L. D. Reynolds, N. G. Wilson Scribes and Scholars A Guide To The Transmission of Greek and Latin Literature PDF
L. D. Reynolds, N. G. Wilson Scribes and Scholars A Guide To The Transmission of Greek and Latin Literature PDF
L. D. Reynolds, N. G. Wilson Scribes and Scholars A Guide To The Transmission of Greek and Latin Literature PDF
AND SCHOLARS
A Guide to the Transmission
ofGreek and Latin Literature
BY
L. D. REYNOLDS
Fellow and Tutor of Brasenose College, Oxford
AND
N. G. WILSON
Fellow and Tutor of Lincoln College, Oxford
THIRD EDITION
This book is sold subject to the condition that it shall not, by way
of trade or otherwise, be lent, re-sold, hired out or otherwise circulated
without the publisher s prior consent in anyform of binding or cover
other than that in which it is published and without a similar condition
including this condition being imposed on the subsequent purchaser
3 5 7 9 8 6 4 2
Printed in the United States of America
on acid-free paper
PREFACE
L.D.R.
N.G.W.
January lggo
CONTENTS
i. ANTIQUITY
i.^ Ancient books
ii. The library of the Museum and Hellenistic scholarship
iii. Other Hellenistic work
iv. Books and scholarship in the Roman Republic
v v. Developments under the early Empire
vi. Archaism in the second century
vii. The compendium and the commentary
viii. From roll to codex
ix. Paganism and Christianity in the fourth century in the
Western Empire
x. The subscriptions
4. T H E RENAISSANCE 122
i. Humanism
ii. The first humanists
iii. The consolidation of humanism: Petrarch and his
generation
iv. Coluccio Salutati (1331-1406)
v. The great age of discovery: Poggio (1380-1459)
vi. Latin scholarship in the fifteenth century: Valla and
Politian
vii. Greek studies: diplomats, refugees and book collectors
viii. Greek scholarship in the fifteenth century: Bessarion and
Politian
ix. The first printed Greek texts: Aldus Manutius and Marcus
Musurus
x. Erasmus (c. 1469-1536)
ABBREVIATIONS
NOTES
Index of MSS. 2
95
General Index 303
Notes to the Plates 3X7
Plates
I
ANTIQUITY
I. ANCIENT BOOKS
that the first works to reach even a modest public were either the
writings of the Ionian philosophers and historians or those of the
sophists. There must also have been a certain demand for copies of
the poetic texts that formed the basis of school education. It is not
until the middle of the fifth century or a little later that a book trade
can be said to have existed in Greece: we find references to a part of
the Athenian market where books can be bought (Eupolis fr. 327
K.-A.), and Socrates is represented by Plato as saying in his Apology
(26D) that anyone can buy Anaxagoras' works for a drachma in the
orchestra. All details of the trade, however, remain unknown.
Of the appearance of the books that were produced in classical
Greece not much can be said with certainty. The number of books
or fragments surviving from the fourth century is so tiny that it
would not be reasonable to regard them as a representative sample.
The general statements that follow are therefore based primarily on
Hellenistic material, but it may be inferred with some plausibility
that they are true also for the classical period. An attempt will be
made to show how the physical differences between ancient and
modern books affected the ancient reader in his relation to literary
texts.
The form of the book was a roll, on one side of which the text was
written in a series of columns. The reader would unroll it gradually,
using one hand to hold the part that he had already seen, which was
rolled up; but the result of this process was to reverse the coil, so
that the whole book had to be unrolled again before the next reader
could use it. The inconvenience of this book-form is obvious,
especially when it is remembered that some rolls were more than
ten metres long. Another disadvantage was that the material of
which it was composed was by no means strong, and damage easily
ensued. It is not difficult to imagine that an ancient reader faced
with the need to verify a quotation or check a reference would rely
if possible on his memory of the passage rather than go to the
trouble of unwinding the roll and perhaps thereby accelerating the
process of wear and tear. This would certainly account for the fact
\L W h e n n e a n c i e n t author quotes another there is so often a
substantial difference between the two versions.
