Tsiamp ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 ISSN 2159-3159 Article copyright held by the author(s) and made available under the Creative Commons Attribution License CC-BY https://creativecommons.org/licenses/by/4.0/ Between Scholarly Tradition and Didactic Innovation: Maximus Planudes’ Dialogue on Grammar Theofanis Tsiampokalos HROUGHOUT the whole period from the early sixth to the mid-fifteenth century1 knowledge of a high-register variety of Greek—one modelled on a selection of canoni- cal Attic or Atticizing authors, along with the Church Fathers, and therefore differing from the vernacular—served a quite pragmatic purpose in the Eastern Roman Empire: highly de- veloped skills in reading and writing this variety would mark a young adult out for a promising career in the administration of the state or church,2 in much the same way as, in later times, 1 For the beginning and end of the Byzantine period I use the conventional dating scheme suggested by K. Krumbacher, Geschichte der byzantinischen Litteratur (Berlin 1897) 1, which places the beginning on 22 September 529, the day Justinian closed the School of Athens, and the end on 29 May 1453, when Constantinople fell to Mehmed II. Cf. also F. Pontani, in F. Montanari et al. (eds.), Brill’s Companion to Ancient Greek Scholarship (Leiden 2015) 297–455, at 298. 2 On this social-mobility function of education see esp. A. Markopoulos, in S. Steckel et al. (eds.), Networks of Learning. Perspectives on Scholars in Byzantine East and Latin West (Münster 2014) 3–15, at 5; Pontani, in Companion 301; F. Nousia, Byzantine Textbooks of the Palaeologan Period (Vatican City 2016) 36–38; A. M. Cuomo, “Medieval Textbooks as a Major Source for Historical Socio- linguistic Studies of (high-register) Medieval Greek,” Open Linguistics 3 (2017) 442–455, at 443–444, 452, and “Late Byzantine Scholia on the Greek Classics,” in B. van den Berg et al. (eds.), Byzantine Commentaries on Ancient Greek Texts, 12th–15th Centuries (Cambridge 2022) 304–338, at 304–305 (with fur- T THEOFANIS TSIAMPOKALOS 327 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 knowledge of Latin and of Greek and Roman history would mark someone out for a post in the British Empire.3 This was a language that one was expected to start learning from around the age of twelve, at the school of a grammarian (γραµµατικός, διδάσκαλος).4 In his teaching, whether at an ele- mentary or an advanced level, this grammarian would use various scholarly works, such as editions of classical texts and commentaries, as well as other textbooks, including systematic works of grammar, examples of sentence-parsing known as Epimerismoi,5 paradigms for declension and conjugation, and, needless to say, an assortment of dictionaries and word lists, the main function of which was to juxtapose low-register Greek words with high-register (or at least higher-register) Greek words.6 In the twilight of the Byzantine age, the Palaeologan ther bibliography). On the basic outlines of Byzantine education, which re- mained unchanged throughout this period, see e.g. Markopoulos 3, Pontani 302, Nousia 29–48, Cuomo 306 n.11. 3 See e.g. N. Vance, The Victorians and Ancient Rome (Oxford 1997) 13–14. For acquaintance with classical authors as part of a gentlemanly culture in Britain up to the late nineteenth-century, see C. Stray, Classics Transformed: Schools, Universities, and Society in England, 1830–1960 (Oxford 1998) 1–82. 4 Cf. Nousia, Byzantine Textbooks 33, 47. 5 On the method of µερισµός, ὑποµερισµός, and ἐπιµερισµός see A. Garzya, “Per l’erudizione scolastica a Bisanzio,” Byzantino-Sicula III (Palermo 2000) 135–140. Cf. E. Dickey, Ancient Greek Scholarship: A Guide (New York 2006) 27– 28; Nousia, Byzantine Textbooks 53. 6 For precisely this reason, such texts—especially the dictionaries and word lists—themselves constitute important testimonies for the study of medieval Greek; they are suitable for a historical sociolinguistic analysis that could shed light on the social and cultural factors that influenced linguistic variation in medieval Greek. The reason for this is that, in order to explain high-register Greek, they systematically use lower registers, including variants of the ver- nacular. Thus e.g. schol. Mosch. Soph. El. 606, where the word κήρυσσε (<κηρύσσω) is glossed not only as διαδίδου (<διαδίδωµι), but also as γυβέντιζε (<γυβεντίζω), a medieval French loanword which at some point entered the Greek vernacular. For all this see Cuomo, Open Linguistics 3 (2017) 450. Needless to say, as far as classical scholarship is concerned, refining our 328 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 period,7 a new type of textbook began to gain popularity: the so- called Erotemata (Questions), which emerged with Manuel Mos- chopoulos and subsequently, thanks to Manuel Chrysoloras, made their way into Italy and the rest of Western Europe.8 These texts were exegetical works structured according to a question-and-answer format that dealt with various grammatical topics, a form of text whose roots can of course be traced back to the ancient tradition of philosophical questions (Quaestiones, Zetemata).9 knowledge about the high-register Greek as taught by medieval scholars may also result in a historically more accurate perspective on the transmission of certain ancient authors. I do not intend to analyze this topic here, since it deserves separate treatment, but will investigate it in a future study. 7 The period is often described as “Renaissance,” which is not entirely accurate, since contrary to the Latin West, the Greek-speaking East never lost contact with the classical Greek sources. On this see e.g. Pontani, in Companion 404–405. 8 The earliest specimen is Manuel Moschopoulos’ Erotemata; see A. Rollo, “Gli Erotemata di Manuele Moscopulo e i suoi precedenti,” AION(filol) 41 (2019) 235–252. Chrysoloras’ Erotemata represents another popular specimen, one which was additionally designed to meet the needs of Latin speakers who wished to learn Greek; see the critical edition by A. Rollo, Gli Erotemata tra Crisolora e Guarino (Messina 2012) 301–338. On Chrysoloras (ca. 1350–1415) cf. J. E. Sandys, A History of Classical Scholarship I (Cambridge 1903) 421, 573; U. von Wilamowitz-Moellendorff, History of Classical Scholarship (Baltimore 1982 [1921]) 21; I. Thompson, “Manuel Chrysoloras and the Early Italian Renaissance,” GRBS 7 (1966) 63–82, esp. 73–74; L. Thorn-Wickert, Manuel Chrysoloras (ca. 1350–1415). Eine Biographie des byzantinischen Intellektuellen vor dem Hintergrund der hellenistischen Studien in der italienischen Renaissance (Frankfurt am Main 2006), esp. 147–209 on his philological activities while staying in Florence. 