by Cyril Mango
In antiquity books were written on rolls of papyrus. The writing, usually on one side only, was in capital letters without division between words or accentuation (which in Greek can lead to confusion). For reasons of convenience each roll could not be unduly long: in the case of the Iliad, for example, no ‘book’ exceeded about 900 verses, corresponding to some thirty printed pages, so that the entire poem took up twenty-four numbered rolls. Papyrus, produced uniquely in Egypt, was relatively cheap and durable, but the ancient system had several disadvantages: it required too much space for storage, the individual rolls contained too little text and it was difficult to look up a particular passage.
Between the first and fourth centuries AD the roll was gradually replaced by the codex, i.e. the bound book as we still know it—one of the greatest advances in the entire history of letters. A codex could contain much more text than a roll, especially if its leaves were made of parchment, which could carry writing on both sides, so that now the whole Iliad could comfortably fit into one volume. It saved shelf space. It also made it much easier to locate a particular passage, which is probably the reason why the codex was favoured by Christians for their sacred books. To find the prescribed reading of the day all you had to do was to insert a marker. Besides, the Bible was an exceedingly long text: the Old Testament alone would have taken up several dozen rolls. On the other hand, the codex, especially if it was made of parchment, was decidedly expensive, a consideration that was to acquire increasing importance.
Greek majuscule book script as exemplified by the codex Sinaiticus (fourth century), one of the two oldest Greek manuscripts of the Bible. Here the beginning of St John’s Gospel.
The conquest of Egypt, first by the Persians, then by the Arabs, restricted, if it did not cut off entirely, the supply of papyrus, but another century and more was to elapse before steps were taken to tackle the problem of book production. At a date that cannot be determined exactly, but surely before the end of the eighth century, books began to be copied in minuscule: the earliest dated specimen that has survived, a product of the Studius scriptorium, is of the year 835. Minuscule cannot be described as an invention, since cursive writing had long been used for documents, but it had remained a specialized script, the preserve of trained notaries. It needed some adaptation for the production of books and the reading public had to get used to it. That said, minuscule offered obvious advantages. It was more compact and so required less parchment; it was also quicker to write, since the scribe did not have to lift up his pen after each letter. The process of converting to minuscule, called transliteration, was not purely mechanical, but was more akin to that of a new edition. Words had to be correctly divided and accented. Often punctuation was introduced. Besides, the model could well have been damaged or indistinct in places. In other words, the editor had to interpret the text. Down to about the middle of the tenth century minuscule and capital (uncial) scripts ran side by side. Thereafter the uncial all but disappeared, except in headings. Once transliteration had taken place, the uncial model, whether it was a scroll or a codex, tended to be discarded.
Not surprisingly, transliteration was spread over several decades, even centuries. There is some evidence to suggest that it proceeded thematically, starting with scientific and philosophical treatises, going on to orators and historians, and ending with poets. Now, if we try to imagine the total volume of Greek writing from the earliest times until Late Antiquity, there can be no doubt that it underwent progressive erosion: much more was in circulation in the fifth to sixth centuries AD than survived into the ninth, and of the latter group a good portion has since been lost. We cannot express this shrinkage even in approximate figures, but some examples are suggestive. Discoveries of Egyptian papyri in locations that were not major centres of learning have given us many hitherto unknown literary texts, the most famous case being that of the playwright Menander, the foremost exponent of the New Comedy, who fell into oblivion in the Byzantine period. John Stobaeus (fifth century) in his Anthology quotes about 385 authors, most of whom are mere names to us. Photios (of whom more presently) had the complete text of Stobaeus: we have only part. Roughly the same may be said of the vast geographical gazetteer called Ethnica by Stephanus of Byzantium (sixth century).
The earliest dated manuscript in Greek minuscule script is the Uspensky Gospels, St Petersburg, Public Library, cod. gr. 219 of AD 835. It was copied by the monk Nikolaos of the Studius monastery.
