PG 88 Κοσμάς Cosmas Indicopleustes, Christian Topography. Preface to the online edition. This text was transcribed by Roger Pearse, Ipswich, UK, 2003. All material on this page is in the public domain - copy freely. Greek text is rendered using the Scholars Press SPIonic font, free from here.
Cosmas Indicopleustes ('India-voyager') of Alexandria was a Greek sailor in the early 6th century who travelled to Ethiopia, India and Sri Lanka. He then became a monk, probably of Nestorian tendencies, and around 550 AD wrote a strange book, copiously illustrated, which is the text presented here.
There can be few books which have attracted more derision, mixed with wonder, than the Christian Topography of Cosmas Indicopleustes. It advances the idea that the world is flat, and that the heavens form the shape of a box with a curved lid. The author cites passages of scripture which he distorts wildly in order to support his thesis, and attempts to argue down the idea of a spherical earth by stigmatizing it as 'pagan.' The approach to scripture is discreditable, and the conclusion made simply wrong.
The book is often cited as evidence that Christianity introduced the idea of the flat-earth into the world, and brought in the age of ignorance. This is hardly fair, since Cosmas does not represent a mainstream of any kind, personally or spiritually. The latter pages of his work are devoted to rebutting the criticism of his fellow-monks, that what he was saying was wrong.
As far as we can judge from the surviving literature, Christians and pagans did not as such hold different views about the shape of the world. Some of the philosophers had supposed a spherical earth, and even calculated its size, and some Christians followed them, particularly if they were educated men. Other philosphers had derided the idea, and some Christians did likewise, such as Lactantius and Cosmas. Naturally Christian writers of this school turned to scripture to illustrate their theme, but their methods of exegesis attracted severe criticism from other Christian authors such as Photius. In short, it was a subject on which there was no certain knowledge in the ancient world. In the absence of the Victorian culture of science which we enjoy today, there was also no means for a proto-scientist to publish his discovery in such a way that it would be clear to all that he was not simply a crank idly speculating. It seems unreasonable to condemn Cosmas for reflecting the era in which he lived, when our incredulity reflects only the better-informed era in which we were wise enough to be born.
The book is not without value, however. 'Indicopleustes' means 'Indian voyager'. We learn from stray scraps in classical literature that there was some trade between the Roman empire and India. But Cosmas was one of the rare souls who had actually made the journey. Indeed we learn from his book that he had travelled over much of the Red Sea coast, and as far as Ceylon (modern Sri Lanka), and he describes some of what he saw, and even drew pictures of the strange animals in his autograph manuscript. Some of these have been copied into the existing manuscripts. Away from his daft theory, Cosmas proves to be an interesting and reliable guide. He happened to be in Ethiopia at the time when the King of Axum was preparing a military expedition to attack Jewish Arabs in the Yemen. He records now vanished inscriptions. In short, he gives us a window into a fascinating world of which we would otherwise know nothing. This is the main value of his work.
The work was originally in 5 books. Objections led to him adding first book 6 and then further books up to book 10. Books 11 and 12 seem to have no connection with the main portion of the work, and may have been added by a later editor from other works by Cosmas, such as his work on Geography addressed to a certain Constantinus. This includes his description of the island of Taprobane, Ceylon or Sri Lanka. The Vatican manuscript was copied from a text of only 10 books.
Cosmas' own name is not absolutely certain. Two of the three Mss. give no name for the author -- only the Laurentian Ms. names him. However a portion of book 5 appears in the marginal commentaries ('catenae') on the Psalms, giving the name Cosmas Indicopleustes as the author.
Cosmas tells us that he was a native of Egypt, probably of Alexandria. He never received a complete education (II, 1). He was a merchant (II, 54 and 56) in early life, perhaps importing spices. He made many voyages. He knew Palestine and the area around Mt. Sinai (V, 8, 14, 51-52), had been to Socotra (III, 65), and had navigated in the Mediterranean, Red Sea and Persian Gulf (II, 29). He had rounded cape Gardafui and sailed off Somalia (II, 30). Later in life he settled in Alexandria, developed indigestion, ophthalmia and other ills (II, 1). That he was a monk is supposed, as the Laurentian manuscript calls him Kosmaj monaxoj, and indeed it is likely enough. Cosmas even mentions in book 2 another merchant, Menas, a friend of his, who also became a monk.