9 On the question-and-answer format (Erotapokriseis) see A. Garzya, “Appunti sulle erotapocriseis,” VetChr 29 (1992) 305–314; A. M. Ieraci Bio, “L’ἐρωταπόκρισις nella letteratura medica,” in C. Moreschini (ed.), Esegesi, parafrasi e compilazione in età tardoantica (Naples 1995) 187–207; V. Law, “The Mnemonic Structure of Ancient Grammatical Doctrine,” in P. Swiggers et al. (eds.), Ancient Grammar: Content and Context (Leuven 1996) 37–52; A.-L. Rey, “Des textes anciens et production de nouveaux textes,” in A. Volgers et al. THEOFANIS TSIAMPOKALOS 329 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 The present article seeks to examine the scholarly develop- ment underlying the use of this latter text-type for teaching Greek grammar, as well as to identify the purpose (in terms of didactic efficiency) that a question-and-answer format might have been likely to serve. To do this, I will focus on the Dialogue on Grammar, a literary dialogue10 written by Maximus Planudes, one of the leading scholars of the earlier Palaeologan period.11 The work was used in an educational context, as the content of codices transmitting this text indicates.12 Unfortunately, the text is not available in any other edition than an editio princeps, which—as is often the case with such texts—relies on the authority of a single manuscript, which is not even the oldest (ed.), Early Christian Question-and-Answer Literature in Context (Leuven 2004) 165– 189; Y. Papadoyannakis, “Instruction by Question and Answer,” in S. F. Johnson (ed.), Greek Literature in Late Antiquity: Dynamism, Didacticism, Classicism (Aldershot 2006) 91–105; Nousia, Byzantine Textbooks 82–83. 10 Generally speaking, the choice of a literary dialogue to serve as a medium of knowledge transmission is of course neither novel (e.g. Plato, Plutarch) nor random; see S. Föllinger, “Lehren im Gespräch: Der litera- rische Dialog als Medium antiker Wissensvermittlung,” Gymnasium 113 (2006) 455–470. 11 For general accounts of Planudes (PLP no. 23308) see e.g. Sandys, A History of Classical Scholarship I 417; C. Wendel, “Planudes, Maximos,” RE 20 (1950) 2202–2253; H. Hunger, Die hochsprachliche profane Literatur der Byzantiner (Munich 1978) I 129–130, II 67–71, 246–246; N. G. Wilson, Scholars of Byzan- tium2 (London 1996) 230. On Planudes’ exceptional position among his con- temporaries see Pontani, in Companion 409–421. 12 The earliest MSS. preserving the text consist for the most part of texts whose content is either grammatical/rhetorical or moral/philosophical or mythographical: e.g. Vat.gr. 113, of the 13th/14th cent. (which includes gram- matical texts by Planudes and Moschopoulos alongside Aesop’s fables and further texts mainly by authors such as Basil, Libanius, Plato, Synesius), Vat.Urb.gr. 151, 14th cent. (which consists exclusively of various grammatical works and Aesopian fables), Vat.Urb.gr. 152, 14th cent. (texts of grammatical content alongside Philostratus’ Images), and even Vat.gr. 15, 14th/15th cent. (for the most part grammatical works, lexica, Gregoras’ scholia on Synesius, and speeches by Aelius Aristides). 330 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 surviving one.13 It presents an experienced teacher of Greek by the name of Palaitimos ( “the long-esteemed one”)14 explaining various grammatical phenomena to a young student called Neophron ( “novice in thinking”).15 The analysis presented here is not intended to cover the whole dialogue, which in Bachmann’s old edition runs to over 100 pages and deals with several grammatical topics. For the sake of coherence and brevity, the scope is restricted to the first section of the text (ending at Dial. 15.11), since this is already identified as a separate section in most of the earliest manuscripts.16 This 13 The same holds true for Planudes’ treatise On Syntax. The only edition available of both texts is still L. Bachmann, Anecdota Graeca II (Leipzig 1828) 1–101 (the Dialogue) and 103–166 (On Syntax), which is based on the authority of Paris.suppl.gr. 70 (15th cent.). That said, however important this MS. may be, it is neither the only nor the oldest one to document these texts. More specifically, for the Dialogue on Grammar there seem to be at least 49 MSS. available (according to the Pinakes), the earliest, Vat.gr. 113, dating to the 13th/14th cent. On Syntax seems to be transmitted in 45 MSS. (once again, ac- cording to the Pinakes), of which the earliest is the same Vat.gr. 113. The text is also transmitted in many 14th/15th cent. MSS. Bachmann (153–154) also used some excerpts from Paris.gr. 2669 (17th cent.). For the phenomenon of early editions being based on the authority of a single MS., which is not always of the best-possible quality, see e.g. S. Timpanaro, The Genesis of Lachmann’s Method (Chicago 2005) 45 (with further references). Nonetheless, it should be noted that in recent years there has been an attempt to re-edit both texts of Planudes through collation of a selection of additional witnesses. This work was conducted in the context of a master’s thesis at Aristotle University of Thessaloniki but remains unpublished: A. N. Kakali, Η κριτική έκδοση των έργων του Μάξιμου Πλανούδη “Διάλογος περί γραμματικής” και “περί συν- τάξεως” (1998). More recently, K. Oikonomakos, “Pour une nouvelle édition du Περὶ γραµµατικῆς διάλογος de Maxime Planude,” RCCM 50 (2008) 139– 155, announced a forthcoming new edition. However, to the best of my knowledge, this work was never completed. 14 Lit. the one who is esteemed for a long time; hence, very esteemed. Cf. LSJ s.v. παλαιένδοξος “of old renown.” 15 LSJ s.v. νεόφρων “childish in spirit”; LBG “von neuartiger Gesinnung.” 16 E.g. Vat.gr. 113, at 47r (towards the end of the folium); Vat.Urb.gr. 152, at 276r; Monac.gr. 499 (15th cent.), at 282v. In the context of my project on the THEOFANIS TSIAMPOKALOS 331 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 is in keeping with the aim of the article, as the section in question covers topics related to the Greek tenses, with a particular focus on the much-discussed problem of voice (on which more below). Hence, it enables us to see Planudes as belonging to a broader tradition dealing with these specific topics. Moreover, the article will focus mainly on how the grammatical exposition is arranged in this dialogue. As we shall see, the arrangement is of a sort that could enable a thirteenth- or fourteenth-century student of Greek to develop a more fundamental understanding of the language, one that decidedly goes beyond the “mechanical,” productive, and descriptive form of grammatical exposition to which students of Greek would have been accustomed. The article is divided into two main sections. The first, shorter section provides a brief overview of the standard sources that were available to Byzantine students of Greek and comments on the usual way in which grammatical knowledge is organised in those sources, especially with regard to the topics of Greek tenses and voice. Although this section does not deal directly with the text of the Dialogue, the information it provides is essential for understanding and contextualizing Planudes’ innovation, which is then presented in the longer second section. This has two parts, which focus on Planudes’ text and discuss two charac- teristic cases in which innovative approaches are used to explain the grammatical topics—approaches that would not have been possible if the pattern of grammatical exposition had followed that of more authoritative, traditional grammar textbooks in- stead of a question-and-answer format. structures of Maximus Planudes’ works on grammar and syntax (see n.41 below), I had the opportunity to selectively collate several earlier manuscripts of the Dialogue and On Syntax. However, my aim was not to prepare a new critical edition—though such an edition is indeed much needed—but rather to explore how non-textual elements, such as indications of paragraph or section changes, rubrics, and marginal comments, contribute to ‘information management’ and interact with grammatical rules taught in Planudes’ text- books on the Greek language. 