Eusebius: the Manuscripts of the "Demonstratio Evangelica"

The Demonstratio Evangelica or Proof of the Gospel, originally in 20 books, followed on from his Praeparatio Evangelica and continued his lengthy apology for the Christians with substantial quotes from scripture and other authors. Books 1-10 and a fragment of book 15 have survived; the rest is lost.  The passage most often referred to in modern times is the citation of the so-called Testimonium Flavianum of Josephus from Antiquities 18, in book 3.  The work was probably written between 314-318.

These excerpts from FERRAR, then some details from the GCS. 

§ 8. MSS., ETC.

The earliest MS. of the Demonstratio is the Codex known as the Medicean or "Parisinus 469," of the twelfth century, registered in the Catalogue of the Library of Paris, vol. ii. p. 65. It is deficient at the beginning and end, beginning with the words h9 paidi/skh soi, p. 17, and ending at th~j swth~roj h9mw~n parakeleu&sewj, p. 688. These deficiencies were supplied by J. A. Fabricius in his Delectus argumentorum et syllabus scriptorum, qui veritatem religionis Christiana: adversos atheos . . . asseruerunt, who used a copy that had been made by Stephen Bergler, at Hamburg, in 1725, from a MS. in the possession of Nicholas Mavrocordato, Prince of Wallachia, who collected many Greek MSS. from Mount Athos and other monasteries. The MS. was unfortunately lost at the death of the Prince. Bergler gave no information about its age or condition. It was almost certain that it was either derived from Parisinus 469 before its mutilation, or from a MS. of the same family.

There are four other MSS. of the Demonstratio at Paris, parchments of the sixteenth century numbered 470, 471, 472 and 473 in the Catalogue, vol. ii. pp. 65, 66. And there is at St. John's College, Oxford, a parchment MS. of the fifteenth century (No. 41 in the Catalogue of O. Coxius, p. 12). As all these have the same deficiencies, there is little doubt that they come from the common source, Parisinus 469.

There is a sixth MS. in the Ambrosian Library, at Milan, of the fifteenth century, of the same family (Montfaucon in Bibliotheca Bibliothecarum, vol. i. p. 527). And a seventh was possessed by T. F. Mirandola, and was used by Donatus of Verona for his Latin version, first published at Rome in 1498.

Of the four later Paris MSS., 473 bears the date 1543, and was written at Venice (or 1533 according to Montfaucon, Diario Italico, p. 408) by Valeriano of Forli. One of the four was no doubt the foundation of Stephen's Paris edition of 1548.

The Oxford MS. was collated by Gaisford with this edition of Robert Stephen in 1548 with the minutest care. But in the opinion of Dindorf his work added little to the elucidation of the text, beyond the correction of a few slight mistakes of copying, the divergencies in the quotations from the LXX being probably changes made by later scribes in order to bring the quotations into agreement with the accepted text.

Dindorf's conclusion is that a satisfactory text is secured by the use of the Parisinus 469, on which his own edition (Teubner series) is based. It is, he says, comparatively free from the errors of transcribers, with the exception of some lacuna; (pp. 195 d, 210 a, 217 b), and from the frequent interpolations of the Praeparatio and the History, because the Demonstratio, having fewer readers, was seldom copied. There is, therefore, little room in the study of the text for conjectural emendation.

The first Edition of the Greek was that of Robert Stephen, 1548.

Viguier's Praeparatio was published at Paris in 1628, with the Demonstratio and other works of Eusebius, and the Latin translation of Donatus.

Gaisford's edition (2 vols., Oxford) appeared in 1852 with critical apparatus and the same Latin translation.

The Demonstratio forms vol. xxii. of the Greek Patrology of Migne (1857), who uses the Paris edition of 1628 with the same translation.

The most recent text is W. Dindorf's in the Teubner Series (Leipzig, 1867), from whose Preface the data of the above are drawn.

The Latin version of Donatus (Rome, 1498) was reprinted at Basle in 1542, 1549, 1559 and 1570, and with the Scholia of J. J. Grynaeus at Paris in 1587. It is remarkable for its omissions and alterations of passages doctrinally suspected.

The present translation [FERRAR] is made from the text of Gaisford (Oxford, 1852), with reference to Migne.

Manuscripts of Books 1-10



Shelfmark & Notes

Date /

P Paris, BNF Codex Parisinus Graecus 469 (once Mediceus).  Damaged at front and back.

Parchment.  Pages 34 x 24 cm; text area 25x15 cm.  31 lines per page.  322 pages numbered and written. 

