Short URL for this page:

[image ALT: Much of my site will be useless to you if you've got the images turned off!]
Bill Thayer

[image ALT: Cliccare qui per una pagina di aiuto in Italiano.]

[Link to a series of help pages]
[Link to the next level up]
[Link to my homepage]

[image ALT: link to previous chapter]
Book VI
Chapter 7

This webpage reproduces a section of
Italy and Her Invaders

Thomas Hodgkin

published by the Clarendon Press

The text, and illustrations except as noted,
are in the public domain.

This page has been carefully proofread
and I believe it to be free of errors.
If you find a mistake though,
please let me know!


[image ALT: link to next chapter]
Book VI
Chapter 8

Vol. IV
Note F
The Letters of Pope Gregory I

These letters, the importance of which is recognized by every student of the history of the Middle Ages, have lately been made the subject of an exceedingly minute and critical examination by Paul Ewald, the scholar who was employed to edit them for the Monumenta Germaniae Historica. Unhappily he died before even the first section of the letters was published, but the work is being carried on upon the lines indicated by him, and will probably soon be completed.

Meanwhile the indispensable guide for every student who would thoroughly explore the documentary history of these important letters is an article written by Ewald, which appeared in the third volume of the Neues Archiv der Gesellschaft für ältere Deutsche Geschichtskunde (1878). The article is long, occupying nearly 200 pages of the Neues Archiv. It is not quite as clear as it might be, and it is certainly not easy reading, but, as before said, it must be studied by any one who wishes to form an independent judgment of the chronology of Gregory's pontificate. All that I propose to do here is to summarize the chief results of Ewald's investigations, and to erect a few guide-posts which may direct future students through the jungle.

The edition of Gregory's letters which has been most extensively used up to the present time is that put forth in 1705 by the Benedictine congregation of St. Maur.1 The soul of this enterprise was Dom Denis de Ste. Marthe (Dionysius Sammarthanus), who, however, unfortunately handed over the actual editing of the letters to a greatly inferior scholar, Dom Guillaume Bessin, whose work sometimes incurred his censure. Pope Clement XI sent Ste. Marthe twenty-four gold medals in token of his approval. Ste. Marthe sold them all, and gave the proceeds to the poor, nor would he allow his name to appear on the title- p334 page of the work; so nobly unselfish was the spirit in which these Benedictine scholars laboured. The text adopted in this edition does not come up to the requirements of modern criticism, not being founded on an accurate collation even of all the MSS. to which the editors had access (and which were chiefly French and Italian codices). It has also suffered somewhat from the anxiety of the good fathers to make the Pope always write classical Latin. But on the whole it is a fair working text, and, as far as differences of readings go, it would not be necessary to remark here upon its defects. The really important question is as to the order of the letters, and before we can discuss that question we must follow Ewald through a long, perhaps tedious, analysis.

The sheet-anchor of the Benedictine editors, and the MS. which they mainly follow in arranging the order of the letters, is Codex Vaticanus A, a beautifully written MS. of the fifteenth century, with gilt initials at the beginning of each letter, and careful paintings at the beginning of each book. The collection of the Gregorian letters herein contained was apparently made by order of John IV, archbishop of Milan (1485‑8), an intimate friend of Galeazzo Maria Sforza. This arrangement is always spoken of by Ewald as the Milanese Codification.

Of course, in itself, a fifteenth century MS., however beautifully executed, is not a first-rate authority for letters at the end of the sixth century; and we therefore naturally enquire on what foundations this Milanese Codification rests. Ewald's researches prove that it rests on three separate collections of Gregorian letters, dating from a very early period, which had never been all combined before the fifteenth century, and the manner of whose arrangement in the Milanese Codification is certainly erroneous.

