Trends in Classics –

Supplementary Volumes
Edited by
Franco Montanari and Antonios Rengakos

Scientiic Committee
Alberto Bernabé · Margarethe Billerbeck · Claude Calame
Philip R. Hardie · Stephen J. Harrison · Stephen Hinds
Richard Hunter · Christina Kraus · Giuseppe Mastromarco
Gregory Nagy · Theodore D. Papanghelis · Giusto Picone
Kurt Raalaub · Bernhard Zimmermann

Volume 9

De Gruyter
From Scholars to Scholia
Chapters in the History
of Ancient Greek Scholarship

Edited by
Franco Montanari
Lara Pagani

De Gruyter
 ISBN 978-3-11-025162-3
e-ISBN 978-3-11-025163-0
ISSN 1868-4785

Library of Congress Cataloging-in-Publication Data:

From scholars to scholia : chapters in the history of ancient Greek scholarship

/ edited by Franco Montanari and Lara Pagani.
p. cm. -- (Trends in classics. Supplementary volumes ; v. 9)
Includes bibliographical references and index.
ISBN 978-3-11-025162-3 (acid-free paper)
1. Greek philology--History. 2. Scholia. 3. Greek literature--Criticism,
Textual. I. Montanari, Franco. II. Pagani, Lara.
PA53.F76 2011
480--dc22
2010050354

Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche
Nationalbibliograie; detailed bibliographic data are available in the Internet
at http://dnb.d-nb.de.
© 2011 Walter de Gruyter GmbH & Co. KG, Berlin/New York
Typesetting: Michael Peschke, Berlin
Printing: Hubert & Co. GmbH & Co. KG, Göttingen
∞ Printed on acid-free paper
Printed in Germany
www.degruyter.com
 Pioneers of Grammar.
Hellenistic Scholarship and the Study of Language
Lara Pagani

The study of language represents only a minimal portion of the vast field
of knowledge and scholarship that in the Hellenistic period was desig-
nated by the term “grammar” (γραμματική, scil. τέχνη), which today
we use to define the study of the normative conventions of a linguistic
system on phonological, morphological, and syntactic levels. As some
ancient sources demonstrate, “grammar” in its full and complete sense1
was understood to be knowledge2 of literary compositions, accompanied
only in some cases by knowledge of what was said and thought in Greek
according to common usage.
We know that for Eratosthenes (ap. sch. Vat. D.T., in GG I/III 160.10-11)
grammar was a «complete hexis in letters, where “letters” signify literary

English translation by Johanna Hanink.

I would like to thank Stephanos Matthaios for carefully reading my typescript
and for his valuable advice, which allowed me to make improvements to these
pages and to avoid some inopportune omissions. I am also grateful to my col-
league Serena Perrone for discussing with me some crucial points. It is of
course the case that I alone am responsible for the content and opinions ex-
pressed here.
1 In this sense, it was distinct from the more elementary and restricted discipline
that taught reading and writing, for which the name γραμματιστική is attested
(see Ph. De congr. erud. gr. 146ff. = Chrysipp. Stoic. fr. 99 von Arnim; S.E. M.
1.44, 1.47, 1.52, 1.53, 1.54, 1.56; sch. Vat. D.T., in GG I/III 114.22-28; sch.
Lond. D.T., in GG I/III 448.12-16); on this point see Steinthal 1890-18912, II
175; Frede 1977, 52; Lallot 1995b, 74; Blank 2000, 402.
2 The nature of this knowledge, i.e. whether empirical or technical, was the
subject of differing opinions and even explicit polemics in antiquity (see just
below): see Pecorella 1962, 59-62; Siebenborn 1976, 116-139; Lallot 1995b,
78-79; Swiggers-Wouters 1995; Robins 1996, 6-10; Lallot 19982, 70-72; Lam-
bert 2000, 390-391.
18 Lara Pagani

compositions» (… γραμματική ἐστιν ἕξις παντελὴς ἐν γράμμασι,

γράμματα καλῶν τὰ συγγράμματα)3.
About a century later we have the famous definition of Dionysius
Thrax (ap. S.E. M. 1.57), according to which grammar is «empirical
knowledge, to the greatest extent possible, of things said by poets and prose
authors» (γραμματική ἐστιν ἐμπειρία ὡς ἐπὶ τὸ πλεῖστον τῶν παρὰ
ποιηταῖς τε καὶ συγγραφεῦσι λεγομένων); the phrase ὡς ἐπὶ τὸ πλεῖστον
has been interpreted as indicating grammar’s goal to be as exhaustive as
possible or, by a minority of scholars, as a limitation of the term
γραμματική: «grammar is, for the most part, etc.»4. The manuscripts of the
Techne grammatike attributed to Dionysius5 report this definition with a
slight but important variation (GG I/I 5.1-2): the adverbial phrase, here ὡς
ἐπὶ τὸ πολύ, occurs immediately before λεγομένων so as to indicate
«things usually said by poets and prose authors»6. According to Uhlig, Sex-
tus’ version to some extent distorts Dionysius’ original definition preserved
in the Techne, while Di Benedetto has argued that the definition found in
the Techne cannot be genuine, given that the restriction of what is «usually»
said by poets and prose authors undermines one of the purposes of gram-
mar indicated by Dionysius, namely to explain glosses and rare words: an
appeal to prudence on this point has been formulated by Lallot7.
Criticism of the empirical nature that Dionysius attributed to grammar
may have first been voiced by Ptolemy the Peripatetic (ap. S.E. M. 1.60-

3 On the historical bases, the theoretical and conceptual foundations of this defi-
nition, and the cultural backdrop against which it was formulated see Matthaios
2010a, who among other things conducts a meticulous investigation of the
meaning of the term ἕξις with regard to its usage in philosophical writings.
There it signifies the general concept to which τέχνη is subordinated: accord-
ing to Eratosthenes γραμματική was an epistemic condition that resulted from
or referred to the acquisition and mastery of a special science, i.e. that of writ-
ten compositions. I have proposed an Italian translation of the word ἕξις in this
context in PAWAG (www.aristarchus.unige.it/pawag), s.v.
4 Advocates of the first position have been Di Benedetto 1958-1959, 196-198;
Blank 1998, 124-125; Lallot 19982, 69; Di Benedetto 2000, 395; for the second
see Uhlig 1882, 73.
5 On the debate about the work’s authenticity see infra, 30-37.
6 See Lallot 1995b, 75 and n. 6; Lallot 19982, 43, 69-70 on the other hand, Patil-
lon 1990, 693-694 maintains that Dionysius was referring to «texts, most often
those of poets and prose authors».
7 Uhlig (in GG I/I 5.1-2, ad loc.); Di Benedetto 1958-1959, 196-198; Lallot
19982, 69. A discussion of the various possible forms, positions and functions of
the syntagm ὡς ἐπὶ πολύ / ὡς ἐπὶ τὸ πλεῖστον can be found in Ventrella
2004, who attributes a temporal value to the phrase (i.e. something similar to
πολλάκις), and therefore translates Dionysius’ definition as follows: «la gram-
matica è pratica ripetuta/esercitata di ciò che si dice presso poeti e prosatori».
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 19

61), about whom we know little, but who should probably be assigned to a
period shortly after Dionysius himself8.
Openly critical of Dionysius’ position was a certain Asclepiades, gener-
ally agreed to be the scholar from Myrlea who lived between the second
half of the IInd and the Ist centuries (see infra, 20 and n. 17). Asclepiades
maintained that grammar was «a techne of things said by poets and prose-
authors» (ap. S.E. M. 1.74); he thus advanced beyond the notion that
grammar was an inherently empirical art (conjectural and subject to acci-
dents, in the same way as e.g. navigation and medicine); his definition also
established that grammar was comprehensive, and thus could account for all
utterances of the literary authors (ap. S.E. M. 1.72-73). In this different
conception of grammar Di Benedetto has recognised hints of the beginning
of a profound renewal of the grammatical discipline in the course of the Ist
century9: we shall return to this point.
Sextus (M. 1.76) attributes another definition to Chares (Χάρης), who
should probably be identified with the Chairis (Χαῖρις) active in about the
Ist century B.C. and cited in the scholia to the Techne (sch. Vat. D.T., in
GG I/III 118.9-11) for the same definition10. This definition identified
grammar «in its complete form» as a «hexis that, beginning from a techne, al-
lows one to determine as precisely as possible those things said and thought
by the Greeks, except for as far as the other arts are concerned» (... τὴν
τελείαν ... γραμματικὴν ἕξιν εἶναι ἀπὸ τέχνης διαγνωστικὴν τῶν παρ᾿
Ἕλλησι λεκτῶν καὶ νοητῶν ἐπὶ τὸ ἀκριβέστατον, πλὴν τῶν ὑπ᾿ ἄλλαις
τέχναις), thus eliminating grammar’s restriction to literature11.
Similarly, Demetrius Chlorus (era unknown, perhaps the Ist century
B.C.) spoke of grammar as a «techne of the things <said> by poets and
<prose-authors and> knowledge of the words which belong to common
usage» (τέχνη τῶν παρὰ ποιηταῖς τε καὶ <συγγραφεῦσι λεγομένων
καὶ>12 τῶν κατὰ κοινὴν συνήθειαν λέξεων εἴδησις).
Another definition, which is of an Aristotelian character and goes back
to Tyrannion13 (Ist century B.C.), should be mentioned in this context: this
definition described grammar as the «theory of mimesis» (γραμματική ἐστι
θεωρία μιμήσεως, sch. Vat. D.T., in GG I/III 121.17 = fr. 57 Haas). While

8 So Siebenborn 1976, 105 n. 1; less convincing are the arguments of Dihle 1959
in favour of a date in the IInd-IIIrd centuries A.D.
9 Di Benedetto 1958-1959, 199-200.
10 So first Blau 1883, 56-57, followed by Di Benedetto 1958-1959, 198 n. 2;
Lallot 1995b, 79 n. 15; Blank 1998, 137-138 and n. 105; contra Berndt 1902, 3-
18 and 25-28.
11 See Blank 1998, 137-139; for the meaning of ἕξις, see supra, n. 3 regarding
Eratosthenes.
12 The integration goes back to Di Benedetto 1966, 322 and is accepted by Lallot
1995b, 79 and n. 16, but not by Blank 1994, 146 n. 23, who considers it su-
perfluous.
13 On his role in the study of language in antiquity see infra, 21 n. 20 and 61.
20 Lara Pagani

this formulation was once considered suspect on the grounds that the text
of the scholion may be damaged or incomplete14, Tyrannion’s idea was
later explained by means of comparison with Arist. Rh. 1404a 21 (τὰ γὰρ
ὀνόματα μιμήματά ἐστιν): if language is a μίμησις, then grammar is the
theory of this μίμησις15.
As a result, only a few of the ways in which the grammatical discipline
manifested itself were of a linguistic nature: of the six μέρη of grammar
identified by Dionysius Thrax16, only the «study of etymology» and «ac-
counting of analogies» can be considered grammar as such. The later
tripartite model of Asclepiades of Myrlea, on the other hand, dedicated a
specific section of grammar (μέρος τεχνικόν) to rules concerning letters,
parts of speech, orthography, and hellenismos (S.E. M. 1.91, 1.252)17; it is
upon this and the historical part (focussed on individuals, geographical
and chronological information, myths, glosses, and proverbs) that the
grammatical part relies (S.E. M. 1.252) for carrying out exegesis, textual
criticism, and judgments about the authenticity of texts. We know,
moreover, that Tauriscus, a disciple of Crates, also dedicated a section of
the κριτική (as he called the science of letters) to language and to
grammatical features (μέρος λογικόν, S.E. M. 1.248)18. Finally, a strictly

14 So Steinthal 1890-18912, II 177.

15 Haas 1977, 167-168. For a complete study of the epistemological constitution
of γραμματική by means of analysing ancient definitions of and observations
relating to it, see Prencipe 2002.
16 According to Dionysius (S.E. M. 1.250; cf. GG I/I 5.3-6.3) the six parts of
grammar are: reading aloud masterfully and in accordance with prosody; expla-
nation of poetic expressions in the text; interpretation of glosses (λέξεις is here a
synonym for γλῶσσαι: see Blank 1998, 263) and of histories; discovery of
etymology; accounting of analogies; and critical evaluations of literary works.
17 At § 252 Sextus explains the Asclepiadean tripartition (μέρος τεχνικόν,
ἱστορικόν and γραμματικόν), discussing at length the subdivision of the μέρος
ἱστορικόν; a synthesis of the aspects involved in each individual part occurs at §
91: although here there is no mention of Asclepiades, it is usually agreed that
the content of this section is owed to him, given the consistent correspon-
dences (the only discrepancy is purely terminological: the μέρος ἰδιαίτερον of §
91 is evidently the same as the μέρος γραμματικόν of § 252): see Di Benedetto
1958-1959, 203 and n. 3; 1973, 806; Blank 1998, 148, 265. It is generally ac-
cepted that the Asclepiades cited three times in the Adversus mathematicos, al-
ways without ethnic epithet, is the scholar from Myrlea (this identification is
oftentimes not even questioned; arguments in its favour are found in Wentzel
1896, 1630 and Rispoli 1988, 179 n. 19, who has moreover refuted [183-184
and n. 33] the hypothesis proposed only by Slater 1972, 331-332 that this As-
clepiades should be identified with the Bithynian doctor).
18 The other two parts identified by Tauriscus were the μέρος τριβικόν, concern-
ing dialects and the differences of styles and registers, and again the μέρος
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 21

technical component also figures in the quadripartite model of gram-

matike reconstructed by H. Usener on the basis of traces found in the
scholiastic literature on the Techne grammatike19 and in Latin authors; this
model was attributed – in the hypothesis by Usener, which was later
rejected – to Tyrannion20. According to that hypothesis, grammar has
four μέρη or ἔργα (ἀναγνωστικόν, ἐξηγητικόν, διορθωτικόν and
κριτικόν) and as many ὄργανα (γλωττηματικόν, ἱστορικόν, μετρικόν
and precisely τεχνικόν)21.
In this “restricted” sense of attention to linguistic phenomena,
grammar’s roots in the Greek world lie deep in the philosophical tradi-
tion22. There the study of grammar began with discussions, first held by
the Presocratics, concerning the question of whether language exists
naturally (φύσει) or by convention (θέσει or νόμῳ), as well as about the
related problem of the “accuracy” of names, that is whether they corre-
spond to the reality that they signify23.

ἱστορικόν, which had to do with the material that was not systematically or-
ganised. For a hypothesis as to the relationship between the divisions made by
Asclepiades and Tauriscus see Blank 1998, 264-265.
19 Usener 1892 (= 1913): the references to the passages taken into consideration
are on pp. 582 (= 1913, 266) n. 1, 584 (= 1913, 267) n. 9, 587 (= 1913, 269)
n. 15.
20 The system outlined by Usener was based at all levels upon a quadripartite
arrangement. It was thought plausible that Tyrannion could indeed have been
the creator of a theoretical framework such as this because, in addition to the
suitability of his epoch and education, of the evidence provided by his distinc-
tion – deemed artificial – of four types of accent (cf. fr. 59 Haas). The objec-
tions rested on the fact that such a distinction not only would not have been
arbitrary, but indeed was not Tyrannion’s invention; moreover he made use, at
least for the typology of the name (fr. 56 Haas), of a tripartite structure: see
Wendel 1943, 1818; Haas 1977, 171-172; Fehling 1979, 489, who attributes
the establishment of this model rather to Trypho. See also Ax 1987, 30-32.
21 For a comparison of the various ancient divisions of grammar and a discussion
of these, see Steinthal 1890-18912, II 181ff.; Müller 1903, 29-46; Barwick
1922, 215ff.; Mette 1952, passim; Ax 1982, 96-97 and recently Bravo 2006,
248-254, with bibliographical references.
22 For this phase of ancient grammar here it should suffice to mention only a few
fundamental points; for a detailed overview see Schmitter 1991, 57-272; Hen-
nigfeld 1994, 4-124; Arens 2000; Blank 2000, in particular 400-404; Schmitter
2000; Sluiter 2000a; Law 2003, 13-51; Frede/Inwood 2005 with further bibli-
ography. A well-documented, though dated, synthesis can be found in Gude-
man 1912, 1781-1791.
23 Even if not yet formalised, the issue of the relationship between language and
reality may be glimpsed in a few fragments of Heraclitus (e.g. 22 B23, 32, 48,
67 D.-K.), Parmenides (28 B8.38-41 and 19 D.-K.), Anaxagoras (59 B17 and
22 Lara Pagani

An intense interest in the area of linguistics was also demonstrated

by the sophists, who looked to assess the relationships between word,
thought, and reality, and in doing so both gave rise to a few morpho-
logical classifications and laid the foundations for the art of rhetoric: in
particular, Protagoras affirmed the conventional nature of language and
distinguished the three genders of substantives (80 A27 D.-K.), along
with four verbal moods (optative, subjunctive, indicative and impera-
tive) corresponding to the types of phrases (80 A1 D.-K.; cf. A29).
Prodicus is well known as the founder of the scientific study of syno-
nyms (84 A16, 17 and 19 D.-K.), and fundamental to Gorgias’ art of
rhetoric was the notion of logos as a μέγας δυνάστης capable of manipu-
lating listeners’ opinions (82 B11.49ff. D.-K.).
The controversy surrounding the accuracy of names was later seized
upon and examined in detail by Plato in his Cratylus, and to Plato are
also owed a number of other important contributions in the area24.
These include phonetic classifications and certain observations on the
structure of logos, such as the distinction between ὄνομα and ῥῆμα (Sph.
261d): this distinction represents the original nucleus of analysis of the
parts of speech which, as we shall see, constitutes one of the fundamen-
tal themes of Greek grammatical studies. As to phonetics, we know that
Plato identified three classes of letters (vowels, consonants, and mutes:
Cra. 424c) and two types of accents (acute and grave: Cra. 399b).
Aristotle was then responsible for the systematisation of various lin-
guistic concepts and terms25: worth mentioning is Poetics §§ 20-21,
where we find his extensive classification of the parts of linguistic ex-
pression (letter, syllable, conjunction, noun, verb, connecting word,
inflection, discourse: 1456b 20-1457b 30) and of word forms (normal,
composite, gloss, metaphorical: 1458a 18-1459a 14), all formulated with
a view to determining language’s, and especially poetic language’s,

19 D.-K.), Empedocles (31 B8 and 9 D.-K.), Democritus (68 A37, B5, 9, 26

and 125 D.-K.), to whom Thrasyllus (68 A33 D.-K.) also attributed works on
orthoepia, onomastics, and ῥήματα.
24 The investigation of language pervades the Platonic oeuvre, but it is the specific
subject of (in addition to Cratylus) the Theaetetus and Sophist.
25 As we shall see, recent studies emphasise the Aristotelian influence on Alexan-
drian scholars, even when it came to linguistics in particular and not only in
general terms, as has now been established, in contrast with Pfeiffer 1968, by
Montanari 1993, 262-264 (with bibliography at nn. 63 and 67) and Montanari
1994 (in partic. 7-28 [= Richardson 1994], 29-38 [Discussion] and 361-364
[Conclusioni]). More recently see Schironi 2009, with bibliography.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 23

merit26. Also noteworthy are his studies of inflection and the nature of
the verb, which is defined as something that carries an idea of time and
is a sign of predication (Int. 16b 6ff.); his introduction of the concept of
«proposition» (λόγος ἀποφαντικός) in the De interpretatione; his elabora-
tion of the idea of ἑλληνίζειν (Rh. 1407a 19ff.), i.e. correct expression in
Greek (involving e.g. exact placement of conjunctions, use of appropri-
ate nouns, avoidance of ambiguity, use of correct agreement of gender
and number, production of texts that are easy to read and pronounce), as
opposed to σολοικίζειν (i.e. the use of inappropriate terms); as well as
his development of a model of communication in the Rhetoric.
Stoic philosophy, too, played a decisive role in the study of language
and was responsible for investigations of aspects of phonetics, semantics,
and syntax, as well as for the establishment of a good deal of terminol-
ogy that would persist in use. In particular, Stoicism’s contributions
comprised certain observations on φωνή (thanks especially to Diogenes
of Babylon), the doctrine of inflection and of tenses, and the relationship
between linguistic form and concept, including the introduction of the
notion of anomaly to indicate the discrepancy between signifier and
signified (Chrysippus)27.
At a certain point, alongside this philosophy’s tradition of linguistic
observation, the study of grammar in the field of philology and scholar-
ship came into being and the two entered into a relationship of mutual
influence. With regard to the latter, modern critics took up antithetical
positions as to the question of whether to accept the idea that a system
of grammatical rules (even if still in nuce) developed as early as with fig-
ures such as Aristophanes of Byzantium (ca. 265/57-190/80), Aris-
tarchus of Samothrace (ca. 215-144) and Crates of Mallus (first half of

26 See Lersch 1838-1841, II 257-280; Ax 1987, 35-37 and Swiggers/Wouters

2002c. A general account of Aristotle’s linguistic observations can be found in
Ax 1993, 12-15.
27 The canonical reference work on the Stoic theory of grammar is Schmidt
1839; a good deal of space was also devoted to this topic by Steinthal in his
overview of ancient linguistics (1890-18912, I 271ff.); the importance of this
philosophical current to the foundation of Greek grammatical theory was par-
ticularly emphasised by Pohlenz 1939. For more recent studies of the Stoic ap-
proach to language see the bibliography of Sluiter 2000a, to which should be
added Barwick 1957; Pinborg 1975, 77-98 and, on φωνή, Ax 1986; see also
Schenkeveld 1990a and Ax 1993 for the Peripatetic influences on the linguistic
observations traceable to the ancient Stoa. Here I avoid an account (even a
merely informative one) of Stoic advances in the study of linguistics: when
necessary the relevant aspects will be discussed in the course of the paper.
24 Lara Pagani

the IInd century)28. The differences of critical opinion are owed largely
to the difficulty of interpreting the surviving evidence, which is second-
hand, preserved fragmentarily, and generally in need of serious critical
examination29, which sometimes involves evaluations of authenticity30.
For these reasons and in the light of the recent studies that have organ-
ised an important portion of the ancient material and contributed inter-
esting developments to the subject31, now appears to be an opportune
moment to propose a synthesis of current problems and critical de-
bates32.
As a matter of preliminaries, it seems appropriate to state that the
linguistic arguments of the Hellenistic scholars occur along two primary
lines: on the one hand, there is the doctrine of the constitutive elements
of language, with the crucial question being that of the parts of speech;
on the other there is the issue of linguistic correctness (ἑλληνισμός) in
relation to individual words (their orthography, prosody, inflection, and
meaning) and to constructions (by means of the criteria of analogy, ety-
mology, dialect, usage, and the literary tradition)33.
The first seeds of this type of study can be detected, beginning in the
XIXth century, in overviews of the origins of the study of grammar in
the ancient world and in studies, albeit partial and non-systematic ones,
of Aristarchus’ approach to language, according to what could be
gleaned from the fragments of his work on Homer. As to studies with a
more general scope, at least those of J. Classen (1829) and L. Lersch
(1838-1841) should be mentioned here. Classen addressed – from a

28 Crates was a contemporary, though probably a slightly younger one, of Aris-

tarchus: it is plausible that their work overlapped and that contacts between the
Alexandrian and Pergamene schools were more substantial than their traditional
rivalry has led us to believe (see Broggiato 2001 [= 2006], XVII-XIX).
29 See the apt formulation of Ax 1982, 98 («… es scheint, als würde die Beleglage
jeden Schluß zulassen»); for a close examination of the difficulties specific to
this area of study, see also Ax 1991, 276-277.
30 I am of course alluding to the vexata quaestio relating to the Techne grammatike
attributed to Dionysius Thrax: I shall provide an account of the problem
shortly, making certain adjustments concerning its importance to my own ar-
gument.
31 This is above all a reference to S. Matthaios’ work (which I address below) on
the doctrine of the parts of speech in Aristarchus, and more generally amongst
the Alexandrians.
32 The bibliography is fairly vast: here I retrace a selection of that which I deem
most relevant (unless I am guilty of omissions) to the history of scholarship on
the subject.
33 See Barwick 1922, 227ff.; Siebenborn 1976, 32ff.; Ax 1982, 97; 1991, 277-
278; Matthaios 1999, 15-16.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 25

diachronic perspective – both the philosophical and philological côtés

(pp. 79-85) of linguistic study in antiquity, hinging the latter on a sup-
posedly sharp and polemical contrast between the Alexandrian and Per-
gamene schools. He considered the content of the Techne to be genu-
inely Dionysian and therefore a witness to the doctrine of Aristarchus,
and he further provided an account of the evolution of a grammatical
terminological apparatus that, beginning from an imperfect and nascent
form in Aristotle, was enriched by the Stoics and thence reached the
Alexandrian grammarians to whom – especially Aristarchus – most of its
merit was due. Lersch’s volume, on the other hand, combined an his-
torical treatment (discussing Greek, Roman and later philosophers and
grammarians) of the ancient debate between linguistic analogy and
anomaly34 with both a taxonomy of the parts of speech considered in
terms of their development and a history of etymology among the
Greeks and Romans.
An interest, albeit an incidental one, in the linguistic aspects of Alex-
andrians’ philology is evident in the studies of K. Lehrs (1837 and 18823
[Ist ed. 1833]), where space is accorded to observations made by the
grammarians and Aristarchus on prosodic issues in Homer35. Also strictly
related to work on Homer was the research of L. Friedländer, the author
of a collection of Fragmenta schematologiae Aristarcheae (1853), a study of
the systems that Aristarchus adopted so as to manage the particularities
and peculiarities of Homer’s language36; closely connected to Homer as
well was the work of J. La Roche, to whom we owe a rich collection –
although one that does not yet attempt a methodical illustration of lin-
guistic principles and argumentation – of evidence for the Alexandrians’
explanations of prosodic problems and specific forms in Homer’s text
(1866)37. The ambition to conduct a systematic analysis of the grammar