332 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 1. Scholarly tradition At the transition from the thirteenth to the fourteenth century —approximately the time when the dialogue was composed— certain textbooks existed for the teaching and learning of high- register Greek that had been in use at schools for centuries and were thus considered standard.17 A short text from the first reign of Emperor Isaac II Angelos (1185–1195), edited by Max Treu and known as the “byzantinisches Schulgespräch,”18 provides a glimpse into the character of these textbooks. This work is a pseudo-dialogue between a teacher and a student (“pseudo” because only the teacher speaks), presenting a general overview of the knowledge that one could acquire in the course of sys- tematic education, part of which was training in the use of high- register Greek.19 In this pseudo-dialogue, however, the grammatical textbooks that the teacher mentions are those of Dionysius Thrax (ob- viously the Τέχνη γραµµατική attributed to him) and Theodosius —the latter was the author of a set of paradigms for declensions and conjugations dating to late antiquity,20 although a work on grammar is also mistakenly attributed to him21—alongside the works of the so-called commentators, Oros, Herodian, Heliodo- rus, and George Choeroboscus.22 Of these five, (ps.-)Dionysius23 17 Cf. Nousia, Byzantine Textbooks 51. 18 M. Treu, “Ein byzantinisches Schulgespräch,” ByzZeit 2 (1893) 96–105. For the date, however, see C. M. Mazzucchi, “Ambrosianus C 222 inf. (Graecus 886): Il codice e il suo autore,” Aevum 77 (2003) 263–275, at 275. Treu (103) dated the text to the second half of the eleventh century. 19 On all these points, and especially on whether the tract should be con- sidered as a pseudo-dialogue, see Cuomo, Open Linguistics 3 (2017) 446–447. 20 See Dickey, Ancient Greek Scholarship 83. 21 Text: C. G. Göttling, Θεοδοσίου γραμματικοῦ περὶ γραμματικῆς (Leipzig 1822) 1–197. 22 On the ‘standard’ literature for the teaching of grammar and rhetoric see Pontani, in Companion 301. 23 V. Di Benedetto, “Dionisio Trace e la Techne a lui attribuita,” AnnPisa THEOFANIS TSIAMPOKALOS 333 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 and Choeroboscus are of particular significance for the present study, because (along with certain passages from Apollonius Dyscolus) they constitute our main source for the so-called ‘orthodox’ approach to the problem of voice, the main gram- matical topic that Planudes addresses in the first two sections of the dialogue. In an early but still very influential article, Albert Rijksbaron outlined the treatment of voice, especially the middle voice, by Greek and Latin grammarians.24 As we know, the Greek middle voice occupies an intermediate space between the active and the passive—the passive “borrows all its forms, except the future and the aorist, from the middle,” and the middle “usually denotes that the subject acts on himself or for himself, as λούοµαι wash myself, ἀµύνοµαι defend myself (lit. ward off for myself ).”25 Rijksbaron ex- plains that this intermediate space occupied by the middle voice posed certain difficulties for ancient grammarians, difficulties which already from the early days of ancient scholarship resulted in a mixed and rather superficial approach to classifying the voices, one whose influence may be traced down to the Byzan- tine era. In the Τέχνη (ps.-)Dionysius Thrax identifies three voices: the active, the passive, and the middle. However, he does not define SER. II 27 (1958) 169–210, esp. 210, has made a case for questioning the authenticity. The discussion is considered to be open. For an overview of various problems related to the Τέχνη see S. Matthaios, “Aristarch, Dionysios Thrax und die Tekhne Grammatike,” in E. Karamalengou et al. (eds.), Ἀντιφίλησις. Studies of Classical, Byzantine and Modern Greek Literature and Culture in Honour of John-Theophanes A. Papademetriou (Stuttgart 2009) 386–400. Cf. S. Matthaios, Untersuchungen zur Grammatik Aristarchs: Texte und Interpretatione zur Wortartenlehre (Göttingen 1999) 18–20, and M. Callipo, Dionisio Trace e la tra- dizione grammaticale (Acireale 2011) 12–13. 24 A. Rijksbaron, “The Treatment of the Greek Middle Voice by the Ancient Grammarians,” in Form and Function in Greek Grammar: Linguistic Con- tributions to the Study of Greek Literature (Leiden 2018 [1986]) 357–369. Cf. also Matthaios, Untersuchungen 320–326. 25 H. W. Smyth and G. M. Messing, Greek Grammar2 (Cambridge [Mass.] 1984) 107, §356.a-b. 334 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 them in terms of morphology alone, but by means of a hybrid approach that considers both morphological and semantic char- acteristics. More specifically, Rijksbaron (361–365) identifies three different conceptions of the middle voice in (ps.-)Dionysius and his commentators (mainly Heliodorus), namely that it “com- prises verbs that either have active endings but passive meaning (πέπηγα [I am fixed]) or passive endings but active meaning (ἐποιη- σάµην [I made for myself ],” or, alternatively, that “it comprises verbs with passive endings only, that may have active as well as passive meaning (βιάζοµαι) [which may mean to be hard-pressed or overpowered as well as to overpower by force, press hard ],” or, even, that “it comprises verbs with passive endings only, of which a portion has an active meaning (ἐγραψάµην [I wrote for myself ]) and another has a passive meaning (ἐτριψάµην [I was worn out]).” These conceptions appear, in turn, to have exerted an in- fluence in one way or another over Apollonius Dyscolus and George Choeroboscus, as well as over the Latin grammarians Macrobius, Charisius, and Priscian, whether directly or in- directly. However, as demonstrated by Rijksbaron, in all these works the discussion of Greek voice involves a fairly “mechani- cal” approach that stands in contrast to the more comprehensive discussions of the concepts of “agency,” “patiency,” and “change” in the works of Aristotle, Simplicius, and Plotinus, who also occasionally touch upon topics related to Greek voice.26 Be that as it may, this “mechanical” mode of exposition is typical of ancient and medieval works on Greek grammar. It may be conceptualized as a “productive” mode, in the sense that it pro- gresses from the general to the particular. (Ps.