Printing in Greek Before Aldus Manutius

The present paper examines the history, circulation and use of the earliest Greek books ever printed (1471-1488). In particular, it focuses on the publishing enterprises of Bonus Accursius in Milan, who issued the first complete set of books to learn Greek, and of Laonicus & Alexander, the first Greeks to actively engage with the art of printing, who operated out of Venice but clearly had a double readership in mind: Westerners and, for the first time, the Greek communities of Venice and elsewhere.

Greek printing, however, Aldus' editorial, technical and aesthetic contributions are surely outstanding, despite Proctor's disapproval of the "wiry thinness and nerveless imbecility" of his Greek types. 2 On this, however, I agree with Nicolas Barker: "his success created a European market for Greek texts. His own programme [...] converted the publishing of Greek from an occasional to a regular part of the book trade. In turn [...] this stabilized the shape of Greek letters, not only as printed but also as written". 3 For all these reasons, right and wrong, Manutius and his Greek production have been the object of so much examination, that little attention has been given to those who paved the way before him. This paper thus examines the production of Greek before Aldus, and, most importantly, its circulation and use. I wish I had the time here to address some typographical issues too, since this is a period of extreme technical and aesthetic experimentation, but this would require more space, and it will thus have to be discussed on a separate occasion.
It is useful to break down the pre-Aldine production of Greek books into two distinct periods. A first, early period, running from 1471 to 1488, and a second, intermediate period, running from 1488-89, when the first complete edition of Homer, wholly in Greek and without commentary (i.e. for advanced readers) was issued in Florence, to 2 Proctor,Printing of Greek,17. 3 Barker, Aldus Manutius, 2. 1496, when the magnificent experiment of the Greek press founded by Janus Lascaris and Franciscus de Alopa, also in Florence, came to an end. If until the publication of the Florence Homer the production of Greek books had consisted almost exclusively of texts clearly intended for beginners, from that moment the publication of advanced Greek literature without Latin translation became increasingly widespread. This intermediate period, and, with it, the experimental years of early Greek printing, was already declining when Aldus Manutius opened shop, in Venice, in 1495.

Pre-Aldine Editions: Early Period
In this paper I focus on the early period (1471-1488). The texts that were published in this period are ten, in nineteen distinct editions that survive today in just under five hundred copies. 4 All but one of these editions were produced in Italy, and more specifically in Northern Italy. Interestingly, though, it is still largely a non-Venetian production: only four editions were issued in Venice, with the remaining being produced in Milan (8), by far the capital of early Greek printing, Vicenza (3), Brescia (1), Parma (1) and Florence (1). Thematically, these nineteen editions can be divided into five categories: grammars (7), dictionaries (4), grammatical pamphlets (2), 5 liturgy (2) and literature (3). The chart below represents the breakdown of such categories, though it is essential to bear in mind that these texts all remained, primarily, educational tools.
4 At present, out of 494 known surviving copies, I have examined 149 in detail, while data has already been gathered on approximately 160 more copies, in the process of being inserted in Material Evidence in Incunabula (MEI).
5 Among grammars, I further distinguish these texts as grammatical pamphlets: very short texts that offer additional, advanced, and very limited information on technical aspects of the Greek grammar. The two most popular grammars, by Emanuel Chrysoloras and Constantinus Lascaris, were especially written in the 15th century by Byzantine scholars to help Westerners approach the Greek language. Erotemata means 'questions', and the term represents the way in which these texts were organised, that is, Question: 'How many declensions are there in Greek?'; Answer: 'In Greek there are ten declensions'; 8 Question: 'How many conjugations?' and so on, in a fashion not too dissimilar from today's 'frequently asked questions' or 'FAQ'. Chrysoloras' grammar was simpler and thus more popular, and by 1488 it had already been published at least four times. Lascaris' text was more advanced, and consequently it took a while longer to gain in popularity. 8 It is useful to bear in mind that, by opting for a genitive rather than a nominativebased system, Chrysoloras had already dropped the number of declensions down from the original fifty-six (!).