34 On this subject see infra, 27 and nn. 41 and 44, 28 and n. 46, 29, 38-39, 51-52,
54.
35 Lehrs 1837, 35-166 (Dissertatio II: Capita selecta ex Alexandrinorum doctrina de
prosodia Homerica); 18823, 247-327 (Dissertatio IV, De prosodia: 1. De accentibus; 2.
De interaspiratione; 3. De spiritu vocabulorum principali).
36 The reconstruction of the system can be found in the collection of the Iliadic
fragments of Aristonicus: Friedländer 1853, 1-35. According to Erbse 1980,
242-244, this type of approach to Homeric language, of which he studies a few
illustrative cases, involved a search for regular repetitions amongst alleged ex-
ceptions and presupposed grammatical knowledge such as of the paradigms of
declination and conjugation, as well as an awareness of syntactic functions,
word forms, and particle usage.
37 La Roche 1866, 175-432. This is effectively a collection in alphabetical order
of individual Homeric terms, which in each case investigates the available in-
formation relating to textual-critical decisions made by the Alexandrians, on
26 Lara Pagani

of Aristarchus became evident for the first time in a piece by W. Rib-

bach (1883), which right from its title presented itself as a study of Aris-
tarchus’ ars grammatica. The material that he examined was nevertheless
restricted to Aristarchus’ application of the principle of analogy and to
his observations on orthography, inflection (verbal and nominal) and
prosody that could be derived from the fragments of his Homeric exe-
gesis38; it thus excluded matters related to the doctrine of the parts of
speech39.
It is within this climate of study that we should situate the overview
of the history of Greek and Roman linguistics outlined by H. Steinthal
(1890-18912 [Ist ed. 1863]), in which attention was also paid to Hellenis-
tic philology, both Alexandrian and Pergamene. Regarding Zenodotus,
Steinthal wrote of his «less than finely honed grammatical awareness»
and maintained the absence from his work of fixed rules for the con-
struction of word forms, as well as for differences between dialects, and
that which was characteristically Homeric40. He did however attribute a
more important role to Aristophanes of Byzantium, on the basis of the
ancient evidence that associated him with Crates of Mallus and Aris-
tarchus when it came to the elaboration of grammar (S.E. M. 1.44). In
particular, Steinthal cited Aristophanes’ collection of Λέξεις as the be-
ginning of a methodical process, and examined his approach to the prin-

the basis of their discussions of grammatical phenomena. The nature and scope
of this study are sensibly evaluated by Matthaios 1999, 26.
38 It was to Ribbach’s merit that he both rigorously arranged and commented
upon some of the more important testimonia on the topic (including the frag-
ments transmitted by Varro). For analogy, he sought chiefly to clarify the set of
criteria of similarity attributed to Aristophanes and Aristarchus; when it came to
inflection he concluded that a complete doctrine on the matter was achieved
only by the pupils of Aristarchus, who according to Ribbach used the principle
of analogy mostly in determining the correctness of forms.
39 According to Di Benedetto 1958 (206 n. 6) and 1959 (118), such a limitation
of content was symptomatic of the fact that Aristarchus’ grammatical work did
not advance any further: Di Benedetto maintains that in fact it was centred ex-
clusively on the study of Homer and therefore did not allow for extrapolation
of an autonomous grammatical system. Nevertheless, this lacuna has been filled
in recent years by Matthaios 1999, who, beginning with a documented collec-
tion of sources, has conducted a reconstruction of the Aristarchean doctrine of
the parts of speech, as shall be discussed below (see Matthaios 1999, 26-27 and
192-193 as well for an evaluation of Ribbach’s work).
40 Steinthal 1890-18912, II 73-77 («… weil sein [scil. Zenodots] grammatisches
Bewusstsein noch wenig geschärft war, weil er noch keine feste Regeln über
den Bau der Wortformen, über die Unterschiede der Dialekte, über das eigen-
tümlich Homerische hatte, um nach ihnen zu bestimmen, was richtig oder
falsch ist»).
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 27

ciple of analogy as a grammatical rule41. Steinthal’s principal interest,

however, was in Aristarchus, whom he credited with a vast knowledge
of language, but not yet with a complete grammatical view nor with a
complete theory of forms and constructions, which would be established
only later by his pupils42. Indeed, in Steinthal’s opinion, Aristarchus
provided these pupils with little more than the principle upon which
they could found their grammatical studies, namely the criterion of
analogy43. This criterion regulated the system of declination and conju-
gation, which according to Steinthal Aristarchus knew only in its sim-
plest form, i.e. the comparison of two members, and not in the com-
plete form of four-member proportions44. Steinthal also reconstructed

41 Steinthal 1890-18912, II 78-82. Steinthal however denied that Aristophanes

was aware of the conditions of analogy (II 151 n.; cf. II 81 n.) and rejected (81-
82 n.) the thesis of Nauck 1848, 264-271, according to which Aristophanes de-
fended the regularity of the doctrine of inflection and formulated conditions for
the retrieval of analogous inflectional forms in a specific monograph Περὶ
ἀναλογίας. Because Nauck’s hypothesis is not supported by ancient evidence it
has then been largely rejected: see Pfeiffer 1968, 202-203; Callanan 1987, 107;
Ax 1990, 12; 1991, 282.
42 Steinthal 1890-18912, II 82-111. Steinthal’s discussion of Aristarchus’ linguistic
competence concerned accentuation, the doctrine of forms, syntax, and the
parts of speech.
43 Steinthal 1890-18912, II 112-113.
44 Steinthal 1890-18912, II 103; so later also Siebenborn 1976, 71-72; contra Erbse
1980, 237-240; Ax 1991, 284 (see infra, 53-54). Scholars usually distinguish dif-
ferent applications of the analogical method, on the basis of the degree of re-
finement and elaboration in the juxtaposition of an uncertain form with a nor-
mative model of reference. The comparison of two members provides for the
simple juxtaposition of two words: when a form is uncertain – whether in
terms of orthography, prosody, sequence of phonemes, or inflection – one can
quote as a model a form that is known with respect to that particular aspect, i.e.
that is “analogous” to the first according to specific criteria (about the condi-
tions on the basis of which two words are considered “analogous”, see infra,
49-50). The grammatical features of the first are then established by means of
the features of the second (Siebenborn 1976, 64 mentions as an example of this
type of analogy sch. Lond. D.T., in GG I/III 454.20-21: καὶ τὸ πηρός
ὀξυτόνως δεῖ ἀναγινώσκειν, ὡς τὸ πηλός). A more sophisticated procedure al-
lows for the comparison of two “analogous” base words with two respective
inflected or derived forms of the words, where the result is a four-member
proportion: one of these members is normally unknown and must be deter-
mined on the basis of the others (Siebenborn’s example, loc. cit., is S.E. M.
1.197: ζητουμένου γὰρ τοῦ πῶς δεῖ λέγειν, χρῆσθαι ἢ χρᾶσθαι, φασὶν ὅτι
χρᾶσθαι, καὶ ἀπαιτούμενοι τούτου τὴν πίστιν λέγουσιν, ὅτι χρῆσις καὶ
κτῆσις ἀνάλογά ἐστιν· ὡς οὖν κτᾶσθαι μὲν λέγεται, κτῆσθαι δὲ οὐ λέγεται,
οὕτω καὶ χρᾶσθαι μὲν ῥηθήσεται, χρῆσθαι δὲ οὐ πάντως: the correct form
28 Lara Pagani

Aristarchus’ position regarding the doctrine of the parts of speech on the

basis of the content of the Techne grammatike, which he believed to be a
genuine work by Dionysius Thrax (ca. 170-90), an Aristarchean, and
which he thus treated as a valid point of reference45. Finally, Steinthal
appraised Crates of Mallus in antithesis to Aristarchus, pigeonholing him
as a supporter of anomaly in the dispute with the analogists46: counting
Crates more a Stoic philosopher interested in literary history than a phi-
lologist, Steinthal hesitated to ascribe to him any serious or even occa-
sional interest in grammatical subjects, and evaluated his competence in
this area by comparison with Aristarchus47.
The notion that the earliest Alexandrians did not elaborate a system
of parts of speech but rather that Dionysius Thrax was the first to do so
in the Techne also formed the basis of K. Barwick’s study on the Roman
Ars grammatica (1922). This work outlined, among other things, a pic-
ture of a bipartite grammatical tradition, consisting on the one hand in
the Alexandrian techne and on the other the Roman ars, both stemmed
from a Stoic-Pergamene model which itself derived for Barwick from
Diogenes of Babylon’s techne48.

between χρῆσθαι and χρᾶσθαι is established thanks to the following propor-

tion: κτῆσις : κτᾶσθαι = χρῆσις : χρᾶσθαι). The next step is towards greater
abstraction and generalisation: when confronted with a form’s dubious ortho-
graphical, prosodic, or inflectional features, one subsumes the form under the
rule to which it belongs (κανών) and as a result can extract a conclusion about
the problematic grammatical feature (cf. An.Ox. IV 331.32-332.1: ... ἐν τῷ
ἡμερινὸς ἣ νυκτερινός· τὸ ρι ἰῶτα· ἐπεὶ τὰ διὰ τοῦ ῥινὸς καιροῦ
παραστατικὰ διὰ τοῦ ι γράφεται: faced with uncertainty as to whether
ἡμερινός and νυκτερινός are written with ι or with ει, recourse is made to the
rule that temporal indicators terminating in -ρινος are written with an ι: see
Siebenborn 1976, 67).
45 See e.g. Steinthal 1890-18912, II 211. We shall soon see how the discussion of
the Techne’s authenticity has been one of the critical points in the debate as to
the linguistic notions of Aristarchus and his contemporaries.
46 Steinthal posed this debate as a central factor in the development of ancient
linguistic theory (see also Colson 1919), but it has since been drastically put
back into perspective (Fehling 1956, 264-270 has put the historicity of the dis-
pute in doubt, arguing that it may have been constructed by Varro). A history
of the analogist/anomalist controversy in modern scholarship can be found in
Pinborg 1975, 106-110; Siebenborn 1976, 2-13; Blank 1982, 1-4 and Taylor
1987, 6-8, who support Fehling’s thesis; Ax 1991, 289-295; Blank 1994;
Schenkeveld 1994, 283-287; Broggiato 2001 (= 2006), XXXIIII-XL, with
bibliography, and Blank 2005.
47 Steinthal 1890-18912, II 121-126.
48 This thesis was revised by Calboli 1962 and refuted in favour of an unitarian
perspective according to which the Alexandrian techne descended from the
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 29

Precisely in reference to the Pergamene context, H. J. Mette (1952)

conducted an analysis of the linguistic theories of Crates of Mallus, em-
ploying a rather broad notion of what constituted a fragment. Mette
used large passages from Varro and Sextus Empiricus in his reconstruc-
tion of Crates’ ideas49. According to Mette, first, Crates saw linguistic
anomaly as the fundamental irregularity of κλίσις, subject to no system
of rules, and so set himself in opposition to the Alexandrian view that
favoured linguistic analogy; and second, he identified observation of
linguistic usage (παρατήρησις τῆς συνηθείας) as the means for estab-
lishing hellenismos50. For Mette, the methodological foundations of these
positions could be derived from the medical empirical school, fact that
would locate the quarrel between the analogists and anomalists within
the broader framework of the opposition between τέχνη and ἐμπειρία
in Hellenistic science51. As to the doctrine of the parts of speech, he
credited Crates with a complete and systematic subdivision, but on a
purely speculative basis52.

Stoic Diogenes of Babylon, and thus the Roman artes did so directly as well (so
too Pinborg 1975 [see infra, n. 84] and Siebenborn 1976, 138-139, 162-163).
Along the same lines as Barwick should be placed Frede 1977 (see infra, n. 93)
and in essence Ax 1986, yet with adjustments (see infra, 44), and Matthaios
2002, in particular 190-191 and n. 118, with restriction to the doctrine of the
parts of speech.
49 Mette 1952 presented as Crates’ fragments everything that might complete and
clarify the material strictly attributable to the ancient grammarian (e.g. Varro
ling. 7.109-10.84 [fr. 64a Mette] and S.E. M. 1.148-154 and 175-247 [fr. 64e
M.]). An evaluation on the nature of this work can be found in Broggiato 2001
[= 2006], XIII-XIV.
50 Mette 1952, 9-11 and 31-45. Crates’ presumed exclusive use of the terms
παρατηρεῖν and παρατήρησις is a controversial claim of Mette’s study (see
Broggiato 2001 [= 2006], with earlier bibliography).
51 Mette 1952, 45-48. This connection, too, is not without its difficulties (see
Siebenborn 1976, 118ff. and Broggiato 2001 [= 2006], XXXVII-XXXVIII).
52 Mette 1952, 20-21: the presumed Cratetean system was deduced by Varro ling.
8.44-84 and should have included ὄνομα in a broad sense (further divided into
a more restricted ὄνομα [in turn separated into ὡρισμένον = κύριον ὄνομα and
ἀοριστῶδες ὄνομα = προσηγορία] and ἄρθρον [this too divided into
ὡρισμένον ἄρθρον = the article and ἀοριστῶδες ἄρθρον = ἀντωνυμία]),
ῥῆμα, ἐπίρρημα (μεσότης), σύνδεσμος. More recently, Janko 1995a has pre-
sented, as dating back to Crates, a collection of parts of speech and other
grammatical terms that may demonstrate a close proximity with Diogenes of
Babylon’s doctrine of φωνή: it covers accent terminology (περισπᾶσθαι,
ἄνεσις, ἐπίτασις); the separation of letters into φωνήεντα, ἡμίφωνα and
ἄφωνα; the definition of features of letters, syllables and words as πρόσπνευσις
and ψιλότης, ἔκτασις and συστολή, πρόθεσις and πτῶσις; the parts of the
30 Lara Pagani

The significance of the analogists/anomalists dispute was de-

emphasised by D. Fehling (1956), according to whom Crates simply
dismissed the value of analogy for determining hellenismos, and substi-
tuted συνήθεια in its place53. Moreover, Fehling maintained that the
Alexandrians themselves, in their earliest stages, fully developed a sys-
tematic doctrine of inflection: grammatical science in general reached its
peak in the age of Aristarchus, with its principal tenets then rapidly be-
coming fixed in the hands of his disciples54. More than twenty years
later, Fehling recanted on this formulation (1979), which had been
firmly criticised by E. Siebenborn in his study on the doctrine of linguis-
tic correctness (1976)55. Fehling now embraced the thesis of a later de-
velopment of grammar, especially in the wake of the suggestion raised in
the meantime by the work of V. Di Benedetto that the Techne attributed
to Dionysius Thrax was spurious (1958-1959 and 1973).
The contested issue of the Techne grammatike’s authenticity is consid-
ered a decisive point for our understanding of the origins and first de-
velopments of grammatical theory in the Greek world: the Techne is a
handbook which contains a systematic description of language that, if
really the work of one of Aristarchus’ pupils, would testify to an already
fully evolved awareness of grammatical features in the period. Neverthe-
less, Dionysius’ authorship of the Techne, which was both accepted and
rejected by scholars of the XVIIIth and beginning of the XIXth century56
and then canonised by M. Schmidt (1852-1853)57 and by G. Uhlig’s

speech ὄνομα, ῥῆμα, πρ[όθεσις] and σύνδεσμος). These are attested in the so-
called “Treatise B” of Philodemus (particularly in fragments of PHerc. 460 and
1073), which seems to summarise and critique an important adversary’s doc-
trine of euphony: that this adversary was Crates was no more than a conjecture
by Janko himself (Janko 1995b, 89-92, where he proposes among other things
[pp. 73-87] a new ordering of the papyrus fragments in the various books of
the Περὶ ποιημάτων on the basis of an innovative method of reconstruction
[pp. 70-73]), which he later abandoned (Janko 2000, 182-187) so as to assign
all of the material to Pausimachus, one of the «critics» cited by Philodemus.
That these texts do not have to do with Crates is assumed by the edition of
Broggiato 2001 [= 2006], in which they do not appear.
53 Fehling 1956, 268-269. See supra, n. 46.
54 Fehling 1956, 214, 247-248 n. 1, 260-261.
55 Siebenborn 1976, 11-12, 71. See infra, 38-39.
56 There is a summary in Di Benedetto 1958-1959, 169-170.
57 Schmidt 1852-1853: the study, based on an analysis of both the manuscript and
indirect tradition, sealed the work’s judgment as authentic for over a century.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 31

edition in Grammatici Graeci (1883)58, was first put into serious doubt by
V. Di Benedetto in 1958-195959. It must initially be stated that § 1 of
the work is not suspected of being a later addition, since it is also cited
by Sextus Empiricus (M. 1.57, for the definition of grammar and 250 for
its subdivision into parts), nor presumably are the next sections up to
and including 4, if they are indeed connected to the first section (their
subject is in fact ἀνάγνωσις, which is mentioned in Sextus’ testimo-
nium as the first part of grammar)60. Section 5 contains a discussion on
rhapsody, which is perhaps interpolated and is in any case out of con-
text61; at § 6 the technical exposition begins, which concerns letters,

58 Uhlig proposed an extensive series of Greek and Latin testimonia which dem-
onstrated overlaps with or resemblances to the Techne so as to prove its influ-
ence on grammar from the IInd century B.C. onwards.
59 Di Benedetto continued to revisit the subject, responding to criticisms and
objections (1973, 1990) and later offering synthesised overviews (1998, 2000).
On the debate raised by Di Benedetto, see the outline sketched by Pinborg
1975; Siebenborn 1976, 69 n. 2; Kemp 1991, who embraced the idea that
grammatical theory became fully developed, with a formal approach to linguis-
tic analysis, only in the Ist century B.C.; Lallot 1998, Robins 1998.
60 See Di Benedetto 1958-1959, 181-182; 1959, 115; Schenkeveld 1994, 267 n.
13; 1998, 51. This idea is supported by the fact that the content of § 3 (περὶ
τόνου) is also confirmed by a citation, explicitly attributed to Dionysius by
Varro (fr. 282.5-8 Funaioli). Erbse 1980, 246 (who did not link the issue to the
question of the work’s authenticity) argued that the ἀνάγνωσις ἐντριβής of the
proem did not coincide with that addressed at §§ 2-4, and therefore that these
were conceptually distinct from § 1 and cohered rather with the remainder of
the discussion. Di Benedetto 1990, 38-39, explicitly rejected this interpretation,
but later modified his own opinion (1998, 152-153 and 2000, 396-397), main-
taining that the two concepts of ἀνάγνωσις expressed in the proem and §§ 2-4
respectively were not indeed coincident (according to this opinion, the first in-
dicates the philological practice of determining the correct accentuation of a
word, while the second refers to the way in which written texts should be read
aloud with due attention paid to accents, punctuation and literary genre, and
should therefore be viewed in the context of schooling). Di Benedetto there-
fore extended his judgment of inauthenticity to cover all of §§ 2-20. By con-
trast, Lallot 19982, 25-26 (Ist ed. 1989, seconded by Swiggers/Wouters 1995,
91) proceeded in the opposite direction, suggesting that §§ 6-10, on phonetics,
should also be included amongst the material of Dionysian origin.
61 See e.g. Di Benedetto 1958-1959, 181; Pfeiffer 1968, 269, who hypothesises
that this discussion may not have been entirely out of place in the original,
given Dionysius’ interest in Homer and the fact that the rhapsodes were the
first “interpreters” of poems. Di Benedetto 1973, 812 instead thought that the
presence of this paragraph went back to the scholastic function of the treatise,
which was intended for youths learning to read precisely by means of the Ho-
32 Lara Pagani

syllables, and parts of speech and upon which doubts of authenticity rest.
Di Benedetto demonstrated the existence of internal inconsistencies
regarding content and arrangement within the material as transmitted,
and these led him to hypothesise that a compiler had intervened. This
compiler would have been responsible for the large gap between the
first part, which contains Dionysius’ definition and subdivision of
grammar and is certainly an Alexandrian product (in that it is dedicated
to interpretation and textual criticism) and the bulk of the work, which
contains a later technical-grammatical discussion62. Di Benedetto saw
confirmation of his reconstruction in the fact that Sextus introduces
Dionysius’ writings with the title Παραγγέλματα (M. 1.57), differently
from manuscripts’ Τέχνη or Τέχνη γραμματική63. There are further-
more three relevant inconsistencies, recognised already in antiquity (Pro-
legomena Vaticana, in GG I/III 124.7-14 and 161.2-8)64, between the

meric texts (see also Di Benedetto 2000, 397); Erbse 1980, 247; Schenkeveld
1994, 267 n. 13; 1998, 47.
62 See Di Benedetto 1958-1959, 179-185, who also noticed (p. 181) that, of the
six parts of grammar mentioned in § 1, only the first is discussed (§§ 2-4).
There is no trace of the others, not even the one which Dionysius himself
judged most important, i.e. κρίσις ποιημάτων (in his re-elaboration of 2000,
397, Di Benedetto differently identifies the interruption as occurring after § 1:
see supra, n. 60). Barwick 1922, 13 had hypothesised that in the first section of
the Techne Dionysius expounded his own view of grammar, and in subsequent
paragraphs reproduced a Stoic source (see the response by Di Benedetto 1958-
1959, 182). Erbse 1980, 245-246 proposed that the apparent anomaly could be
explained if the μέρη of grammar expounded at § 1 were understood not as a
kind of plan for the work, but rather as a preliminary presentation of the pur-
pose, nature, and task of philology.
63 Di Benedetto 1958-1959, 182 and n. 1. It is actually true that Sextus’ formula-
tion does not oblige us to think of a title (ἐν τοῖς παραγγέλμασι = «in his pre-
cepts»), but the parallel of the Παραγγέλματα ῥητορικῆς αʹ attributed to
Theophrastus (D.L. 5.47.27) may suggest that Dionysius composed
Παραγγέλματα γραμματικῆς τέχνης or γραμματικά: see Fraser 1972, II 680;
Di Benedetto 1973, 806 n. 1; Schenkeveld 1998, 42. For Di Benedetto (1958-
1959, 182-185), examination of the indirect tradition confirmed the distinction
between the first, ancient, part – the part that is in fact presupposed by authors
such Asclepiades of Myrlea (IInd-Ist centuries B.C.) and Varro (116-27) (the ref-
erence to Ptolemy the Peripatetic is on the other hand less significant, given his
chronological fluctuation; see supra, 18-19 and n. 8) – and the linguistic sec-
tion, which is only attested beginning at the end of the Vth century A.D.
(Timotheus of Gaza, Ammonius the Philosopher, Priscian).
64 Di Benedetto’s analysis (1958-1959, 171-178) of the origin of the materials
contained in the two Prolegomena seeks to demonstrate that the thesis of inau-
thenticity, which in them is presented as a well-established fact, was not the
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 33

content of the Techne and what is known from other sources for Diony-
sius’ thought concerning categories such as pronoun, article, noun and
verb.
The discrepancies are the following:
1) The Techne retains pronouns and articles as distinct (GG I/I 61.1-
62.5 and 63.1-69.5), whereas Apollonius Dyscolus attests that Dionysius,
like Apollodorus of Athens and in the tradition of Stoic theory, defined
pronouns as a special type of article (ἄρθρα δεικτικά, A.D. Pron., in GG
II/I 5.18-19; cf. sch. Vat. D.T., in GG I/III 160.28 = fr. 54 Linke);
2) The Techne assimilates ὄνομα and προσηγορία into only one part
of speech (GG I/I 23.2-3; cf. 33.6-34.2), while the τεχνικοί (the expres-
sion that the scholia to the Techne use to refer generally to Herodian and
Apollonius and which here may rather indicate Apollonius for reasons of
content65) testify that Dionysius separated them (sch. Vat. D.T., in GG I/III
160.25-28 = fr. 54 Linke), like the Stoics (cf. Diogenes of Babylon, fr.
22.1-5 von Arnim);
3) The Techne proposes a definition of the verb as an indeclinable ex-
pression, which admits of tense, person and number and has an active or
passive value (λέξις ἄπτωτος, ἐπιδεκτικὴ χρόνων τε καὶ προσώπων καὶ
ἀριθμῶν, ἐνέργειαν ἢ πάθος παριστᾶσα, GG I/I 46.4-5), a notion con-
ceptually distant from the definition ascribed to Dionysius by Apollonius,
who spoke of «an expression that signifies a predicate» (λέξις κατηγόρημα
σημαίνουσα, sch. Vat. D.T., in GG I/III 161.6-7 = fr. 55 Linke), thus
drawing near to the Stoic position (a very similar pronouncement is attrib-
uted to Diogenes of Babylon [fr. 22.6 von Arnim]).
These contradictions were set aside as non-fundamental by Pfeiffer66,
who maintained that they were «minor controversial points» on which
Dionysius’ opinions may have wavered. Pfeiffer’s thesis, which Fehling
later deemed hasty67, was the subject of a response by Di Benedetto68, who
showed that such alleged evolutions of Dionysius’ thought are not docu-
mented elsewhere and would imply radical changes of historical and cul-
tural positioning.
Objections to Di Benedetto’s framework were then raised by Erbse69,
who dispelled with the first supposed inconsistency by calling attention to a
misunderstanding of Apollonius’ text: Dionysius and Apollodorus did not
call all pronouns ἄρθρα δεικτικά, but only demonstrative pronouns. Re-

hypothesis of an anonymous scholiast but rather the conviction of an entire

scholarly school.
65 So Di Benedetto 1958-1959, 207 n. 2.
66 Pfeiffer 1968, 271, followed by Linke 1977, 10-11.
67 Fehling 1979, who writes of «oberflächliche Ablehnung» (488 n. 1).
68 Di Benedetto 1973, in particular 798-801 and 1990, 26-29; see also Pinborg
1975, 105-106.
69 Erbse 1980, in particular 252-258.
34 Lara Pagani

cently the issue has been investigated by Matthaios (1999)70 who, in his re-
constructed outline of the history of the pronoun (pp. 491-515), explains
the expression ἄρθρα δεικτικά as a terminological variant used by the two
grammarians as an alternative to ἀντωνομασία / ἀντωνυμία for designat-
ing the category of pronouns, and thereby bringing to light their quality of
deictically indicating the person to whom they refer71. Erbse then dismissed
the second problem on the basis of the scant credibility of the source72,
while he mitigated the third by arguing that the first definition of the verb
explained its function in a sentence, with the other explaining its formal
characteristics73.
This picture was then completed by means of comparison with techni-
cal-grammatical papyri from the Ist to the Vth centuries A.D.: Di
Benedetto, who detected in the most ancient papyri perceptible discrep-
ancies with the Techne when it came to terminology, content, and the
order of discussion, recognised on the other hand some similarities in
the texts from the IIIrd-IVth centuries onwards; he also pointed out that
the first papyrus to contain a piece of the Techne (PSI I 18) dates to the
Vth century, the same period suggested by the indirect tradition (see
supra, n. 63)74.