I. The Hadrianic Register (R). Much the largest and most important of these partial collections is that which Ewald calls the Hadrianic Register, and denotes by the symbol R. To this belong 686 out of the total 851 letters of Gregory which are still extant. It is a collection, or, more properly speaking, a selection of the great Pope's letters made in the time of Pope Hadrian I (772‑795), and probably by his order. This Pope gave the Emperor Charles the Great a copy of the Decretals and a Sacramentarium of Gregory; and it seems highly probable  p335 that he may have accompanied these gifts by a copy of selected Epistles of the Pontiff. It seems clear, however, that the Hadrianic Register cannot have been a complete transcript of the great libri Charticii which once existed at the Lateran, and which are described by Gregory's biographer Joannes Diaconus. This is proved: —

(1) By the express language of Joannes Diaconus (evidently describing the Hadrianic Register). He tells us (IV.71) that 'from the multitude of the books of Pope Gregory's letters certain decretal letters were extracted, according to their several Indictions, and collected into two volumes as is at present seen.'

(2) By the title borne by some of the MSS., 'Epistolae ex registro b. Gregorii,' showing that it is but a selection.

(3) By the existence of certain letters (165 in all) in the other collections which we do not find in this. We must therefore carefully distinguish between the great Lateran Register on sheets of papyrus, which has long since perished, and the Hadrianic Register, which, though far the most complete collection that we possess, is only an approximation towards that now unattainable ideal.

The great characteristic of this Collection R, and that which makes it especially valuable for our present purpose, is that it is arranged 'under Indictions,' that is, in fourteen books corresponding to the fourteen years of Gregory's pontificate.

Moreover, as stated by Joannes Diaconus, it was in two volumes. The first of these contained the letters of seven years (Sept. 590 to Aug. 597, or from the ninth to the fifteenth Indiction inclusive). These letters were 393 in number. The second also contained the letters of seven years (Sept. 597 to Mar. 604, or from the first to the seventh Indiction inclusive). These letters were 293 in number. Total 686. These two volumes have been often separated from one another by transcribers, and in some cases have drifted far apart, and undergone strange combinations with the other collections; but with these phenomena, though they must have enormously increased Ewald's labour in tracing the documentary history of the collection, we need not here concern ourselves.2

 p336  It will suffice to give the results, showing the distribution of the letters through this, the only satisfactory chronological collection.

Volume I

Indiction IX : September 1, 590 – August 31, 591 , contains 82 letters.
" X " 591 " 592 " 41 "
" XI " 592 " 593 " 65 "
" XII " 593 " 594 " 44 "
" XIII " 594 " 595 " 56 "
" XIV " 595 " 596 " 63 "
" XV " 596 " 597 " 42 "
Total in Volume I 393

Volume II

Indiction I 597‑598 = 36 letters.
" II 598‑599 = 95 "
" III 599‑600 = 21 "
" IV 600‑601 = 56 "
" V 601‑602 = 18 "
" VI 602‑603 = 50 "
" VII 603‑604 = 17 "
Total in Volume II 293

(The list from which the above table has been formed, and which is one of the most important factors in determining the dates of the letters, is to be found on pp462‑464 of the article in the Neues Archiv.)

Some of the letters included in this list are put by the Benedictine editors into their Appendix; and one or two trifling adjustments, which need not be particularized here, have to be made in order to reconcile the above number 686 with the 679 numbered letters which are included in the Benedictine edition.

II. The Two Hundred Letters (C). To this second source of Gregorian Epistles, consisting of exactly 200 letters, Ewald gives the symbol C (CC = 200.)

Of these 200 letters, 56 are possessed in common by C and R. Only 144 therefore remain which are peculiar to C.

This collection is generally found in combination with that  p337 next to be mentioned (P), but there are some fine MSS. which contain P without C, and a careful examination of the question leaves us without doubt that the two collections were, in their origin, distinct.

One peculiarity of C is its singularly unchronological character. Only one letter, the twenty-fifth in the collection,3 bears a date (Mense Maio, Indictione II = May, 599). It contains no other reference to month or year of Indiction, and had therefore already, in the twelfth century, earned the name of 'the Indictionless Register.'