Even allowing for the possibility that authors like Stobaeus and Stephanus cribbed many of their references to obscure authors from older compendia, we cannot help asking ourselves where so many books could still be consulted in the age of Justinian. At Constantinople a public library of secular authors was set up at the Basilica by Constantius II (before 357). It burnt down in 476 with the loss (so it was alleged) of 120,000 volumes—not an impossible figure if those were mostly rolls. There is no clear evidence that the collection was reconstituted, nor do we know where an obsessively pedantic antiquarian like John Lydus (also of Justinian’s time) found all the books he quotes, although on one occasion he tells us that he picked up in Cyprus a copy of the ‘Judaean Sibyl’. A patriarchal library did exist, but was probably limited to religious literature and (as we happen to know for the eighth century) was not too well stocked. A palace library is briefly mentioned in the ninth century as containing a book of prophecies. It was expanded by Constantine VII, but filled no more space than the mezzanine of one small building. When, in 814, the emperor Leo V ordered a bibliographic search to be made of patristic testimonia favourable to Iconoclasm, the commission he appointed for the purpose sat in the palace (perhaps for reasons of secrecy), but books had to be fetched from various monasteries and churches. The impression one gets is that old manuscripts had come to rest in a variety of places and that a diligent search had to be made for them before the process of transliteration could be undertaken. Some of them may even have been found in Arab-held territory. Hunain ibn-Ishaq (ninth century), the famous translator of scientific and philosophical texts from Greek into Arabic and Syriac, travelled to Alexandria, Damascus, and Harran in search of Greek manuscripts.
Photios, twice patriarch of Constantinople (858–67, 877–86), is our next witness. Under circumstances that remain disputed he compiled his reading notes covering 280 volumes corresponding to 386 works (since one volume may have contained several different works). Conventionally known as the Bibliotheca, these notes range from a few lines to several dozen pages per book and usually include a summary of its content and a stylistic judgement. The majority of the works are Christian (233) as against 147 pagan or secular ones. School textbooks are excluded as are poetry and drama. Otherwise, there is the greatest diversity. Characteristically, the great majority of books reviewed belong to Late Antiquity, with the subsequent period very sparsely represented.
A little over half the books read by Photios are either totally lost or preserved only in fragments, which suggests that he consulted them in old copies and that the works in question (or most of them) were not considered of sufficient interest to be transliterated. To our great regret, he does not inform us where he had found them. If they formed his private collection, he must have owned a far bigger library than is recorded for any other individual of the Byzantine period. To add to the mystery, not a single book bearing his ex-libris has survived, whereas in the case of his disciple Arethas of Caesarea (c.860 —c.939) we still possess six manuscripts that he caused to be copied and can attribute to his library another twenty-five with greater or lesser probability.
The next stage of the Byzantine revival corresponds to the reign of Constantine VII (912–59) and has been labelled ‘the age of encyclopaedism’, although the age of compilation may be a more accurate description. Bookish by nature, like his father, Constantine was excluded from power by the usurper Ro-manos Lekapenos (‘an ignorant fellow’) and had plenty of leisure to devote to his studies. Once back in control, if not before, he set about instigating a series of
bibliographic enterprises that were carried out by a team of collaborators. For this historians must be grateful: if we did not have the Book of Ceremonies, the De administrando imperio (a treatise on the empire’s foreign relations), the De themat-ibus (a survey of the empire’s provinces), we would know much less than we do about Byzantium. The above three titles, practical in intention, but largely antiquarian in content, are all concerned with the exercise of the imperial office. Others are of a miscellaneous character: the Geoponica is a collection of late antique treatises on agriculture; the Hippiatrica a similar compilation concerning veterinary science. Constantine also seems to have commissioned a medical encyclopaedia and another on zoology. But by far the biggest of his projects was the one conventionally called Excerpta historica. This was an anthology of extracts culled from a wide selection of historians ranging in date from Herodotus to George the Monk (ninth century), arranged thematically under headings such as ‘The proclamation of emperors’, ‘Victory’, ‘Public speeches’, ‘Hunting’, ‘Marriages’, ‘Inventions’, etc. Of the fifty-three named sections only one thirty-sixth has been preserved and fills six printed volumes, so that the entire work would have been the equivalent of over 200. The only possible beneficiary of such a monstrous collection would have been the emperor himself, whom we can imagine looking up precedents for this or that, be it conspiracies or embassies or deeds of valour.