70 Matthaios 1999, 509-515; see already Schoemann 1862, 119-121.

71 Parallels for this kind of alternation in terminology have been found by Mat-
thaios in Dionysius of Halicarnassus [Th. 37.33-34] and in PBerol. inv. 9917, on
which see Wouters 1997.
72 But see the objections of Matthaios 2009, 399 on this point.
73 Responses to these arguments can be found in the previously cited Di
Benedetto 1990, in particular 20-29. For a summary overview see Lallot 1998;
19982, 19-26; Robins 1998 and Di Benedetto 2000.
74 Di Benedetto 1958-1959, 185-196; to PSI I 18 should be added PHal. inv. 55a
(Vth-VIth centuries): see also Di Benedetto 1973, 801 and n. 1; Wouters 1979,
109-119 (no. 4), 120-124 (no. 5). Pfeiffer 1968, 270 cautioned against this rea-
soning, since it constitutes a hazardous argumentium e silentio; so too Erbse 1980,
247, who advanced the further objection (p. 248) that the papyrological texts
considered by Di Benedetto, inasmuch as they are remnants of schoolbooks,
are not suited to comparison with the Techne, which is thought to have been
intended for scholars rather than students. Di Benedetto 1973, 801-803; 1990,
29-32; 1998, 152; 2000, 398 refuted these criticisms, arguing that the complete
silence until the Vth century of the tradition for the Techne as the work of Dio-
nysius Thrax would be inexplicable if it had really been composed in the IInd
century B.C. (we would have to posit a sort of conspiracy of silence on the part
of authors such Apollonius Dyscolus, Herodian, Sextus Empiricus and Quintil-
ian: Di Benedetto 1973, 803; 1990, 31; 2000, 398). Apollonius Dyscolus’ atti-
tude would seem equally strange, since he treats Dionysius as a grammarian of
the second rank, citing him only very rarely (his point of reference is rather
Trypho) and certainly not crediting him with the first technical handbook on
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 35

This combination of factors would lead us to attribute the Techne, in

the form in which we now know it, to the IIIrd-Vth centuries, and thus
to locate it not at the origins, but rather at the end of the development
of Greek linguistic science, which only got underway in the Ist century
B.C. with figures such as Tyrannion and Asclepiades of Myrlea and
reached the first stages of systematisation shortly afterward with Trypho
and Habro75.
A more extensive analysis of the issues relating to the authenticity of
the Techne is beyond the scope of this contribution76, but the general

grammar (see Di Benedetto 1958-1959, 209-210; 1990, 31-32; Kemp 1991,

310; Di Benedetto 1998, 152; 2000, 398-399). Nevertheless, Erbse 1980, 247
and then Schenkeveld 1994, 267-268 and 1998, 42-43 rightly urge us to keep
the problem of Techne’s authenticity separate from that of its authority (on the
fact that the model reproduced in the work for centuries was anything but the
sole standard handbook, see Swiggers/Wouters 1995 and Wouters 1998). The
typological chronology of the τέχναι on papyrus reconstructed by Di
Benedetto and retraced by Pinborg 1975, 104-105 and Kemp 1991, 311-315,
was then called into question by Swiggers/Wouters 1995 and Wouters 1998:
these studies in fact established, on the basis of new evidence, that some papyri
from the Ist-IInd centuries A.D. already show notable resemblances to §§ 6-20 of
the Techne; that certain models of τέχνη may have already existed in the Ist cen-
tury B.C. and perhaps earlier; that in the Ist century A.D. grammatical science
already shows a degree of systematisation such that it justifies unproblematically
the Techne as we now have it; and that, as a whole, it does not seem that the
papyrological documentation can be used to support a late date for the treatise.
For grammatical papyri, the standard work of reference is Wouters 1979 (see in
particular 38ff. on aspects connected to the Techne; for an update see Swig-
gers/Wouters 2000). See also TMil.Vogl. inv. 8, A, B (first half of the IInd cen-
tury A.D., Bastianini/Lundon 2000) and the Berlin tablet inv. P. 10511-10512
(IInd century A.D.) presented by F. Reiter and F. Montanari at the 26e Congrès
international de papyrologie (Genève, August 2010).
75 Di Benedetto 1958-1959, 196-210; 1959; 1998, 151-152; 2000, 399. The birth
of grammatical science in the Ist century B.C. was hypothetically linked by Di
Benedetto (1958-1959, 202) to factors such as the need to preserve classical
Greek usage; contact with Roman culture, which brought with it confronta-
tion on a linguistic level; and the role of Rhodes, where mutual contact be-
tween rhetoric (Apollonius Molon) and grammar (Dionysius Thrax) may have
favoured the establishment of the two disciplines’ respective boundaries, with
linguistic instruction falling under the domain of the second. Pinborg 1975,
133, added to these factors the general tendency in the period to collect obser-
vations made by predecessors in systematic handbooks (see also Fuhrmann
1960, 154-155).
76 In recent years its authenticity has been defended by Janko 1995a, 215-216, on
the basis of the observation at § 12 of the Techne that only the neoteroi con-
structed matronymics, and not Homer: since it fully coheres with the Alexan-
drian philological spirit, this observation seems in Janko’s opinion to support
36 Lara Pagani

picture does in any case advise against using the treatise as proof that an
evolved system of grammar existed amongst the earliest Alexandrians77.
By contrast, the fact that the genuineness of the Techne has proven prob-
lematic does not itself preclude the existence of linguistic theorisation in
the early Hellenistic period78. In this respect a crucial corrective has, in
recent years, been brought to light: even if the current text of the lin-
guistic discussion in the Techne is spurious, the initial section, with its
definition and division of grammar, is certainly by Dionysius; as a result,
it is true that he did not write that work, but did nevertheless compose a
work on a grammatical subject. We should recognise that the content of
those sections which are positively authentic does not suggest a com-
plete and technical exposition of morphology and the parts of speech; it
does however allow us to infer a structure that was to some extent sys-
tematic, as well as to identify an incipient awareness of the autonomy of
the discipline, which in true Alexandrian spirit was understood as a

ascribing the entire work to that context and thereby to save it from suspicions
of inauthenticity. Dionysius would at first have adhered to the Stoic doctrines
derived, perhaps by means of Apollodorus, from Diogenes of Babylon, then
modified them until they reached the form that we find in the Techne (on
Apollodorus’ role as an intermediary between Stoic advances and Alexandrian
scholarship see Frede 1977, 52 and Schenkeveld 1984, 348). To this Janko
2000, 178 and n. 1 has added, in support of the Techne’s authenticity, the pres-
ence within it of remarks on euphony (GG I/I 11.5-12.4), the importance
which it attributes to κρίσις ποιημάτων (GG I/I 6.2-3), and the presence of
influences from Stoic grammar. On the other hand, without taking a position
on the work’s authenticity, Matthaios 1999, 265, 282-283, 623 has demon-
strated how at least part of the material that it contains is congruous, on a theo-
retical level, with Aristarchean doctrines and therefore may plausibly have an-
cient origins (see infra, 59). Alongside this group of points of contact – a sort of
common heritage of ancient grammatical thought – there also emerge (Mat-
thaios 2009) substantial conceptual and taxonomic differences between the
Techne and the positions held by Aristarchus (in particular pp. 395-398), not to
mention discrepancies between the doctrines relating to the system of parts of
speech attributed to Dionysius Thrax by testimonia external to the Techne and
that which is present within it (in particular pp. 398-399). In its conception the
systematic section of the Techne is thus later than, though still in the tracks of a
tradition leading back (in that it has to do with the parts of speech) to the Aris-
tarchean framework. The grammatical theory presupposed by Dionysius was
on a different front, and may represent the first attempt to instill linguistic de-
velopments of the Stoic school within the Alexandrian scholarly tradition.
77 The grammatical theory displayed in the Techne is regarded as representative of
the linguistic knowledge of Alexandrian scholarship by Robins 1957.
78 As we shall see, related conclusions should instead be drawn from that which is
definitively known, even in a fragmentary and indirect way, about the ap-
proach of ancient scholars to linguistic matters.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 37

combination of philological and linguistic study aimed at the interpreta-

tion of literary texts79.
The arguments against the authenticity of the Techne were consid-
ered of little substance by Pfeiffer (1968)80, who saw in this work the
peak, with noteworthy influence from the Stoic tradition, of grammati-
cal science as a stand-alone discipline in the Hellenistic period. For him
this discipline arose out of general philology, which initially applied it
for purposes of textual criticism and exegesis. The first steps in this di-
rection consisted in the “preparatory work” represented by Aristo-
phanes’ Λέξεις81 and by his investigations into recurring patterns of in-
flection, which led him to recognise the fundamental principle of
regularity known as analogy that was later both practiced and improved
by Aristarchus82. Within these lines of development, which viewed

79 The issue, profiled already by Ax 1986, 228-229, is well-recognised by Mon-

tanari 1993, 255-256, Schenkeveld 1994, 269; Montanari 1994, 302-303 (Dis-
cussion of the contribution by Schenkeveld). Entirely dedicated to the linguistic
content of the Παραγγέλματα is Schenkeveld 1998, according to whom Dio-
nysius’ treatise should have also contained something like a formal description
of language (resulting from a fusion of the findings of predecessors’ scholarship
and Stoic observations on the subject), probably in the section entitled «exege-
sis according to poetic modes of expression» (Schenkeveld 1994, 291-292). On
the other hand, Di Benedetto 2000, 396 judges it inconceivable that something
so important as specific discussion of the science of grammar could have been
included in Dionysius’ work without also being cited in the subdivision of
γραμματική. Di Benedetto further maintained that there was no reason to lo-
cate that discussion within the study of poetic modes of expression, rather than
within the treatment of any other of grammar’s μέρη. On the subject see also
Blank 2000, 407-408.
80 For his stances on individual points, see supra, 33. Pfeiffer 1968, 272 however
recognised that the modern arrangement of the Techne does not correspond to
the original one, and hypothesised a lacuna after the present § 4, where, in his
opinion, an editor had later intervened in an attempt to rewrite what remained
of the original.
81 Within this lexicographical collection Pfeiffer highlighted in particular a section
Περὶ τῶν ὑποπτευομένων μὴ εἰρῆσθαι τοῖς παλαιοῖς (frr. 1-36 Slater), from
which he inferred an interest in the historical aspects of words, particularly with
regard to chronological distinction between ancient and modern usages (Pfeif-
fer 1968, 197-200; but see infra, 48 and n. 121). On the other hand, the con-
tent of some fragments, especially those relating to greetings and dialectal
forms, suggests attention to contemporary spoken language (Pfeiffer 1968, 202).
These two aspects of Aristophanes’ work are considered in relation to his prac-
tice of analogy by Ax 1990 (see in particular 13-14).
82 For Pfeiffer these studies were strictly connected to philological investigations,
and did not trespass onto the philosophers’ disputes about the relationship be-
38 Lara Pagani

grammatical science as the final achievement of Hellenistic philology,

Pfeiffer found confirmation for his highly limiting notion of Aristotle’s
role in the birth and evolution of Alexandrian scholarship, which he
argued would have been much more directed towards linguistic studies
had it indeed been influenced by the philosopher83.
The evolutionary path of grammar traced by Di Benedetto was ac-
cepted by J. Pinborg (1975), who outlined a conceptual distinction be-
tween accumulations of individual linguistic observations, abundant ever
since Aristotle and the Stoa, and the incorporation of these into a coher-
ent grammatical system, something which would not occur until the Ist
century B.C.: the Alexandrian scholars hardly aimed to establish a lin-
guistic system, but rather sought to interpret the poets. In the course of
doing so, however, they may have occasionally made use of and perhaps
even described some grammatical categories84.
The view that the philologists of the IIIrd-IInd centuries B.C. were
interested exclusively in the study of literature was also sustained by E.
Siebenborn in his study of the doctrine of hellenismos (1976)85, which
investigated among other things the application of the principle of anal-
ogy to determining the correct forms of words86. In Siebenborn’s opin-
ion, Aristarchus used in most cases this criterion as a heuristic method
for establishing correct prosody, and not for determining inflectional

tween words and things (see Pfeiffer 1968, 203). Apropos of Aristophanes’ sup-
posed monograph Περὶ ἀναλογίας see supra, n. 41.
83 So Pfeiffer 1968, 272. For recent adjustments made to this interpretation, see
supra, n. 25.
84 Pinborg 1975, 103-106 (on the Techne), 106-110 (on analogy and anomaly,
with reference to the opinion of Fehling 1956, 264-270, on which see supra,
30), 110-114 (on the development of grammatical theory in the Hellenistic pe-
riod). Pinborg considered Di Benedetto’s results comparable to Fehling’s, cer-
tainly not in terms of their chronologies of grammar’s beginnings, but because
of the similarity of approach that had led both to maintain that Greek gram-
matical theory represented a unitary system equipped with uniform criteria and
methods. In fact, according to both Di Benedetto and Fehling the existence of
conflicting opinions as to individual points did not require that the ancients
create completely different versions of grammar.
85 Siebenborn also accepted Di Benedetto’s position on the inauthenticity of the
Techne (see in particular Siebenborn 1976, 27 n. 2 and 69 n. 2).
86 The three principal criteria for linguistic correctness lay in analogy (Siebenborn
1976, 56-84), the literary tradition (85-89), and usage (90-97); complementing
these were etymology (140-146), διάλεκτος (146-151) and, in the Roman
world, natura (151-154) and euphonia (154-155).
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 39

endings87. On the other hand, Siebenborn maintained that the set of

conditions of analogy observed by Aristarchus was restricted to simple
comparisons of words, and did not go so far as to constitute a system of
concepts and rules for declension and conjugation88. For Siebenborn
Aristarchus and his contemporaries thus did not get beyond recognising
a certain regularity in the morphological structures of language; it was
only at the end of the IInd century B.C. that the grammarians began to
concern themselves first with demonstrating the scientific nature of their
discipline and then with founding a system of concepts and rules89.
A reduced role in founding the grammatical science was also as-
signed to Alexandrian scholarship by M. Frede (1977), who rejected that
the Techne, which he took to be authentic, could represent an elabora-
tion – let alone codification – of a grammatical system assumed by phi-
lologists in the period immediately before Dionysius90: in fact, ancient
sources do not attribute any of these philologists with a work on linguis-

87 This conclusion was based on a study of the various passages from the scholia
(Siebenborn 1976, 70ff.), all of which, however, go back to Herodian: accord-
ing to Matthaios (1999, 28-29), a similar selection of material produces a partial
result, which should not form the basis of generalisation: it is clear that the ma-
terial from Herodian which was gathered within the scholia preserves fragments
on prosody, while other sources, such as Aristonicus and Didymus, suggest a
different picture (on this point see frr. 48-53 Matthaios and the overview of
Aristarchus’ concept of case at pp. 287-289: see infra, 57ff., for more detail on
Matthaios’ arguments).
88 In fact, according to Siebenborn (1976, 71-72 and 76-77), the Aristarchean
tradition is dominated by the so-called two-term analogy, while there is no
trace of four-member proportion (see supra, n. 44); it is nevertheless important
to consider that the formula «x is like y» does not necessarily presuppose a
comparison of words connected by a formal relationship of analogy, but may
indicate that the accentuation or vocalism of word x, unknown in this regard,
is the same as that of term y, which is known (so Callanan 1987, 116, and later
Matthaios 1999, 29-30). A sizable portion of presumed Aristarchus’ two-term
analogies would thus have acted as “examples”, exempt from the formal restric-
tions of analogy (see Matthaios 1999, 30). In this way cases where the rule of
equal syllable-number is not observed (such as φυλακούς-φρουρούς [sch. Hdn.
Il. 24.566d1] and φωριαμῶν-κιβωτῶν [sch. Hdn. Il. 24.228a]), and in which
Siebenborn 1976, 77 saw «eine Lockerung der Rigorosität», might be ex-
plained. On the other hand, the idea that Aristarchus also knew of the four-
member proportion and employed it to deduce grammatically-correct forms
has been convincingly demonstrated by Erbse 1980, 237-238 (see infra, 40-42).
89 Siebenborn 1976, 70, 84, and 97.
90 Frede 1977, 56; he makes cursory reference (p. 52) to the issue of authenticity
raised by Di Benedetto 1958-1959, dismissing the problem with the remark
that «nowadays the text is generally accepted as genuine».
40 Lara Pagani

tic matters, and at most Aristophanes and Aristarchus might be credited

with a certain interest in the rules of inflection, a subject which in any
case is not addressed by the Techne91. Frede instead emphasised the role
of Stoic dialectic in the development of the theory of language, high-
lighting how the Stoics, although they never composed a grammar, had
attempted to develop an elaborated theory such as could determine what
should be accepted as correct Greek92. In fact, he considered the influ-
ence of the Stoics’ linguistic project to be so great that he described the
work of later grammarians as a revival of some well-defined aspects of
its93.
On the other hand, H. Erbse (1980) contested Di Benedetto’s and
Siebenborn’s reconstructions, attributing to the Alexandrians knowledge
of a notable system of grammatical rules that was already fairly advanced,
albeit one still in need of improvement: without this they would have
been unable to solve the linguistic problems posed by the poetic texts
which they were investigating94. Erbse supported this conclusion with a
study of the Aristarchean testimonia in the Homeric scholia that 1) made
recourse to the concept of analogy, 2) had to do with the so-called sche-
matologia Aristarchea, and 3) demonstrated awareness of grammatical ter-
minology.
1) Recourse to analogy presupposes recognition of linguistic regularities,
according to Erbse95, who on this point cited the following fragments: sch.
Hdn. Il. 24.8a (fr. 92A Matthaios), which offers a direct citation of Aris-
tarchus concerning the accentuation of πείρων, determined on the basis of

91 According to Frede 1977, 56, in order to salvage the assumption that the ori-
gins of grammatical science were Alexandrian, one would have to imagine that
it had been developed by Dionysius himself (though in the same pages Frede
then warned against doing so, pointing out that the Techne is a very elementary
handbook, in which a rich system has been reduced to definitions, classifica-
tions, and basic examples – certainly not the kind of text that would have been
written so as to introduce a new discipline of such interest and importance).
92 Frede 1977, 76.
93 Frede 1977, 76-77. In this regard the figure of Diogenes of Babylon is consid-
ered particularly significant, since his influence is seen by Frede as lying at the
origin of both the Pergamene (Crates of Mallus and later the Roman artes) and
the Alexandrian (Apollodorus of Athens, Dionysius Thrax) traditions.
94 Erbse 1980, 241; he therefore shared the original position of Fehling (1956, in
particular 214, 247-248 n. 1, 260-261), who however had in the meantime
been persuaded by the interpretation of Di Benedetto and Siebenborn (see also
Erbse 1980, 237).
95 Erbse 1980, 239.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 41

the comparison ἔκειρε : κείρων = ἔπειρε : πείρων96; sch. Hdn. Il. 9.150a,
sch. ex. Il. 14.464a, sch. Hdn. Il. 16.123c (fr. 27A M.), 16.390e, which use
terms such as ἀναλογία and ἀνάλογος -ον (o -ως) in connection to ex-
planations made by Aristarchus; sch. Hdn. Il. 1.493a, 2.592b, 3.198a,
12.158, sch. Did. Il. 12.231a, which testify to cases in which Aristarchus has
chosen readings that neglected or violated the analogy (we find expressions
such as παραλόγως or παρὰ τὴν ἀναλογίαν)97.
2) The schematologia Aristarchea denotes the set of systems which Aris-
tarchus used to account for the unusual features of Homeric language, and
according to Erbse suggests a search for regularities by which to understand
apparent irregularities98. The examples which he considers in this respect
are: sch. Ariston. Il. 1.175a, where the expression κε … τιμήσουσι is justi-
fied either by a pleonasm of the modal particle or by the use of the future
indicative in place of an aorist optative; sch. Ariston. Il. 2.286, on the use of
the present tense in place of the future; Pap. II Erbse ad Il. 785 (POxy.
1086 l. 55), where the genitive of the phrase διέπρησσον πεδίοιο is ex-
plained by the presumed omission of the preposition διά; sch. Ariston. Il.
4.331a, on the use of a middle form in place of an active one.
3) Regarding terminology, relevant evidence includes sch. Hdn. Il.
24.8a (fr. 92A M.; see just above), which attests the term παρατατικός for
the imperfect; and A.D. Pron., in GG II/I 62.16-17 (fr. 120A M.), on Aris-
tarchus’ coinage of the name ἐπιταγματική [scil. ἀντωνυμία?] for forms of
αὐτός used alongside the tonal forms of personal and possessive pronouns.
This evidence is, however, not sufficient to claim that Aristarchus was al-
ready aware of the terminology known to us from the Techne of Diony-
sius99: other sources indicate that many expressions used in later grammar
(and also in the Techne) were in fact unknown to Aristarchus, e.g. those
used for indicating verb tenses100.
On the basis of this overview Erbse concluded that the paradigms of
declension and conjugation, the syntactical functions of words, and the
uses of particles were all known to Aristarchus101. The picture recon-
structed as such was then upheld by Erbse’s examination of the value of

96 This confirms that Aristarchus used the more developed form of analogy,
which Siebenborn (see supra, n. 88), who did not consider this passage, denied
to have been the case.
97 To these examples Ax 1982, 101 n. 19 proposed adding sch. Ariston. Il. 2.397a
and sch. Did. Il. 2.397b-c (frr. 81B, 81A1 and 81A2 M.), on the Homeric use of
plural verb forms with neuter plural subjects.
98 See supra, 25 n. 36.
99 So Erbse 1980, 238, on the basis of the first testimonium.
100 See Matthaios 1999, 31, 327 and 340-351.
101 These conclusions have been criticised by Callanan 1987, 17-19; see also Tay-
lor 1987, 11-12; Schenkeveld 1984, 349 and n. 157; 1994, 285; 1998, 41 and
Matthaios 1999, 30-31.
42 Lara Pagani

the Techne, whose authenticity he defended against the arguments of Di

Benedetto102. Furthermore, in accord with Pfeiffer (1968), he placed this
handbook within the history of ancient scholarship «als Spätform der
hellenistischen Fachschrifstellerei»103.
On the same front but in a less radical manner, W. Ax produced a
series of studies on various aspects of the problem (1982, 1986, 1990,
1991). He in no way denied the practical-philological orientation of
Alexandrian scholarship, but demonstrated how the Alexandrians’ ap-
proach to the linguistic difficulties of texts implied knowledge, albeit
empirical in nature, of an apparatus of linguistic description and an
awareness of the existence of regularities, though not the foundations of
a complete and articulated system. So as to describe this type of gram-
matical competence, Ax spoke, with regard to Aristarchus, of a «Gram-
matik im Kopf»104. In his first work on the subject (1982), he considered
various fragments of Aristarchus transmitted by Apollonius Dyscolus105
in order to determine his place within the general picture of grammar’s
development. Given that in Apollonius, too, Aristarchus appears as a
Homeric exegete (not surprising at all), some passages have to do with
linguistic observations relating to Homeric usage. These consist in notes
on individual words and on the general use of certain parts of speech
and these remarks both took as their starting point and had as their ob-
jective Homeric linguistic usage – usage that was carefully observed,
structured by means of rules, and consulted as the paradigm when it
came to doubtful points of textual criticism106.
The fragments taken into consideration have to do with:

102 The objections raised by Erbse as to individual points made by Di Benedetto

are mentioned supra, 33-34.
103 Erbse 1980, 258; see also Pfeiffer 1968, 272: «Very late, and under the influ-
ence of Stoic doctrines, an Alexandrian scholar constructed from observation
(ἐμπειρία) of the language of poets and prose writers a ‘system of γραμματική’
that is a τέχνη. The lateness of its appearance, often regarded with surprise, is
in harmony with the line of development we have traced from the third to the
first century B.C.» (see also supra, 37).
104 See in particular Ax 1982, 109 and Ax 1991, 288, where this apt formulation
appears.
105 Ax chose to concentrate on Aristarchus because the testimonia for him are the
most numerous and explicit; by programmatically excluding scholiastic sources
from his research he aimed to avoid a constrictive preponderance of Homeric
text and language. Apollonius Dyscolus was favoured, instead of e.g. Herodian,
so that fragments of the most general or most generalisable nature might be
used (1982, 101).
106 Ax 1982, 102-104.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 43

1) the pronoun: a) on a specific level: ἐμοῖο, σοῖο, οἷο are Thessalian

forms (A.D. Pron., in GG II/I 109.4-5 [fr. 133 M.]); 2) on a general level:
third-person pronouns, when accented, have a reflexive meaning (ib.
42.17-25 [fr. 123B1 M.]); possessive pronouns are usually reflexive (ib.
48.7-11 [fr. 130B M.]); Homer is infallible in his use of pronouns (ib.
109.20-23 [fr. 104B M.]); Homer avoids using overlapping forms for per-
sonal and possessive pronouns (A.D. Synt., in GG II/II 222.12-223.15 [fr.
128B M.]);
2) the article: the article is generally omitted (ib. 6.4-11 [fr. 99A1 M.]
and 106.1-107.8 [fr. 99A2 M.]);
3) the adverb: ὧδε, usually used in a locative sense, is in Homer used
only modally (A.D. Adv., in GG II/I 178.25-27 [fr. 157C M.]);
4) the conjunction: Homer has the habit of putting in first position
causal phrases containing the conjunction γάρ (A.D. Conj., in GG II/I
239.23-27 [fr. 173B M.]).
A second group of fragments contains general remarks by Aristarchus on
the Greek language; these unequivocally demonstrate an interest (also)
in linguistic issues unconnected with the Homeric texts, in relation both
to hellenismos and to the parts of speech107.
The fragments contain observations concerning:
1) the noun: nouns are declinable and therefore subject to agreement
(A.D. Adv., in GG II/I 145.5-10 [fr. 136 M.]);
2) the pronoun: a) on a specific level: plural forms of third-person
“composite”, i.e. reflexive, pronouns [scil. ἑαυτῶν, ἑαυτοῖς…] are unac-
ceptable for reasons of syntactic tolerability, inflectional analogy, and com-
parison with Homeric language (A.D. Pron., in GG II/I 71.20-29 [fr. 125A1
M.] and Synt., in GG II/II 244.10-245.5 [fr. 125A2 M.])108; the pronoun
αὐτός can also be called ἐπιταγματική [scil. ἀντωνυμία?], i.e. a “subsidi-
ary” (or “supporting”) pronoun, inasmuch as it can be used alongside every
accented personal pronoun and possessive pronouns in the genitive (A.D.
Pron., in GG II/I 62.16-17 [fr. 120A M.: see supra, 41]); b) on a general
level: description of pronouns as «expressions that form a series according to
person» (λέξεις κατὰ πρόσωπα σύζυγοι, A.D. Pron., in GG II/I 3.12-13
[fr. 103A1 M.] and Synt., in GG II/II 137.9-138.9 [fr. 103A2 M.]);

107 Ax 1982, 104-108. Callanan 1987, 19 and nn. 41, 42 gave less importance to
this second group of evidence, highlighting how Ax himself attributed to Aris-
tarchus the possession of an elaborated system of rules governing Homeric lan-
guage but spoke only of an “interest” in general issues of grammar (Ax 1982,
108: «Aristarch war nicht nur im Besitz eines komplexen Regelwerks zur
homerischen Sprache, sondern war darüber auch an gemeingrammatischen
Frage interessiert»); nevertheless, see Ax’s wording at 1991, 287: «… daß Aris-
tarch … Fragen der Morphologie als Gegenstand sui generis behandelte».
108 See Siebenborn 1976, 30-31.
44 Lara Pagani

3) the adverb: a) on a specific level: the word ἄνεῳ is an adverb (A.D.