But in age the collection C (which on this point must be spoken of in conjunction with P) is certainly not inferior to R: perhaps it is even older, anterior that is to the end of the eighth century.4 It was evidently this collection, which he calls Liber Epistolaris, that Alcuin consulted when he wanted to find a certain well-known letter to Leander of Seville, and when he could not find it here he was disposed to give up his belief in its genuineness. Had he been able to consult the Hadrianic Register, his doubts would have been set at rest, for it appears there (and figures as I.43 in the Benedictine edition).

It is from the manner in which the Milanese codification has interpolated these dateless letters into the last six books of R that almost all the confusion has arisen. This interpolation has added —

36 letters to the IInd Indiction — 598‑599
36 " IIIrd " 599‑600
24 " IVth " 600‑601
39 " Vth " 601‑602
6 " VIth " 602‑603
3 " VIIth " 603‑604

that is to say, it has sprinkled the majority of the letters nearly equally over the four books from September 1, 598 to August 31, 602.

This interpolation, which seems to have been made purely at haphazard, led to some such obviously erroneous results that the Benedictine editors took some timid steps towards its  p338 rectification. They assign 38 of the letters peculiar to C to the IInd Indiction, 37 to the IIIrd, 22 to the IVth, and 35 to the Vth: scattering the remaining 12 over five other years.

All this, however, is, as has been said, purely haphazard work, and the whole method of procedure is discredited by Ewald's great discovery that all the 200 letters in C, with one unaccountable exception,5 really belong to the IInd Indiction, and that C is therefore in fact an extract from the original Register for the year 598‑599.

It would not be possible here to do any justice to the long and laborious process by which, as I must think, Ewald proves this proposition, a proposition which (with Weise) I was disposed to resist to the utmost of my power, since it has the effect of allotting 240 out of the 851 letters of Gregory all to one year. But the a priori probability of it will be seen when the reader compares the two following lists, which in themselves will be found useful by a student of the Gregorian Epistles.

List of letters common to C and R arranged in the order of the Benedictine edition: —

A.D. 592‑3 XIth Indiction III.11, 37, 64 =  3
" 594‑5 XIIIth " V.46 =  1
" 595‑6 XIVth " VI.34 =  1
" 598‑9 IInd " IX.1, 6, 8, 13‑17, 24, 25, 27‑31, 34, 36, 37, 47‑49, 52, 60, 63, 64, 70, 74‑76, 78, 81, 84, 85, 87, 88, 91, 92, 97, 106‑112, 115, 116, 121, 122, 125 = 50
" 599‑600 IIIrd Indiction X.42 =  1

Here we see that where we are able to check the arrangements by that which we surely know, from the fact of the letters being found in the dated R collection, as well as in the dateless C, 50 out of 56 letters, by the confession of the Benedictine editors themselves, belong to the IInd Indiction. And if we examine a little further we find that of the remaining six letters,  p339 all with the single exception of III.11 (an admitted interloper) are by R itself assigned to the same IInd Indiction, and have been for no sufficient reason placed by the Milanese codification and the Benedictine edition in other years.

List of the letters peculiar to C arranged in the order of the Benedictine edition (a very different one from the order of C itself): —

Indiction Book
A.D. 590‑591 IX I.85 =  1
" 594‑595 XIII V.45, 47 =  2
" 596‑597 XV VII.45 =  1
" 597‑598 I VIII.17, 19, 32 =  3
" 598‑599 II IX.2, 5, 18‑23, 26, 32, 40, 42‑45, 50, 51, 54‑56, 62, 66, 71, 73, 82, 83, 89, 90, 93, 94, 101, 102, 104, 117‑120, 124 = 38
" 599‑600 III X.2, 3, 5‑7, 9, 12‑15, 20, 25, 26, 28, 30, 32, 33, 40, 41, 43, 44, 46‑48, 50, 52‑60, 64, 66, 67 = 37
" 600‑601 IV XI.6, 7, 9‑11, 17‑21, 23, 24, 38, 39, 41‑43, 48, 49, 70, 73‑75 = 22
" 601‑602 V XII.2‑6, 9‑11, 13‑23, 25‑27, 34‑37, 39‑46, 49 = 35
" 602‑603 VI XIII.24, 25, 43, 47 =  4
" 603‑604 VII XIV.15 =  1

