Dio Cassius: the Manuscripts of "The Roman History"

Cassius Dio (or Dion Cassius as he is known in Greek) wrote his Roman History in 80 books in Greek, sometime in the early 3rd century under Severus or Caracalla, both of whom he knew.  Dio exerted no appreciable influence on his immediate successors in the field of Roman history. But among the Byzantines he became the standard authority on the subject, a circumstance to which we doubtless owe the preservation of such a large portion of his work.  Most of the remainder is extant in the 'condensed book' format, or 'epitome' so favoured by the Byzantine.

"About one third of Dio's History has come down to us intact. The extant portions are: 

(a) Books XXXIV-LX (in large part), contained in eleven Mss.; 
(b) Book LXXVIII with part of LXXIX (or XXXIX with part of LXXX according to Boissevain's division), preserved in a single Ms. ; 
(c) the Paris fragments describing events of the years 207-200 B.C., recovered from the binding of a Strabo Ms.

For our knowledge of the lost portions of Dio's work we have two kinds of sources: 

(1) Excerpts contained in various Byzantine collections, together with brief quotations made by lexicographers and grammarians; and 
(2) Epitomes by Zonaras and Xiphilinus, supplemented by occasional citations in other historical writers. 

The quotations of the first class may be supposed to give, as a rule, the very words of Dio, subject of course to necessary changes in phraseology at the beginning, and sometimes at the end, and to occasional omission elsewhere of portions unessential to the excerptor's purpose. These constitute the Fragments of our author in the strict sense of the term. 

The Epitomes, on the other hand, while they often repeat entire sentences of Dio verbatim, or nearly so (as may readily be seen by comparing extant portions of the histories with Zonaras or Xiphilinus), must, nevertheless, be regarded as essentially paraphrases." (Cary)

The account of the revolt of Boudicca, in book 62, is for instance only extant in the Epitome of Xiphilinus.

Earnest Cary's introduction, which discusses the Mss., is online, together with his English translation.  I have preferred data from Freyburger as more up to date, where available.

Books 34-60

There are 11 Mss. which contain books 34-60, or portions of this.  L and M are the main witnesses: V, P and A are useful where these are missing portions of the text.



Shelfmark & Notes

Date /


Venice: San Marco Codex Marcianus Graecus 395.  Containing Books XLIV, 35, 4-LX, 28, 3 ; but numerous leaves and even whole quaternions have been lost.  It came from Constantinople and was brought to Italy by Cardinal Bessarion.  He in turn bequeathed it to Venice in 1468.  

end of 9 - start of 10


Florence: Laurentian Library Codex Laurentianus 70, 8. Containing Books XXXVI, 18, 1-L, 6, 2.  This may be a copy of a portion of M made at a time when M was part of a then-complete set of the Roman History.  It only ever contains books 30-60, and is a little later than M in date.  However Boissevain believed that L and M were both copied from the same exemplar.



Rome: Vatican Codex Vaticanus Graecus 144.  Containing Books XXXVI-LIV.  Direct copy of L.  Dated the same day as the Union of Florence of the Greek and Roman churches organised by Bessarion.

5th July 1439


Paris: Bibliothèque Nationale. Codex Parisinus Graecus 1689.  Containing Books XXXVI-LX. Used by Stephanus in his edition of 1548.



Florence: Laurentian Library Codex Laurentianus 70, 10. Containing Books XLII-LX.  Probably copied in Constantinople from L, M and a copy of Xiphilinus (see below).  This scribe went so far as to modify the text of Dio at the end of the battle of Actium (50, 35, 6) and replace it with that of Xiphilinus.


B Paris: Bibliothèque Nationale. Codex Parisinus Graecus 1689.  Possibly from the same scriptorium as A.  Contains also books 36-60. 15
C Venice: San Marco Codex Marcianus Graecus 396.  Probably a copy of B ordered by Bessarion at Constantinople.  Contains likewise books 36-60. 15
D Rome: Vatican Codex Vaticanus Graecus 993.  Copied from C. end of 15 - start of 16
T Turin Codex Taurinensis 76. Copied from C. end of 15 - start of 16
S Madrid: Escorial Codex Scorialensis Y-I-4.  Copied from C. end of 15 - start of 16
P Paris: Bibliothèque Nationale. Codex Parisinus Graecus 1690.  Copy of V. 16
Z Besancon  Codex Vesontinus 846 (once Z 68.80).  Copy of M for books 44-56. 16 (start)

"It has been conclusively shown by Boissevain that V is a copy of L, made, however, while L was in a completer state than at present; that A is in the main a copy of M, but with additions from L; and that P is derived from L for the earlier books and from A for the later. ...