Adv., in GG II/I 144.11-145.10 [fr. 136 M.]); b) on a general level: adverbs
are indeclinable and therefore not subject to agreement (ib. 145.5-10);
4) the article: ὦ before a vocative is not an article (A.D. Synt., in GG
II/II 71.10-72.2 [fr. 98 M.]).
Ax recognised that it was not possible to determine what weight Aris-
tarchus attributed to such matters, nor to ascertain whether he had
somehow combined them thematically or considered them coincidental,
secondary results of his philological work. In any case, the evidence
indicates that his linguistic observations were advanced enough in terms
of both level and degree of autonomy to make it plausible that even his
immediate students sought to systematise grammar109.
Ax also achieved further results concerning the history of the origins
of grammatical science in a study dedicated to the section περὶ φωνῆς
(1986), i.e. to language’s subdivision starting from sound: both this and
the model of linguistic description connected to it prove to be a sort of
“guide fossil”110 that can aid in recovering the relationships of depend-
ence which existed within the Greek and Roman grammatical tradition.
He adopted the bipartite evolutionary model of Barwick (1922)111, but
adjusted the lines of development: both the techne of Diogenes of Baby-
lon and the “Roman” techne (he avoids the designation “Pergamene
techne”) had split off from an ancient Stoic original; the Alexandrian
techne on the other hand derived from that of Diogenes, but was also
influenced by the “Roman” one, from which the Roman ars then di-
rectly descended – which in turn would also be influenced by the Alex-
andrian type (with Varro and Remmius Palaemon)112.
Later studies by Ax (1990 and 1991) also addressed Aristophanes of
Byzantium’s approach to language, which in the meantime had been the
specific subject of a piece by C. K. Callanan (1987). Callanan conducted
an analysis of Aristophanes’ surviving fragments in order to reconstruct

109 Ax 1982, 109.

110 Ax 1986, 252.
111 See supra, 28 and n. 48. For Ax the Alexandrian model was excluded as a direct
antecedent of the Roman artes, since (like the model of Diogenes) it differed
from them in its definition of φωνή; its division of φωνή into four types (there
were only two in the Roman model); and its presentation of the parts of lan-
guage as two groups dependent upon the two respective types of φωνή, the
first being subordinate to the second: λέξις and λόγος (instead of a continuous
sequence as in Roman grammar): see Ax 1986, 242-243 and 249-250.
112 Ax 1986, 249-252. The validity of this reconstruction has been called into
question by Schenkeveld 1990b, 298-306 (see also Schenkeveld 1990a).
46 Lara Pagani

terms ἄρσεν, θῆλυ and οὐδέτερον. Various other cases document attempts
by Aristophanes to determine the correct gender of a word, especially in
the context of his critical work on the poems of Homer;
2) he knew of various types of nominal derivation, even if he cannot
be credited with devising a system in this respect (pp. 42-50): according to
Eust. ad Il. 777.58-60 (cf. fr. 241 Sl.; the idea that the material in this pas-
sage dates back to Aristophanes belonged to Nauck 1848, p. 153), Aristo-
phanes had a notion of “original” and “derived” nouns, though we cannot
determine whether he actually called them πρωτόθετον and παράγωγον;
3) he knew of three numbers for nouns, even if we can only guess that
he used the terms ἑνικόν, πληθυντικόν and δυικόν for these (pp. 50-56): it
is here that the questions of agreement posed by Aristophanes should also
be located (e.g. sch. Did. Il. 10.349a [p. 180 Sl.], sch. Did. Il. 15.301a [p.
186 Sl.], sch. Od. 6.74 [p. 198 Sl.] and 11.174 [p. 199 Sl.]), although what is
admittedly at stake here is the relationship between vox and res, rather than
real grammatical agreement: this is recognised only in sch. Od. 2.45 (p. 193
Sl.) where the grammarian read κακά in place of κακόν so as to construe it
with δοιά at l. 46, a term which others, such as Aristarchus, explained as
adverbial (τὸ δοιά ἀντὶ τοῦ διχῶς). The treatment of problems relating to
the Homeric use of the dual is attested both by the number of times that
Aristophanes read a dual instead of a plural in instances when the text was
referring to two people or objects (sch. Did. Il. 6.121 [p. 178 Sl.], 8.290c
[p. 179 Sl.], 10.4b [p. 179 Sl.], 11.103a [p. 181 Sl.], 11.135 [p. 181 Sl.],
12.127-138a [p. 182 Sl.], 13.613b [p. 184 Sl.], 17.721 [p. 188 Sl.], 18.526c
[p. 188 Sl.]), and by the passage from which we infer that Aristophanes did
not allow second and third person dual verb forms to be switched in sec-
ondary tenses (sch. Did. Il. 13.613b [p. 184 Sl.]);
4) he had an idea of the regularity of declension (pp. 57-61), as can be
deduced from the explanation that he gave for the origin of the heteroclitic
dative γερόντοις (Eust. ad Il. 279.38-42 [fr. 25 Sl.]), in which we are able
to glimpse his conviction that nominatives of the same type are inflected in
the same way: Aristophanes in fact maintained that the Aetolians had con-
fused the genitive (γενική) γέροντος with a nominative (εὐθεῖα) of -ο de-
clension, and on this basis had constructed the erroneous dative plural
γερόντοις. In addition to the previously cited εὐθεῖα and γενική
[πτώσεις], Aristophanes should moreover have been aware of the case
known as κλητική (see Eust. ad Il. 1118.8 [fr. 241 Sl.]) and likely also of
the δοτική and αἰτιατική cases (as Callanan 1987, 61, inferred, also consid-
ering the fact that these cases never received other technical names).
The material is not as extensive for the verb as it is for the noun, but
does still allow for some observations:
1) Aristophanes examined irregular forms. This may be inferred from
two of the notes in the Λέξεις (pp. 62-64) which, because they assemble
several examples of the same phenomenon, appear to imply a sort of theo-
retical framework for verbal inflection and which seem not to be motivated
by the empirical aim of determining or explaining the text of a literary
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 47

work: fr. 19 Sl. (Eust. ad Od. 1761.30-31 and ms. M) attests that Aristo-
phanes registered the use of the ending -σαν for some imperfects
(ἐσχάζοσαν, ἐλέγοσαν and ἐφεύγοσαν), believing it to be a Chalcidian
feature, while Eust. ad Od. 1761.38 (fr. 28 Sl.) attests his recording of the
imperative forms ἀπόστα and κατάβα with their respective “regular”
forms ἀπόστηθι and κατάβηθι;
2) he was aware of the rules of contraction (p. 65), as can be deduced
from sch. Od. 2.50 (p. 193 Sl.), which cites Aristophanes for the form
ἐπέχρων in place of ἐπέχραον, and sch. E. Or. 1287 (fr. 389 Sl.), where
the grammarian is mentioned for his reading ἐκκεκώφωνται in place of
ἐκκεκώφηται, which implies a contract -ο stem;
3) he had a stance on the construction and use of verb forms in the
Homeric poems, as well as on the augment’s presence or absence in historic
tenses (pp. 66-71). Examples of textual constitution based on the interac-
tion between Aristophanes’ research in Homericis and linguistic methods in-
clude: sch. Did. Il. 1.298c (p. 175 Sl.) for the future form μαχήσομαι, sch.
Did. Il. 3.57a (p. 177 Sl.) for the pluperfect ἕσσο, with -σσ- (for the de-
termination of this form the scholion attests, without referring it to Aristo-
phanes, the use of an analogy with the future ἕσσω at Od. 16.79 and the
aorist ἕσσας at Od. 14.396) and sch. Od. 14.522 (p. 201 Sl.) for the infini-
tive εἴνυσθαι (in place of ἕννυσθαι), on the basis of comparison with
καταείνυον at Il. 23.135. A couple of cases document his awareness of a
relationship between morphology and semantics when it came to the use of
verb forms in Homer: sch. Did. Il. 13.51a (p. 183 Sl.), where Aristophanes’
position in favour of σχήσουσιν in place of ἕξουσιν is recorded, with
mention of the parallel at Il. 13.151 (what led Callanan to argue that the
grammarian understood the different semantics of the two future forms of
ἔχω, where σχήσειν signified to “restrain”, “check”, certainly its meaning
at Il. 13.151); and Eust. ad Od. 1680.23ff. (fr. 22 Sl.), where Aristophanes
notices that in Attic a single form has two functions, while Homer uses a
morphologically different form for every semantic function (ἴσθι in place of
γίνωσκε, ἔσο in place of ὕπαρχε). As to the augment, Callanan high-
lighted the following fragments: sch. Did. Il. 11.686b (p. 182 Sl.: χρεῖος
ὀφείλετο in place of χρέως ὠφείλετο), 17.234a (p. 188 Sl.: ἔλπετο with
initial ε, instead of η), 15.601a (p. 187 Sl.: μέλλε in place of ἔμελλε, «Ioni-
cally»: here however it is possible to read Aristarchus’ name rather than
Aristophanes’: see Erbse’s apparatus ad loc.);
4) he may have had (though this is a mere hypothesis) an idea of the
durative character of the imperfect as compared with the aorist (p. 72). In
reality, the available evidence only gives vague indications which do not al-
low us to draw a clear conclusion on this point: at Il. 13.443, Aristophanes
read πελέμιζεν in place of πελέμιξεν (sch. Did. Il. 13.443b, p. 183 Sl.),
while at 20.306 he preferred ἤχθαιρε to ἤχθηρε (sch. Did. Il. 20.306b, p.
189 Sl.), but the reasons for his choices are not transmitted. There also exist
other examples of Aristophanes’ textual interventions when it came to vari-
ants based on verb tenses, but according to Callanan these are owed to the
48 Lara Pagani

demands of exegesis rather than to his knowledge of the theory of tenses:

sch. Did. Il. 2.436a, p. 176 Sl.: ἐγγυαλίζει instead of ἐγγυαλίξει; 14.474a,
p. 186 Sl.: ἔοικεν in place of ἐῴκει; and 19.86, p. 189 Sl.: νεικ<ε>ίουσιν for
νεικείεσκον;
5) he knew of moods, even if it is impossible to tell whether he devel-
oped any theories about them (pp. 73-74). In this case as well, the tradition
preserves no remark made by Aristophanes which is unrelated to textual in-
formation: at Od. 5.168 (sch. ad loc., p. 197 Sl.) Aristophanes preferred the
optative ἵκοιο to the subjunctive ἵκηαι, just as at Od. 14.328 (sch. ad loc., p.
201 Sl.) he opted for ἐπακούσαι instead of the ἐπακούσῃ defended by
Aristarchus; at Il. 17.264 (sch. Did. Il. 17.264b, p. 188 Sl.) he chose the sub-
junctive βεβρύχῃ, which is more common in the tradition than the indica-
tive βέβρυχεν: exactly what motivations guided these choices and whether
those motivations can be traced back to a general speculative framework
were, for Callanan, questions destined to remain open.
An Aristophanic text which Callanan perhaps did not emphasise
enough from this perspective is A.D. Synt., in GG II/II 443.7-13 (fr. 382
Sl.), from which we learn that Aristophanes considered ἀναστροφή a de-
fining characteristic of prepositions: the fragment was indeed included in
the section on accents – since it discusses the Aeolic barytonesis, i.e. reces-
sive accentuation. A separate discussion, however, would have underscored
its status as a witness to a notion about a part of speech118, without however
implying the unwarranted conclusion, rightly rejected by Callanan, that
Aristophanes elaborated, or was even aware of, a doctrine concerning the
ἴδια of the parts of speech119.
When it came to the formal description of language, Callanan identified
a more substantial contribution by Aristophanes in the area of semantics,
the field of study to which his collection of Λέξεις belongs120. Callanan
investigated the nature and purpose of this work, eventually rejecting
Pfeiffer’s thesis that Aristophanes had taken a diachronic approach to the
description of language121. On the other hand, it turns out that Aristo-

118 Callanan 1987, 28-30. See also Ax 1991, 279; Schenkeveld 1994, 275; Lallot
1997, II 286-287; Matthaios 1999, 588 and 608; Matthaios 2010c.
119 Callanan 1987, 30.
120 Callanan 1987, 75-96.
121 Callanan 1987, 75-82. For Pfeiffer’s position see supra, n. 81. Even if Callanan’s
disagreement with Pfeiffer’s conjectures of a structural, chronological subdivi-
sion into two groups of words and a primarily historical-linguistic purpose of
the beginning section of the Λέξεις is understandable, his rigorous denial of the
diachronic aspects in both the Λέξεις and ancient linguistic theory in general
seems excessive (see also Ax 1990, 14-15, who recognised among other things
that various Homeric scholia indicate that Alexandrian philology had an aware-
ness of diachrony [15 and n. 39]).
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 49

phanes rarely made recourse to etymology122, and when he did so it was

usually to clarify rare or obsolete expressions, often Homeric ones, or
glosses in the strict sense; he also used etymology to argue for certain
forms and orthographies123, though without going so far as to establish
general rules124.
Callanan illustrated the recourse made to etymology in order to explain
unusual words with the cases of δμῶες and δμωίδες (mss. M and L, s.v., frr.
322-324 Sl.), οἰκότριβες (mss. M and L, s.v.; Eust. ad Il. 1327.22-23, fr.
328 Sl.), κεράδες (Eusth. ad Od. 1625.45, fr. 162 Sl.), πτώξ, δασύπους,
ταχίνας (ms. M, s.v. λαγώς, frr. 188-190 Sl.), βόαξ (Ath. 7.287a, fr. 409
Sl.), μασχαλίσματα (Phot. μ 249, 19, s.v., Sud. μ 275, s.v., fr. 412 Sl.),
τρίγλη (Artemid. 2.14, fr. 377 Sl.). For clarification of glosses he adduced
the examples of: τηλύγετος (ms. M, s.v., fr. 234 Sl.), μολοβρός (Eust. ad
Od. 1817.19-23, fr. 197 Sl.); and for the determination of forms and or-
thography he cited (pp. 23-25): ἄαπτος / ἄεπτος (sch. Hdn. Il. 1.567b, fr.
418 Sl.), εἰσίθμη / εἰσίσθμη (sch. Od. 6.264, p. 198 Sl.), πῆξε / πτῆξε (sch.
Did. Il. 14.40b, p. 184 Sl.), παράνυμφος / παρανύμφιος (ms. P, s.v., Eust.
ad Il. 652.41ff., frr. 280-284 Sl.: see supra, 45).
Finally, Callanan dedicated an elaborate discussion to the analogical
method that is discernable in Aristophanic fragments (pp. 107-122),
identifying within these the use of four-term proportion and the devel-
opment of a catalogue of conditions of analogy, though no set of rules
for inflection, nor any normative purpose.
The principal evidence cited by Callanan for four-term proportion (pp.
115-119) is sch. Hdn. Il. 15.606b (p. 187 Sl.: see supra, 45), in which he re-
tained the reading Ἀριστοφάνης against Erbse’s conjecture (1960, 401-402)
Ἀρίσταρχος. He moreover presented (pp. 119-121) a series of references
by Aristophanes to similarities or analogies between expressions, formulated
without any formal structure and introduced simply by terms such as ὡς,
ὥσπερ, οὕτως.
For Aristophanes, the conditions which allowed for the insertion of
words into analogical relationships125 (pp. 57-58) were coincidences of
gender, case, ending, number of syllables, and accent; to these Aristarchus

122 Callanan 1987, 97-102, in particular 97; see also Nauck 1848, 268-269; Pfeiffer
1968, 201 with n. 4, 260; Slater 1986, 19.
123 Callanan 1987, 99-102.
124 Callanan 1987, 97-98; the idea that Aristophanes developed etymological rules
was upheld by Reitzenstein 1897, 184, perhaps on the basis of the evidence of
Varro (ling. 6.2, fr. 372 Sl.), who seems to attribute to Aristophanes a theory of
κλίσις in an etymological context: see also Slater 1986, 138; Schenkeveld
1990b, 297-298.
125 For the types of analogical relationships, see supra, n. 44.
50 Lara Pagani

added the prohibition against comparing simple with compound words

(Char. gramm. 149, 26ff. = fr. 375 Sl.: «huic [scil. analogiae] Aristophanes
quinque rationes dedit vel ut alii putant, sex: primo, ut eiusdem sint generis
de quibus quaeritur, dein casus, tum exitus, quarto numeri syllabarum, item
soni. Sextum Aristarchus, discipulus eius, illud addidit ne umquam simplicia
compositis aptemus». Nauck 1848, 269-270, emended the text of Cha-
risius, attributing to Aristophanes another condition of analogy: he inserted
quinto between numeri and syllabarum and corrected sextum to sexto, which
he read at the end of the sentence, thus modifying the punctuation:
«fourth, [words] having the same number; fifth, having the same number of
syllables, and likewise sixth, having the same accent»)126.
Callanan proceeded with extreme caution in inducing from this docu-
mentation Aristophanes’ premises, methods and objectives of linguistic
analysis127: he credited Aristophanes with a rich and advanced knowl-
edge of grammar, though one which operated according to a pragmatic
approach. This approach was guided by evidence and usage rather than
by theories or presumed laws, and was for the most part based on de-
scription as well as upon an understanding of language derived from
empirical observation, and was often aimed at Homeric textual criticism:
within such a framework, elaboration of a developed abstract system
would have been inconceivable128. Nevertheless, Callanan identified at

126 This evidence was called into question by Steinthal 1890-18912, II 81 n.; on
the subject see also Barwick 1922, 179ff.; Fehling 1956, 240-250; Siebenborn
1976, 72ff. and Ax 1982, 98 n. 8. The identification of the conditions of anal-
ogy was a much-debated issue in Greek and Roman antiquity: according to
Varro (ling. 10.8-10), Parmeniscus fixed them at eight (so also Caesar [fr. 11
Funaioli] and perhaps Varro [ling. 10.21-26; see also Fehling 1956, 248-249 and
nn. 1 and 2; 1957, 74-75 for the interpretative difficulties of this passage]),
Dionysius of Sidon at 71, and Aristocles at 14. Reconstructing Herodian’s posi-
tion in this respect (in GG III/II 634, 5-8) is difficult (see Colson 1919, 28 and
n. 4; Fehling 1956, 246 and n. 2; Siebenborn 1976, 73).
127 The indirect tradition poses clear obstacles to this type of reconstruction: it is
indeed difficult to establish whether the transmitted terminology is original or is
owed to some intermediary (all the more so given that the introduction of spe-
cialised grammatical concepts seems to have been one of the most frequent
types of interpolation introduced into the works of the earliest Alexandrians:
Callanan 1987, 37). Moreover, the fact that the scholia merely cite an Aristo-
phanic reading but fail to provide the reasons for that reading often renders un-
realistic any attempt at grasping any general theoretical background that may
lay behind it (pp. 51, 62, al.).
128 It is possible to reconstruct this synopsis from Callanan’s scattered observations:
see e.g. pp. 27, 30, 38, 40, 43, 72, 74, 122. The absence from Callanan’s
monograph of an overview, in addition to the caution of his argumentation has
led to varied critical receptions of his thought: e.g. Janko 1995a, 214 n. 1 wrote
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 51

least a few cases in which Aristophanes did not limit himself to practical
and analytical work, but also extracted general rules and theoretical con-
structs (see e.g. his discussion in general terms of questions of prosody [p.
32], his use of practical observations on the genders of nouns as prelimi-
nary work for further researches [p. 40], and his discussions of the verb
on the basis of his theoretical knowledge of inflection, which were in-
tended to lead to the creation or application of abstract categories or
structures [p. 64]). On the other hand he also de-emphasised Aristo-
phanes’ image as an “advocate of analogy” in polemical contrast with
the supporters of anomaly, an image that Callanan rejected as un-
founded129: he rather saw Aristophanes as a neutral observer of language,
who was descriptive and never prescriptive (pp. 103-106 and passim),
and who recognised and accepted anomalies in the doctrine of word
construction without claiming to regulate them in the manner of the
analogists (pp. 110-112)130.
With reference to this last point, Ax (1990) has called attention to a
piece of evidence that is found in Varro and not considered by Callanan
and which presents Aristophanes as an innovator, at least in some cases,
of the old linguistic consuetudo on the basis of the rules of analogy131: this

that «Callanan … goes too far in denying that Aristophanes had a grammatical
system», while according to Matthaios 1999, 22, Callanan did indeed, though
with every caution, credit him with such a system («… einen
Sprachbeschreibungsapparat, d.h. ein System grammatischer Regeln und Be-
griffe … Das Vorhandensein eines solchen Systems hat Callanan für Aristopha-
nes von Byzanz, wenn auch mit aller Vorsicht, angenommen»).
129 See Callanan 1987, 97-122.
130 Some useful correctives to the use of analogy in Aristophanes have been intro-
duced by Schenkeveld 1990b, 290-297, who has lucidly posed the question in
these terms: how much awareness of linguistic categories is necessary for apply-
ing an analogical method in a philological discussion? That is, what are the as-
sumptions of this method? Schenkeveld demonstrates, also thanks to the so-
called Donatiani Fragmentum (in GL VI 275.13-276.9), not considered by Cal-
lanan, that Aristophanes must have had both an awareness of regularities in lan-
guage and a tendency towards abstraction on the basis of accidental observa-
tions and, ultimately, that he was able to make use of grammatical classifications
to a greater extent than Callanan recognised.
131 Varro ling. 9.12 = fr. 374 Sl. («… artufices egregii non reprehendundi, quod
consuetudinem … superiorum non sunt secuti, Aristophanes improbandus, qui
potius in quibusdam veritatem [veteritatem codd.] quam consuetudinem secu-
tus?»). The passage contains textual uncertainties discussed by Ax 1990, 7-11,
on the basis of a parallel with Cic. orat. 155-162 (which contains an overview
of interventions made by the analogists of Cicero’s era on the numerous vari-
ants of consuetudo). The testimonium’s meaning is, however, accepted as be-
yond doubt.
52 Lara Pagani

would imply that Aristophanes both was aware of the diachronic devel-
opment of language and used analogy to make prescriptive interventions
on contemporary language, in sharp contrast with the interpretation
suggested by Callanan’s data. Ax hypothetically explained such a contra-
diction between the testimony found in Varro and the rest of the docu-
mentation as resulting from the chance nature of the tradition, which
may have done away with other fragments that would have helped to
clarify the picture. He moreover demonstrated prudent scepticism when
it came to the idea that Aristophanes abstained from applying his own
knowledge in the field of analogy (indicated by frr. 373 and 375 Sl.
too132) also to the regulation of doubtful cases of consuetudo133.
An overview of the contribution of Hellenistic philology to gram-
matical science was offered by Ax in 1991, with a summary of the
known evidence and principal positions taken by scholars in this area. In
the case of Aristophanes of Byzantium Ax concluded that, although his
conceptual and terminological contributions are only vaguely detectable,
he must have already had a notable grasp of morphology. While it is
certainly true that he never composed something like a compendium of
the parts of speech, he did concern himself with problems related to the
linguistic rules of etymology and analogy and used these rules in his role
as both textual critic and “guardian” of language; it cannot however be
determined whether he joined in the controversy between anomaly and
analogy134. In Aristarchus, on the other hand, Ax saw the first clear indi-
cations of the μέρος τεχνικόν becoming autonomous, although he rec-
ognised that Aristarchus’ grammatical competences functioned first and
foremost as an aid to his philological aims. According to Ax, in fact,
Aristarchus had a highly developed descriptive apparatus of language,
especially of morphology, at his disposal and began to shape the Alexan-
drian system of the eight parts of speech and their accidentia, even if he
was not yet able to rely upon a set of rules governing inflection. In his
observations about language Aristarchus had not only descriptive, but
also prescriptive, objectives, and to these ends above all applied the prin-
ciple of analogy. Taken as a whole, his contribution to the development

132 These are, respectively, Varro ling. 10.42.47, where it is said that Aristophanes
and others wrote about “perfect” analogies (where the compared words are also
similar in terms of meaning), such as bonus : malus = boni : mali («tertium genus
[scil. analogiae] est illud duplex quod dixi, in quo et res et voces similiter pro
portione dicuntur, ut bonus malus, boni mali, de quorum analogia et Aristo-
phanes et alii scripserunt»); and the fragment on the conditions of analogy (for
which see supra, 49-50).
133 See Ax 1990, 17.
134 Ax 1991, 277-282.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 53

of Hellenistic grammar should therefore be judged, in Ax’s opinion, as

extraordinarily significant135. Ax also considers Crates of Mallus briefly
for his definition of στοιχεῖον (whose original context is unknown),
which goes something like this: the element is the smallest part of φωνή,
and this functions in relation to the entire system of ἐγγράμματος
φωνή136. This definition presumes an awareness of the meaning of
φωνή, in addition to, if the attribution of the second part of the frag-
ment is correct137, the Stoic distinction between ἐγγράμματος and
ἀγράμματος φωνή, perhaps as well as the subdivision of written lan-
guage into the sequence στοιχεῖον, συλλαβή, λέξις, λόγος. This con-
ceptual background led Mette to compare the thin testimonium for
Crates with the content of systematic treatises on the parts of speech
such as the Techne of Diogenes of Babylon and the one ascribed to Dio-
nysius Thrax138. A parallel to the text in question is furthermore recog-
nisable in a papyrus fragment of a grammar (POsl. 13, col. I, 9-10: fr. 95
Broggiato, in the apparatus). The hypothesis that the definition of
στοιχεῖον appeared in a work dedicated to the doctrine of φωνή and to
its constituent parts, were it able to be substantiated, would contribute
evidence in favour of the “Stoic-Pergamene techne” postulated by Bar-
wick139. Nevertheless, the transmitted data are not sufficient to support
the idea140: in fact, an entirely different context for this fragment, namely
Crates’ work on poetics, has been proposed on the basis of Philodemus’
reference to Crates’ theories concerning στοιχεῖα, upon which «he
claims judgments of good poetic works are founded» (PHerc. 1425 and
PHerc. 1538, col. XXIX, 7-15 = fr. 101b Broggiato)141. Finally, Ax’s

135 Ax 1991, 282-288.

136 Sch. Marc. D.T., in GG I/III 316.24 (fr. 52 Mette = 95 Broggiato): ὁρίζεται δὲ
τὸ στοιχεῖον ὁ μὲν Κράτης οὕτω, φωνῆς μέρος {τὸ} ἐλάχιστον· μέρος
ἐλάχιστον εἶπεν ὡς πρὸς τὸ ὅλον σύστημα τῆς ἐγγραμμάτου φωνῆς …; Aris-
totle’s definition and another four anonyms follow. The words μέρος
ἐλάχιστον εἶπεν - φωνῆς, reported by the two mss. of the Scholia Marciana (V
and N) after the citation of Aristotle have been transposed to the end of the
Cratetean fragment by Hilgard, a transposition accepted by Crates’ editors. The
passage is discussed by Ax 1991, 288-289; see also Mette 1952, 5-6 and 67-68
(who also removed the expression τῆς κατὰ σύνταξιν from Aristotle’s defini-
tion so as to put it at the end of the citation of Crates; contra Broggiato 2001 [=
2006], 251); Ax 1986, 218-223; Janko 2000, 123-124 n. 6; Broggiato 2001 (=
2006), XXXVI and 250-253.
137 See the preceding note.
138 Mette 1952, 2-6.
139 See supra, 28.
140 See Ax 1986, 223; 1991, 289; Broggiato 2001 (= 2006), 253.
141 Broggiato 2001 (= 2006), 252-253, with bibliography.
54 Lara Pagani

study sought to reconstruct the arguments of the debate between Crates’

approach to the defence of anomaly (that is, the principle of inflectional
irregularity, with the resulting assumption that linguistic usage consti-
tutes a normative criterion) and Aristarchean analogy (which established
regularity as a rule in the process of derivation). Ax saw no reason to
doubt the existence of this conflict of ideas as to how language
worked142, though he declared himself uncertain as to its actual magni-
tude and influence in the ancient world. According to him, the problem
of analogy must not have been a purely academic matter of marginal
importance, but could have been a question that arose from attempts to
establish the correct texts of the classics and the ever-growing need for
regulating language. That need concerned not only the limited field of
philology, but also the world of education, with important repercussions
for rhetoric as well – in short, this was high-profile debate that swept
through all levels of the educational system and in which the most im-
portant figures of the age had participated143.
The same years saw the publication of other overviews of the sub-
ject. D. J. Taylor (1987) had read into the positions of scholars such as
Di Benedetto, Pinborg, Siebenborn, Frede, and Fehling (1979) a new
interpretative model of the history of ancient linguistics, with which he
agreed144. This model no longer treated grammar as the result of a series
of cumulative acquisitions145; it also eliminated the analogy/anomaly
debate146 and avoided creating an explicit dichotomy between technical
and philosophical grammar. As to the Alexandrian scholars, they pos-
sessed, for Taylor, only a few grammatical ideas and did not always have
known how to manage them or what to do with them: their work was
exclusively philological, and the remarks on analogy attested for Aristo-
phanes and Aristarchus should be interpreted strictly in this context and
not in relation to any putative knowledge of the rules of declension,
inflection, paradigms and similes – an idea which, in Taylor’s opinion,

142 This was contested by Fehling 1956, 264-270 (see supra, n. 46, with bibliogra-
phy).
143 Ax 1991, 289-295.
144 Taylor 1987, who nevertheless recognised (see in particular p. 16) that the new
model was not yet finalised and warned of the necessity of constructing a more
structured and distinct architecture for it before trusting in its ability to com-
pete successfully with the traditional model that it was intended to replace.
145 See in particular Taylor 1987, 4-6 and 13.
146 The history of scholarship on the subject is traced by Taylor 1987, 6-8 (see
immediately supra and n. 46).
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 55

ought to be entirely rejected147. Thus, for Taylor, in this phase grammar

had not yet emerged as an independent field of study, something which
would not take place until the Ist century B.C.148
The account of the status quaestionis outlined by D. M. Schenkeveld
(1990b) was avowedly a continuation of Taylor’s work. In it Schen-
keveld made a number of useful observations, particularly about Ax’s
study of ancient ideas about φωνή and its connections to the concept of
language (1986)149 and Callanan’s research on the linguistic fragments of
Aristophanes of Byzantium (1987)150. Schenkeveld’s careful analysis of
Ax’s findings raised doubts as to the plausibility of a bipartite model of
exposition περὶ φωνῆς within the Greek and Roman grammatical tradi-
tion; he also introduced a substantial corrective to Callanan’s results by
establishing with greater certainty that Aristophanes had made use of
grammatical classifications.
Later, Schenkeveld (1994) sketched a picture of the techne gram-
matike’s relationship with Alexandrian philology, addressing the role of
the handbook attributed to Dionysius Thrax151, the doctrine of the parts
of speech, the theory of hellenismos, and the question of syntax152. He
highlighted how Aristophanes and Aristarchus had made a number of
strictly grammatical observations and had at their disposal a considerable
apparatus of linguistic distinctions thanks to their drawing upon Stoic
categories and to their transformation of these into grammatical word-