-)Dionysius Thrax provides us with a good example that helps make clear how this mode of exposition works. The Τέχνη consists of twenty distinct sections that deal with various topics, beginning with a general definition of grammar, then continuing 26 Form and Function 365–366, where the following references are given: Arist. Ph. 253a10–20, 259b11–31; Plotinus Enn. 6.1.19–20; Simpl. In Cat. 8.296, 299, 300, 310, 318. THEOFANIS TSIAMPOKALOS 335 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 with aspects of phonology and morphology, before finally pass- ing to the parts of speech (noun, verb, participle, article, pro- noun, preposition, adverb, conjunction).27 It then proceeds to examine each of these parts separately. In each section, the text follows the same pattern. In the section dedicated to the verb, for instance, (ps.-)Dionysius first gives a short definition of what a verb is, then lists its eight accidents, namely moods, disposi- tions (i.e. voices), species, forms, numbers, tenses,28 persons, and conjugations, before finally dealing with each of these and cer- tain other characteristics separately. The overall progression from the general to the particular is striking, while, at the same time, extreme brevity and precision are the two basic virtues of the style of this tract. Usually the text predominantly makes use of lists, as is evident in the following lines—the above-mentioned definition of the middle is also in- cluded here (Gram.Gr. I.1 47–53):29 There are five Moods: Indicative, Imperative, Optative, Subjunc- tive, and Infinitive. There are three Dispositions: Activity, Passivity, and Mediality30 – Activity, as τύπτω (I strike); Passivity, as τύπτοµαι (I am struck); Mediality, marking partly activity and partly passivity, [as πέπηγα (I am fixed),31] διέφθορα (I waste), ἐποιησάµην (I became), ἐγραψάµην (I registered). […] There are three Tenses: Present, Past, Future. Of these, the Past has four sub- 27 For a more detailed overview of the structure of (ps.-)Dionysius’ text see Callipo, Dionisio Trace 17–19. 28 For (ps.-)Dionysius’ treatment of the tenses see especially Callipo, Dionisio Trace 188–192. 29 Transl. T. Davidson, “The Grammar of Dionysius Thrax,” The Journal of Speculative Philosophy 8 (1874) 326–339, at 335. 30 For more on the term διάθεσις denoting also the grammatical voice (cf. LSJ s.v. 3) see Matthaios, Untersuchungen 303–312. 31 This is a minor alteration I made to match the more recent Greek text of Uhlig’s edition, Gram.Gr. I.1 (Leipzig 1883) 5–101. Davidson’s translation has here “as πέποιθα (I trust),” because it is based on Bekker’s old edition (Anecdota Graeca II [Berlin 1816] 638). The reading πέπηγα I am fixed is at Uhlig 48. In this passage the text in the recent edition by Callipo, Dionisio Trace 74, 76, is entirely consistent with that of Uhlig. 336 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 species – Imperfect, Perfect, Pluperfect, and Aorist – which stand in three respective relations: the Present is related to the Im- perfect, the Perfect to the Pluperfect, and the Aorist to the Future. Choeroboscus’ commentary on Theodosius’ Κανόνες is not significantly different in this respect. The Κανόνες is an intro- ductory textbook for the study of Greek grammar containing inflectional paradigms for nouns and verbs, which was intended to replace (ps.-)Dionysius’ Τέχνη. Choeroboscus adopts the or- ganizational paradigm of his source, but adds a significant amount of further information. However, in terms of organiza- tion, this information displays the same pattern as in (ps.-)Dio- nysius. I will not quote the relevant passage here, as it is much longer than the one in the Τέχνη. Instead, I confine myself to mentioning that Choeroboscus’ treatment of the verb begins with a definition of the verb, which he then breaks down into its components to allow for individual commentary on each part. After that, he lists the eight accidents of the verb (Gram.Gr. IV.2 4.28–31) and then proceeds by commenting on each one of them separately (4.31 ff.). The only difference from the approach found in (ps.-)Dio- nysius’ Τέχνη is that Choeroboscus’ method typically exhibits a more pronounced explanatory tone. When introducing the tenses, for example, he does not simply list them, as the Τέχνη does, but instead explains that they relate to the flow of time and attempts to provide a clearer definition by means of the adverbs “yesterday,” “today,” and “tomorrow” (3.29–4.4). Later, when discussing tense as an attribute of the verb, he attempts to define each tense more systematically based on its temporal aspects and whether the action expressed is presented as continuous or repeated (11.23–12.36). Likewise, when dealing with the voices, he attempts to explain why the middle constitutes a separate category, since it includes verbs whose meanings could be con- sidered as better suited to either the active or the passive voice (4.4–18). That said, it should be emphasized that Choeroboscus does not elaborate at length on the explanations he provides. Once THEOFANIS TSIAMPOKALOS 337 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 he is able to offer a clear and succinct descriptive answer to a particular problem, he promptly moves on to the next point, an approach that sets him apart from Planudes. 2. Didactic innovation As already mentioned, Planudes’ Dialogue on Grammar em- bodies the text-type of the literary dialogue, the earliest examples of which can be found in the tradition of ancient philosophy. The very first topic addressed in the Dialogue is the tenses (3.1– 7.23). From there he moves on to the topic of voice (7.24–26), which in turn opens the way for the treatment of activity and passivity (7.27–12.19). After that, the focus shifts back to the tenses, particularly to the interconnections among four of them: the present, imperfect, perfect, and pluperfect (12.20–23). Palai- timos, the teacher in the Dialogue, suggests that these tenses are linked to each other in pairs in two different ways. First, the imperfect is related to the present in terms of time and aspect, in the same way as the pluperfect is related to the perfect (13.1–6), which is, in effect, the same classification as the one suggested already by (ps.