"It is clear, therefore, that only L and M are of value except where passages now lost in one or both appear in the derived Mss. Thus V and P are our only Mss. for XXXVI, 1-17; V takes the place of L for the greater part of L-LIV; and similarly A serves instead of M for LII, 5, 2-20, 4; LX, 17, 7-20, 2, and LX, 22, 2-26, 2, being the sole Ms. to give the last two passages. Unfortunately M has several extensive gaps in books LV-LX which cannot be filled out from the later Mss." (Cary)

The tradition splits into groups: MVP, and ABCD.

Books 78-79

A single manuscript preserves this portion of the text:



Shelfmark & Notes

Date /

V Rome: Vatican Codex Vaticanus Graecus 1288, vellum Ms. of fifth or sixth century, in uncial characters. It teems with errors, many of which, however, were corrected by a second hand, apparently with the aid of another Ms. V' belonged to Fulvio Orsini, who published the contents in 1582 (Excerpta Ursiniana, pp. 416-47). 5/6

Paris Fragments

"These are found on five parchment leaves which have been used in patching up a Strabo Ms. (Parisinus 1397 A). They evidently belonged to a Ms. of Dio written about the eleventh century, and describe events of the years 207-200 B.C. (Frgs. 57, 53-60, 63-71, 76, 81, 83-86; 58, 1-6). Haase first published them in the Rheinisches Museum for 1839, pp. 445-76." (Cary)

Indirect Tradition

1.  The Excerpts

"The Excerpts De Virtutibus el Vitiis (V) are found in a Ms. of the tenth century, the Codex Peirescianus, now in the library of Tours. It was first published in 1634 by Henri de Valois, whence the fragments are sometimes called Excerpta Valesiana, as well as Peiresciana. The collection consists (at present) of quotations from fourteen historians, extending from Herodotus to Malalas. From Dio alone there are 415 excerpts, and the Ms. originally contained still more.

"The Excerpts De Sententiis (M) are contained in a Vatican palimpsest (Vaticanus Graecus 73) of the tenth or eleventh century. The Ms. is in very bad condition; numerous leaves were discarded and the others disarranged when the Ms. was used for the second writing. Angelo Mai, who first published the collection in 1826, employed chemical reagents to bring out the letters and even then had to despair of many passages. Since his use of the Ms. the letters have naturally faded still more, and parts of some leaves have been covered in the work of repair. The excerpts attributed to Dio are drawn from nearly all periods of Roman history, and fall into two groups, the first extending down to 216 B.C., the other from 40 B.C. to the reign of Constantine ; between the two portions several leaves, and probably entire quaternions, have been lost from the Ms. That the former set of fragments is taken from Dio none will deny. The later collection, however, extends much beyond the reign of Alexander Severus, where Dio ended his history; furthermore, the style and diction are considerably different from Dio's own. It is now generally agreed that all the excerpts of this second set were the work of one man, whom Boissevain, following Niebuhr, would identify with Petrus Patricius, a historian of the sixth century. Nevertheless, though not direct quotations from Dio, they are of value in filling out both his account and that of Xiphilinus.

"The Excerpts De Legationibus, Embassies (a) of Foreign Nations to the Romans (UG), and (b) of the Romans to Foreign Nations (UR), appear in nine Mss., all derived from a Spanish archetype (since destroyed by fire) owned by Juan Paez de Castro in the sixteenth century. First published by Fulvio Orsini in 1582, and hence called Excerpta Ursiniana.

"The three collections thus far named are known collectively as the Excerpta Constantiniana. They formed a small part of a great encyclopedia of more than fifty subjects, compiled under the direction of Constantine VII. Porphyrogennetus (A.D. 912-59). They have recently been reedited by Boissevain, de Boor, and Biittner-Wobst (Berlin, 1903-06).

"The Florilegium (Flor.) of Maximus the Confessor contains excerpts from various authors, arranged under seventy-one categories, the first of which is Virtue and Vice. Mai first published a number of fragments of Dio from this collection (from a Vatican Ms.), but inserted several which have since been rejected. There are at least six Mss. of the Florilegium containing excerpts from Dio. From one of these (Parisinus 1169, of the fourteenth or fifteenth century) Boissevain adds to the previous fragments No. 55, 3a and 3b.

"The Excerpla Planudea, a collection made by the monk Maximus Planudes (1260-1310) and published by Mai, have been shown by Boissevain and others to have no place among the fragments of Dio. A unique exception is the fragment at the beginning of Book XXI (Vol. ii, p. 370). 

"The short syntactical lexicon ( Περὶ Συντάξεως) published in Bekker's Anecdota Graeca (vol. i. pp. 117-180) contains nearly 140 brief citations from Dio, nearly all of which are assigned to their several books, though unfortunately many of the numbers have been corrupted. On the basis of these citations, compared with the epitomes, von Gutschmid and Boissevain independently attempted to determine the points of division between the lost books of Dio, and reached essentially the same results. Yet in several places the evidence is insufficient to constitute more than a reasonable probability.

