THE TEXT OF THE COMMENTARIES

EVERY one who can read the Commentaries with interest will want to know how far the manuscripts in which they have been handed down to us correspond with what Caesar wrote; for if he will think, he will see that none of them correspond with it exactly, and that although scholars have been trying ever since 1469, when the first printed edition was published, to remove the errors, some must still and always will remain. The oldest of the extant manuscripts was written fully 900 years after the book was first put into circulation. Now, however careful a scribe may be, he can hardly avoid making some mistakes in copying out a written book; the scribe who copies his copy will make more; and so on. Even contemporary copies of Caesar's original manuscript doubtless contained mistakes. Cicero' complains that books sold by the booksellers of Rome had been carelessly copied; and, notwithstanding all the care of proofreaders, few modern books are entirely free from printers' errors. Besides, a manuscript might pass into the hands of a reader who would make notes on the margin ; and if another copy were to be made from the one which contained these notes, the copyist might be misled into incorporating them in the text. Thus two kinds of mistakes would gradually find their way in. An example of the latter kind-nullam . . . intermisso-will be found in i, 26, § 5. An example of the other shows how even a very careful copyist might go astray. In viii, 32, § 2 the famous stronghold, Uxellodunum, is mentioned for the first time. Uxellodunum was only written by the copyist

in two of the good manuscripts: the rest have auxilio dunum, which, as every one will see, is nonsense. Can you imagine how this curious blunder was made? In this way. In some manuscript a reader wrote either in the margin or above uxellodunum (not Uxellodunum, for even proper names were written with small initial letters) the words a. uxillodunum, and by a., which was an abbreviation, he meant aliter, 'otherwise'. He wished to show that besides uxellodunum there was another spelling, uxillodunum. This manuscript passed into the hands of a copyist who misunderstood the abbreviation a. and wrote auxillo dunum, and as I might easily be mistaken for i, somebody else wrote auxilio dunum.

A great many manuscripts of Caesar exist; but only nine or ten of them are now considered good. They are divided into two groups, known as a and ẞ, and generally believed to be derived from a common original, or archetype, which is called X. Each manuscript is called by a letter, which is here prefixed to the full name :

A codex Bongarsianus (or Amstelodamensis 81) of the ninth or tenth century.

B= Parisinus I (Paris, Bibliothèque nationale, 5763, ninth or tenth century).

M= Vaticanus (Vatican, 3864, tenth century).

Q= Moysiacensis (Paris, Bibl. nat., 5056, twelfth century).

S = Ashburnhamianus (Bibl. Laurent. R. 33, tenth

century).

a = Parisinus II or Thuaneus (Paris, Bibl. nat., 5764, eleventh century).

ƒ= Vindobonensis I (Bibl. Vindob. [Vienna], 95, twelfth century).

h = Ursinianus (Vatican, 3324, eleventh century).

7 = Riccardianus (Bibl. Riccard. [Florence],541, eleventh or twelfth century).

xviii THE TEXT OF THE COMMENTARIES

H. Meusel traces the pedigree of these MSS. as follows: X

To may be added the best manuscript in the British Museum (Add. MSS. 10,084), which is known as Lovaniensis and referred to as L.

Professor A. Klotz (Rheinisches Museum, 1910, pp. 22434) thinks that the foregoing pedigree, which has been generally accepted, is incorrect. He believes, with Professor B. Kübler, that the archetype of all the extant MSS. was a copy belonging to B, and that a is descended from a copy belonging to the same group, in which readings from a manuscript of the sixth century, published by two editors-Julius Celsus Constantinus and Flavius Licerius Firminus Lupicinus-were inserted. Accordingly Klotz has constructed this pedigree, which, in the opinion of Meusel, may possibly be right :

The two groups, a and ẞ, differ from each other about 1,500 times; and an editor cannot do without either. But when they differ and neither is obviously wrong, how is he to decide between them? Simply, in most cases, by considering the context or by carefully noting Caesar's use of language in passages in which the two groups agree. This laborious task has been performed

by various critics, notably by three German scholars, Rudolf Schneider, Meusel, and Alfred Klotz. Let me give one or two examples. In v, 35, § 5 a has (cum a prima luce ad horam octavam) pugnaretur; ß has pugnassent. The former is preferable because Caesar in describing the duration of a battle almost always uses the passive. In vii, 64, § 2 ah have (peditatu quem ante) habuerat (se fore contentum dicit); while the rest of the ẞ MSS. have habuerit, which is certainly right, because the relative clause is part of what Vercingetorix said, and therefore the subjunctive is necessary.

There is also a considerable number of passages in which, though all the manuscripts agree, the text is obviously wrong, and has been corrected with more or less success. Some of these emendations are certainly right. For instance, in i, 40, § 9 the MS. reading is (cui rationi contra homines barbaros . . . locus fuisset) ac (ne ipsum quidem sperare nostros exercitus capi posse); and the obvious correction, hac, appeared just four centuries ago in the Aldine edition. Again, in vii, 3, § 2 the MSS. have (Nam) ubique (maior atque inlustrior incidit res, clamore. . . significant): the emendation ubi quae is selfevident. Other emendations are highly probable; and fortunately those doubtful or corrupt passages which are important for history are very few.

In this book it would be useless to give a list of the various readings of the manuscripts, or to explain the reasons that have led me to adopt one reading in preference to another. I must ask my readers to believe that I have done my best to produce a sound text; but if any one desires to test my conclusions, he may refer to the complete edition. Generally, however, when I have enclosed words in square brackets, I have given the reasons for doubting whether they were written by Caesar, because I have found that even beginners are inquisitive about such things.

THE CREDIBILITY OF CAESAR'S

NARRATIVE

But

For the history of the first seven years of Caesar's conquest of Gaul our principal authority is Caesar himself. It is, indeed, impossible to grasp the full meaning of his narrative without the help of the modern scholars who have contributed so much to the task of solving the problems which the Commentaries present. It is true, moreover, that Cicero's writings illustrate certain phases of the war, and that later writers, such as Suetonius, Plutarch, and Dion Cassius, make certain statements, true or false, which are not to be found in Caesar. Caesar is the authority; and thoughtful readers will want to know how far his narrative is trustworthy. According to Suetonius,1 Pollio, who served under Caesar in the civil war, thought that the Commentaries were written carelessly and with little regard for truth; that Caesar had accepted without due inquiry the reports of his officers; and that, either intentionally or from failure of memory, he was inaccurate in describing what he had done himself. Now it is almost certain that Pollio was referring to Caesar's Commentaries on the Civil War, with which we are not here concerned; 2 but suppose that he had the same opinion of the Gallic War. If so, as he did not serve in Gaul, he could only have repeated what he had heard from others; and their opinions would have been of no value

1 Diuus Iulius, 56.

2 According to Suetonius, Pollio thought that Caesar would have rewritten and corrected his narrative if he had had an opportunity (existimatque rescripturum et correcturum fuisse). A. Klotz (Rh. M., 1911, p. 81) remarks that Pollio could hardly have said this in regard to the Commentaries on the Gallic War, which were published by Caesar himself.