-)Dionysius (see above). But then, there is also a second way of looking at things that Palaitimos points out, according to which the imperfect is related to the pluperfect and the perfect to the present (13.7–10). Throughout the remainder of the first (13.10–15.11) and the entirety of the second (15.12– 16.28) and the third sections (16.29–18.14), a total of five reasons are provided—with the fifth covering the whole second and third sections—to validate this alternative classification. In discussing these reasons, Palaitimos also touches on several other key sub- jects connected to verbs, such as augments, reduplications, and the sigmatic stems in the future and the aorist. While there are clear precedents in the earlier scholarly literature for most of the grammatical explanations that Pla- nudes presents to his readers in this part of his work, the arrangement is striking. More specifically, this unusual, selective arrangement can be better understood as “inductive” (in contrast to the “productive” arrangement described above) and seems to be closely tied to the dialogical form of the work, insofar 338 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 as new topics of discussion are constantly and spontaneously introduced into the text by the student’s requests for further clarification from the teacher. Not long after the beginning of the Dialogue, it is indeed Palaitimos, in his capacity as teacher, who emphatically encourages Neophron, the student, not to hesitate to interrupt and ask questions whenever he feels that he has not fully grasped the teacher’s explanations (5.14–32). Thus, the reader understands immediately that the structure will not be linear. At this point, one might also consider that, within the framework of a literary conversation between an in- structor and a pupil, the discourse is meant to advance—in order to maintain a sheen of authenticity—at a tempo and in a manner that aligns with the presumed cognitive abilities of an actual Greek learner. Following this rationale, readers of Planudes who were genuine Greek learners themselves could similarly acquire insights into the language that they might not have obtained had the arrangement followed the conventional “productive” ap- proach.32 In order to demonstrate this point, I will discuss two char- acteristic examples where the Dialogue offers insights that are notably distinct from what is found in other textbooks, such as those examined in the previous section. 2.1. Classifying the tenses through periphrastic constructions (pp.3.12–5.5 Bachmann) The very first question Neophron asks Palaitimos at the be- ginning of the text is: What differences can one discern between the various tenses, especially those referring to the past and the future respectively (3.1–3)? Palaitimos does not immediately proffer the usual explanation that a specific tense corresponds to a specific point in the flow of time and to the way in which the action expressed with each tense is presented as ongoing or re- peated. This is the approach one finds, say, in Choeroboscus (see again Gram.Gr. IV.2 11.23–12.36). Instead, Palaitimos first ex- plains how third-person singular forms of the verbs “to be” (εἰµί) 32 On this see also Papadoyannakis, in Greek Literature 100–101. THEOFANIS TSIAMPOKALOS 339 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 and “to become” (γίνοµαι) are constructed in various tenses: ἦν (imperfect), ἐστί (present), ἔσται (future); ἐγένετο (aorist), ἐγίνετο (imperfect), γίνεται (present), γενήσεται (future). Next, he takes a specific verb (πληροῦµαι) and derives from it both an adjective (πλήρης) and a present active participle (πληρῶν), in order to be able to construct a number of periphrastic forms that correspond semantically to the principal forms of the same verb. More specifically, Palaitimos uses the adjective to construct a number of forms that correspond to the middle-passive voice paradigm: πλήρης ἦν is thus linked with ἐπεπλήρωτο, the middle pluperfect; πλήρης ἐστί is linked with πεπλήρωται, the middle per- fect; πλήρης ἔσται with πληρωθήσεται, the passive future; πλήρης ἐγένετο with ἐπληρώθη, the passive aorist; πλήρης ἐγίνετο with ἐπληροῦτο, the middle imperfect; πλήρης γίνεται with πληροῦται, the middle present; and πλήρης γενήσεται with πεπληρώσεται, the middle future perfect. Next he uses the participle to construct those forms that correspond to the active-voice paradigm: πλη- ρῶν εἰµι is linked with πληρῶ, the active present; πληρῶν ἦν with ἐπλήρουν, the active imperfect; πεπληρωκώς εἰµι with πεπλήρωκα, the active perfect; πεπληρωκὼς ἦν is ἐπεπληρώκειν, the active pluperfect; πληρώσας εἰµί with ἐπλήρωσα, the active aorist; and πληρώσων εἰµί with πληρώσω, the active future. These are all periphrastic constructions, some of which involve the verb “to be” (εἰµί), while the participle potentially represents forms that are also attested in classical Greek texts as equivalents to finite forms of the corresponding verbs (see Cam- bridge Grammar of Classical Greek 52.51 with further references). Moreover, the middle-passive perfect in the third-person plural indicative occurs as a synthetic form only in certain cases, whereas, in all other cases, the expected form is a periphrastic construction (see Cambridge Grammar 19.8). Palaitimos does not, however, mention these constructions as actual equivalents. Rather, juxtaposing them with the cor- responding synthetic forms appears to serve an exclusively didactic function. Using only seven easily memorizable forms, namely the imperfect, the present and future forms of εἰµί, along 340 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 with the aorist, imperfect, present and future forms of γίνοµαι, which refer to the past, present, and future respectively, Palai- timos creates a paradigm that clearly and reliably demonstrates the semantic differences between the Greek tenses. This para- digm illustrates the position of actions along the timeline and indicates whether a given action is presented as ongoing or re- peated, which applies to both the active and the middle-passive voice. The need to convey a finite and concise set of information, structured in a way that allows the reader to derive from it a much larger range of insights—and in some cases even an infinite number of possibilities—is a concept that we find already in the ancient Fachliteratur. Both Plutarch and Pollux for instance, two authors whom Planudes certainly knew very well, mention —in works providing the reader with systematized sets of infor- mation—that their purpose in these texts is essentially to estab- lish paradigms, which, if used thoughtfully, can assist the reader in obtaining additional knowledge that may not have been ex- plicitly discussed in these works.33 What is new in Planudes’ treatment is that the forms he uses to explain his paradigm are all artificial. Perhaps there is a superficial influence from schedography in all this, since it is not the linguistic variation in high-register Greek that Palaitimos is most interested in, but rather establishing a paradigm that in- volves the pairing of standard forms with a set of variants which are to be assessed not in terms of correctness or stylistic elegance, but from the point of view of didactic efficiency. Schedodraphy was a method for the elementary teaching of Greek that emerged in the eleventh century and that focused primarily on 33 See Plut. Praec.ger.rei. 804C, περὶ µὲν οὖν τῆς τοῦ λόγου παρασκευῆς καὶ χρείας ἱκανὰ ταῦτα τῷ δυναµένῳ τὸ ἀκόλουθον προσεξευρίσκειν; Pollux Onom. 4.1, Ἰούλιος Πολυδεύκης Κοµµόδῳ Καίσαρι χαίρειν. οἶµαι καὶ σέ, εἰ καὶ νέος εἶ, πολλὰ προσεξευρήσειν οἷς ἔγραψα. Cf. Hippoc. VM 4; Auct. ad Her. 3.40; Diod. 5.74.6–75.1; Philo De aetern. mundi 16–17; Clem. Al. Strom. 