Cosmas is often referred to in the literature as a Nestorian. He tells us that he was a pupil of Patricius, Theodore of Mopsuestia and Diodorus of Tarsus, and a friend of Thomas of Edessa. All of these were Nestorians. He highlights the churches planted in the East, all again Nestorian. One passage only gives the opposite impression, that where he uses the Chalcedonian term theotokos, Mother of God, for Mary; but this passage is not found in the Vatican manuscript, suggesting it is a later addition.
Interestingly he refers to Marcionites and Montanists in book 5, which suggests that these groups were still active in his part of the world at this time.
Date of the work
The date of the work is fairly certain. In book 2, Cosmas tells us that it is 25 years since he was in Axum, and he was there when Elesbaas was preparing his expedition against the Homerites. That expedition probably took place in 525 AD, or possibly 522 AD. At the beginning of book 6, he refers to two eclipses, giving the dates as Mechir 12 and Mesori 24: these would seem to be the eclipses of 6 Feb. 547 and 17 Aug. 547. The logical inference is that the work was written around 550 AD.
McCrindle's translation is given in the following pages. This was based on the only available Greek text, that of the great Montfaucon, which Migne had reprinted. Nevertheless this was a pre-critical text. In 1909 E.O.Winstedt published a critical edition of the text, without a translation but with English preface and notes. I considered adding his preface to this collection also, but it is difficult to scan and not really relevant in large portions. I have drawn on it for various points in this preface, and also made use of the preface to Wanda Wolska-Conus edition, Sources Chrétiennes 141 (1968).
Montfaucon's text was primarily based on the Laurentian manuscript (L). Those who have read his Diarium Italicum will appreciate the difficulties this venerable scholar, the father of bibliography, faced in obtaining access to manuscripts in Italy. Winstedt suggests that his text does not follow the manuscript very closely. One serious error occurs in book 5 (248B-249B), where a leaf is missing in L. Montfaucon for whatever reason only restored a small part of this text from V. Unfortunately this passage is the one used in the catenae, and the difference has led to confusion over whether Cosmas wrote some other work containing the words in the catena. All of them are in fact from the Christian Topography.
Winstedt also included monochrome plates of some of the illustrations from the Vatican manuscript (V). I have added these plates as an appendix, after some hesitation, since the originals were probably made by the author. All the manuscripts are magnificently illustrated. Those in the Vatican Ms. are probably close copies of an antique original. It is likely that Cosmas was his own artist, since some of them show subjects unfamiliar to the average monastic copyist; at any rate, he must have directed the execution. The first picture in V (f.12) has no connection with the text, but shows an antelope between two banana trees. Native words for the banana, and the mention of the little known port of Adulis, Gabaza, in the picture on f.12v, all tend to suggest the involvement of the author.
I have added a translation of the summary of contents to the front of McCrindle's translation. Wolska-Conus describes McCrindle's version is being a little hasty, particularly in the theological portions, and a little prone to force the text in the direction of the droll and amusing in others.
Three complete manuscripts exist:
Rome: Vaticanus Graecus 699 (V). An uncial manuscript of the 9th century, written in Constantinople. It comprises 123 folios written in two columns, each of 32 lines except where miniatures appear. It contains only books 1-10, and as the index of books at the front is written in the same hand, and lists only 10, clearly it never had more. It omits the introductory prayer, first prologue, and start of the second prologue. Various leaves are missing. The manuscript is the best of them, but corrupt in point of copying accuracy. General production quality is top-class. Accents have been added by a subsequent corrector. The codex is illustrated 'magnificently'.
No. 1186 of the Greek Mss. of the Monastery of St. Catherine at Sinai (S). 11th century, 209 leaves, each of 30 lines, folio. Written in Cappadocia. Very correctly written and accented. It contained 12 books, but the end is lacking. Commences with the first prologue, then the summary of 12 books, then the second prologue starting where V does. At the end of the 5th book (i.e. the end of the original work), it has a prayer not found in L. L and S are generally very closely related. It contains practically the same illustrations as L.