Dictionaries
Essential tools to approach the language were of course Greek to Latin dictionaries and vice versa and, since for obvious reasons these were not typical of the Byzantine tradition, they were especially produced with Western learners in mind. The first to take up this task, on behalf of Bonus Accursius, on whom more to follow, was the Carmelite Johannes Crastonus (c. 1415-after 1497) of Piacenza, unsung hero of early Greek editing, tirelessly active throughout the second half of the 15th century. The Latin to Greek dictionary, commonly referred to as Vocabulista, is of particular interest, since it shows how, during the Renaissance, being able to read Greek was clearly not sufficient: as a living language, learners were also expected to learn to write and speak in Greek. It is probably not by coincidence that the first edition of the Vocabulista (c. 1480), in particular, was printed in octavo, a rather unusual format, particularly for early Greek editions, suggesting a more portable use. Two small grammatical pamphlets were also published, but of rather different nature. The first, Saxolus Pratensis, on Greek accents, is occasionally found bound with Accursius' Vocabulista, suggesting that they were probably intended for joint circulation, again, as a portable aid to Greek writing. The second pamphlet is a basic explanation of the system of Greek verbs, particularly remarkable because it may be the only 15th century Greek book produced outside Northern Italy, and more specifically in Deventer, in the Netherlands. The few surviving copies are almost exclusively found in small cities of Northern and Western Germany, many of which had ties with the Hanseatic League. The copy now in Manchester JRUL, interestingly, shows early English provenance, probably from Lincolnshire, though this should not necessarily be taken as evidence of bulk trade. As already in Byzantium, the Psalter served a double function, liturgical and educational. Since the Latin version was often known by heart by anyone wishing to learn Greek, the Psalter was a perfect tool to practice on both Greek grammar and vocabulary. These are texts that were also used as educational tools, and indeed they were already part of the Byzantine curriculum. Aesop and the Batrachomyomachia had both been, for centuries, popular texts among beginners. On Theocritus and Hesiod, clearly more advanced, see below.

Printing Endeavours
Though initial experiments in the early period of Greek printing are fairly erratic, there are nonetheless at least two discernible attempts at producing sets. It does not take an expert to see that Accursius' plan was to create the first Greek 'reading list', inclusive of all the appropriate learning tools: one grammar (bilingual) and two dictionaries, accompanied by one elementary text, Aesop, 10 one intermediate text, the Psalterium (also bilingual), and, finally, Theocritus with Hesiod. This latter publication is of course the most ambitious, and clearly intended for advanced students, in that it is the only one without Latin translation, not to mention the complexity of the text. We know that Poliziano lectured on Theocritus in Florence in 1482, and used Accursius' edition as textbook. 11 Particularly interesting, though, is the fact that Theocritus' Idyllia are followed by Hesiod's Opera et Dies. 12 The current statistics at my disposal on this edition are unequivocal: 75% of copies show evidence of use, but half of these in Hesiod only, suggesting that readers were substantially less interested in Theocritus, or, possibly, that they found it too hard to read. The text of the Opera et Dies, incidentally, escapes easy categorisation: it is a poem that re- flects the austere pessimism of a Great-Depression-style Greek agrarian crisis of the 7th century B.C., but it also blends in a bizarre selection of recommendations dispensed by Hesiod to his brother -a brother whom, judging by the text, the poet clearly did not hold in much esteem. However that may be, interspersed between high advice on how to properly worship the gods and base guidance such as "μηδ᾽ ἄντ᾽ ἠελίου τετραμμένος ὀρθὸς ὀμιχεῖν" ('do not urinate while facing the sun'), 13 we find a number of references to the most various and obscure husbandry tools. I was very pleased to find, in at least two copies (Prague NL and Athens NL), hand-drawn explanatory sketches of such tools with Greek technical terminology. 14 13 Hesiodii Carmina, 90 (720).
14 Prague, National Library, XLII. E. 19 (MEI 02127813); Athens, National Library, ΕΦ. 12164 (MEI 02018625). The similarity of the drawings may suggest an interconnec- The only titles that do not make an appearance in this list are the three minor texts: the Batrachomyomachia, Saxolus' De accentibus, and the Deventer Coniugationes verborum graecorum, which of course was printed shortly after the end date of de Madiis' ledger. Conversely, all the major Greek titles published before 1488 feature in the ledger, though the identification of the exact edition is not always straightforward. Three can be identified with relative ease: Aesop, Theocritus, and the Psalterium, all from Bonus Accursius' Milanese set. The first two were printed only once, making the identification certain. The Psalterium was indeed published a second time on 15 November 1486 in Venice by Laonicus & Alexander, but by that date de Madiis had already sold 7 out of 10 copies, so it is most likely that the remaining three were also part of Accursius' print-run. Laonicus & Alexander are two fascinating characters. They set up the earliest press owned and run by Greeks, publishing two books, both in 1486, and with a most peculiar typeface. If the average number of sorts produced per early Greek type-case ranged between approximately 60-240, they managed to produce in excess of tion between the two copies. I recently came across another set of drawings in a copy of the 1495-96 Aldine edition of Hesiod.
1,300. 15 They printed the Batrachomyomachia and the Psalterium, though it appears that they intended to print more. According to Proctor, the Batrachomyomachia was probably "a trial or specimen, intended to test the type and the method of printing in two colours, preparatory to the issue of a series of service-books, beginning with the Psalter, for which red and black printing would be indispensable. Of this series, if it were in truth ever projected, only the Psalter was completed". 16 Incidentally, the fact that de Madiis did not sell a single copy of any edition of the Batrachomyomachia buttresses the impression that Laonicus & Alexander relied on other channels to market their books. Aside from the predictable copies found in Italy and in large international collections, the fact that a concentration of them is today found in Southern Germany, and in small centres of the German Upper Rhine Valley in particular, 17 suggests that whoever sold their books probably had trading links in the area. Interestingly, however, some copies display unusual early provenances, including one from Portugal. 18 15 Needless to say, the number of punches and matrixes actually produced must have been somewhat smaller, and this staggeringly high number of sorts was obtained by means of excision of accents and breathings, a fairly common practice among early cutters of Greek types.