7.16.103.6–7; Artem. Oneir. 4.65; Gal. De comp. medic. per genera XIII 503.5–6 K., De crisibus IX 739.7–13; Marcellin. De pulsibus 55–58; Lucian Salt. 61; Basil. Ep. 150.4. THEOFANIS TSIAMPOKALOS 341 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 orthography, making use of irregular words and phrases as variants on standard forms, albeit without placing significant emphasis on the quality of the examples or, at times, even their meaning.34 This method had been criticized by the leading scholars of the Comnenan era,35 although it subsequently en- joyed increasing acknowledgment and recognition, reaching its peak of popularity in the time of Manuel Moschopoulos, a for- mer pupil of Planudes, who also made use of it.36 As soon as Palaitimos concludes his explanation, Neophron, although initially displaying enthusiasm for this mode of ex- 34 On this characteristic of schedography see Pontani, in Companion 416. For further accounts of schedography see P. Agapitos, “Grammar, Genre and Patronage in the Twelfth Century: Redefining a Scientific Paradigm in the History of Byzantine Literature,” JÖB 64 (2014) 1–22, and “Learning to Read and Write a schedos,” in S. Efthymiadis et al. (eds.), Vers une poétique de Byzance: Hommage à Vassilis Katsaros (Paris 2015) 11–34; L. Silvano, “Schedo- grafia bizantina in Terra d’Otranto,” in A. Capone (ed.), Circolazione di testi e scambi culturali in Terra d’Otranto tra Tardoantico e Medioevo (Vatican City 2015) 121–167; Nousia, Byzantine Textbooks 49–92; I. Vassis, S. Kotzabassi, and I. Polemis, “A Byzantine Textbook of the Palaeologan Period. The Schedo- graphic Collection of MS Laurentianus 56.17,” Parekbolai 9 (2019) 33–182, at 35–46. 35 See Tzetz. Chil. 9 hist. 280.704–706; Anna Comn. 15.7.9 (485.19–23 K.-R.). Cf. P. Agapitos, “Anna Komnene and the Politics of Schedographic Training and Colloquial Discourse,” Nea Rhome 10 (2013) 89–107; Pontani, in Companion 384; P. Agapitos, “John Tzetzes and the Blemish Examiners: A Byzantine Teacher on Schedography, Everyday Language and Writerly Dis- position,” MEG 17 (2017) 1–57. That said, it should be emphasized that other influential scholars of the same period, such as Prodromus and Eustathius— and even Tzetzes himself—occasionally made use of schedography. On this see P. Agapitos, “Literary Haute Cuisine and its Dangers: Eustathius of Thessa- lonike on Schedography and Everyday Language,” DOP 69 (2015) 225–241. 36 On Moschopoulos’ schedography and its reception see F. Nousia, in F. Ciccolella et al. (eds.), Teachers, Students, and Schools of Greek in the Renaissance (Leiden 2017) 1–25, esp. 1–10. On Moschopoulos in general see e.g. Wilson, Scholars of Byzantium 244–247, and E. B. Fryde, The Early Palaeologan Renaissance (Leiden 2000) 295–298. On the date of his death see N. Gaul, “Moscho- poulos, Lopadiotes, Phrankopulos(?), Magistros, Staphidakes,” in E. Trapp et al., Lexicologica Byzantina (Bonn 2007) 163–196, at 169–171. 342 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 position, admits that it is complex and expresses the need for a simpler explanation (5.6–13). This reaction leads to the intro- duction of a second, more formal exposition of the system of Greek tenses delivered by Palaitimos (5.33–7.23), one which more closely resembles expositions found in more conventional textbooks (e.g. Choeroboscus). Does this imply that the former method of teaching based on the systematic use of purely tech- nical variants was inappropriate for the student? It is certainly a possibility. But even if this were the case, it would be important to emphasize that both approaches are present in the text, enabling Planudes’ readers (and students) to benefit from them. 2.2. The hidden Aristotle (pp.8.24–9.25 Bachmann) The second example I would draw attention to is found some lines later. In his preceding analysis, when using periphrastic forms to construct the middle-passive voice paradigm of πληρῶ/ πληροῦµαι, Palaitimos mentioned only the passive forms πληρω- θήσεται and ἐπληρώθη as the future and the aorist respectively. As stated above, the Greek passive voice borrows its forms from the middle voice, with the exception of two tenses, the future and the aorist, for which there are both middle and passive forms available. However, Palaitimos’ analysis does not recognize the “middle” voice as a distinct category; instead, it identifies only the “active” and “passive” voices—referred to (e.g. 4.30–31) as the categories of “active” (ἐνεργητικά) and “passive verbs” (πα- θητικὰ ῥήµατα). Within this framework, the category “passive verbs” subsumes all middle forms, except for those in the future and the aorist, since for these tenses there are distinct passive forms. By adhering to a terminology involving this strict dichoto- my between “active” and “passive” voices, Palaitimos neglects the middle forms for both the future and the aorist. Thus, as soon as he has finished with the subsequent, more traditional treatment of the tenses mentioned above, Neophron comes back to the middle-passive voice paradigm and asks specifically about the middle future and middle aorist forms (7.24–26). Palaitimos has not mentioned them, and he responds that there is no reason to pay any special attention to these two tenses, THEOFANIS TSIAMPOKALOS 343 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 since the middle future and the middle aorist have an active meaning (ἐνεργητικὴν σηµασίαν), even if their forms correspond to the “passive” paradigm (προφορᾶς παθητικῆς) (7.30–35). Ac- cording to Palaitimos, the main contribution of the “middle”— specifically referring to the middle future and the middle aorist, excluded from the “passive” voice paradigm—is to generate some additional nuances for the active, as is evident from the examples he then provides (7.35–36): ἀκούσοµαι I will hear carefully, θήσοµαι I will put for myself, ἠνεγκάµην I brought for my own use, and ἐδρεψάµην I plucked for myself. This is the sort of ex- planation that an author like Choeroboscus would have declared sufficient, before moving on to the next subject. This however is not the case with Planudes, and it is here that his text most obviously follows a different path from other text- books. Rijksbaron (n.24 above) points out that the “orthodox” approach of ancient and medieval scholars to the problem of the middle voice is quite “mechanical” and generally lacks the insightfulness of the discussions