147 Taylor 1987, 12-13. A similar position was rightly judged to be excessive by
Matthaios 1999, 31, who on the other hand also demonstrates the limits of
Erbse’s thesis, which ran to the other extreme (see supra, 40-42).
148 Taylor 1987, 11.
149 See supra, 44.
150 See supra, 44-51.
151 Schenkeveld 1994, 266-269, 291-292; he would later address the subject in
greater detail (Schenkeveld 1998: see supra, n. 79).
152 This area of linguistic analysis was discussed by Schenkeveld, beginning with
the systematic study devoted to it by Apollonius Dyscolus, before whom it
seems to have received only minor attention (Schenkeveld 1994, 293-298). On
the issue, which requires some conceptual distinctions, see recently Swig-
gers/Wouters 2003, as well as Matthaios 2003 and 2004 for a restatement of the
role of Trypho, whom some critics had wanted to see as a predecessor of Apol-
lonius in this field. On the contrary, from Matthaios’ examination it becomes
clear that Trypho’s work implies theoretical discussion, however rudimentary,
of syntactical data; nevertheless it does not demonstrate a developed analysis of
the topic, to the extent that it fails to address the crucial questions of the cor-
rectness of constructions and of their linguistic components.
56 Lara Pagani

classes153. In the beginning these tools were used primarily in the context
of philological research and exegetical work, though Schenkeveld did
not rule out that the earliest Alexandrians might have discussed, at least
orally, some theoretical aspects of the grammatical ideas available to
them, and may have made original contributions in this area154. When it
came to hellenismos, the conditions of analogy set by Aristophanes and
Aristarchus cannot demonstrate, to Schenkeveld’s mind, extensive
awareness of paradigms or of an elaborate system of declension and con-
jugation, nor can they indicate the institution of a theory of κανόνες, as
would later be the case. Rather, these conditions may permit a glimpse
of initial attempts to compile lists of inflections155. Given these premises,
then, the first steps towards definition and description of the discipline
have been taken by Dionysius Thrax, while genuine systematisation of
grammar as a techne have only occurred in the next generation, with
Asclepiades of Myrlea and above all Trypho156.
An entirely different point of view was adopted by F. Ildefonse
(1997) who, sharing Di Benedetto’s interpretation157, investigated the
problem of grammar’s first origins in antiquity by programmatically ex-
cluding from her study the philological approach of figures such as Aris-
tophanes and Aristarchus158. Their contribution to linguistic analysis,
made before the great turning point (which she agreed with Di
Benedetto to have taken place in the Ist century B.C.) was purely acci-
dental, simply a means to a philological end and not an aim pursued for
its own sake. Ildefonse’s work therefore started from “philosophical”

153 Doubts persist as to the actual origins of the terminology attested by the sources
for the parts of speech, which may be later (see however Matthaios 1999, 43-
46), but it is highly likely that the distinctions, apart from their nomenclature,
had already been made (Schenkeveld 1994, 276-278, 280). The Stoic influence
becomes even more pronounced in the next generation (see supra, 33) with
Dionysius Thrax and Apollodorus of Athens (Schenkeveld 1994, 280-281).
154 The fact that systematic technical treatises are not attested for the Alexandrians
does not itself constitute sufficient grounds for refuting this hypothesis (Schen-
keveld 1994, 278).
155 Schenkeveld 1994, 283-287.
156 Schenkeveld 1994, 278-281. These results are taken up again in Schenkeveld
1998, 46-47.
157 See supra, 31-34.
158 Ildefonse 1997, in particular 27. The study begins with an introduction to the
concept of grammar in Greek antiquity, its relationship with philosophy, its
various ancient definitions, and its nature and its constitutive parts; it also al-
ludes to the problem of the late emergence of the discipline. Aristophanes and
Aristarchus’ contributions are summarised in the chapter on empirical vs tech-
nical approaches (pp. 20-23), but then explicitly set aside.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 57

grammar (Plato, Aristotle, the Stoics) and came directly to the “techni-
cal” grammar of Apollonius Dyscolus, confining to an appendix a note
on the Techne attributed to Dionysius, which she dated to a period later
than Apollonius’. This sort of method, however, prevents us from view-
ing properly all of the factors which contributed to the development of
theories of language, in that it obliterates an entire intermediate stage
which a more patient investigation would reveal to have been of crucial
importance159.
New impetus in this direction arrived, under the auspices of W. Ax,
from the work of S. Matthaios (1999), who gathered and examined all
of the texts of the Aristarchean tradition having to do with the doctrine
of the parts of speech: a limited selection but one of vital importance for
ancient grammatical theory160. For the first time, a study of Aristarchus
did not advance on the basis of a limited sampling, but instead proposed
a reconstruction based on an ordered and exhaustive collection of evi-
dence in order to establish Aristarchus’ position and role in the devel-
opment of ancient linguistic science161. This extensive study allowed for

159 See the reservations expressed by Schenkeveld 1999 and Matthaios 2002, 165
n. 16.
160 Matthaios 1999. The number of passages relevant to Aristarchus’ approach to
language as a whole amounts to about 5,000. A half of these concerns the doc-
trine of parts of speech and only this group has consequently been took into
consideration by Matthaios: roughly a third has been analysed in detail in the
theoretical section and 225 pieces have been registered as fragments (in the text
Matthaios prints the passages which are most important and most certainly
traceable to Aristarchus’ own formulations, whereas other loci similes of Aris-
tarchean origin are referred to in the apparatus of the testimonia) (Matthaios
1999, 33, 59ff.). The sources consist in the scholia to Homer, Hesiod, Pindar,
and Apollonius of Rhodes; the lexicon of Apollonius Sophistes; and in the
writings of Apollonius Dyscolus, Herodian, Priscian, Maximus Planudes, Cha-
risius-Donatian, Varro, and Quintilian (pp. 36-59). The risks posed by the indi-
rect tradition, especially when it comes to terminology, are well highlighted,
even if trust is placed in the credibility of Aristonicus (pp. 43-48): Aristarchus’
knowledge in the field of grammar can in fact be gleaned from his application
of linguistic ideas to textual criticism, though we cannot be sure of accurately
recovering his theoretical premises, nor that Aristarchus moved from empirical
observation to theoretical description of grammatical phenomena; potential
conclusions are also restricted by the chance nature of this material’s selection
in the course of transmission (pp. 33-36).
161 The material is arranged by content according to a structure that reproduces
the eight parts of speech, knowledge of which is attributed to Aristarchus by a
passage of Quintilian that has been debated (inst. 1.4.17 = fr. 1 M., see imme-
diately below): noun, verb, participle, article, pronoun, adverb, conjunction,
and preposition.
58 Lara Pagani

confirmation of Quintilian’s account of Aristarchus’ knowledge of the

eight parts of speech162, as well as for the attribution to Aristarchus of a
conceptual apparatus of linguistic description involving a certain degree
of abstraction, differentiation, and systematisation of grammatical doc-
trine. Since it is neither possible nor necessary to retrace in detail the
entire set of evidence analysed by Matthaios, here I shall discuss only a
few of the most important results that he achieved163.
In the case of nouns (frr. 1-53 M.) we find investigations of ὄνομα as the
technical term for proper names, common nouns, and adjectives164, as well
as of the opposition between ἰδίως and κοινῶς λεγόμενον, and the catego-
ries κύριον ὄνομα, προσηγορία and ἐπίθετον; for semantics there are dis-
cussions of homonymy, synonymy, nouns of genus and species, collective
nouns, nouns characterised by an ἔννοια περιεκτική, and nouns with active
meaning (δραστικὴ ἔννοια); for constructions Matthaios examines frag-
ments relating to composition, derivation, patronymics, denominatives, and
comparative forms of adjectives; regarding case there are discussions of de-
clension and examples of case-switching. As to verbs (frr. 54-93 M.), ob-
servations emerge regarding diathesis, with particular attention to the mid-
dle voice, of which Aristarchus must have known something165, as well as
instances of passives in place of actives and vice versa. For tenses there are
fragments on the oppositions present/future, present/past, and imper-
fect/aorist (παρατατικός [scil. χρόνος] / συντελικόν [scil. ῥῆμα / σχῆμα])
and on the tenses of participles and infinitives; as to moods, Aristarchus
must have been aware of the infinitive and imperative, of the optative and
its use, and of the subjunctive and indicative. About number we hear of the
use of the dual in Homer, agreement of subject and predicate (plural verb
with plural neuter subject, constructions «according to sense» [σχῆμα πρὸς

162 The evidence’s reliability has often been doubted, and the introduction of the
eight-part system of speech has been assigned to Trypho’s generation. An ac-
count of the question can be found in Matthaios 1999, 191-198. Aristarchus’
tradition is viewed as concluding the processes of formation and expansion of
the grammatical system of parts of speech, signs of which are already evident in
Aristophanes of Byzantium (pp. 621-622).
163 Pontani 2002b presents a series of observations on and proposals for improve-
ment regarding individual points, especially when it comes to readings of the
scholia to the Odyssey, as well as the suggestion that some statements might be
mitigated (e.g. that regarding the presence of «eine Tendenz zur grammatischen
Abstraktion und Normierung» with respect to conjugations, p. 410). These are
admittedly remarks on individual points, and are not intended to invalidate the
importance of the work overall.
164 On which see the clarification of Pontani 2002b, 149-150 as to the correct
textual arrangement of sch. Od. 4.221 [fr. 4 M.], which would eliminate the at-
testation of the use of ὄνομα for adjectives.
165 Pace Steinthal 1890-18912, II 106.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 59

τὸ νοητόν], Ἀλκμανικὸν σχῆμα [in a sentence with two subjects, the verb
occurs after the first but in number conforms to both, in either a plural or
dual from], Πινδαρικὸν σχῆμα [in a sentence with a plural masculine or
feminine subject the verb is singular], and interchanges between one num-
ber and another). Aristarchus’ concepts of πρόσωπον for grammatical per-
son and συζυγία for conjugations are also reconstructed. As to the partici-
ple (frr. 94-95 M.), Aristarchus was as far as we know the first to isolate it
as a distinct part of speech, which he called μετοχή. With regards to articles
and pronouns (frr. 96-135 M.), Aristarchus is also responsible for the most
ancient known attestation of the term ἄρθρον for the article (a category di-
vided into ἄρθρα προτακτικά, i.e. ὁ, ἡ, τό, and ἄρθρα ὑποτακτικά, i.e.
ὅς, ἥ, ὅ) and the definition of pronouns, which Aristarchus probably called
ἀντωνυμίαι, such as λέξεις κατὰ πρόσωπα σύζυγοι. Intense work on this
last part of speech is also documented: there are notes on agreement of per-
son in personal and possessive pronouns, the coinage of the term
ἐπιταγματική (scil. ἀντωνυμία?) for αὐτός used in conjunction with the
tonic forms of personal and possessive pronouns, and observations on per-
sonal and reflexive pronouns as well as on personal and possessive ones. For
adverbs (frr. 136-167 M.) there is a collection of morphological, syntactic,
and semantic observations, with discussion of this part of speech’s history
and its designation as μεσότης. There are no attested theoretical comments
about conjunctions (frr. 168-178 M.) and prepositions (frr. 179-225 M.),
but only observations regarding Homeric usage (on the subjects of coordi-
nating, causal, and expletive conjunctions; on the interchanging of preposi-
tions and adverbs, the interchanging of prepositions amongst themselves, as
well as prepositions’ omission, pleonastic use, government, and accentua-
tion), in addition to the respective appellations σύνδεσμος and πρόθεσις.
Among the doctrine’s theoretical components it is possible to identify
on the one hand Stoic thought (relating to verb tenses, the concept of
case, the term μεσότης for the adverb)166, and on the other the Aristote-
lian-Peripatetic tradition (the conception of the noun and its internal
divisions, the theory of the pronoun, and the concept of grammatical
person), which here is counted as significant even though it had previ-
ously been little considered167. Furthermore, various affinities with the
contents of the Techne allow us to recognise the antiquity, at least in
some places, of the grammatical theory crystallised in the latter168. Mat-
thaios’ conclusion is that the work of Aristarchus and his contemporaries

166 Though also relevant are the departures from the Stoic approach, e.g. the con-
ception of the nature of the article and of the relationship between proper and
appellative names. See the previous page and infra, n. 173.
167 Matthaios 1999, 623-625; see supra, n. 25.
168 Matthaios 1999, 265, 282-283, 623; see also Matthaios’ position in 2009, men-
tioned supra, n. 76.
60 Lara Pagani

laid the essential foundations which made later technical elaborations of

linguistic science possible: Aristarchus had indeed taken part in the con-
temporary debate about language, participated in the process of “gram-
maticising” philosophical and rhetorical-literary concepts, and coined
new linguistic categories169. The primacy of his philological/exegetical
work is by no means denied; rather this area is upheld as the one in
which the first Alexandrians exercised the apparatus of linguistic descrip-
tion, i.e. the system of grammatical rules and concepts, at their dis-
posal170.
On the basis of these results Matthaios has developed, in the course
of the last decade, a broader view of Hellenistic approaches to linguistic
phenomena. In the first place he presents a comprehensive synopsis of
the history of the system of parts of speech from its first origins to its
reception in the Roman world and demonstrates that Aristarchus, as
well as his predecessors and contemporaries, played a decisive role in the
development of this theory171. The evidence that he has gathered allows
us to see that the Aristotelian-Peripatetic tradition had a conspicuous
effect on Alexandrian description of language; it also allows us to hy-
pothesise, in contrast with the traditional theory of a linear evolution
(Plato-Aristotle -> Stoa -> Alexandrians), as to parallel developments of
Alexandrian philology and philosophy, especially Stoicism, which in
some cases exerted mutual influence upon each other172. The influence
of the Stoic doctrine intensified in the generation following Aristarchus,

169 Matthaios 1999, 625.

170 Matthaios 1999, 21-22.
171 Matthaios 2001 (= 2005) and 2002. Matthaios sees the dawn of this process as
having occurred in the generation before Aristarchus: Aristophanes of Byzan-
tium spoke of prepositions as an independent category (see supra, 48), and the
concept of ἀντωνυμία for pronouns must have been in existence before Co-
manus (an older contemporary of Aristarchus), who proposed to replace it with
ἀντωνομασία; the nature of the participle was investigated by both Comanus
himself and Callistratus (Matthaios 2001 [= 2005], 69-71; 2002, 166-168).
172 Matthaios 2001 (= 2005), 80-83; 2002, 188-191. For traces of Aristotle’s influ-
ence, see immediately above and Matthaios 2001 (= 2005), 73-76. Matthaios
bases this theory of parallel development primarily upon the history of the con-
sideration of prepositions and participles as independent parts of speech (see re-
spectively Matthaios 2001 [= 2005], 76-78 and 78-80; 2002, 179-184 and 185-
187): in this process the grammarians beat a path independent from philosophi-
cal theorising, yet one which eventually came to represent the theoretical
work’s point of reference within the philosophical tradition. A more direct
Stoic influence can however be recognised in the case of the adverb (Aris-
tarchus’ term for adverbs, μεσότης, seems to belong to the context of Stoic
logic: see Matthaios 2001 [= 2005], 80; 2002, 187-188).
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 61

partly for historical reasons connected to the secessio doctorum that had led
both to the reestablishment of philological study in centres such as Rho-
des, Pergamum, Athens, and later also at Rome, and to the Alexandri-
ans’ encounter with Stoic-Pergamene ideas: Demetrius Ixion worked at
Pergamum, Apollodorus also worked there and then at Athens, acting as
an intermediary between the doctrine of Diogenes of Babylon and the
grammarians; Dionysius Thrax was engaged at Rhodes, where the Stoic
school was active first with Panaetius and later with Posidonius. Signs of
a change in the Alexandrian tradition due to Stoic linguistic influence
can also be detected in the work of Dionysius Thrax himself173 and of
Tyrannion, who was the author of a specific treatise Περὶ μερισμοῦ τῶν
τοῦ λόγου μερῶν174.
In the surviving evidence for Tyrannion we can recognise clear signs of Al-
exandrian influence (e.g. the classification of προσηγορικὰ ὀνόματα not as
a separate category, but in subordination to nouns [fr. 56 Haas]; and use of
the term κύριον ὄνομα for the proper noun – for the Stoics simply ὄνομα
[ibid.]), as well as other elements of Stoic derivation (e.g. the concept of
κύρια ὀνόματα as ἄτομα, i.e. “individuals” [ibid.], which presupposes a re-
duction of the concept of κύριον ὄνομα to that of the personal name – not
the case with Aristarchus175; identification of the participle as a subdivision
of the noun and its characterisation as “non-thematic”, i.e. derived [ibid.];
and the definition of pronouns as σημειώσεις because they refer to specific
persons (ὡρισμένα πρόσωπα), a notion which reflects the Stoic idea of
πρόσωπον as an actual individual, identifiable by means of deixis or
anaphora [fr. 58 H.])176.
This process of advancement led to a first codification by Trypho (sec-
ond half of the Ist century B.C.), in whose work we find both a clear
rejection of the Stoic view – especially when it came to the arrangement
of ὄνομα and προσηγορία and the autonomy of the participle – and
first theoretical foundations of the Alexandrian system177. After Trypho,

173 Stoic elements in his thought include the notion of nouns and appellatives
constituting two distinct categories, his definition of ῥῆμα, and the fact that he
called pronouns ἄρθρα δεικτικά: these points are discussed in detail supra, 33.
See Matthaios 2001 (= 2005), 84; 2002, 192-193.
174 On this work and its original title, see Haas 1977, 167-169.
175 See Matthaios 1996.
176 See Matthaios 2001 (= 2005), 85-86; 2002, 193-195.
177 See Matthaios 2001 (= 2005), 86-87; 2002, 195-197. Trypho’s work in this
area is documented by the rich catalogue of titles found in the Suda (τ 1115,
s.v. Τρύφων Ἀμμωνίου; see Velsen 1853, 3-4). For Trypho’s idea of ὄνομα and
προσηγορία, see sch. Marc. D.T., in GG I/III 356.21-357.26; for the partici-
ple, see fr. 39 von Velsen. Trypho’s adherence to the Alexandrian tradition is
62 Lara Pagani

however, this evolution was prolonged for another couple of centuries

or so, during which there was a great deal of vacillation and fluctuation,
until the system achieved canonisation with Apollonius Dyscolus. He
dedicated specific writings to every part of speech and its accidentia; he
also synthesised the philological and philosophical traditions into a uni-
tary construct, creatively integrating the Stoic theory into the Alexan-
drian doctrine and finally overcoming the two-pronged evolutionary
line of the doctrine of the parts of speech178.
Something that the history of this development, which lasted for
roughly 400 years, throws into sharp relief is that the role played by
Aristarchus and by his immediate predecessors and contemporaries was
decisive for the codification and establishment of concepts and termi-
nology concerning the system of language. This system was then studied
on a theoretical level and gradually improved over the course of a long
tradition of scholarship179.
Studies of and opinions, often divergent, about the comprehensive
model of ancient grammar have continued to flourish on the interna-
tional scene180: the miscellany volume edited by P. Swiggers and A.
Wouters (2002a) served precisely as a kind of catch-basin for these. In
the introduction to this book and in a later article (2005), they outline a
reconstruction of the process of grammar’s development into a separate
discipline within the Greek world, emphasising its connections with
philosophy, rhetoric, and poetics, and proposing a handful of guidelines
for tracing the evolution of the model of parts of speech. The grammar
of the philologists, i.e. their study of language with a view to textual
criticism181, appears to constitute a stage within this evolution.

also demonstrated by his distinction of article (frr. 22-27 V.) and pronoun (frr.
28-37 V.), his assignment of the adverb (ἐπίρρημα) to a separate category (frr.
62-77 V.), and by his identical handling of the preposition (πρόθεσις: fr. 40
V.).
178 Matthaios 2001 (= 2005), 88-89; 2002, 197-199.
179 The issue is well summarised in Matthaios 2001 (= 2005), 70, 90; 2002, 168-
169, 212-213.
180 To cite just a couple of recent examples, Law 2003, 54ff. dedicates a chapter in
her survey of the history of linguistics to the description of language in antiq-
uity, in which she examines the work of figures such as Aristophanes and Aris-
tarchus as indications of the existence of some meta-linguistic notions begin-
ning in the IInd century B.C., but not – or better, not necessarily – as evidence
of a complex system of grammatical knowledge. On the other hand Matthews
2007 took the Alexandrian scholars’ work alongside the expertise of the Stoic
philosophers into consideration as an aspect of grammar’s transformation into a
technical science.
181 Swiggers/Wouters 2002b, in particular 11-17; 2005, in particular 8ff.
 Pioneers of Grammar. Hellenistic Scholarship and the Study of Language 63

Thus in the debate’s current state, the critical points hinge on one’s
choice of perspective: the ancient evidence, at least the evidence relating
to the main figures in this history, has now been collected and studied en
bloc and therefore no longer allows for observations to be made on the
basis of a few examples picked from the heap of material. Nevertheless,
disagreement persists regarding the nature of the first Hellenistic philolo-
gists’ linguistic interests, a debate which may be ultimately traced back
to the ancient dispute as to the epistemological value of grammatike:
whether empirical in nature or a techne182. For one group of critics, the
fact that the observations which the scholars – especially Aristarchus –
made about language both arose from the context of their work on liter-
ary texts and were made to that end remains something that strongly
prejudices evaluation of their nature. For these critics the philologists’
linguistic interests, inasmuch as they originated as an offshoot of textual
criticism and exegesis and never held the study of language as an aim in
itself, did not have any independent speculative meaning: thus even if
they recognise that these individuals had grammatical competence, they
indeed still deem it unacceptable to credit them with a grammatical
theory183.
It is a fact that we do not find in Aristarchus and his contemporaries
traces of (aspiration towards) comprehensive linguistic systematisation,
nor of any specific treatises on such subjects; it is moreover certainly true
that Alexandrian grammar was born, to use Ax’s felicitous phrase, «im
Kopf», as a basis and tool for understanding and interpreting literary
texts. Nevertheless it is also possible to outline a framework in which
philological research and linguistic discussion, far from being entirely
separate fields or in conflict with each other, reveal themselves to be
closely and productively interrelated, without the value of the latter
being diminished as a consequence. The two areas have now been well
synthesised by Matthaios (2010b), who sets out primarily to verify how
the observations of language made by the Alexandrians appeared in
practice, what factors prepared and influenced their linguistic-
grammatical argumentation, in what way grammatical rules were
formed, and in what measure the text under consideration could prove
decisive for the recording of a grammatical phenomenon and the recog-
nition of a rule. He also investigates the ways in which the Alexandrians
made abstractions on the basis of literary contexts so as to identify a

182 See Matthaios 2010b, discussed at greater length below.

183 This kind of approach underlies e.g. F. Schironi’s Playing with language. Homeric
grammar according to Aristarch, a paper delivered at the 2nd Trends in Classics inter-
national conference (Thessaloniki, December 2008).
64 Lara Pagani

grammatical phenomenon, describe it, formulate a rule based upon it,

and expand by means of it their apparatus of linguistic description. In
particular, it is possible to demonstrate on these grounds that at least
some of Aristarchus’ linguistic observations effectively suggest a certain
degree of abstraction, and presuppose some theoretical reference points
(definitions, technical names, judgments as to a certain category of
forms’ correctness)184. The fact that these general formulations born out
of the demands of the work of diorthosis of literary texts could then find
concrete application precisely in that work completes the picture of the
interaction – and not opposition – between grammar and philology.
This seems to be the view that best illuminates the conditions be-
hind the Hellenistic philologists’ first approaches to linguistics, and the
means by which those approaches were realised, since it assigns just im-
portance to “grammatical” interventions on texts, without however
failing to recognise textual criticism and exegesis as the original motiva-
tion. In this way, we can set aside the idea that there was a clear gap in
the field of linguistic study between Aristarchus’ era and that of the next
generations in favour of a view of a process marked by evolution and
successive refinements, developed over the course of centuries thanks
also to comparison with the work of the predecessors, pioneers in this
field of study.

184 The examples mentioned by Matthaios pertain to the category of the pronoun:
these have to do with the description of pronouns as «words that form series
according to person» (fr. 103 M.), the coinage of the name ἐπιταγματική (scil.
ἀντωνυμία?) for the form αὐτός when used alongside tonic forms of personal
and possessive pronouns (fr. 120 M.), and criticism of the correctness of com-
posite pronouns’ third-person plural forms (fr. 125 M.).
 Bibliography

Abbamonte, Giancarlo (2004), „Tipologie esegetiche nei commenti di

Alessandro di Afrodisia: la parafrasi“, in: Giancarlo Abbamonte / Ferruccio
Conti Bizzarro / Luigi Spina (eds.), L’ultima parola. L’analisi dei testi: teorie e
pratiche nell’antichità greca e latina. Atti del terzo Colloquio italo-francese
coordinato da Luigi Spina e Laurent Pernot, Napoli 13-15 marzo 2003,
Napoli, 19-34.
Adams, Charles D. (ed.) (1919), The Speeches of Aeschines, with an English
translation, London-New York.
Alberti, Giovan Battista (ed.) (1972), Thucydidis Historiae, I, Libri I-II, Roma.
Allen, Thomas W. (1931), „The Homeric scholia“, in: Proceedings of the British
Academy 17, 179-207 (as abstract: London 1931).
Alpers, Klaus (1991), „Eine byzantinische Enzyklopädie des 9. Jahrhunderts“,
in: Guglielmo Cavallo / Giuseppe De Gregorio / Marilena Maniaci (eds.),
Scritture, libri e testi nelle aree provinciali di Bisanzio. Atti del Seminario di
Erice (18-25 settembre 1988), Spoleto, I 235-269.
Andò, Valeria (2005), L’ape che tesse. Saperi femminili nella Grecia antica, Roma.
Andorlini, Isabella (2000), „Codici papiracei di medicina con scoli e
commento“ in: Marie-Odile Goulet-Cazé (ed.), Le commentaire entre
tradition et innovation. Actes du colloque international de l’Institut des
traditions textuelles (Paris et Villejuif, 22-25 septembre 1999), Paris, 37-52
(Pl. 1-7).
— (2003), „L’esegesi del libro tecnico: papiri di medicina con scoli e
commenti“, in: Papiri filosofici. Miscellanea di studi, IV, Firenze, 9-29 (Pl. I-
VII).
Arata, Luigi / Bastianini, Guido / Montanari, Franco (eds.) (2004), „Aeschylus
3. P.Oxy. XX 2256. Hypotheseis di varie tragedie“, in: CLGP I.1.4 (Aris-
tophanes-Bacchylides), 34-51.
Arens, Hans (2000), „Sprache und Denken bei Aristoteles“, in: Sylvain Auroux
/ Ernst F. K. Koerner / Hans-Joseph Niederehe / Kees Versteegh (eds.),
History of the Language Sciences. An International Handbook on the Evolution of
the Study of Language from Beginnings to the Present / Geschichte der Sprachwis-
senschaften. Ein internationales Handbuch zur Entwicklung der Sprachforschung
von den Anfängen bis zur Gegenwart / Histoire des sciences du langage. Manuel
international sur l’évolution de l’étude du langage des origines à nos jours, I, 1, 1,
Berlin-New York, 367-375.
Ascheri, Paola (2009), „Demetrius [5]“, in: LGGA.
164 Bibliography

— (2010), „Aristodemus [3]“, in: LGGA.