Florence: Laurentianus Plutei IX.28 (L). 11th century, written in minscule hand. 12 books. Written probably in the Iviron monastery on Mt. Athos. 279 folios, of 22 lines. Omits the summary of contents. Differs from V at the point where the two both begin, with an untranslatable sentence. A leaf or more is missing after fol. 104 (Montfaucon 200D), which is absent also in V and S only partly contains the text, so a row of dots appears in the printed text. Since McCrindle had no access to S, at this point his text is incorrect. Illustrations are not so fine as those in V. Corrected throughout by another hand.
Phillips 2581. A paper copy of L, made 21 June 1682. 326 pages, in quarto, paper. This was at Cheltenham when Winstedt wrote, but the Phillips collection is now dispersed. Winstedt had not seen it, however, but relied on a description by H. Omont. Wolska-Conus simply quotes Winstedt.
Other partial manuscripts also exist, mainly containing illustrations:
Paris Suppl. Gr. 844. 18th century. Contains only copies of some of the pictures in L.
Paris Gr. 2426 (P). 16th century. On ff. 112 ff, contains a copy of most of book 11, copied by Nicholas de la Torre, possibly from the archetype of Z although it contains more of this book than Z did. The text is handled freely, and seems to relate to the Smyrna manuscript.
Smyrna B-8 (Z). ca. 1100AD. Described by Papadopoulos-Kerameus in an 1877 catalogue. Selections appear on pp.156-192, under the name of Maximus (written over a shorter name which has been erased). Just a collection of pictures with short bits of text attached to it.
Vienna Theol. 9 (W). Selections. Bought in Constantinople by A. Busbeck. Copied from S, or more likely an Ms. similar to S.
Finally a bunch of Mss. of the Psalter and Gospels are listed by Winstedt, which prefix extracts from book 5 as catenae. These he describes as all worthless, but he lists a few:
Vat. Gr. 363 (R2). 10th century.
Oxford, Bodleian Library: Ms. Cromwell 15. 11th century. Bought on Mt. Athos in 1727.
Bodleian Arch. Selden 29. AD 1338. Fol. 116 has a catena on Luke, ascribed to Cosmas Indicopleustes.
There are also numerous Mss. which use a section from book 5 as a catena on the Psalms, again of limited value:
Vat. Gr. 342. 12th century. f. 7v.
Vat. Gr. 525. 12th century. f. 1.
Venice, Marcianus Gr. 498. 14th century. f. 270.
Bodleian, Baroc. 15. 12th century. f. 22.
Turin B. I. 10.
Milan, Ambrosian. B. 106. 10th century.
Moscow 358. 11th century.
Vat. Gr. 1747.
Paris Gr. 2743. Once Colbertinus 1476, 16th century, copied by J. Diassorinos. This is mentioned by Montfaucon, and also by Omont, both of whom lead the reader to suppose Cosmas was the author of a commentary on the Psalms preserved herein. In fact it contains only the usual chunk of book 5, followed by material from other authors.
Paris Gr. 169 (Mazarin-Reg. 3450). 14th century. A similar Ms., with the paragraph expanded by adding a following section from other authors.
Vallicellianus C. 4. 16th century. ff. 434-5. Also with the expanded paragraph from book 5.
Paris Gr. 3179. 16th century. Copied by Bigot. Also with the expanded paragraph from book 5.
Vat. Gr. 711. fol. 196.
The three main Mss. fall into two families: V and the others. V has 10 books; S and L have 12. In V the paragraphs on the major and minor prophets are in Septuagint order, as aso in the parallel passage in the Chronicon Paschale; in S and L, they are disarranged into the order of Theodore of Mopsuestia, as in McCrindle's translation. Winstedt suggests this shows that Cosmas' literary executor added two books from his papers, and 'corrected' the order into that favoured among Nestorians, and perhaps added other notes at various points, found only in L and S. Cosmas wrote his own notes, so supplementary paragraphs cannot be distinguished from the original with ease. S cannot be a copy of L, and is more reliable as L has been corrected by an educated man who corrected Cosmas' bad Greek at various points. V is undoubtedly the best manuscript.