of the notions of “agency,” “patiency,” and “change” in Aristotle, Plotinus, and Simplicius (the latter commenting on Aristotle). But here is what we find in Planudes’ text (Dial. 8.24–9.3): P. So which of the two categories shall we say the word λούοντα [he who washes] falls under? N. Under the category of doing (ποιεῖν), dear Palaitimos. P. And what about the word λουόµενον [he who is being washed]? N. Under the category of being affected (πάσχειν). P. But is it not the case that he who washes himself is also being washed? N. Certainly. P. Is it, therefore, that the category of doing (ποιεῖν) the same thing as the category of being affected (πάσχειν)? N. How the same? P. Because both conditions can be seen as impacting the same man. N. So what do you suggest? P. Both doing (ποιεῖν) and being affected (πάσχειν), Neophron, are generated by the difference of persons (προσώπων ἑτερότητα). For the one who strikes will be said to be striking, if he strikes someone 344 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 else, hence he will fall under doing (ποιεῖν). But the one who is struck will be said to be struck, if he is being struck by someone else, hence he will fall under being affected (πάσχειν). And of course, as far as he who is washed is concerned, if he is washed by someone else, it is most obvious that he is subjected to something. The overall didactic and explanatory character of the discussion between Palaitimos and Neophron is obvious. The teacher tries to guide his student through the notions of “activity/passivity” by asking questions that will subsequently lead the latter to understand the classification of the voices that the former had earlier proposed. However, Palaitimos’ approach to the problem is neither as “mechanical” nor as “descriptive” as the one we find in, say, Choeroboscus (I am referring again to Gram.Gr. IV.1 4.4–18). What catches our eye here is the explanation that both activity and passivity are generated by “the difference of per- sons” (προσώπων ἑτερότης). This term is not attested as a technical term of grammar in ancient authors, but rather is found for the first time in Christological debates about the nature of the Trinity from late antiquity onwards.37 That said, in the present context Palaitimos uses this term to distinguish between “ac- tivity” and “passivity” on the basis of the person—or better the entity—towards which the action expressed each time by the verb is directed. Now, it is not simply that Palaitimos’ approach, which deals here with concepts related to “agency,” can be broadly com- pared to the discussion in the passages from Simplicius (com- menting on Aristotle) that Rijksbaron cites. Rather, Planudes’ text provides us with concrete evidence that he drew on these texts, evidence that may be found in the choice of words used in the passage immediately following the one quoted above (Dial. 9.8–25): N. No. For since the one who washes himself neither undergoes washing by another, nor does he wash someone else, how can the 37 E.g. Basil. Contra Sabell. 31 (PG 31 604–605); ACO I.1.6 36.15–18; Max. Conf. De duabus 2 Levrie; Leo VI Hom. 23.182–185 Antonopoulou; Psellus Theol. 111.60–61 Gautier. THEOFANIS TSIAMPOKALOS 345 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 one claim carry more weight than the other? It won’t make any difference whether we report that this man is doing something or that he is subjected to something. P. I leave aside everything that I ought to say now, which causes us not to accept that it makes no difference if, about the same thing, one time we say this and the next time we say the opposite. But let me ask you one short question: Which do you accept, that there is something naturally warm (φύσει θερµόν) or that there is not? N. The former. P. And are you able to accept that the same thing is naturally cold (φύσει ψυχρόν) too? N. By no means. P. But you would come to that conclusion judging from your pre- vious considerations. For it has happened that you have fallen into the same absurdity by bringing doing (ποιεῖν) and being affected (πάσχειν) together, since you say that it makes no difference which of the two opposites you would accept around the same thing. Despite Palaitimos’ explanation, Neophron apparently still has difficulties understanding what his teacher meant by “difference of persons” (9.8–12). Palaitimos therefore embarks on a second attempt to explain the phenomenon, this time by comparing the opposition “doing/being affected,” that is “activity/passivity,” with a further fundamental opposition, that between what is “naturally warm” (φύσει θερµόν) and what is “naturally cold” (φύσει ψυχρόν) (9.13–21). Now, this appears to be a variant of an argument we find in a very similar context already in Aristotle’s Categories,38 as well as in Simplicius’ commentary on the same passage. I quote here both passages in translation: Aristotle Cat. 11b1–4: Doing (τὸ ποιεῖν) and being-affected (πάσχειν) admit of contrariety and of a more and a less. For heating (θερµαίνειν) is contrary to cooling (ψύχειν), and being heated (θερµαίνεσθαι) to being cooled 38 This passage is followed in the text by an interpolation. For more on this see K. Oehler, Aristoteles. Kategorien2 (Berlin 2006) 133–134, 326. 346 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 (ψύχεσθαι), and being pleased to being pained.39 Simplicius In Arist. Cat. (Comm. in Arist. Graeca VIII.296.14–16): For heating (θερµαίνειν) and cooling (ψύχειν) are contrary, each being in the [category of] acting (ποιεῖν), and being heated (θερ- µαίνεσθαι) [contrary] to being cooled (ψύχεσθαι), each of these being in the [category of] undergoing (πάσχειν), just as are being pleased and being grieved.40 In these passages, the categories of doing (τὸ ποιεῖν) and being- affected (πάσχειν) are defined as contrary to each other by invoking the obvious contrarieties between heating (θερµαίνειν) and cooling (ψύχειν), on the one hand, and between being heated (θερµαίνεσθαι) and being cooled (ψύχεσθαι), on the other. These are, respectively, equivalent to the active and (medium-)passive voice forms of these two verbs. Planudes’ approach is far simpler, however. The message in Palaitimos’ argument is that as long as what is hot is not the same as what is cold, “doing” is not the same as “being-affected.” Hence, a given verbal form cannot fall at the same time under both the category of activity and that of passivity. Although Palaitimos does not need to define activity and passivity as con- trary to each other here—as in the passages from Aristotle and Simplicius—the occurrence of the simile cannot be explained in the present context unless one assumes a familiarity with the Aristotelian and/or Simplician intertext, which counts among the passages that (according to Rijksbaron) the ancient gram- marians did not use in their treatment of the problem of voice, although Planudes apparently did. 3. Conclusion First, some further points deserve emphasis in the broader research context concerning Planudes. The traditional limited scholarly interest in Maximus Planudes’ Dialogue on Grammar has 39 Transl. J. L. Ackrill, Aristotle. Categories and De Interpretatione (Oxford 1963) 31. 