Ax, Wolfram (1982), „Aristarch und die “Grammatik”“, in: Glotta 60, 96-109
(reprinted in: Ax 2000, 128-139).
— (1986), Laut, Stimme und Sprache. Studien zu drei Grundbegriffen der antiken
Sprachtheorie, Göttingen.
— (1987), „Quadripertita ratio: Bemerkungen zur Geschichte eines aktuellen
Kategoriensystems (Adiectio, Detractio, Transmutatio, Immutatio)“, in: Daniel
J. Taylor (ed.), The History of Linguistics in the Classical Period, Amsterdam-
Philadelphia, 17-40 (reprinted in: Ax 2000, 190-208).
— (1990), „Aristophanes von Byzanz als Analogist. Zu Fragment 374 Slater (=
Varro, de lingua Latina 9, 12)“, in: Glotta 68, 4-18 (reprinted in: Ax 2000,
116-127).
— (1991), „Sprache als Gegenstand der alexandrinischen und pergamenischen
Philologie“, in: Schmitter 1991, 275-301 (reprinted in: Ax 2000, 95-115).
— (1993), „Der Einfluss des Peripatos auf die Sprachtheorie der Stoa“, in:
Klaus Döring / Theodor Ebert (eds.), Dialektiker und Stoiker. Zur Logik der
Stoa und ihre Vorläufer, Stuttgart, 11-32 (reprinted in: Ax 2000, 73-94).
— (2000), Lexis und Logos. Studien zur antiken Grammatik und Rhetorik, hrsg.
von Farouk Grewing, Stuttgart.

Baladié, Raoul (1980), Le Péloponnèse de Strabon. Étude de géographie historique,

Paris.
Baltussen, Han (2008), Philosophy and Exegesis in Simplicius. The Methodology of a
Commentator, London.
Barker, Edmund H. (ed.) (1820), Arkadiou peri tonon, e codicibus Parisinis.
Addita est editoris epistola critica ad J. F. Boissonade, Leipzig.
Barwick, Karl (1922), Remmius Palaemon und die römische Ars grammatica, Leipzig.
— (1957), Probleme der stoischen Sprachlehre und Rhetorik, Berlin.
Bastianini, Guido (2001), „Osservazioni sul papiro di Teramene“, in: Ποίκιλμα.
Studi in onore di Michele R. Cataudella, La Spezia, 81-87.
Bastianini, Guido / Gallazzi, Claudio (eds.) (2001), Papiri dell’Università degli
Studi di Milano, VIII, Posidippo di Pella. Epigrammi (P. Mil. Vogl. VIII 309),
con la collaborazione di Colin Austin, Milano.
Bastianini, Guido / Lundon, John (2000), „Coniugazioni verbali e scholia minora
in un polittico scolastico“, in: H. Melaerts (ed.), Papyri in honorem Iohannis
Bingen octogenarii, Leuven, 31-42 (Pl. 4).
Bastianini, Guido / Sedley, David N. (eds.) (1995), „Commentarium in
Platonis «Theaetetum»“, in: CPF III, Firenze, 227-562.
Bayet, Jean (1920), „Les origines de l’arcadisme romain“, in: Mélanges
d’archéologie et d’histoire de l’École Française de Rome 38, 63-143.
Bécares Botas, Vicente (ed.) (1987), Apolonio Díscolo, Sintaxis, introducción,
traducción y notas, Madrid.
 Bibliography 165

Bednarski, Michał (ed.) (2000), Apollonios Dyskolos, O składni, Kraków.

Berndt, Richard (1902), De Charete, Chaeride, Alexione grammaticis eorumque
reliquiis, I, Charetis Chaeridisque fragmenta quae supersunt, Königsberg.
Bickerman, Elias J. (1952), „Origines gentium“, in: Classical Philology 47, 65-
81.
Blank, David L. (1982), Ancient Philosophy and Grammar. The Syntax of Apollo-
nius Dyscolus, Chico, CA.
— (1994), „Analogy, Anomaly and Apollonius Dyscolus“, in: Stephen Everson
(ed.), Language, Cambridge, 149-165.
— (ed.) (1998), Sextus Empiricus, Against the Grammarians (Adversus mathematicos
I), translated with an introduction and commentary, Oxford.
— (2000), „The Organization of Grammar in Ancient Greece“, in: Sylvain
Auroux / Ernst F. K. Koerner / Hans-Joseph Niederehe / Kees Versteegh
(eds.), History of the Language Sciences. An International Handbook on the Evo-
lution of the Study of Language from Beginnings to the Present / Geschichte der
Sprachwissenschaften. Ein internationales Handbuch zur Entwicklung der Sprach-
forschung von den Anfängen bis zur Gegenwart / Histoire des sciences du langage.
Manuel international sur l’évolution de l’étude du langage des origines à nos jours,
I, 1, 1, Berlin-New York, 400-417.
— (2005), „Varro’s Anti-analogist“, in: Frede/Inwood 2005, 210-238.
Blass, Friedrich (ed.) (19082), Aeschines, Orationes, Leipzig (Ist ed. 1896).
Blass, Friedrich / Schindel, Ulrich (eds.) (1978), Aeschinis Orationes, Stuttgart.
Blau, August (1883), De Aristarchi discipulis. Dissertatio philologica, Jena.
Boudreaux, Pierre (1919), Le texte d’Aristophanes et ses commentateurs, Paris.
Brandenburg, Philipp (ed.) (2005), Apollonios Dyskolos, Über das Pronomen,
Einführung, Text, Übersetzung und Erläuterung, München-Leipzig.
Braswell, Bruce Karl (2002), The grammarian Epaphroditus, in: ΠΡΑΚΤΙΚΑ ΙΑˊ
ΔΙΕΘΝΟΥΣ ΣΥΝΕΔΡΙΟΥ ΚΛΑΣΣΙΚΩΝ ΣΠΟΥΔΩΝ. ΚΑΒΑΛΑ 29-30
ΑΥΓΟΥΣΤΟΥ 1999, Athina, II 161-171.
Braswell, Bruce Karl / Billerbeck, Margarethe (eds.) (2008), The Grammarian
Epaphroditus. Testimonia and Fragments, edited and translated with an intro-
duction, notes and commentary, with the collaboration of Simonetta Mar-
chitelli / Orlando Poltera / Christian Zubler, Bern.
Bravo, Benedetto (2006), „Felix Jacoby, Arnaldo Momigliano e l’erudizione
antica“, in: Carmine Ampolo (ed.), Aspetti dell’opera di Felix Jacoby, Pisa,
227-257.
Briquel, Dominique (1984), Les Pélasges en Italie. Recherches sur l’histoire de la
légende, Roma.
Brocquet, Sylvain (2005), ,,Apollonius Dyscole et l’adverbe“, in: Histoire, épis-
témologie, langage 27/2, 121-140.
Broggiato, Maria (ed.) (2001), Cratete di Mallo, I frammenti, edizione, introdu-
zione e note, La Spezia (then: Roma 2006).
166 Bibliography

Buttmann, Alexander (ed.) (1877), Des Apollonios Dyskolos vier Bücher über die
Syntax, überstetz und erläutert, Berlin.

Cadoni, Nicola (2010), „Aristotele e l’omeristica antica: sondaggi su ἦθος“, in:

Fausto Montana (ed.), Aner Polytropos. Ricerche di filologia greca antica
dedicate dagli allievi a Franco Montanari, Roma, 3-39.
Calboli, Gualtiero (1962), La tendenza grammaticale dell’Auctor ad Herennium, in:
Id., Studi grammaticali, Bologna, 139-242.
Callanan, Christopher K. (1987), Die Sprachbeschreibung bei Aristophanes von
Byzanz, Göttingen.
Canfora, Luciano (1995), „Le collezioni superstiti“, in: Giuseppe Cambiano /
Luciano Canfora / Diego Lanza (eds.), Lo spazio letterario della Grecia antica,
I, La produzione e la circolazione del testo (II, L’Ellenismo), Roma, 95-250.
Cassio, Albio Cesare (1993), „Parlate locali, dialetti delle stirpi e fonti letterarie
nei grammatici greci“, in: Emilio Crespo / José Luis García Ramón / Rosa
Araceli Striano (eds.), Dialectologica Graeca. Actas del II Coloquio Interna-
cional de Dialectología Griega (Miraflores de la Sierra [Madrid], 19-21 de
junio de 1991), Madrid, 73-90.
Cavallo, Guglielmo (1980), ,,La trasmissione scritta della cultura greca antica in
Calabria e in Sicilia tra i secoli X-XV. Consistenza, tipologia, fruizione“,
in: Scrittura e civiltà 4, 157-245.
— (1986), „Conservazione e perdita dei testi greci: fattori materiali, sociali,
culturali“, in: Andrea Giardina (ed.), Società romana e impero tardoantico, IV,
Tradizione dei classici. Trasformazioni della cultura, Roma-Bari, 83-172, 246-
271 (reprinted in: Cavallo 2002, 49-175).
— (1992), „La storia dei testi antichi a Bisanzio. Qualche riflessione“, in:
Jacqueline Hamesse (ed.), Les problèmes posés par l’édition critique des textes
anciens et médiévaux, Louvain-la-Neuve, 95-111 (partially reprinted in:
Cavallo 2002, 177-186).
— (1995a), „Qualche annotazione sulla trasmissione dei classici nella tarda
antichità“, in: Rivista di filologia e di istruzione classica 125, 205-219
(reprinted in: Cavallo 2002, 31-47).
— (1995b), „I fondamenti culturali della trasmissione dei testi a Bisanzio“, in:
Giuseppe Cambiano / Luciano Canfora / Diego Lanza (eds.), Lo spazio
letterario della Grecia antica, II, La ricezione e l’attualizzazione del testo, Roma,
265-306 (reprinted in: Cavallo 2002, 195-233).
— (2000), „Una mano e due pratiche. Scrittura del testo e scrittura del
commento nel libro greco“, in: Marie-Odile Goulet-Cazé (ed.), Le
commentaire entre tradition et innovation. Actes du colloque international de
l’Institut des traditions textuelles (Paris et Villejuif, 22-25 septembre 1999),
Paris, 55-64.
— (2002), Dalla parte del libro. Storie di trasmissione dei classici, Urbino.
 Bibliography 167

Cavallo, Guglielmo / Maehler, Herwig (1987), Greek Bookhands of the Early

Byzantine Period: A.D. 300-800, London.
Chantraine, Pierre (1968-1980), Dictionnaire étymologique de la langue grecque:
histoire des mots, I-IV, Paris.
Chiabà, Monica (2003), „Trosia P. Hermonis L. Hilara, lanifica circlatrixs (Ins-
crAq. 69)“, in: Alfredo Buonopane / Francesca Cenerini (eds.), Donna e la-
voro nella documentazione epigrafica. Atti del I Seminario sulla condizione
femminile nella documentazione epigrafica, Bologna, 21 novembre 2002,
Faenza, 261-276.
Chiesa, Paolo (2002), Elementi di critica testuale, Bologna.
Christes, Johannes (1979), Sklaven und Freigelassene als Grammatiker und Philolo-
gen im antiken Rom, Wiesbaden.
Christodoulos, Georgios A. (ed.) (1977), Τὰ ἀρχαῖα σχόλια εἰς Αἴαντα τοῦ
Σοφοκλέους. Κριτικὴ ἔκδοσις, Athina.
Classen, Johannes (1829), De grammaticae Graecae primordiis, Bonn.
CLGP = Bastianini, Guido / Haslam, Michael / Maehler, Herwig / Montanari,
Franco / Römer, Cornelia (eds.), Commentaria et Lexica Graeca in Papyris
reperta (CLGP), München-Leipzig, 2004-.
Cohn, Leopold (1901), „Demetrios (n. 102)“, in: RE 4/2, 2847.
— (1905), „Epaphroditos (n. 5)“, in: RE 5/2, 2711-2714.
Collart, Jean (1954), Varron, grammairien latin, Paris.
Colomo, Daniela (2008), „Osservazioni sullo scriba ossirinchita dell’omega
quadrangolare (Johnson A2)“, in: Segno e testo 6, 3-34.
Colson, Francis H. (1919), „The Analogist and Anomalist Controversy“, in:
The Classical Quarterly 13, 24-36.
Condello, Emma (2002), „Ai margini di Virgilio. Paleografia e codicologia
degli Scholia Veronensia“, in: Vincenzo Fera / Giacomo Ferraù / Silvia
Rizzo (eds.), Talking to the Text: Marginalia from Papyri to Print. Proceedings
of a Conference held at Erice, 26 September-3 October 1998, as the 12th
Course of International School for the Study of Written Records, Messina,
59-80.
CPF = Corpus dei Papiri Filosofici greci e latini (CPF). Testi e lessico nei papiri di
cultura greca e latina, I.1* Firenze 1989; I.1** Firenze 1992; I.1*** Firenze
1999; III Firenze 1995; IV Firenze 2002.
Cribiore, Raffaella (2001), Gymnastics of the Mind. Greek Education in Hellenistic
and Roman Egypt, Princeton.
Cufalo, Domenico (ed.) (2007), Scholia Graeca in Platonem, I, Scholia ad dialogos
tetralogiarum I-VII continens, Roma.

Dahlmann, Hellfried (1935), „Terentius (n. 84)“, in: RE Suppl. 4, 1172-1277.

Daintree, David / Geymonat, Mario (1988), „Scholia non Serviana“, in:
Enciclopedia Virgiliana, IV, Roma, 706-720.
168 Bibliography

Dalimier, Catherine (1991), „Sextus Empiricus contre les grammairiens: ce que

parler grec veut dire“, in: Suzanne Saïd (ed.), Ἑλληνισμός: quelques jalons
pour une histoire de l’identité grecque. Actes du Colloque de Strasbourg, 25-27
octobre 1989, Leiden, 17-32.
— (ed.) (2001), Apollonius Dyscole, Traité de conjonctions, introduction, texte,
traduction et commentaire, Paris.
De Gregorio, Giuseppe (2000), ,,Materiali vecchi e nuovi per uno studio della
minuscola greca tra VII e IX secolo“, in: Giancarlo Prato (ed.), I manoscritti
greci tra riflessione e dibattito. Atti del V Colloquio internazionale di paleogra-
fia greca, Cremona, 4-10 ottobre 1998, Firenze, 83-151.
Delcourt, Anouk (2005), Lecture des Antiquités romaines de Denys d’Halicarnasse.
Un historien entre deux mondes, Louvain-la-Neuve.
Depew, Mary / Obbink, Dirk (eds.) (2000), Matrices of Genre. Authors, Canons,
and Society, Cambidge, Mass.-London.
Devreesse, Robert (1928), „Chaînes exégétiques grecques“, in: Dictionnaire de la
Bible. Supplément I, Paris, 1084-1233.
Di Benedetto, Vincenzo (1958-1959), „Dionisio Trace e la Techne a lui attri-
buita“, in: Annali della Scuola Normale Superiore di Pisa (Classe di Lettere e Fi-
losofia) 27, 169-210; ibid. 28, 87-118 (reprinted in: Di Benedetto 2007, II
381-461).
— (1966), „Demetrio Cloro e Aristone di Alessandria“, in: Annali della Scuola
Normale Superiore di Pisa (Classe di Lettere e Filosofia) 35, 321-324 (reprinted
in: Di Benedetto 2007, II 463-468).
— (1973), „La Techne spuria“, in: Annali della Scuola Normale Superiore di Pisa
(Classe di Lettere e Filosofia) 3, 797-814 (reprinted in: Di Benedetto 2007, II
469-484).
— (1990), „At the Origins of Greek Grammar“, in: Glotta 68, 19-39 (reprinted
in: Di Benedetto 2007, II 485-503).
— (1998), „Afterword“, in: Vivien Law / Ineke Sluiter (eds.), Dionysius Thrax
and the Techne grammatike, Münster, 151-153.
— (2000), „Dionysius Thrax and the Tékhne grammatiké“, in: Sylvain Auroux /
Ernst F. K. Koerner / Hans-Joseph Niederehe / Kees Versteegh (eds.), His-
tory of the Language Sciences. An International Handbook on the Evolution of the
Study of Language from Beginnings to the Present / Geschichte der Sprachwissen-
schaften. Ein internationales Handbuch zur Entwicklung der Sprachforschung von
den Anfängen bis zur Gegenwart / Histoire des sciences du langage. Manuel inter-
national sur l’évolution de l’étude du langage des origines à nos jours, I, 1, 1, Ber-
lin-New York, 394-400 (reprinted in: Di Benedetto 2007, II 521-531).
— (2007), Il richiamo del testo: contributi di filologia e letteratura, I-IV, Pisa.
Di Gregorio, Lamberto (1974), „Sulla tradizione manoscritta degli scholia vetera
alla Teogonia di Esiodo, IV, Conclusioni“, in: Aevum 46, 1-15.
— (ed.) (1975), Scholia vetera in Hesiodi Theogoniam, Milano 1975.
 Bibliography 169

Dickey, Eleanor (2007), Ancient Greek Scholarship. A Guide to Finding, Reading,

and Understanding Scholia, Commentaries, Lexica, and Grammatical Treatises,
from their Beginnings to the Byzantine Period, Oxford-New York.
Dihle, Albrecht (1959), „Ptolemaios (n. 70)“, in: RE 23/2, 1860-1861.
Dilts, Mervin R. (ed.) (1997), Aeschinis Orationes, Stuttgart-Leipzig.
Donini, Pierluigi (1994), „Testi e commenti, manuali e insegnamento: la forma
sistematica e i metodi della filosofia in età postellenistica“, in: Aufstieg und
Niedergang der römischen Welt II 36/7, 5027-5100.
Donnet, Daniel (1967), ,,La place de la syntaxe dans les traités de grammaire
grecque, des origines au XII siècle“, in: L’antiquité classique 36, 22-48.
Dorival, Gilles (1984), „Des commentaires de l’Écriture aux chaînes“, in: Bible
de tous les temps, I, Paris, 361-383.
— (1986), Les chaînes exégétiques grecques sur les Psaumes: contribution à l’étude
d’une forme littéraire, I, Leuven.
Dover, Kenneth (ed.) (1993), Aristophanes, Frogs, Oxford.
Dubuisson, Michel (1984), „Le latin est-il une langue barbare?“, in: Ktèma 9,
55-68.
— (1987), „Homérologie et politique. Le cas d’Aristodémos de Nysa“, in: Jean
Servais / Tony Hackens / Brigitte Servais-Soyez (eds.), Stemmata. Mélanges
de philologie, d’histoire et d’archéologie grecques offerts à Jules Labarbe,
Liège – Louvain-la-Neuve, 1987, 15-24.
Dué, Casey (ed.) (2009), Recapturing a Homeric Legacy. Images and Insights from the
Venetus A Manuscript of the Iliad, Washington, DC.

Eberline, Charles N. (1980), Studies in the Manuscript Tradition of the Ranae of

Aristophanes, Meisenheim am Glan.
Egger, Émile (1854), Apollonius Dyscole. Essai sur l’histoire des théories grammati-
cales dans l’Antiquité, Paris.
Erbse, Hartmut (1960), Beiträge zur Überlieferung der Iliasscholien, München.
— (1965), „Scholien“, in: Carl Andresen et al. (eds.), Lexikon der alten Welt,
Zürich-Stuttgart, 2723-2725.
— (ed.) (1969), Scholia Graeca in Homeri Iliadem (scholia vetera), I, Berlin.
— (ed.) (1971), Scholia Graeca in Homeri Iliadem (scholia vetera), II, Berlin.
— (1980), „Zur normativen Grammatik der Alexandriner“, in: Glotta 58, 236-
258.
Ernst, Nicola (ed.) (2006), Die D-Scholien zur Odyssee. Kritische Ausgabe, Köln
(digital edition: http://kups.ub.uni-koeln.de/volltexte/2006/1831/).
Esposito, Elena (ed.) (2004), „Aeschines“, in: CLGP I.1.1 (Aeschines-Alcaeus),
München-Leipzig, 3-11.

Faraggiana di Sarzana, Chiara (1978), „Il commentario procliano alle Opere e i

giorni, I, Plutarco fonte di Proclo“, in: Aevum 52, 17-40.
170 Bibliography

— (1981), „Il commentario procliano alle Opere e i giorni, II, Destinazione e

fortuna dell’opera nella Scuola d’Atene e dopo la sua chiusura“, in: Aevum
55, 22-29.
— (1987), „Le commentaire à Hésiode et la paideia encyclopédique de
Proclus“, in: Pépin/Saffrey 1987, 21-41.
Fehling, Detlev (1956-1957), „Varro und die grammatische Lehre von der
Analogie und der Flexion“, in Glotta 35, 214-270; ibid., 36, 48-100.
— (1979), review of Siebenborn 1976, in: Gnomon 51, 488-490.
Fernández-Galiano, Manuel / Heubeck, Alfred (eds.) (19934), Omero, Odissea,
VI, Libri XXI-XXIV, trad. di Giuseppe Aurelio Privitera, Milano (Ist ed.
1986).
Ferrary, Jean-Louis (1988), Philhellénisme et impérialisme. Aspects idéologiques de la
conquête romaine du monde hellénistique, de la seconde guerre de Macédonie à la
guerre contre Mithridate, Roma.
Fournet, Jean-Luc / Gascou, Jean (2008), „Un lot d’archives inédit de Lycopo-
lis (Égypte) à l’Académie des Inscriptions et Belles Lettres“, in: Comptes ren-
dus de l’Académie des Inscriptions et Belles Lettres, 1041-1074.
Fournet, Jean-Luc / Montana, Fausto (eds.) (2006), „P.Acad. inv. 3d + Bodl.
Ms. Gr. class. f. 72 (P)“, in: CLGP I.1.4 (Aristophanes-Bacchylides), 48-56.
Fraser, Peter M. (1972), Ptolemaic Alexandria, I-III, Oxford.
Frede, Dorothea / Inwood, Brad (eds.) (2005), Language and Learning. Philoso-
phy of Language in the Hellenistic Age. Proceedings of the Ninth Symposium
Hellenisticum, Cambridge-New York.
Frede, Michael (1977), „The Origins of Traditional Grammar“, in: Robert E.
Butts / Jaakko Hintikka (eds.), Historical and Philosophical Dimensions of Log-
ic, Methodology and Philosophy of Science. Part four of the Proceedings of the
Fifth International Congress of Logic, Methodology and Philosophy of
Science, London, Ontario, Canada 1975, Dordrecht-Boston, 51-79.
— (1978), „Principles of Stoic Grammar“, in: John M. Rist (ed.), The Stoics,
Berkeley-Los Angeles-London, 27-75.
Friedländer, Ludwig (ed.) (1853), Aristonici Περὶ σημείων Ἰλιάδος reliquiae
emendatiores, Göttingen.
Führer, Rudolf / Schmidt, Martin (2001), „Homerus redivivus“, in:
Göttingische gelehrte Anzeigen 253, 1-32.
Fuhrmann, Manfred (1960), Das systematische Lehrbuch. Ein Beitrag zur Geschichte
der Wissenschaften in der Antike, Göttingen.
Funaioli, Gino (ed.) (1907), Grammaticae Romanae Fragmenta, Leipzig.
— (1930), Esegesi virgiliana antica. Prolegomeni alla edizione del commento di Giunio
Filargirio e di Tito Gallo, Milano.

Gabba, Emilio (1963), „Il latino come dialetto greco“, in: Miscellanea di studi
alessandrini in memoria di Augusto Rostagni, Torino, 188-194.
 Bibliography 171

— (1974), „Storiografia greca e imperialismo romano (III-I sec. a.C.)“, in:

Rivista storica italiana 86, 625-642.
— (1976), „Sulla valorizzazione politica della leggenda delle origini troiane di
Roma fra III e II secolo a.C.“, in: Marta Sordi (ed.), I canali della propaganda
nel mondo antico, Milano, 84-101.
— (1982), „La ‘Storia di Roma arcaica’ di Dionigi d’Alicarnasso“, in: Aufstieg
und Niedergang der römischen Welt II 30/1, 799-816.
Gallavotti, Carlo (ed.) (19933), Theocritus quique feruntur bucolici Graeci, Roma (Ist
ed. 1946).
Gammacurta, Tatiana (2006), Papyrologica scaenica. Copioni teatrali nella tradizione
papiracea, Alessandria.
Geerard, Maurits (1980), Clavis Patrum Graecorum, IV, Concilia – Catenae,
Turnhout.
Gerstinger, Hans (1970), Dioscurides. Codex Vindobonensis Med. Gr. 1 der
Österreichischen Nationalbibliothek. Kommentarband zu der Facsimileausgabe,
Graz.
Geymonat, Mario (1985), „Filargirio“, in: Enciclopedia Virgiliana, II, Roma,
520-521.
GG = Grammatici Graeci, I-IV, Leipzig 1867-1901.
Gibson, Craig A. (2002), Interpreting a Classic. Demosthenes and His Ancient
Commentators, Berkeley-Los Angeles.
Giomini, Remo (1953), „Il grammatico Filosseno e la derivazione del latino
dall’eolico“, in: La parola del passato 8, 365-376.
Glock, Andreas (2001), „Scholien“, in: NP 11, 209-214.
Goetz, Georg (1896), „Ateius (n. 11)“, in: RE 2/2, 1910-1911.
— (1900), „Cloatius (n. 2)“, in: RE 4/1, 61-62.
Golitsis, Pantelis (2008), Les commentaires de Simplicius et de Jean Philopon à la
“Physique” d’Aristote: tradition et innovation, Berlin-New York.
Gow, Andrew S. F. (ed.) (19522), Theocritus, I-II, Cambridge (Ist ed. 1950).
Grenfell, Bernard P. / Hunt, Arthur S. (eds.) (1915), „P.Oxy. 1371“, in: The
Oxyrhynchus Papyri. Part XI, London.
Gudeman, Alfred (1912), „Grammatik“, in: RE 7/2, 1780-1811.
— (1921), „Scholien“, in: RE 2A/1, 625-705.