F J. A. Fabricius, Delectus argumentorum et syllabus scriptorum, qui veritatem religionis Christiana: adversos atheos . . . asseruerunt, Hamburg (1725).  The portions missing from P are contained here, obtained from a handwritten copy (1725) of a lost MS ('codex Maurocordati') belonging to Nicholas Mavrocordato, Prince of Wallachia.  Probably from Mt. Athos.  Fabricius said:

"Singulari autem beneficio celsissimi ac sapientissimi Walachiae principis Maurocordati atque illustrrissimi eius propinqui et archicamerarii Johannis Scarlati mihi contingit ut proxeneta C. V. Stephano Berglero licuerit mihi ex integriore codice, quem Graecia servavit principique obtulit, defectum illum ingentem initio libri primi supplere Decimique finem ubi perpauca desiderabantur subnectere".

HEIKEL adds that more information on this MS and its fate after his death (1730) is given by C.B.Hase in his Commentary on Joannes Lydus, p. lxix-lxxi, but he was unable to check this reference.

Paris, BNF Codex Parisinus Graecus 470. parchment.  Derived from Paris 469. 16
Paris, BNF Codex Parisinus Graecus 471. parchment.  Derived from Paris 469. 16
Paris, BNF Codex Parisinus Graecus 472. parchment.  Derived from Paris 469. 16
Paris, BNF Codex Parisinus Graecus 473. parchment.  Derived from Paris 469. Written at Venice by Valeriano of Forli. 1543
S Oxford, St. Johns College Cox's catalogue, #41.  Derived from Paris 469. 15
Milan, Ambrosian Library Codex Ambrosianus L 109 sup.  Derived from Paris 469. 15
[Unknown] Owned by T. Mirandola, and used by Donatus of Verona for the Latin 1498 edition.
O Bologna, University Library Codex Bononiensis University 3644. Written by a monk Nicephorus.  [Details from MRAS edition of the PE, p.xxxv.  See the PE for MS 3643 for details of origin of MS]

Paper.  Copied from P.

Bologna, University Library Codex Bononiensis 2304.  Also contains Tatian Contra Gentes.  Derived from Paris 469.  Written by Valerianus.

HEIKEL mentions (p.xv) that Harnack in Gesch. der Altchristl. Literatur I s. 568, following Montfaucon Diar. Ital. 408, lists an MS in Pisa, but this is a misunderstanding by Harnack, and looking at Montfaucon verifies that the MS is this one in Bologna.

Madrid, Escorial An MS is listed as number 597 by Miller p.370 and reported as lost (HEIKEL, p.xv)

The Fabricius MS was a member of the same family as Paris 469, perhaps a copy taken before the front and back were damaged.  The other MSS are all derivative of Paris 469, as can be seen from a common defect.

Manuscripts of fragments from book 15

The earlier editions contained only fragment 1, derived from A. Mai, script. vet. nov. coll. I 2, p. 173.  The text is present in the Catena MS, Cod. Vatic. 1153-1154.  It is also found in Cod. Ottob. 452, ff. 242-257.  There is also supposed to be a copy in Cod. Chisan. R. VIII 54.   Mai obtained his fragment from the Codex Ottobonianus, as a comparison of readings with the Vaticanus shows.  He may also have used the Cod. 18. Pii. II, a copy of the Vaticanus.

Fragments 2-6 are also to be found only in the Cod. Ottob. 452, another Catena MS.

Contents of the lost books

The introduction to the first chapter of the first book gives a general overview of the contents of the whole work.  From this we can see that in books 11-20, the following themes were treated: the condemnation and death of Christ (continuing the material in book 10), his resurrection, how he revealed himself to his disciples, the baptism of the holy spirit, the ascension, his rule with the father and return at the end of the world.  The fragments of book 15 speak of the Judgement and the reign of Christ, from Daniel.  Jerome mentions in the preface to his commentary on Hosea that Eusebius quotes Hosea in book 18 of the DE.   Photius (Bibliotheca, cod. 10) still knew of 20 books.  

It can be seen from the surviving books that Eusebius rambled away from the subject, and did not in any case arrange his material particularly skilfully - possibly the work was composed in haste (cf. book 2); and indeed this explains why the remaining material required a further ten books.

'Tables of contents', chapter divisions and titles

Each book begins with a summary of the contents, misleadingly called a 'table of contents' or 'table of chapters', although ancient books did not have chapter divisions.  The chapter titles are different and fuller in the MSS, but these are doubtless medieval.  Here are FERRAR's comments on this:

The above list of chapters was given at the beginning of each book. It was lost from the Paris Codex for Book I together with the first pages of that book, and from the copies, one of which Robert Stephen used in his edition of 1545. In the Paris edition of 1628, the editor composed the headings of the first three chapters, and supplied the others from a second catalogue, which is given at the head of each chapter throughout the work. Though no doubt the catalogue was complete in the Mavrocordato Codex, Stephen Bergler omitted to give it in the portion of the work which he supplied for the edition of Fabricius.

The headings of the separate chapters, which are in our translation given in their places and form a second catalogue, are much fuller than the introductory list, being enriched by outlines of the prophetic passages that are used.