"There are so few fragments from Books XXX-XXXV that Boissevain attempts no division within these limits. Between Books XI and XII the proper point of division is particularly uncertain; [Cary] differs from Boissevain.

"The lexicon of Suidas, the Etymologicum Magnum, and a few other compilations of like character are also useful in affording occasional citations from Dio, often by book-number." (Cary)

2.  The Epitome of John Zonaras

"Zonaras was private secretary to the emperor Alexis I. Comnenus in the early part of the twelfth century; later he retired to a monastery on Mt. Athos and devoted himself to literary labours. Among various works which he left is his 'Epitomh_ 'Istoriw~n, a history of the world, in eighteen books, extending from the creation down to the death of Alexis in 1118. It has been satisfactorily shown that for Books VII-IX, in which Roman history is carried down from the landing of Aeneas to 146 B.C., his chief source was Dio, supplemented by Plutarch and a couple of quotations from Herodotus: We are justified, therefore, in recognizing as an epitome of Dio whatever remains after the exclusion of the portions that are derivable from the other two sources. After narrating the destruction of Corinth Zonaras laments that he could find no ancient authorities for the remainder of the republican period ; hence it is inferred that Books XXII-XXXV had even then been lost from all the Mss. He resumes his narration with the time of Sulla, and after relying on various lives of Plutarch for a time, finally follows Dio's account once more, beginning with Book XLIV, 3 ; but for the period subsequent to Domitian's death he used Dio only indirectly, through the epitome of Xiphilinus. Zonaras is therefore of great importance for Books I-XXI, and to a lesser degree for Books XLIV-LXVII, where he occasionally supplements our Mss. of Dio or the epitome of Xiphilinus. There are numerous Mss. of Zonaras, five of which are cited by Boissevain..." (Cary)

[I have been unable to locate any details of the manuscripts, as I have no access to Boissevain or any critical text of Zonaras (if any exists)]

3.  The Epitome of John Xiphilinus

"For Books LXI-LXXX our chief authority is Xiphilinus, a monk of Constantinople, who made an abridgment of Books XXXVI-LXXX at the request of the emperor Michael VII. Ducas. (1071-78). Even in his time Books LXX and LXXI (Boissevain's division), containing the reign of Antoninus Pius and the first part of that of Marcus Aurelius, had already perished. He divided his epitome into sections each containing the life of one emperor, and thus is of no authority as regards Dio's divisions ; furthermore his task was very carelessly performed." (Cary)



Shelfmark & Notes

Date /

V Rome: Vatican Codex Vaticanus Graecus 145 15
C Paris: Bibliothèque Nationale. Codex Parisinus Coislinianus 320 15

"The epitome is found in at least sixteen Mss.; but all the rest are derived from one or the other of two fifteenth century Mss., Vaticanus 145 and Coislinianus 320. Besides these two (abbreviated V and C), we have readings from an unknown Xiphilinus Ms. entered in A of Dio to fill various gaps ; but the scribe of A dealt very freely with such passages." (Cary)

[I have been unable to obtain more precise information]

4.  John Tzetzes and others

"loannes Tzetzes (twelfth century) in his farrago of historical and mythological stories now entitled Chiliads, from the arbitrary division of the work into sections of one thousand verses each, occasionally cites Dio among his various authorities. But he dealt very freely with his material, and it is often difficult to determine exactly how much of Dio underlies his version. The present text omits a few passages printed with some hesitation by Boissevain. Tzetzes also cites Dio a few times in his commentary on Lycophron's Alexandra. 

"Other writers who are similarly of use in supplementing the epitomes are Eustathius, archbishop of Thessalonica in the twelfth century, famous for his commentary on Homer; loannes Antiochenus [John of Antioch], a historian of the seventh century; loannes Damascenus [John Damascene], an ecclesiastical writer of the eighth century; loannes Laurentius Lydus [John the Lydian], of the sixth century, who wrote of the Magistrates of the Roman Republic, and Cedrenus, a historian of the eleventh century." (Cary)

Chapter titles, Summaries, Tables of Contents

There are summaries of the content consisting of numbers followed by a text at the start of each book.  In addition the consuls are listed.   However, these summaries cannot be authorial, as in one case (book 56, ch. 27) the compiler has misunderstood a faulty reading in the copy before him. 


E. CARY, Dio's Roman History ... in Nine Volumes, Loeb edition (1914 ff). Checked.
Marie-Laure FREYBURGER & Jean-Michel RODDAZ, Dion Cassius: Histoire Romaine. Livres 50 et 51. Paris: Belles-Lettres (1991).  Checked.

Constructive feedback is welcomed to Roger Pearse.

This page has been online since 25th October 2002

Return to the Manuscripts Pages             Return to Roger Pearse's Pages