Haas, Walter (ed.) (1977), Die Fragmente der Grammatiker Tyrannion und Diokles,
(SGLG 3), Berlin-New York.
Hagen, Hermann (ed.) (1902), „Appendix Serviana. Ceteros praeter Servium et
Scholia Bernensia Vergilii commentatores continens“, in: Georg Thilo /
Hermann Hagen (eds.), Servii grammatici qui feruntur in Vergilii carmina
commentarii, III/2, Leipzig.
Haslam, Michael W. (ed.) (1986), „POxy. 3710“, in: The Oxyrhynchus Papyri.
Part LIII, London, 89-112.
172 Bibliography

Hatch, William H. P. (1937), „A Redating of Two Important Uncial

Manuscripts of the Gospel. Codex Zacynthius and Codex Cyprius“, in:
Robert P. Casey / Silva Lake / Agnes K. Lake (eds.), Quantulacumque.
Studies presented to Kirsopp Lake, London, 333-338.
Hennigfeld, Jochem (1994), Geschichte der Sprachphilosophie. Antike und Mittelal-
ter, Berlin-New York.
Hertz, Martin J. (ed.) (1855-1859), Grammatici Latini, II/III, Prisciani grammatici
Caesariensis Institutionum grammaticarum libri XVIII, Leipzig.
Hilgard, Alfred (ed.) (1901), Grammatici Graeci, I/II, Scholia in Dionysii Thracis
artem grammaticam, Leipzig.
Hill, Herbert (1961), „Dionysius of Halicarnassus and the Origins of Rome“,
in: The Journal of Roman Studies 61, 88-93.
Hillscher, Alfred (1892), „Hominum litteratorum Graecorum ante Tiberii
mortem in urbe Roma commoratorum historia critica“, in: Jahrbücher für
classische Philologie, Suppl. 18, 355-440.
Hoekstra, Arie (ed.) (19934), Omero, Odissea, IV, Libri XIII-XVI, trad. di Giu-
seppe Aurelio Privitera, Milano (Ist ed. 1984).
Hoffmann, Philippe (2000), „Bibliothèques et formes du livre à la fin de
l’antiquité. Le témoignage de la littérature néoplatonicienne des Ve et VIe
siècles“, in: Giancarlo Prato (ed.), I manoscritti greci tra riflessione e dibattito.
Atti del V Colloquio Internazionale di Paleografia Greca (Cremona, 4-10
ottobre 1998), Firenze, II 601-632.
Holtz, Louis (1984), „Les manuscrits latins à gloses et à commentaires“, in:
Cesare Questa / Renato Raffaelli (eds.), Il libro e il testo. Atti del Convegno
Internazionale, Urbino, 20-23 settembre 1982, Urbino, 141-167.
— (1995), „Glosse e commenti“, in: Guglielmo Cavallo / Claudio Leonardi /
Enrico Menestò (eds.), Lo spazio letterario del Medioevo, I, Il Medioevo latino
(III, La ricezione del testo), Roma, 59-111.
— (2000), „Le rôle des commentaires d’auteurs classiques dans l’émergence
d’une mise en page associant texte et commentaire (Moyen Âge
occidental)“, in: Marie-Odile Goulet-Cazé (ed.), Le commentaire entre
tradition et innovation. Actes du colloque international de l’Institut des
traditions textuelles (Paris et Villejuif, 22-25 septembre 1999), Paris, 101-
117.
Householder, Fred W. (ed.) (1981), The Syntax of Apollonius Dyscolus, translated
and with commentary, Amsterdam.
Hug A. (1924), „Lanipendia“, in: RE 12/1, 689-690.
— (1937), „Pensum“, in: RE 19/1, 498-499.
Hunger, Herbert (1961), „Überlieferungsgeschichte der griechischen
klassischen und hellenistischen Literatur“, in: Geschichte der Textüberlieferung
der antiken und mittelalterlichen Literatur, I, Antikes und mittelalterliches Buch-
und Schriftwesen. Überlieferungsgeschichte der antiken Literatur, Zürich, 207-283.
Hunt, Arthur S. / Johnson, John (eds.) (1930), Two Theocritus Papyri, London.
 Bibliography 173

Ihm, Sibylle (2002), Clavis commentariorum der antiken medizinischen Texte,

Leiden-Boston-Köln.
Ildefonse, Frédérique (1997), La naissance de la grammaire dans l’antiquité grecque,
Paris.
Ioannidou, Grace (ed.) (1996), Catalogue of Greek and Latin Papyri in Berlin
(P.Berol. inv. 21101-21299, 21911). BKT IX, Mainz am Rhein.
Ippolito, Antonella (2005), „Heracleon [1]“, in: LGGA.
Irigoin, Jean (1952), Histoire du texte de Pindare, Paris.
— (1984), „Livre et texte dans les manuscrits byzantins de poètes. Continuité et
innovations“, in: Cesare Questa / Renato Raffaelli (eds.), Il libro e il testo.
Atti del Convegno Internazionale, Urbino, 20-23 settembre 1982, Urbino,
85-102.
— (1994), „Les éditions de textes“, in: Montanari 1994, 39-93 (reprinted in:
Jean Irigoin, La tradition des textes grecs. Pour une critique historique, Paris
2003, 133-173).

Jacob, Christian (1999), „Du livre au texte: pour une histoire comparée des
philologies“, in: Quaderni di storia 50, 61-93 (= Christian Jacob / Juliet Vale
(1999), „From Book to Text: Towards a Comparative History of Philolo-
gies“, in: Diogenes 47, 4-22).
Janko, Richard (1995a), „Crates of Mallos, Dionysius Thrax and the Tradition
of Stoic Grammatical Theory“, in: Lewis Ayres (ed.), The Passionate Intel-
lect: Essays on the Transformation of Classical Traditions Presented to Professor I.
G. Kidd, New Brunswick-London, 213-233.
— (1995b), „Reconstructing Philodemus’ On Poems“, in: Dirk Obbink (ed.),
Philodemus and Poetry. Poetic Theory and Practice in Lucretius, Philodemus and
Horace, New York-Oxford, 69-96.
— (ed.) (2000), Philodemus, On poems, Book 1, edited with introduction, trans-
lation, and commentary, Oxford.
Julien, Jacques (1985), ,,Mode verbale et diathesis chez Apollonius Dyscole“,
in: Histoire, épistémologie, langage 7/1, 83-125.

Kaster, Robert A. (1988), Guardians of Language. The Grammarian and Society in

Late Antiquity, Berkeley-Los Angeles-London.
Kemp, Alan (1991), „The Emergence of Autonomous Greek Grammar“, in:
Schmitter 1991, 302-333.
Kirk, Geoffrey S. (1985), The Iliad: A Commentary, I, Books 1-4, Cambridge.
Kleinlogel, Alexander (1964), „Beobachtungen zu den Thukydidesscholien, I“,
in: Philologus 108, 233-246.
— (1965), Geschichte des Thukydidestextes im Mittelalter, Berlin.
— (1998), „Beobachtungen zu den Thukydidesscholien, II”, in: Philologus 142,
11-40.
174 Bibliography

Knaack, Georg (1903), „Aristodemos“, in: RE Suppl. 1, 133.

Koster, Willem J. W. (1963), „Aristophane dans la tradition byzantine“, in:
Revue des études grecques 76, 381-396.
Kouremenos, Theokritos / Parássoglou, George / Tsantsanoglou, Kyriakos
(eds.) (2006), The Derveni Papyrus, edited with introduction and
commentary, Florence.
Krüger, Julian (1990), Oxyrhynchos in der Kaiserzeit. Studien zur Topographie und
Literaturrezeption, Frankfurt am Main.

La Roche, Jacob (1866), Die homerische Textkritik im Alterthum. Nebst einem

Anhange über die Homerhandschriften, Leipzig.
Lallot, Jean (1987), ,,Le grammairien grec et le texte: l’exemple d’Apollonius
Dyscole“, in: Le texte et ses représentations, Paris, 145-157.
— (1994), ,,Le problème des functions syntaxiques chez Apollonius Dyscole“,
in: Jan De Clercq / Piet Desmet (eds.), Florilegium historiographiae linguisticae.
Études d’historiographie de la linguistique et de grammaire comparée à la mémoire de
Maurice Leroy, Leuven, 131-141.
— (1995a), „Analogie et pathologie dans la grammaire alexandrine“, in: Lalies.
Actes des sessions de linguistique et de littérature. 15 (Aussois, 29 août-3
septembre 1994), Paris, 109-123.
— (1995b), „Qu’est-ce que la grammaire?“, in: Lalies. Actes des sessions de
linguistique et de littérature. 15 (Aussois, 29 août-3 septembre 1994), Paris,
73-82.
— (ed.) (1997), Apollonius Dyscole, De la construction / Peri syntaxeos, texte grec
accompagné de notes critiques, introduction, traduction, notes exégétiques,
I-II, Paris.
— (1998), „Grammatici certant: vers une typologie de l’argumentation pro et
contra dans la question de l’authenticité de la Techne“, in: Vivien Law / Ine-
ke Sluiter (eds.), Dionysius Thrax and the Techne grammatike, Münster, 27-39.
— (ed.) (19982), La grammaire de Denys le Thrace, traduite et annotée, 2e éd.
revue et augmentée, Paris Ist ed. 1989).
Lambert, Frédéric (1978), ,,Le terme et la notion de διάθεσις chez Apollonius
Dyscole“, in: Varron. Grammaire antique et stylistique latine (Recueil offert à
Jean Collart), Paris, 245-252.
— (2000), „La linguistique grecque chez les alexandrins: Aristophane de By-
zance et Aristarque“, in: Sylvain Auroux / Ernst F. K. Koerner / Hans-
Joseph Niederehe / Kees Versteegh (eds.), History of the Language Sciences.
An International Handbook on the Evolution of the Study of Language from Be-
ginnings to the Present / Geschichte der Sprachwissenschaften. Ein internationales
Handbuch zur Entwicklung der Sprachforschung von den Anfängen bis zur Gegen-
wart / Histoire des sciences du langage. Manuel international sur l’évolution de
l’étude du langage des origines à nos jours, I, 1, 1, Berlin-New York, 385-394.
 Bibliography 175

Lamberton, Robert (2001), „Schools of Platonic Philosophy of the Roman

Empire: the Evidence of the Biographies“, in: Yun Lee Too (ed.),
Education in Greek and Roman Antiquity, Leiden-Boston-Köln, 433-458.
Lamberz, Erich (1987), „Proklos und die Form des philosophischen
Kommentars“, in: Pépin/Saffrey 1987, 1-20.
Law, Vivien (2003), The History of Linguistics in Europe from Plato to 1600, Cam-
bridge-New York.
Leanza, Sandro (1995), „L’esegesi biblica cristiana antica: scolii e catene“, in:
Claudio Moreschini (ed.), Esegesi, parafrasi e compilazione in età tardoantica.
Atti del Terzo Convegno dell’Associazione di Studi Tardoantichi, Napoli,
209-227.
Lehrs, Karl (1837), Quaestiones epicae, Königsberg.
— (18823), De Aristarchi studiis Homericis, Leipzig (Ist ed. 1833).
Lemerle, Paul (1971), Le premier humanisme byzantin. Notes et remarques sur
enseignement et culture à Byzance des origines au Xe siècle, Paris.
Lersch, Laurenz (1838-1841), Die Sprachphilosophie der Alten, I-III, Bonn.
LGGA = Montanari, Franco / Lapini, Walter / Montana, Fausto / Pagani, Lara
(eds.), Lessico dei Grammatici Greci Antichi (www.aristarchus.unige.it/lgga).
Linke, Konstanze (ed.) (1977), Die Fragmente des Grammatikers Dionysios Thrax,
(SGLG 3), Berlin-New York.
Lobel, Edgar (1952), „P.Oxy. 2258“, in: The Oxyrhynchus Papyri. Part XX,
London.
Lossau, Manfred J. (1964), Untersuchungen zur antiken Demosthenesexegese, Bad
Homburg.
Lowe, Elias A. (1934-1972), Codices Latini Antiquiores. A Palaeographical Guide to
Latin Manuscripts prior to the Ninth Century, I-XI + Suppl., Oxford.
Lundon, John (1997), „Σχόλια: una questione non marginale“, in: Discentibus
obvius. Omaggio degli allievi a Domenico Magnino, Como, 73-86.
Luppe, Wolfgang (2002), „Σχόλια, ὑπομνήματα und ὑποθέσεις zu
griechischen Dramen auf Papyri“, in: Wilhelm Geerlings / Christian
Schulze (eds.), Der Kommentar in Antike und Mittelalter. Beiträge zu seiner
Erforschung, Leiden-Boston-Köln, 55-77.
Luschnat, Otto (1954), „Die Thukydidesscholien. Zu ihrer handschriftlichen
Grundlage, Herkunft und Geschichte“, in: Philologus 98, 14-58.
Luzzatto, Maria Jagoda (1993a), „Itinerari di codici antichi: un’edizione di
Tucidide tra il II e il IX secolo“, in: Materiali e discussioni per l’analisi dei testi
classici 30, 167-203.
— (1993b), „Scholia tardoantichi: il commentario di Marcellino a Tucidide“, in:
Quaderni di storia 38, 111-115.

Maas, Paul (ed.) (1911), Apollonius Dyscolus, De pronominibus. Pars generalis,

Bonn.
176 Bibliography

Maehler, Herwig (1968), „Ein Fragment aus Aristophanes’ Rittern mit

Scholien“, in: Hermes 96, 287-293.
— (1994), „Die Scholien der Papyri in ihrem Verhältnis zu den Scholien-
corpora der Handschriften“, in: Montanari 1994, 95-127.
— (1998), „Élites urbaines et production littéraire en Égypte romaine et
byzantine“, in: Gaia 3, 81-95.
— (2000), „L’évolution matérielle de l’hypomnèma jusqu’à la basse époque. Le
cas du POxy. 856 (Aristophane) et PWürzburg 1 (Euripide)“, in: Marie-
Odile Goulet-Cazé (ed.), Le commentaire entre tradition et innovation. Actes du
colloque international de l’Institut des traditions textuelles (Paris et
Villejuif, 22-25 septembre 1999), Paris, 29-36.
Majcherek, Grzegorz (2008), „Academic Life of Late Antique Alexandria: a
Viewpoint from the Field“, in: Mostafa El-Abbadi / Omnia Mounir
Fathallah (eds.), What Happened to the Ancient Library of Alexandria?, Leiden-
Boston, 191-206.
Manetti, Daniela (1998), „Commento ed enciclopedia“, in: Salvatore Settis
(ed.), I Greci, II/III, Torino, 1199-1220.
Manfredi, Manfredo (2000), „Un frammento di Aristofane, I cavalieri, con
scolî“, in: Henri Melaerts (ed.), Papyri in honorem Johannis Bingen octogenarii
(P.Bingen), Leuven, 95-104.
Maniaci, Marilena (2000), „Stratégies de juxtaposition du texte et du
commentaire dans quelques manuscrits d’Homère“, in: Marie-Odile
Goulet-Cazé (ed.), Le commentaire entre tradition et innovation. Actes du
colloque international de l’Institut des traditions textuelles (Paris et
Villejuif, 22-25 septembre 1999), Paris, 65-76.
— (2002), „«La serva padrona». Interazioni fra testo e glossa sulla pagina del
manoscritto“, in: Vincenzo Fera / Giacomo Ferraù / Silvia Rizzo (eds.),
Talking to the Text: Marginalia from Papyri to Print. Proceedings of a
Conference held at Erice, 26 september-3 october 1998, as the 12th Course
of International School for the Study of Written Records, Messina, 3-35.
— (2006a), „Problemi di mise en page dei manoscritti con commento a cornice.
L’esempio di alcuni testimoni dell’Iliade“, in: Segno e testo 4, 211-297.
— (2006b), „Words Within Words: Layout Strategies in Some Glossed
Manuscripts of the Iliad“, in: Manuscripta 50, 241-268.
Marin, Demetrio (1969), „Dionisio di Alicarnasso e il latino“, in: Jacqueline
Bibauw (ed.), Hommages à Marcel Renard, I-III, Bruxelles, 595-607.
Mársico, Claudia T. (2006), ,,¿Una crítica de Apolonio Discolo a la “Téchne
grammatiké”? Enfoques en torno de la noción de “diáthesis” en la grámati-
ca griega antigua“ in: Cuadernos de filología clásica. Estudios griegos e indoeuro-
peos 16, 23-44.
Martin, Victor / Budé, Guy de (eds.) (1928), Eschine, Discours, II, texte et tra-
duction, Paris.
 Bibliography 177

Marzi, Mario / Leone, Pietro / Malcovati, Enrica (eds.) (1977), Oratori attici
minori, I, Iperide, Eschine, Licurgo, Torino.
Matthaios, Stephanos (1996), „Κύριον ὄνομα. Zur Geschichte eines grammati-
schen Terminus“, in: Pierre Swiggers / Alfons Wouters (eds.), Ancient
Grammar. Content and Context, Leuven, 55-77.
— (1999), Untersuchungen zur Grammatik Aristarchs. Texte und Interpretation zur
Wortartenlehre, Göttingen.
— (2001), „Das Wortartensystem der Alexandriner. Skizze seiner Entwick-
lungsgeschichte und Nachwirkung“, in: Göttinger Beiträge zur Sprachwissen-
schaft 5, 65-94 (reprinted in: Ioannis G. Taifacos (ed.), The Origins of Euro-
pean Scholarship. The Cyprus Millennium International Conference,
Stuttgart 2005, 13-39).
— (2002), „Neue Perspektiven für die Historiographie der antiken Grammatik.
Das Wortartensystem der Alexandriner“, in: Swiggers/Wouters 2002a,
161-220.
— (2003), „Tryphon aus Alexandria: Der erste Syntaxtheoretiker vor Apollo-
nios Dyskolos?“, in: Pierre Swiggers / Alfons Wouters (eds.), Syntax in
Antiquity, Leuven-Paris-Dudley, 97-132.
— (2004), „Die Syntaxtheorie bei den alexandrinischen Grammatikern: Das
Beispiel Tryphons aus Alexandria“, in: Gerda Hassler / Gesina Volkmann
(eds.), History of Linguistics in Texts and Concepts / Geschichte der Sprachwissen-
schaft in Texten und Konzepten, Münster, 87-100.
— (2009), „Aristarch, Dionysios Thrax und die Τέχνη γραμματική. Zur
Echtheitsdiskussion des ersten Lehrbuchs über die Grammatik“, in: Eleni
Karamalengou / Eugenia Makrygianni (eds.), Ἀντιφίλησις. Studies on Clas-
sical, Byzantine and Modern Greek Literature and Culture. In Honour of John-
Theophanes A. Papademetriou, Stuttgart, 386-400.
— (2010a), „Eratosthenes von Kyrene: Lesarten seiner ‘Grammatik’-
Definition“, in: Matthaios/Montanari/Rengakos 2010.
— (2010b), „Textinterpretation und grammatische Argumentation im Kreis der
alexandrinischen Philologen. Konsequenzen für die ἐμπειρία-τέχνη-
Diskussion“, in: Nikolai Kazansky / Vladimir I. Mazhuga / Larissa
Stepanova / Pierre Swiggers / Alfons Wouters (eds.), Ancient Grammar and
Its Posterior Tradition, Leuven-Paris-Dudley, forthcoming.
— (2010c), „Von Aristoteles zu den Grammatikern. Die εἰκών-Definition
(Aristot. Rh. 3, 10) und die Entstehung des grammatischen Terminus
πρόθεσις“, in: Rivista di filologia e di istruzione classica 138, forthcoming.
Matthaios, Stephanos / Montanari, Franco / Rengakos, Antonios (eds.) (2010),
Ancient Scholarship and Grammar. Archetypes, Concepts, and Contents. Pro-
ceedings of the 2nd Trends in Classics international conference (Thessaloni-
ki, 5th-7th December 2008), Berlin-New York.
Matthews, Peter H. (2007), „The Ancient Grammarians“, in: Anastasios-
Phoivos Christidis / Maria Arapopoulou / Maria Chriti (eds.), A History of
Ancient Greek. From the Beginnings to Late Antiquity, Cambridge-New York
178 Bibliography

(revised and expanded translation of: Istoria tis Ellinikis glossas: apo tis arches
eos tin ysteri archaiotita, Thessaloniki 2001), 1193-1200.
Matthias, Theodor (1883), ,,De Apollonii Dyscoli epirrhematici et syndesmici
forma genuina“, in: Leipziger Studien zur classischen Philologie 6, 1-92.
Mazal, Otto (ed.) (1998), Der Wiener Dioskurides. Codex medicus Graecus 1 der
Österreichischen Nationalbibliothek, I, Graz.
McNamee, Kathleen (1977), Marginalia and Commentaries in Greek Literary
Papyri, Diss. Duke University, Durham.
— (1981), „Greek Literary Papyri Revised by Two or More Hands“, in:
Proceedings of the Sixteenth International Congress of Papyrology, Chico,
CA, 79-91.
— (1995), „Missing Links in the Development of Scholia“, in: Greek, Roman
and Byzantine Studies 36, 399-414.
— (1997), „An Innovation in Annotated Codices on Papyrus“, in: Bärbel
Kramer / Wolfgang Luppe / Herwig Maehler / Günther Poethke (eds.),
Akten des 21. Internationalen Papyrologenkongresses, Berlin, 13-19
August 1995, Stuttgart, II 669-678.
— (1998), „Another Chapter in the History of Scholia“, in: Classical Quarterly
n.s. 48, 269-288.
— (2007), Annotations in Greek and Latin Texts from Egypt, [New Haven,
Conn.].
McNelis, Charles (2002), „Greek Grammarians and Roman Society During the
Early Empire: Statius’ Father and his Contemporaries“, in: Classical Antiqui-
ty 21, 67-94.
Merkelbach, Reinhold (1959), „Kritische Beiträge, III, Ein Aeschines-
Papyrus“, in: Hellfried Dahlmann / Reinhold Merkelbach (eds.), Studien
zur Textgeschichte und Textkritik (Günther Jachmann zur fünfzigsten
Wiederkehr seiner Promotion gewidmet), Köln, 162-164.
Messeri Savorelli, Gabriella / Pintaudi, Rosario (1997), „Frammenti di rotoli
letterari laurenziani“, in: Zeitschrift für Papyrologie und Epigraphik 115, 171-
177.
— (2002), „I lettori dei papiri: dal commento autonomo agli scolii“, in:
Vincenzo Fera / Giacomo Ferraù / Silvia Rizzo (eds.), Talking to the Text:
Marginalia from Papyri to Print. Proceedings of a Conference held at Erice,
26 september-3 october 1998, as the 12th Course of International School
for the Study of Written Records, Messina, 37-57.
Mette, Hans J. (1952), Parateresis. Untersuchungen zur Sprachtheorie des Krates von
Pergamon, Halle.
Monaco, Matteo (2000), „La tradizione manoscritta di Eschine tra i papiri e i
codici medievali“, in: Aegyptus 80, 3-98.
Montana, Fausto (2000), „Le note marginali ai Cavalieri di Aristofane nel papiro
Oxford Bodl. Ms. Gr. Class. f. 72 (P)“, in: Eikasmós 11, 85-96.
 Bibliography 179

— (2001a), „I marginalia di P.Bingen 18 e l’interpretazione di Ar. Eq. 1008“, in:

Eikasmós 12, 161-172.
— (2001b), „Uno scolio del tipo “οὐκ ἀργῶς” ad Aristoph., Equ. 551 in
P.Berol. 21105?“, in: Zeitschrift für Papyrologie und Epigraphik 135, 21-22.
— (2004), „I grammatici alessandrini nei papiri di Aristofane“, in: Roberto
Pretagostini / Emanuele Dettori (eds.), La cultura ellenistica. L’opera letteraria
e l’esegesi antica. Atti del Convegno COFIN 2001, Università di Roma
“Tor Vergata”, 22-24 settembre 2003, Roma, 371-384.
— (2005), „L’esegesi ad Aristofane su papiro“, in: Fausto Montana (ed.),
Interpretazioni antiche di Aristofane, Sarzana (then: Roma 2006), 1-53.
— (ed.) (2006a), Aristophanes, in: CLGP I.1.4 (Aristophanes-Bacchylides), 3-
240.
— (2006b), „Didymus [1]“, in: LGGA.
— (2006c), „L’anello mancante: l’esegesi ad Aristofane tra l’antichità e
Bisanzio“, in: Guido Avezzù / Paolo Scattolin (eds.), I classici greci e i loro
commentatori. Dai papiri ai marginalia rinascimentali. Atti del convegno,
Rovereto, 20 ottobre 2006, Rovereto, 17-34.
— (2009a), „Nicanor [3]“, in: LGGA.
— (2009b), „Storici, filologi, storici-filologi: intersezioni nella cultura
ellenistica“, in: Francesca Gazzano / Gabriella Ottone / Luigi Santi
Amantini (eds.), Ingenia Asiatica. Fortuna e tradizione di storici d’Asia Minore.
Atti della prima giornata di studio sulla storiografia greca frammentaria,
Genova, 31 maggio 2007, Roma, 157-181.
— (2010), „Due note di lessico scoliastico: παρακείμενον σχόλιον e
παραγράφεσθαι“, in: Fausto Montana (ed.), Aner polytropos. Ricerche di
filologia greca antica dedicate dagli allievi a Franco Montanari, Roma, 185-
195.
Montanari, Franco (1979), „I prolegomena del codice Rom. Bibl. Naz. gr. 6
(C)“, in: Id., Studi di filologia omerica antica, I, Pisa, 43-75.
— (1993), „L’erudizione, la filologia e la grammatica“, in: Giuseppe Cambiano
/ Luciano Canfora / Diego Lanza (eds.), Lo spazio letterario della Grecia anti-
ca, I, La produzione e la circolazione del testo (II, L’Ellenismo), Roma, 235-281.
— (ed.) (1994), La philologie grecque à l’époque hellénistique et romaine: sept exposés
suivis de discussions. XL Entretiens sur l’antiquité classique – Fondation
Hardt, Vandœuvres-Genève.
— (1997), „The Fragments of Hellenistic Scholarship“, in: Glenn W. Most
(ed.), Collecting Fragments / Fragmente sammeln, Göttingen, 273-288.
— (1998), „Zenodotus, Aristarchus and the Ekdosis of Homer“, in: Glenn W.
Most (ed.), Editing Texts / Texte edieren, Göttingen, 1-21.
— (2000), „Ripensamenti di Aristarco sul testo omerico e il problema della
seconda ekdosis“, in: Maria Cannatà Fera / Simonetta Grandolini (eds.),
Poesia e religione in Grecia. Studi in onore di G. Aurelio Privitera, Napoli,
479-486.
180 Bibliography

— (2002), „Alexandrian Homeric Philology. The Form of the ekdosis and the
variae lectiones“, in: Michael Reichel / Antonios Rengakos (eds.), Epea Pte-
roenta. Beiträge zur Homerforschung. Festschrift für Wolfgang Kullmann zum
75. Geburtstag, Stuttgart, 119-140.
— (2004), „La filologia omerica antica e la storia del testo omerico“, in: Anton
Bierl / Arbogast Schmitt / Andreas Willi, Antike Literatur in neuer Deutung.
Festschrift für Joachim Latacz, München, 127-143.
— (2006), „Glossario, parafrasi, ‘edizione commentata’ nei papiri“, in: Guido
Avezzù / Paolo Scattolin (eds.), I classici greci e i loro commentatori. Dai papiri
ai marginalia rinascimentali. Atti del convegno, Rovereto, 20 ottobre 2006,
Rovereto, 9-15.
— (2009a), „A new papyrus commentary on the Iliad“, in: Franco Montanari /
Serena Perrone (eds.), Fragments of the Past. Ancient Scholarship and Greek
Papyri, Trends in Classics 1/2, 173-178.
— (2009b), „Ekdosis alessandrina: il libro e il testo“, in: Manuel Sanz Morales /
Myriam Librán Moreno (eds.), Verae Lectiones. Estudios de crítica textual y
edición de textos griegos, Huelva-Cáceres, 143-167 (Pl. 1-8).
— (2009c), „Un papiro di Eschine con correzioni (P.Oxy. 2404):
considerazioni sull’ekdosis alessandrina“, in: Archiv für Papyrusforschung 55,
401-411.
Most, Glenn W. (ed.) (1999), Commentaries / Kommentare, Göttingen 1999.
Mountford, James F. (ed.) (1934), The Scholia Bembina, London.
Müller, Bruno A. (1903), De Asclepiade Myrleano, Diss. Leipzig.