Minuscule copy of Euclid, written in AD 888 for Arethas of Patras (later archbishop of Caesarea in Cappadocia), who paid 14 gold pieces for it.
Christ crowning the emperor Constantine VII Porphyrogenitus. Ivory plaque.
Facing: Moses receiving the tablets of the Law on Mount Sinai. Miniature reproducing an original of the fifth or sixth century in the Bible of the patrician Leo of the first half of the tenth century.
Constantine’s mania for systematization also extended to the religious sphere. We know for certain that he commissioned an extensive hagiographical calendar known as the Synaxarion, consisting of potted biographies of the saints commemorated each day, and may have been the originator (though that is far from certain) of an even vaster collection of saints’ Lives in extenso (148 in all), paraphrased into acceptably elegant Greek by Symeon the Logothete, known as the Metaphrast in remembrance of his efforts. Finally, there is the nearest Byzantine equivalent to an encyclopaedia, mysteriously called Souda (literally ‘ditch’), which consists of about 30,000 entries arranged alphabetically: difficult words, historical and literary notices, proverbs, mostly confined to antiquity. This appears to date from the end of the tenth century and is certainly in Constantine’s spirit, although there is no evidence linking him directly to it.
What are we to make of the cultural activity that Constantine inspired? The absence of legal science is explained by the fact that an ‘epuration’ of the law, i.e. of Justinian’s Corpus in Greek translation, had already been carried out, starting in the reign of Basil I and completed under Leo VI: this was the Basilica in 60 books, more of an antiquarian than a practical compendium. Military science had also been tackled by Leo VI. For the rest, one has the impression that Constantine was trying to collect and boil down all useful knowledge, most of which dated to the period of Late Antiquity. His hagiographic programme was destined for a wide diffusion, but his secular programme could not have been meant to reach much beyond the palace. It was to serve as a vast reference collection for future emperors and their courtiers.
Byzantium did not experience another dark age after Constantine Porphyrogenitus. Whilst not all the texts that were available to him have survived, the bulk of what was transliterated and excerpted in the ninth and tenth centuries continued to be copied (if only occasionally) and, therefore, to be read. The Byzantine revival not only salvaged many of the ancient classics, but also—and this needs to be stressed—determined the content of the surviving corpus, which is the product of Byzantine selection. Old texts were not copied at random; they were copied because they were considered useful either for practical ends (e.g. medicine, agriculture, military science, astrology) or as storehouses of curious knowledge or as exemplars of literary style. Besides, there was a financial factor. Arethas, who sometimes noted down the sums he expended on manuscripts he commissioned, informs us that his Euclid (now at Oxford) cost 14 gold pieces (nomismata), his Plato (471 folios) 13 for copying and 8 for parchment, his corpus of Christian Apologists (now in Paris) 20 for copying and 6 for parchment. These were considerable sums, beyond the reach of all but the richest individuals. By way of comparison, the yearly salary of a medium-ranking court dignitary was 72 nomismata. In today’s prices (to the extent a comparison may be made) Arethas’ Plato cost something like £5,000. Institutional sponsorship does not appear to be recorded.