40 Transl. R. Gaskin, Simplicius: On Aristotle Categories 9–15 (London 2014) 20. THEOFANIS TSIAMPOKALOS 347 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 primarily been due to an old view that such late works are overly derivative of ancient literature. This perspective has often con- fined scholarly interest to instances where texts provide direct echoes of lost ancient works through quotations, paraphrases, allusions, and the like. Now, as we have seen, Planudes, in his engagement with earlier sources, does not faithfully reproduce their content nor does he directly cite them, revealing a more complex interaction with the transmitted material. At the same time, it is evident that his Dialogue, much like the numerous yet largely overlooked grammatical textbooks from the Palaeologan period, may serve as a very important testimony to teaching practices in the Middle Ages. However, precisely assessing the Dialogue’s impact on sub- sequent scholars and teachers of Greek, as well as its effective- ness, is still challenging due to the absence of direct quotations or clear evidence (e.g. marginal notes) of critical engagement with Planudes’ grammatical exposition in later works—a signifi- cant gap that ongoing research would hope to fill. Despite these obstacles, the survival of Planudes’ Dialogue in at least 49 manu- scripts, coupled with the educational implications suggested by their content, significantly underscores its value as a didactic tool among late Byzantine grammarians. This evidence strongly im- plies that the Dialogue was a widely-used teaching resource, one that can now provide us with new insights, particularly into the development of dialogue as a teaching form in late Byzantium. Examination of the first section of Maximus Planudes’ Dialogue on Grammar, which deals mainly with tenses and the problem of voice, has shown not only that, in the Palaeologan period, ad- vanced knowledge of high-register Greek was an educational desideratum of the highest sort that attracted the attention of one of the leading scholars of the time, but also that the literary output produced to cover the needs related to this teaching did not blindly copy earlier works. This is not, of course, to deny that Planudes’ approach largely depends on earlier scholarship—the foregoing analysis of the passages dealing with the classification of tenses and the problem of the middle voice did not overlook 348 MAXIMUS PLANUDES’ DIALOGUE ON GRAMMAR ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 this aspect. Nevertheless, the analysis has also made clear that, at the same time, Planudes also adopted a critical stance towards older practices and tried to move forward by supplementing this tradition with further elements borrowed from other, non-con- ventional traditions. The literary form of the dialogue, a form which Planudes’ text in effect shares with the text-type of the grammatical Erotemata— the earliest example of which we find in Manuel Moschopoulos, a student of Planudes—seems to have played a significant role in all this. Of course, the question whether the Erotemata of Mos- chopoulos and, later on, Chrysoloras show a degree of sophisti- cation comparable to Planudes’ text needs to be examined separately. Be that as it may, Planudes’ case demonstrates that the dialogical form definitely allows for greater flexibility when it comes to the arrangement of information, so that one could abandon the traditional “productive” mode of grammatical exposition in favor of a more flexible, “inductive” mode, one which is perhaps less systematic and exhaustive in terms of the grammatical phenomena covered, but which could simul- taneously appear particularly useful to advanced students look- ing to acquire a deeper insight into all the terms, divisions, and categorizations that they had delved into up to that point. Thus, contrary to the established practice, as attested in the standard textbooks of Antiquity and the Middle Ages, in Planudes’ dia- logue the exposition proper can begin with the classification of the verbal tenses, and from that point onward the text moves from one particular topic to another, occasionally furnishing insights that are not at all common in other texts, such as the discussion about “activity/passivity” that draws on Aristotelian/ Simplician intertexts, or even using examples, as is the case with the periphrastic constructions, that are purely technical and therefore need not be acceptable from the point of view of syn- tactic correctness or stylistic elegance. Nonetheless, there is still much fundamental work to be done, in order to fully appreciate the sophistication of these textbooks, as well as to arrive at sound results. My choice in the present THEOFANIS TSIAMPOKALOS 349 ————— Greek, Roman, and Byzantine Studies 64 (2024) 326–349 paper to focus exclusively on the first section of Planudes’ Dialogue was not random after all. As said in the introductory remarks, one can assume relatively safely that it is representative of the work. That said, both the text of this first section, as well as the rest of the work, must first be constituted on the founda- tion of a collation of all available manuscripts, before we can embark on a further discussion of matters related to the arrange- ment of information. I only hope that with the present paper on Planudes’ Dialogue on Grammar, I have managed to provide the reader with a solid idea of what sort of new knowledge could be gained from a more systematic study of the surviving Byzantine grammatical textbooks.41 February, 2024 Universität Trier FB II - Klassische Philologie tsiampokalos@uni-trier.de 41 This article was composed as part of my postdoctoral research in the ERC-funded project MELA: The meaning of language. A digital grammar of the Greek taught at schools in Late Constantinople (https://www.mela. ugent.be/), hosted at Ghent University. An earlier version was presented at the OIKOS day on 26 May 2023 in Leusden. Thanks are due to the participants in the event for their valuable remarks during the discussion. Moreover, I am particularly indebted to Andrea Cuomo (Ghent) for ex- tensive assistance and feedback, as well as to Grigory Vorobyev (Ghent) for our extremely helpful discussions about the transmission of Planudes’ text. Thanks are also due to Stephanos Matthaios (Athens), Chiara Monaco (Ghent), Ugo Mondini (Oxford), and Christiane Schwind (Trier), who made valuable remarks on earlier drafts and provided me with literature. Finally, I extend my gratitude to the editorial board and the anonymous referee for their insightful comments, which significantly improved the previous version of the article.