Here it will be seen that even the Benedictine editors refer all the letters but twelve to the years between 598 and 602; but we may go further than this. There is no internal evidence which requires us to place these twelve in the books to which the editors have assigned them. Comparing this list of letters peculiar to C with the previous list of letters common to C and R, we see at once the improbability that the former should belong to four (or rather ten) years, while the latter (with which  p340 they are promiscuously blended) all6 belong to one, namely, the IInd Indiction. A strong probability is thus raised that the whole of C is really an extract from the great registers for that year: and this probability is, I think it may be said, converted into certainty by the extremely minute analysis to which Ewald subjects the two lists C and R in the fifth chapter of his essay.7 It is a triumphant vindication of his method that the one solitary date in the whole of C, 'xxv. Mense Maii Indict. II. Gregorius Venantio Episcopo Lunensi,' comes, on Ewald's principles of reconstruction, exactly where it ought to come: that is to say that the 25th C letter (X.44 in the Benedictine edition) ought to come at the beginning of the month of May, 599.

The reassertion of the true character of the C collection, and the restoration of the letters contained in it to their right place, was well worth all the labour which Ewald had bestowed upon it, since without it a true chronological arrangement of the Gregorian Epistles was impossible. But it cannot be said that these letters are in themselves of any especial interest, except that which the Benedictines make I.85, a letter of introduction for Droctulf to Gennadius the Patrician of Africa, and IX.42, 43, the two very interesting and important letters which the Pope addressed to Agilulf and Theudelinda in connection with the conclusion of the great peace. Many of the other letters are connected with the internal affairs of the dioceses of Campania and Sicily; and though they throw some valuable light on the social and religious condition of Italy, they can hardly be considered of great political importance.

We now come to —

III. The Collectio Pauli (P), containing 53 letters, of which only 21 are peculiar to P, the remaining 32 being also found in R.

The reason for the name given to this collection is that in one MS. (Codex S. Germani), dating from the eighth century, there was to be found a short preface addressed by a certain Paulus to his dearest brother and lord Adalard [abbot of Corbie], apologizing for having failed to visit him in the preceding summer, and sending him the desired letters of Gregory, of which, however,  p341 he regrets that he has only been able to correct 34 to his satisfaction. Sickness, poverty, the absence of his amanuensis (clericulus) have prevented him from doing more.

Who this Paulus was is by no means clear. Naturally our minds recur to the great Lombard historian and biographer of Gregory: but if this be the man, it is singular that none of the three letters quoted in his history (IX.42, 43 and XII.21) should be found in this collection, but only in C. On the whole, as Paulus was a very common name among ecclesiastics, it is safer to leave this point undecided.

It will be observed that I have said that the prefatory note was to be found in the Codex S. Germani. That MS. was much relied upon by the Benedictine editors, who have given us many valuable notices of its readings; but unfortunately it was stolen during the troubles of the French Revolution, and has never since reappeared.

The following are the lists of the letters peculiar to P and common to P and R (using here, as always, the Benedictine numbering): —

Letters peculiar to P Letters common to P and R
A.D. Indiction Book Book Total
590‑1 IX I.10 =  1 1
591‑2 X II.3, 4, 5, 28 29, 30, 31, 32, 34, 53 = 10 II.33, 36, 37, 41 =  4 14
593‑4 XII IV.47 =  1 1
594‑5 XIII V.16, 30, 31, 32, 37 =  5 V.10, 18, 21, 23, 39, 41, 42, 43, 53, 54, 55, 57, 58 = 13 18
598‑9 II IX.41 =  1 IX.52 =  1 2
600‑1 IV XI.1, 14, 65 =  3 XI.2, 12, 13, 28, 45, 55, 56, 59, 62, 63, 66, 67, 69, 71 = 14 17
21 32 53

It will be seen from these lists that, according to the Benedictine arrangement, the great majority of the letters in P belong to the Xth, XIIIth, and IVth Indictions (591‑2, 594‑5, and 600‑1), and Ewald shows that the apparent exceptions have been either misplaced by the Benedictines, or are probably interpolations into the original collection; in short, that we  p342 may safely treat P as a series of extracts from the above-named three years of the Gregorian Register.