Trade
Initial considerations on trade and circulation can rely on approximately 30% of all known surviving copies, therefore these figures should be taken with a pinch of salt, but they are already quite revealing. The first element is that, of all the copies currently examined, only one bears unequivocal marks of early Milanese provenance, 19 a highly unexpected result, in light of the substantial role of Milan in the earliest production of Greek books, possibly ascribable to a looser practice of ownership inscriptions in Milan, or to the fact that these editions were primarily intended for export. The second element is, on the contrary, a remarkably high number of early provenances from Florence and Tuscany in general. A large number of early Greek editions, particularly those from Milan, bear evidence from this area, suggesting that there clearly was high demand and a strong trading link, not implausible in light of Accursius' background and teaching history in Tuscany.

Learning Greek: Annotations and Use
It is also important to understand how many copies show extensive signs of use, and how many, on the contrary, appear to have been abandoned halfway through. Bearing in mind the same overall figures as above, I've here intentionally left out editions that survive in too few copies, pamphlets, and dictionaries, since all of these categories may produce misleading results. This said, the average number of 'giver-uppers' is 42.5% among beginners' editions against 34% among Greek-only (i.e. advanced) editions. Vice versa, it is clear that readers of Greek-only books engaged substantially more with the text, leaving extensive annotations behind, often in Greek, than those reading texts for beginners, in either Latin or Greek. This information may of course have many interpretations, but it seems to suggest that Greek-only books were more likely used under the guidance of a teacher, whereas bilingual editions might also have been purchased by over-optimistic individuals. Another question worth trying to answer is whether these books were used in classrooms or not. It is clear that Greek teachers adopted the grammars that most suited their teaching needs, and indeed several of these were already available by 1470, at least in manuscript form. But it would be interesting to understand the extent to which their choices were then affected by the availability of printed grammars. With regards to these, there is scattered evidence, mostly in letters, that teachers would select and occasionally even provide schoolbooks for their pupils, though mostly at a slightly later date. 20 Although this investigation is still ongoing, the current impression is that the majority of copies of the earliest production of Greek books were not yet used in classrooms. Two elements point in this direction. The first is that the vast majority of de Madiis' sales are individual copies, there are no bulk sales and no apparent clusters of purchases, which however do occur with other titles listed in his ledger. The second is that the percentage of 'give-uppers', particularly among readers of grammars and the Psalters, is so remarkably high (almost 50%) that, had teachers been more frequently involved, one would expect, the number of quitters should have been somehow more contained.
20 In 1500, Daniele Clario, an Italian who taught Greek in Ragusa (Dubrovnik), ordered copies of the Aldine editions of Gaza and Lascaris for academic purposes; in 1501, Erasmus switched from Lascaris' to Gaza's grammar because cheap printed copies of the latter had become available on the market; upon his arrival in Paris, in 1508, Girolamo Aleandro wrote to Manutius that he had encouraged his students to use Manutius' Lascaris, but they preferred to use the 1507 Paris-printed Gourmont Chrysoloras. Botley, Learning Greek, 10 and 20-1. 14 (MEI: 02126528)