HEIKEL has the following to say (pp.xxiv-xxv):

Nichts spricht gegen die Annahme, daß die vor den einzelnen Büchern stehenden Inhaltsangaben von Eusebius selbst herrühren. Von diesen aber weichen die in den Text selbst eingefügten öfter ab. Vor allem enthalten sie viel mehr als die zuerst genannten Indices. Es ist fast sicher, daß sie nicht Eusebius zum Verfasser haben. S. 57, 7 sagt Eusebius, daß er nur die Bibelstellen anführen will ohne jedwede Interpretation (di/xa pa&shj e9rmhnei/aj). Als eine solche muß man indessen die zu jedem Capitel hier (im Gegensatze zu dem, was in dem vor dem ganzen Buche stehenden Inhaltsverzeichnis der Fall ist) vorkommenden Inhaltsangaben betrachten. Noch deutlicher tritt dies hervor bei den Citaten S. 60,27 ff, betreffs deren Eusebius S. 60, 23 ff ebenfalls sagt, daß er zu ihnen keine Erklärung beifügen will (di/xa pa&shj th~j ei0j au0th_n qewri/aj), denn die ausführlichen Inhaltsangaben sind doch wirkliche Erklärungen. -- Die Capitelindices im III. Buche sind sehr spärlich; die erste Inhaltsangabe (S. 94) ist mit zwei Zeilen des Textes identisch. Das Citat aus Porphyrius ist mit einer Inhaltsangabe versehen.

S. 346,25 sagt Eusebius: ge/noito d0 a@n e0k periousi/aj kai\ tou&tou parastatiko_n to_ para_ Mwsei lo&gion, touton e1xon to_n tro&pon. Dann müßte das Bibelcitat (Gen. 49,8-10) unmittelbar folgen, aber in der HS wird zuerst angegeben, was die Bibelstelle sagen will. Einen ganz ähnlichen Fall haben wir S. 351, 34 ff. Die Inhaltsangaben haben also nicht ursprünglich im Texte gestanden, sondern sind vom Rande dahin gekommen. -- Im X. Buche bestehen die Inhaltsangaben zu den einzelnen Capiteln z. T. aus lästigen Wiederholungen, die nichts Neues bringen -- im Gegensatz zu den beiden ganz angemessenen Angaben, die im Index vor dem Texte des X. Buches stehen.

which may be summarised as:

There is nothing to suggest that the indices standing at the head of each book are not by Eusebius himself.  However the versions inserted at intervals into the text itself are not necessarily authorial.  Above all, they contain a great deal more than the indices.  It is almost certain that Eusebius did not write them.  Eusebius states on p.57, 7 that he is quoting the bible passages without any interpretation. But the chapter/section titles (as opposed to the summaries at the start of each book) are just such interpretation.  The same point is made in p. 60, 27 ff, and p. 60, 23 ff, where he again says that he does not want to attach an explanation to the text, while the detailed titles in the text are neverthless real explanations.  -- The chapter titles in  book 3 are very sparing; the first of them is identical to two lines of the text.  The citation from Porphyry is provided with a title.

On p. 346, l. 25 Eusebius says: ge/noito d0 a@n e0k periousi/aj kai\ tou&tou parastatiko_n to_ para_ Mwsei lo&gion, touton e1xon to_n tro&pon.  The bible citation (Gen. 49, 8-10) must thus follow immediately, but in the MS we are first told what the biblical citation will say.  A very similar case appears on p. 351, 34ff.  Thus the chapter titles were not originally part of the text, but in the margin.  In book 10, the chapter titles for individual chapters are simply mindless repetition, bringing nothing new -- as compared with the real and appropriate text in the index at the start of the book.

Other notes

It is possible to evaluate the quality of P, the primary manuscript, by comparing the text of works which Eusebius cites also in other works.  In general P seems to have suffered from errors in transmission.  The citation of Josephus is not verbatim, but does demonstrate that P does contain errors of transmission.  Comparision of quotes from Porphyry compared against the PE indicates the same, as does a quote from Julius Africanus used in the Ecl. Proph.  (GCS p.xviii).  The consequence is that the indirect transmission of the text of the DE is also of interest, to correct errors in P.  

The DE is cited extensively in Procopius of Gaza in his Commentary on the Octateuch; also by John Damascene, De imagin. Or., and in Biblical Catena MSS.

This topic is dealt with in detail in the GCS introduction.


W. J. FERRAR, The Proof of the Gospel, being the Demonstratio Evangelica of Eusebius of Caesarea. London: SPCK, New York: Macmillan (1920). 2 vols.  Reprinted, and currently in print in a single volume edition.  Checked. This is the only English translation.

Ivar A. HEIKEL, Eusebius Werke VI: Die Demonstratio Evangelica.  Griechische Christlichen Schriftsteller, Leipzig (1913).  Checked.

Erich LAMBERZ, Johann Albert Fabricius und der Codex Maurocordati, Byzantinische Zeitschrift 99, (2006), 515-522. Not checked.

Constructive feedback is welcomed to Roger Pearse.

This page has been accessed by  ****** people since 10th August 2002

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