Considering the constraints, it is remarkable how much was salvaged, and we can hardly complain that we would have made a different selection. Contrary to our tastes, the Byzantines had least interest in drama, more in prose than in poetry, and most of all in rhetoric: hence the endless volumes that have few readers today of Aelius Aristides (represented by over 200 manuscripts), Lucian, Himerius, Libanius, Themistius, Choricius and many more of that ilk. Rhetoric prevailed even over religious odium, which is why we still have the Orations and Letters of Julian the Apostate, except, of course, his Contra Galilaeos. Surprisingly, the earliest Christian authors attracted little attention, perhaps because they were considered outdated and some were doctrinally suspect. The textual tradition of Justin Martyr, Athena-goras, Tatian, and Clement of Alexandria goes back to a single manuscript commissioned by Arethas in 914. The Epistle to Diognetus was preserved in one late manuscript, now destroyed. The Shepherd of Hermas, one of the oldest pieces of Christian writing, widely read in Late Antiquity and present in the famous Codex Sinaiticus of the Bible (fourth century), is found in only one other Greek manuscript as late as the fifteenth century.
Facing: Busts of the Minor Prophets in a luxurious manuscript now at Turin, Biblioteca Nazionale, cod. B.I.2 of the latter part of the tenth century. A damaged subscription of the year 535 provides the date of the original from which the Turin manuscript was copied.
Below: Tenth-century copy of a medical treatise on the setting of bones by Apollonius of Citium (first century BC). The human figures probably reproduce an ancient original, but have been incongruously placed within ornate arches.
One of the very few secular products of the ‘Macedonian Renaissance’ is the illustrated copy of Nicander’s Theriaca, a learned poem of the second century BC on snakes and other poisonous animals.
The criterion of utility may be tested by a somewhat unusual example. What we call the Greek Anthology is a vast collection of about 4,000 verse epigrams ranging in date from the sixth century BC to the tenth AD. Originally an epigram was meant to be inscribed on stone, but it outgrew its in-scriptional function to embrace such topics as love, drinking, and jests. It was particularly cultivated in the Hellenistic period, when the first anthologies of epigrams began to be compiled. Their production continued until the sixth century AD : the last anthology before the Dark Age, the Cycle of Agathias, published soon after 565, brought together the compositions of some twenty-five contemporary authors, most of whom were lawyers and civil servants. George of Pisidia, a notable poet, wrote epigrams in the reign of Heraclius, and then the genre died. It reappeared in the early ninth century, when a group of iconoclastic epigrams was inscribed at the entrance of the imperial palace as a kind of manifesto. Theodore the Studite took the trouble of refuting them in verse and himself composed over one hundred other epigrams on religious topics. From then on epigrams continued to be written by Byzantine litterati, very seldom in elegiacs (which is the proper metre for epigrams), usually in twelve-syllable iambics that are better attuned to the medieval pronunciation of Greek.
The Greek Anthology originated in the milieu of the palace school and was first compiled by a clergyman called Constantine Kephalas, attested in 917. His manuscript has not survived, but we are fortunate in
having one that is not much later, the Palatinus 23 (at Heidelberg), usually dated 930-50. It is divided thematically into fifteen books, opening with ‘Christian epigrams’, perhaps to reassure the public, but going on to amatory (Book 5) and even pederastic ones (Book 12). Their imagery is predominantly pagan.
The Anthology has seldom been considered from a Byzantine viewpoint, although it is a Byzantine creation that raises some interesting questions. How is it that Kephalas and his small group of collaborators were able to lay their hands on such a mass of old epigrams? We are told that one member of the team, Gregory by name, copied verse inscriptions from stones in Greece and Asia Minor, and there are several Christian epigrams from Constantinople that were copied directly from monuments. The bulk of the poems, however, must have come from manuscripts. That means that Kephalas had access not only to the Cycle of Agathias, but to a series of earlier anthologies, going back to the Garland by Meleager of Gadara (first century BC). Furthermore, why did he go to so much trouble? The only plausible answer is that he did so to provide models for imitation. But why, we may ask, the sexual epigrams? Perhaps the good Kephalas enjoyed them or was loath to leave out any curious remnants of antiquity that came his way. In any case he deserves our gratitude even if, in the event, the elegiac epigram had a limited future in the Byzantine world.
