetic (from asyndeton: not bound together), and such absence of connectives is called asyndeton. The absence of connectives in a language so rich in means of@ coördination as is Greek is more striking than in other languages." (Smyth, id.,p.484.)

*32) Theodectes' rhetoricai technai (Steph.Byz.s.Phaselis); Aristotle's technes tes Theodectou synagoge (D.L.V,24); "Aristotle collected the early books on rhetoric, even going back as far as Tisias, well known as the originator and inventor of@ the art; he made a careful examination of the rules of each author and wrote them out in plain language, giving the author's name, and finally gave a painstaking explanation of the difficult parts. And he so surpassed the original authorities in charm and brevity that no one becomes acquainted with their ideas from their own books, but everyone who wishes to know what their doctrines are, turns to Aristotle, believing him to give a much more convenient exposition. He, then, published his own works and those of his predecessors, and as a result we became acquainted with him and the others as well through his work." (Cicero, De Invent. II,ii,6-7)

*33) Lersch, L., Die Sprachphilosophie der Alten, dargestelt an der historischen Entwickelung der Sprachkategorieen, F.Baaden, Bonn,II, Geschichte der Entwickelung der Sprachkategorien, I. Die Redetheile A. Die Griechen, Theodektes,1840, S.24.

*34) Cf. Cicero, De Or. III,35,141.

*35) Strabo, XIII,1,54. Cf. Düring,id..

*36) Hülser, id., Nr.542.

*37) If we ask who, among the Stoics, succeeded Aristotle's grammatical system by transforming it, Zeno of Citium, just the founder of the school, is to be recalled, having made his approach to Aristotle through Theophrastus' popular handboook of rhetoric On The Diction (peri lexeos), which "was during the whole Hellenistic age a well-known and widely used handbook." Düring, id.,p.39. Cf. also Pohlenz,M., id. Bd.I,S.43-44; and Vogel, C.J.D., Greek Philosophy, III, E.J.Brill, Leiden,1964, p.104; and Pfeiffer, R., History of Classical Scholarship from the beginnings to the end of the Hellenistic age, OUP,1968, p.244.

*38) "The Stoics call the adverb pandectes, because it takes all upon itself like the Saturnian swallowing of everything." Charis. Inst. Gramm. II,13. Cf. Schmidt, R., Die Grammatik der Stoiker, Friedr. Vieweg & Sohn, Braunschweig,1979, S.134. Hülser, Nr.581.

*39) Cf. LIDDELL & SCOTT, id.,.s.v.

*40) Cf.p.10. And cf. also Robins, R.H., Dionysius Thrax and the Western Grammatical Tradition, Transactions of the Philological Society 1957 (pp.67-106), p.104.

*41) Dionysius of Halicarnassus, too, exposes in his On Literary Composition,14, the phonological tripartition, his terminology being absolutely identical with that of Aristotle. And he refers only to Aristoxenus, a pupil of Aristotle majoring in music. Perhaps, this is one, and a second (the first in the chapter 2 about the matter of arthron) of his highly rhetorical ways of implicit reference to Aristotle, the theoretical adversary from his point of view, to his most favorite practical rhetor Demosthenes.

*42) As to hemiphonon (semivowels) Dionysius of Halicarnassus (id.) as well as Dionysius of Thrace enumerates, besides s, r, l, m and n, the so-called double consonants (zd, ks, ps), which being not the Aristotelian semivowels, because zd, ks and ps have a so-called stop (d, k, p) that inhibits them from the continuous nonstop repetition of their whole utterance. And Dionysius of Thrace calls the semivowels: l, m, n and r also hygra (fluids, liquids), which will signify their characteristic competence for nonstop pronounciation. And@ it@ is this quality that makes them possible to adapt to the inflexional circumstances of phonology without losing their own property, whence they are called ametabola (unchangeable alphabet). Cf. DIONYSII THRACIS ARS GRAMMATICA, §6; BEKKER, Anecdota Graeca, p.632.

*43) SMYTH, id.,p.11.

 

##) We are greatly indebted to prof. Rudolf Kassel for the text: Aristotelis De Arte Poetica Liber, OUP, Oxford, 1965, which gave us the foundation for the real approach to Aristotle's grammatical thoughts.

###) Rhet. to Alex. is not by Aristotle, on the grounds of our arguments. For, its notice "the so-called articles" [1435a35, et passim] indicates clearly the definite article. Now, the origin of it is whether here in Rhet.to Alex. or in the Stoa. As a whole, the transformation of Aristotle's arthron (Preposition and Adverb) into the Stoic arthron (The Definite Article and The Relative Pronoun including Pronouns) can be visible in the correlative transformational bearings between the grammatical systems of the parties. Besides, the thoughts of logos (reason or wording) explicit in Rhet.to Alex. [esp. in its prologue] make us recall that of Zeno the Stoic [cf. "Die Logos philosophie" in Die Stoa by Pohlenz, M., S.32sq.], whose lectures open to the public in the public corridor ornamented with beautiful paintings (stoa poichile) might have been delivered to one of the Peripatetics, so abundant in number as to cause Zeno to mention; "The choir of Theophrastus is large, but mine is in better harmony." [SVF,I,280] Then he wrote down his own rhetoric to publish it in anonymity under the name of Aristotle, full of Stoic and Peripatetic ideas, his principal sources of@ the theory.

_______________________________________________________________________________________

© Koji Nihei Daijyo, 1991 - 2012. Tous droits réservés.

inserted by FC2 system