On what principle the collection was made it is impossible to say: but the letters in P are on the whole much richer in political interest than those in C, and there seems to be an especial reference in the collector's mind to the affairs of the Gaulish Churches and the relations of the Pope to the Frankish kings.

Ewald devotes many pages to the discussion of the dates to be found in the 'Collectio Pauli,' a very abstruse and difficult subject. In some MSS. the letters have been dated at the beginning; in others at the end; and the vacillation between the two systems has led the scribes into many perplexities. Ewald comes to the conclusion that upon the whole the MSS. which insert the date in black ink at the end of the letter are more to be relied upon than those which insert it in the beginning in red ink as part of the title of the letter.

Incidentally he touches on an interesting question, why in these dates we sometimes find the old Roman notation by Kalends, Nones, and Ides, and sometimes our present custom of numbering the days of the month continuously. It is clear that Gregory himself used both methods, and that the modern plan, which had not gained full ascendency in his day, gradually after his death became more usual. If he was not the actual introducer of the custom, he seems at any rate to have helped powerfully towards rendering it popular. (And thus we may observe, the First Gregory as well as the Thirteenth had some claim to be considered a Reformer of the Kalendar.) Mommsen makes the interesting suggestion that the custom of numbering the days of the month continuously, originated in Syria, and spread from that country by way of Greece to Italy.

I have now sufficiently indicated the general principles on which the rearrangement of the Gregorian letters in the new edition in the 'Monumenta Germaniae Historica' will be based. Something must still be left to conjecture as to the consolidation of the three collections; and I am not sure that if once the old order had to be departed from, it would not have been better to give us each of the three collections R, C, and P separately, so that students might have used their own judgment as to their combination one with the other: but it cannot be doubted  p343 that we shall have here in the main a chronological series of the letters of the great Pontiff and a scientific basis for the study of a period which was one of the great turning-points in the history of Europe.

(As there is no proper Index or Table of Contents attached to Ewald's article in the third volume of the 'Neues Archiv,' the student may be helped by the following reference): —

(1) List of the 686 letters in R 462‑464

The same list with the dates

This is perhaps the most valuable part of the whole paper, and is independent of all conjectural theories as to reconstruction.


List of the 200 letters in C, arranged in their own order


The same in the order of the Benedictine edition


List of the 144 letters peculiar to C, in their own order


Ewald's reconstruction of the IInd Indiction, showing the way in which he inserts the C letters among the R letters for that year


Result: approximate dates of C letters


List of the 53 letters in P


List of the 21 letters peculiar to P


Dated list of the P letters


'The Milanese Codification,' showing the way in which the C letters (in round brackets) and the P letters [in square brackets] were interpolated into the Hadrianic register R


The Author's Notes:

1 Always called by Ewald 'Die Maurinerausgabe.' For convenience' sake I speak of it as the Benedictine edition.

2 For brevity's sake, Ewald, who styles the whole Hadrianic Register R, labels the first volume r, and the second ρ, and then traces, with the industry of an analytical chemist, the various combinations into which r and ρ have respectively entered with other fragments of the great original collection.

3 X.44 in the Benedictine edition.

4MS. of C + P in the Cathedral Library at Cologne (Codex Coloniensis), the work of a scribe of the eighth century, is in Ewald's opinion by far the finest and most trustworthy of all the MSS. of Gregory's letters.

5 III.11 in the Benedictine edition.

6 With the single exception of III.11.

7 pp522‑531. See also pp573‑580.

[image ALT: Valid HTML 4.01.]

Page updated: 27 Jun 20