Nagy, Gregory (2000), review of West 1998-2000, in: The Bryn Mawr Classical
Review 2000.09.12.
— (2003), review of West 2001a, in: Gnomon 75, 481-501.
— (2010), Homer the Classic, Cambridge, Mass.
Nardelli, Jean-Fabrice (2001), review of Martin L. West (ed.), Homeri Ilias, II,
München-Leipzig 2000, in: The Bryn Mawr Classical Review 2001.06.21.
Nauck, August (1848), Aristophanis Byzantii grammatici Alexandrini fragmenta,
Halle.
Neri, Camillo (ed.) (2003), Erinna. Testimonianze e frammenti, Bologna.
Nickau, Klaus (1972), „Zenodotos (3)“, in: RE 10A, 23-45.
— (1977), Untersuchungen zur Textkritische Methode des Zenodotos von Ephesos,
Berlin-New York.
Nicolai, Roberto (1992), La storiografia nell’educazione antica, Pisa.
Norden, Eduard (1901), „Vergils Aeneis im Lichte ihrer Zeit“, in: Neue Jahrbü-
cher für Philologie und Pädagogik 7, 249-334 (= Id., Kleine Schriften zum klassi-
schen Altertum, Berlin 1966, 358-421).
NP = Cancik, Hubert / Schneider, Helmuth (eds.), Der Neue Pauly. Enzyklopä-
die der Antike, Stuttgart-Weimar, 1996-2005.
 Bibliography 181

Page, Denis L. (ed.) (1962), Poetae melici Graeci, Oxford.

Pagani, Lara (2006a), ,,Apollonius [9] Dyscolus“, in: LGGA.
— (2006b), „Tyrannion [1] Maior“, in: LGGA.
Pantiglioni, Massimo (1998), ,,Il termine διάθεσις nella linguistica classica e
Dionisio Trace“, in: Athenaeum 86, 251-261.
Papathomas, Amphilochios (2003), „Scholien auf literarischen Papyri als
Zeugnisse für philologische Tätigkeit in der Peripherie der griechisch-
römischen Welt“, in: Classica et mediaevalia 54, 255-286.
Parker, David C. / Birdsall, James N. (2004), „The Date of the Codex
Zacynthius (Ξ): A New Proposal“, in: The Journal of Theological Studies 55,
117-131.
Pasquali, Giorgio (1920), Filologia e storia, Firenze.
— (19522), Storia della tradizione e critica del testo, Firenze (Ist ed. 1934).
Patillon, Michel (1990), „Contribution à la lecture de la Technê de Denys le
Thrace“, in: Revue des études grecques 103, 693-698.
Pecere, Oronzo (1986), „La tradizione dei testi latini tra IV e V secolo
attraverso i libri sottoscritti“, in: Andrea Giardina (ed.), Società romana e
impero tardoantico, IV, Tradizione dei classici. Trasformazioni della cultura,
Roma-Bari, 19-81, 210-246.
Pecorella, Giovan Battista (ed.) (1962), Dionisio Trace, Techne grammatike, testo
critico e commento, Bologna.
Pekridou-Gorecki, Anastasia (2009), „Textiles, Production of, II, Classical
Antiquity“, in: Hubert Cancik / Helmuth Schneider (eds.), Brill’s New
Pauly. Encyclopaedia of the Ancient World, Leiden-Boston, XIV 342-346.
Pépin, Jean / Saffrey, Henri D. (eds.) (1987), Proclus lecteur et interprète des
anciens. Actes du colloque international du CNRS, Paris, 2-4 octobre 1985,
Paris.
Perrone, Serena (2006), „Hypsicrates“, in: LGGA.
Pertusi, Agostino (1951), „Intorno alla tradizione manoscritta degli scolii di
Proclo ad Esiodo, IV“, in: Aevum 25, 147-159.
— (ed.) (1955), Scholia Vetera in Hesiodi Opera et Dies, Milano.
Petit, Françoise (ed.) (1977), Catenae Graecae in Genesim et in Exodum, I, Catena
Sinaitica, Turnhout-Leuven.
— (ed.) (1986), Catenae Graecae in Genesim et in Exodum, II, Collectio Coisliniana
in Genesim, Turnhout-Leuven.
Petraccia, Maria Federica (2003), „Catallage, lanipendia apula“, in: Alfredo
Buonopane / Francesca Cenerini (eds.), Donna e lavoro nella documentazione
epigrafica. Atti del I Seminario sulla condizione femminile nella documenta-
zione epigrafica, Bologna, 21 novembre 2002, Faenza, 255-259.
Pfeiffer, Rudolf (ed.) (1949-1953), Callimachus, I, Fragmenta; II, Hymni et
epigrammata, Oxford.
182 Bibliography

— (1968), History of Classical Scholarship from the Beginnings to the End of the Hel-
lenistic Age, Oxford.
Piccirilli, Luigi (ed.) (1985), Storie dello storico Tucidide. Edizione, traduzione e
commento delle Vite, Genova.
Pinborg, Jan (1975), Classical Antiquity: Greece, in: Thomas A. Sebeok (ed.),
Current Trends in Linguistics, XIII, Historiography of Linguistics, The Hague-
Paris, I 69-126.
Pizzocaro, Massimo (1993), „Un profilo di Telesilla, famosa poetessa d’Argo, e
guerriera“, in: Annali dell’Istituto Universitario Orientale di Napoli. Sezione
filologico-letteraria, 15, 88-103.
Pohlenz, Max (1939), „Die Begründung der abendländischen Sprachlehre
durch die Stoa“, in: Nachrichten von der Gesellschaft der Wissenschaft zu Göt-
tingen n.F. 3, 151-198 (reprinted in: Id., Kleine Schriften, ed. by Heinrich
Dörrie, Hildesheim 1965, 19-86).
Pollack, Erwin (1903), „Desultor“, in: RE 5/1, 255-259.
Pontani, Filippomaria (2002a), review of Matthaios 1999, in: Gnomon 74, 255-
257.
— (2002b), review of Matthaios 1999, in: Hellenika 52, 145-155.
— (ed.) (2005a), Eraclito. Questioni omeriche. Sulle allegorie di Omero in merito agli
dèi, Pisa.
— (2005b), Sguardi su Ulisse. Tradizione esegetica greca all’Odissea, Roma.
— (ed.) (2007), Scholia Graeca in Odysseam, I, Scholia ad libros α-β, Roma.
— (ed.) (2010), Scholia Graeca in Odysseam, II, Scholia ad libros γ-δ, Roma.
Porro, Antonietta (1985), „Manoscritti in maiuscola alessandrina di contenuto
profano. Aspetti grafici, codicologici, filologici“, in: Scrittura e civiltà 9, 169-
215.
— (ed.) (2004), „Alcaeus“, in: CLGP I.1.1 (Aeschines-Alcaeus), 75-246.
— (2009), „Forms and genres of Alexandrian exegesis on lyric poets“, in:
Franco Montanari / Serena Perrone (eds.), Fragments of the Past. Ancient
Scholarship and Greek Papyri, Trends in Classics 1/2, 183-202.
Prencipe, Vittoria (2002), „Statuto epistemologico della “grammatica”
nell’antichità greca“, in: Swiggers/Wouters 2002a, 23-68.

Rashed, Marwan (1995), „Alexandre d’Aphrodise et la «magna quaestio». Rôle

et indépendance des scholies dans la tradition byzantine du corpus
aristotélicien“, in: Les études classiques 63, 295-351 (reprinted in: Id.,
L’héritage aristotélicien. Texts inédits de l’Antiquité, Paris 2007, 85-141).
— (ed.) (2009), Alexandre d’Aphrodise, Commentaire perdu à la “Physique”
d’Aristote (livres IV-VIII). Les scholies byzantines, édition, traduction et
commentaire, Berlin-New York.
Razzetti, Francesca (2002), I frammenti del grammatico Aristonico, Tesi di Dottora-
to, Genova.
 Bibliography 183

— (2003a), „Aristonicus“, in: LGGA.

— (2003b), „Philoxenus“, in: LGGA.
RE = Paulys Realencyclopädie der classischen Altertumswissenschaft, Stuttgart-
München, 1893-1978.
Reitzenstein, Richard (1897), Geschichte der griechischen Etymologika. Ein Beitrag
zur Geschichte der Philologie in Alexandria und Byzanz, Leipzig.
Rengakos, Antonios (2002a), review of West 2001a, in: The Bryn Mawr
Classical Review 2002.11.15.
— (2002b), „The Hellenistic Poets as Homeric Critics“, in: Franco Montanari
(ed.), Omero tremila anni dopo. Atti del congresso di Genova, 6-8 luglio
2000, con la collaborazione di Paola Ascheri, Roma, 143-157.
Reynolds, Leighton D. / Wilson, Nigel G. (19913), Scribes and Scholars. A Guide
to the Transmission of Greek and Latin Literature, Oxford (Ist ed. 1968).
Ribbach, Willy E. (1883), De Aristarchi Samothracis arte grammatica, Naumburg.
Richard, Marcel (1950), „Ἀπὸ φωνῆς“, in: Byzantion 20, 191-222.
Richardson, Nicholas J. (1980), „Literary Criticism in the Exegetical Scholia to
the Iliad. A Sketch“, in: Classical Quarterly 30, 265-287.
— (1994), „Aristotle and Hellenistic Scholarship“, in: Montanari 1994, 7-28.
Rispoli, Gioia M. (1988), Lo spazio del verisimile: il racconto, la storia e il mito,
Napoli.
Robert, Louis (1940), „La bibliothèque de Nysa de Carie“, in: Hellenica 1, 144-
148.
Roberts, Colin H. (1935), „The Antinoë Fragment of Juvenal“, in: The Journal
of Egyptian Archaeology 21, 199-207.
Robins, Robert H. (1957), „Dionysius Thrax and the Western Grammatical
Tradition“, in: Transactions of the Philological Society, 67-106.
— (1966), „The Development of the Word Class System of the European
Grammatical Tradition“, in: Foundations of Language 2, 3-19.
— (1996), „The Initial Section of the Tékhne grammatiké“, in Pierre Swiggers /
Alfons Wouters (eds.), Ancient Grammar. Content and Context, Leuven-Paris,
3-15.
— (1998), „The Authenticity of the Techne: the status quaestionis“, in: Vivien
Law / Ineke Sluiter (eds.), Dionysius Thrax and the Techne grammatike,
Münster, 13-26.
Romano, Francesco (1994), „La scuola filosofica e il commento“, in: Giuseppe
Cambiano / Luciano Canfora / Diego Lanza (eds.), Lo spazio letterario della
Grecia antica, I, La produzione e la circolazione del testo (III, I Greci e Roma),
Roma, 587-611.
Russell, Donald A. / Konstan, David (eds.) (2005), Heraclitus, Homeric Problems,
Atalanta.
Russo, Joseph (19934) (ed.), Omero, Odissea, V, Libri XVII-XX, trad. di Giu-
seppe Aurelio Privitera, Milano (Ist ed. 1985).
184 Bibliography

Rutherford, William G. (1905), A Chapter in the History of Annotation, being

Scholia Aristophanica III, London.

Saffrey, Henri D. / Segonds, Alain-Philippe (eds.) (2001), Marinus, Proclus ou

Sur le bonheur, avec la collaboration de Concetta Luna, Paris.
Sandys, John E. (19213), A History of Classical Scholarship, I, From the Sixth
Century B.C. to the End of the Middle Ages, Cambridge (Ist ed. 1903).
Schenkeveld, Dirk M. (1984), „Studies in the History of Ancient Linguistics, II,
Stoic and Peripatetic Kinds of Speech Act and the Distinction of Grammat-
ical Moods“, in: Mnemosyne 37, 291-353.
— (1990a), „Studies in the History of Ancient Linguistics, III, The Stoic Τέχνη
περὶ φωνῆς“, in: Mnemosyne 43, 86-108.
— (1990b), „Studies in the History of Ancient Linguistics, IV, Developments
in the Study of Ancient Linguistics“, in: Mnemosyne 43, 289-306.
— (1994), „Scholarship and Grammar“, in: Montanari 1994, 263-306.
— (1998), „The Linguistic Contents of Dionysius’ Παραγγέλματα“, in: Vi-
vien Law / Ineke Sluiter (eds.), Dionysius Thrax and the Techne grammatike,
Münster, 41-53.
— (1999), review of Ildefonse 1997, in: Mnemosyne 52, 606-607.
— (2000), review of Matthaios 1999, in: The Bryn Mawr Classical Review
2000.07.14.
Schironi, Francesca (2009), „Theory into practice: Aristotelian principles in
Aristarchean philology“, in: Classical Philology 104, 279-316.
Schmidt, Martin (1976), Die Erklärungen zum Weltbild Homers und zur Kultur der
Heroenzeit in den bT-Scholien zur Ilias, München.
— (1997), „Variae lectiones oder Parallelstellen: Was notierten Zenodot und
Aristarch zu Homer?“, in: Zeitschrift für Papyrologie und Epigraphik 115, 1-
12.
Schmidt, Mauricius (ed.) (1854), Didymi Chalcenteri grammatici Alexandrini
fragmenta, Leipzig.
Schmidt, Moriz (1852-1853), „Dionys der Thraker“, in: Philologus 7, 360-382;
ibid. 8, 231-253, 510-520.
Schmidt, Peter L. (1999), „Iunius Filagrius [III 2]“, in: NP 6, 70-71.
Schmidt, Rudolph T. (1839), Stoicorum grammatica, Halle.
Schmitter, Peter (ed.) (1991), Geschichte der Sprachtheorie, II, Sprachtheorien der
abendländischen Antike, Tübingen.
— (2000), „Sprachbezogene Reflexionen im frühen Griechenland“, in: Sylvain
Auroux / Ernst F. K. Koerner / Hans-Joseph Niederehe / Kees Versteegh
(eds.), History of the Language Sciences. An International Handbook on the Evo-
lution of the Study of Language from Beginnings to the Present / Geschichte der
Sprachwissenschaften. Ein internationales Handbuch zur Entwicklung der Sprach-
forschung von den Anfängen bis zur Gegenwart / Histoire des sciences du langage.
 Bibliography 185

Manuel international sur l’évolution de l’étude du langage des origines à nos jours,
I, 1, 1, Berlin-New York, 345-366.
Schneider, Otto (1845), ,,Über die Schlussparthie der Schrift des Apollonius
Dyscolus Περὶ ἐπιρρημάτων“, in: Rheinisches Museum für Philologie 3, 446-
459.
Schneider, Richard (1869), ,,Zu Apollonios Dyscolos, III“, in: Rheinisches Mu-
seum für Philologie 24, 585-596.
— (ed.) (1878), Grammatici Graeci, II/I, Apollonii scripta minora, Leipzig.
— (ed.) (1902), Grammatici Graeci, II/II, Commentarius criticus et exegeticus in
Apollonii scripta minora, Leipzig.
— (ed.) (1910), Grammatici Graeci, II/III, Librorum Apollonii deperditorum fragmen-
ta, Leipzig.
Schneider, Richard / Uhlig, Gustav (eds.) (1878-1910), Grammatici Graeci, II,
Leipzig.
Schoemann, Georg F. (1862), Die Lehre von den Redetheilen nach den Alten, Ber-
lin.
Schwartz, Eduard (1881), „De scholiis Homericis ad historiam fabularem
pertinentibus“, in: Jahrbücher für classische Philologie, Suppl. 12, 403-463.
— (ed.) (1887), Scholia in Euripidem, I, Berlin.
— (1895), Aristodemos (n. 30), RE 2/1, 925-926.
Sciarra, Elisabetta (2005), La tradizione degli scholia iliadici in Terra d’Otranto,
Roma.
Serra, Giuseppe (1995), „Sulla tradizione araba“, in: Giuseppe Cambiano /
Luciano Canfora / Diego Lanza (eds.), Lo spazio letterario della Grecia antica,
II, La ricezione e l’attualizzazione del testo, Roma, 251-261.
Siebenborn, Elmar (1976), Die Lehre von der Sprachrichtigkeit und ihren Kriterien.
Studien zur antiken normativen Grammatik, Amsterdam.
Skrzeczka, Rudolph F. L. (1853), ,,Die Lehre des Apollonius Dyscolus von den
Redetheilen und kritische Bemerkungen zu Apollonius de adverbio“, in: Zur
öffentlichen Prüfung im Kneiphöfischen Stadt-Gymnasium Königsberg, 1-28.
Slater, William J. (1972), „Asklepiades and Historia“, in: Greek, Roman and
Byzantine Studies 13, 317-333.
— (ed.) (1986), Aristophanis Byzantii Fragmenta, (SGLG 6), Berlin-New York.
Sluiter, Ineke (1990), Ancient Grammar in Context. Contributions to the Study of
Ancient Linguistic Thought, Amsterdam.
— (1999), „Commentaries and the Didactic Tradition“, in: Most 1999, 173-
205.
— (2000a), „Language and Thought in Stoic Philosophy“, in: Sylvain Auroux
/ Ernst F. K. Koerner / Hans-Joseph Niederehe / Kees Versteegh (eds.),
History of the Language Sciences. An International Handbook on the Evolution of
the Study of Language from Beginnings to the Present / Geschichte der Sprachwis-
senschaften. Ein internationales Handbuch zur Entwicklung der Sprachforschung
186 Bibliography

von den Anfängen bis zur Gegenwart / Histoire des sciences du langage. Manuel
international sur l’évolution de l’étude du langage des origines à nos jours, I, 1, 1,
Berlin-New York, 375-384.
— (2000b), „The Dialectics of Genre: Some Aspects of Secondary Literature
and Genre in Antiquity“, in: Depew/Obbink 2000, 183-203.
Smith, Ole L. (1975), Studies in the Scholia on Aeschylus, I, The recensiones of
Demetrius Triclinius, Leiden.
— (ed.) (1976), Scholia Graeca in Aeschylum quae exstant omnia, Pars I: Scholia in
Agamemnonem Choephoros Eumenides Supplices continens, Leipzig.
— (1981), „Classification of MSS of the Scholia on Aeschylus“, in: Illinois
Classical Studies 6, 44-55.
Smith, Peter M. (1981), „Aineiadai as Patrons of Iliad XX and the Homeric
Hymn to Aphrodite“, in: Harvard Studies in Classical Philology 85, 17-58.
Smyth, Herbert W. (1933), „Catalogue of the Manuscripts of Aeschylus“, in:
Harvard Studies in Classical Philology 44, 1-62.
Sodano, Angelo R. (ed.) (1970), Porphyrii Quaestionum Homericarum liber I,
Napoli.
Solmsen, Friedrich (1986), „«Aeneas founded Rome with Odysseus»“, in: Har-
vard Studies in Classical Philology 90, 93-110.
Sorabji, Richard (ed.) (1990), Aristotle Transformed. The Ancient Commentators
and their Influence, London.
— (ed.) (2004), The Philosophy of the Commentators: 200-600 A.D.: A
Sourcebook, II, Physics, London.
Steinthal, Heymann (1890-18912), Geschichte der Sprachwissenschaft bei den Grie-
chen und Römern mit besonderer Rücksicht auf die Logik, I-II, Berlin (Ist ed.
1863).
Stroppa, Marco (2009), „Some Remarks Regarding Commentaries on Codex
from Late Antiquity“, in: Franco Montanari / Serena Perrone (eds.),
Fragments of the Past. Ancient Scholarship and Greek Papyri, Trends in Classics
1/2, 298-327.
Susemihl, Franz (1891-1892), Geschichte der griechischen Literatur in der
Alexandrinerzeit, I-II, Leipzig.
Swiggers, Pierre / Wouters, Alfons (1995), „Techne et empeiria: la dynamique de
la grammaire grecque dans l’antiquité à la lumière des papyrus grammati-
caux“, in: Lalies. Actes des sessions de linguistique et de littérature. 15
(Aussois, 29 août-3 septembre 1994), Paris, 83-100.
— (2000), „Grammaires grecques (et latines) sur papyrus“, in: Mario De Non-
no / Paolo de Paolis / Louis Holtz (eds.), Manuscripts and Tradition of
Grammatical Texts from Antiquity to the Renaissance. Proceedings of a Confe-
rence held at Erice, 16-23 October 1997, as the 11th Course of Internation-
al School for the Study of Written Records, Cassino, I 59-88.
— (eds.) (2002a), Grammatical Theory and Philosophy of Language in Antiquity,
Leuven-Paris-Sterling.
 Bibliography 187

— (2002b), „Grammatical Theory and Philosophy of Language in Antiquity:

Introduction“, in: Swiggers/Wouters 2002a, 9-20.
— (2002c), „Grammatical Theory in Aristotle’s Poetics, Chapter XX“, in:
Swiggers/Wouters 2002a, 101-120.
— (2003), „Réflexions à propos de (l’absence de?) la syntaxe dans la grammaire
gréco-latine“, in: Pierre Swiggers / Alfons Wouters (eds.), Syntax in Anti-
quity, Leuven-Paris-Dudley, 25-41.
— (2005), „L’élaboration de la grammaire comme discipline ‘technique’“, in:
Ioannis G. Taifacos (2005), The Origins of European Scholarship. The Cyprus
Millennium International Conference, Stuttgart, 1-12.

Taylor, Daniel J. (1987), „Rethinking the History of Language Science in Clas-

sical Antiquity“, in: Id. (ed.), The History of Linguistics in the Classical Period,
Amsterdam-Philadelphia, 1-16.
Tessier, Andrea (ed.) (2005), Demetrio Triclinio, Scolii metrici alla tetrade sofoclea,
Alessandria.
Theodoridis, Christos (ed.) (1976), Die Fragmente des Grammatikers Philoxenos,
(SGLG 2), Berlin-New York.
Thiel, Helmut van (1992), „Zenodot, Aristarch und Andere“, in: Zeitschrift für
Papyrologie und Epigraphik 90, 1-32.
— (1997), „Der Homertext in Alexandria“, in: Zeitschrift für Papyrologie und
Epigraphik 115, 13-36.
— (2000), „Die D-Scholien der Ilias in den Handschriften“, in: Zeitschrift für
Papyrologie und Epigraphik 132, 1-62 (partially available at:
http://kups.ub.uni-koeln.de/volltexte/2006/1810/).
Tosi, Renzo / Rüpke, Jörg (1998), „Grammatiker“, in: NP 4, 1199-1203.
Trojahn, Silke (2002), Die auf Papyri erhaltenen Kommentare zur Alten Komödie.
Ein Beitrag zur Geschichte der antiken Philologie, München-Leipzig.
Turner, Eric G. (ed.) (1957), „P.Oxy. 2404. Aeschines, In Ctesiphontem 51-
53“, in: The Oxyrhynchus Papyri. Part XXIV, London, 130-132.
— (19802), Greek Papyri. An Introduction, Oxford-New York (Ist ed. 1968).
Turner, Eric G. / Parsons, Peter J. (1987), Greek Manuscripts of the Ancient
World, Second Edition revised and enlarged, London.
Turyn, Alexander (1943), The Manuscript Tradition of the Tragedies of Aeschylus,
New York.
— (1952), Studies in the Manuscript Tradition of the Tragedies of Sophocles, Urbana.
— (1957), The Byzantine Manuscript Tradition of the Tragedies of Euripides,
Urbana.

Uhlig, Gustav (1882), „Zur Wiederherstellung des ältesten occidentalischen

Compendium der Grammatik“, in: Festschrift zur Begrüssung der 36. phi-
188 Bibliography

lologischen Versammlung an der Heidelberger Universität, Heidelberg, 61-

85.
— (ed.) (1883), Grammatici Graeci, I/I, Dionysii Thracis ars grammatica, Leipzig.
— (ed.) (1910), Grammatici Graeci, II/II, Apollonii Dyscoli de constructione libri
quattuor, Leipzig.
Usener, Hermann (1892), „Ein altes Lehrgebäude der Philologie“, in: Sitzungs-
berichte der Bayerischen Akademie der Wissenschaft 4, 582-648 (reprinted in:
Id., Kleine Schriften, II, Leipzig-Berlin 1913, 265-314).

Valk, Marchinus van der (1963-1964), Researches on the Text and Scholia of the
Iliad, I-II, Leiden.
Vallance, John T. (1999), „Galen, Proclus and the Non-submissive
Commentary“, in: Most 1999, 223-244.
Vanotti, Gabriella (1995), L’altro Enea. La testimonianza di Dionigi di Alicarnasso,
Roma.
Velsen, Arthur von (1853), Tryphonis grammatici Alexandrini fragmenta, Berlin.
Ventrella, Gianluca (2004), „Dionisio Trace e la definizione di grammatica“, in:
Kleos 9, 103-110.

Watts, Edwards J. (2006), City and School in Late Antique Athens and Alexandria,
Berkeley-Los Angeles-London.
Wendel, Carl (1920), Überlieferung und Entstehung der Theokrit-Scholien, Berlin.
— (1941), „Philoxenos (n. 27)“, in: RE 20/1, 194-200.
— (1943), „Tyrannion (n. 2)“, in: RE 7A/2, 1811-1819.
— (1948), „Tyrannion (n. 3)“, in: RE 7A, 1819-1820.
Wentzel, Georg (1896), „Asklepiades (n. 28)“, in: RE 2/2, 1628-1631.
West, Martin L. (ed.) (1978), Hesiod, Works and Days, Oxford.
— (ed.) (1998-2000), Homeri Ilias, I-II, Stuttgart-Leipzig.
— (2001a), Studies in the Text and Transmission of the Iliad, München-Leipzig.
— (2001b), Response on Nagy and Nardelli, in: The Bryn Mawr Classical Review
2001.09.06.
White, John Williams (ed.) (1914), The Scholia on the Aves of Aristophanes,
Boston-London.
Wilson, Nigel G. (1967), „A Chapter in the History of Annotation“, in:
Classical Quarterly n.s. 17, 244-256.
— (1971), „Two Notes on Byzantine Scholarship, I, The Vienna Dioscorides
and the History of Scholia“, in: Greek, Roman and Byzantine Studies 12,
557-558.
— (1977), review of Zuntz 1975, in: Classical Review n.s. 27, 271.
— (1983a), Scholars of Byzantium, London.
— (1983b), „Scoliasti e commentatori“, in: Studi classici e orientali 33, 83-112.
 Bibliography 189

— (1984), The Relation of Text and Commentary in Greek Books, in: Cesare
Questa / Renato Raffaelli (eds.), Il libro e il testo. Atti del Convegno
internazionale, Urbino 20-23 settembre 1982, Urbino, 105-110.
— (2004), „La filologia greca nell’antichità“, in: Heinz-Günther Nesselrath
(ed.), Introduzione alla filologia greca, Roma (German ed.: Stuttgart-Leipzig
1997), 111-130.
— (2007), „Scholiasts and Commentators“, in: Greek, Roman and Byzantine
Studies 47, 39-70.
Wouters, Alfons (1979), The Grammatical Papyri from Graeco-Roman Egypt. Con-
tributions to the Study of the «Ars grammatica» in Antiquity, Brussel.
— (1997), „The Grammatical Papyrus P.Berol. inv. 9917“, in: Bärbel Kramer /
Wolfgang Luppe / Herwig Maehler / Günther Poethke (eds.), Akten des
21. internationalen Papyrologenkongresses: Berlin, 13-19 August 1995,
Stuttgart-Leipzig, II 1021-1033.
— (1998), „The Grammatical Papyri and the Techne grammatike of Dionysius
Thrax“, in: Vivien Law / Ineke Sluiter (eds.), Dionysius Thrax and the
Techne grammatike, Münster, 95-109.

Xenis, Georgios A. (ed.) (2010), Scholia vetera in Sophoclis Trachinias, Berlin-

New York.

Zetzel, James (1975), „On the History of Latin Scholia“, in: Harvard Studies in
Classical Philology 79, 335-354.
Zuntz, Günther (1965), An Inquiry into the Transmission of the Plays of Euripides,
Cambridge.
— (1975), Die Aristophanes-Scholien der Papyri, Berlin (reprinted, with a
Nachwort, from: Byzantion 13, 1938, 631-690; 14, 1939, 545-614).
Zuretti, Carlo O. (1892), Analecta Aristophanea, Torino.