** standard writing material was papyrus (Plate I), prepared by
Antiquity 3
cutting thin strips from the fibrous pith of a reed that grew freely in
the Nile delta; in the first century A.D. there were also minor centres
of production in Syria and near Babylon. Two layers of these strips,
One laid at right angles over the other, were pressed together to
Sform sheets (Pliny, N.H. 13.680.). The sheets could then be glued
together in a long row to make a roll. Many sizes of sheet were
niade, but the average book allowed a column of text between eight
and ten inches high, containing between twenty-five and forty-five
lines. As there was only one large source of supply the book trade
Was presumably exposed tofluctuationsarising from war or a desire
by the producers to exploit their virtual monopoly. Some such diffi
culty is implied by Herodotus' remark (5.58) that when writing
material was in short supply the Ionians had used sheep and goats*
skins as a substitute. In resorting to this expedient they seem to
have followed the practice of their Oriental neighbours. But leather
as a writing material compared unfavourably with papyrus, and
was no doubt used only in emergency. In the Hellenistic period, if
Varro can be trusted (cf. Pliny, NH 13.70), the Egyptian govern
ment placed an embargo on the export of papyrus, which seems to
have stimulated the search for an acceptable alternative. At
Pergamum a process was devised for treating animal skins to give a
better writing surface than leather, the result being what is now
called parchment (otherwise known as vellum); the word owes part
6f its etymology to the name Pergamum, and the derivation can be
seen more clearly from the Italian form pergamena. But if this tradi
tion is true the experiment was atfirstshort-lived; one must assume
that the Egyptian embargo was soon removed, for it is not until the
l^arly centuries of the Christian era that parchment comes into
common use for books; an early example is the fragment of
Euripides* Cretans (P. Berol. 13 217).
To what extent the supply and price of papyrus hindered or
encouraged its use in Greece is impossible to say. But when
employed for the production of a book it was almost invariably
covered with writing on one side only. The form of the book made
this necessary, since a text written on the back of a roll would have
been very easily rubbed away, and perhaps the surface of the
papyrus contributed to the formation of this convention, since
4 Scribes and Scholars
scribes always preferred to use first the side on which the fibres ran
horizontally. On rare occasions we hear of rolls written on both
sides (Juvenal 1.6, Pliny, EpisL 3.5.17), but such books were excep-
tional. A shortage of writing material did, however, sometimes
cause a literary text to be written on the reverse across the fibres: a
famous example is the manuscript of Euripides* Hypstpyk (P. Oxy.
852). It is important to note in this connection that the quantity of
text carried by an ancient book was very small: the maximum
capacity was a substantial dialogue of Plato or a book of Thucyd-
ides, and Books I and XVII of the late Hellenistic historian Diodorus
Siculus, which occupy 167 and 177 pages in a modern printed
edition, had to be subdivided.
Finally it should be emphasized that the text as arranged on the
papyrus was much harder for the reader to interpret than in any
modern book. Punctuation was usually rudimentary at best. Texts
were written without word-division, and it was not until the middle
ages that a real effort was made to alter this convention in Greek or
Latin texts (in a few Latin texts of the classical period a point is
placed after each word). The system of accentuation, which might
have compensated for this difficulty in Greek, was not invented
until the Hellenistic period, and for a long time after its invention it
was not universally used; here again it is not until the early middle
ages that the writing of accents becomes normal practice. In
dramatic texts throughout antiquity changes of speaker were not
indicated with the precision now thought necessary; it was enough
to write a horizontal stroke at the beginning of a line, or two points
one above the other, like the modern English colon, for changes
elsewhere; the names of the characters were frequently omitted.
The inaccuracy of this method, and the state of confusion to which
texts were soon reduced by it, may be seen from the condition of
the papyri containing Menanders' Dyscolus (P. Bodmer 4) and
Sicyonius (P. Sorbonne 72, 2272, 2273). Another and perhaps even
stranger feature of books in the pre-Hellenistic period is that lyric
verse was written as if it were prose; the fourth-century papyrus of
Timotheus (P. Berol. 9875) is an instance, and even without this
valuable document the fact could have been inferred from the tradi-
tion that Aristophanes of Byzantium (c. 257-180 B.C.) devised the
Antiquity 5
colometry which makes clear the metrical units of the poetry (Dion.
Hal, de comp. verb. 156, 221). It is to be noted that the difficulties
facing the reader of an ancient book were equally troublesome to
the man who wished to transcribe his own copy. The risk of mis-
interpretation and consequent corruption of the text in this period
is not to be underestimated. It is certain that a high proportion of
the most serious corruptions in classical texts go back to this period
and were already widely current in the books that eventually
entered the library of the Museum at Alexandria.
The increase of the book trade made it possible for private indi-
viduals to form libraries. Even if the tradition that sixth-century
tyrants such as Pisistratus and Polycrates of Samos possessed large
collections of books is discounted (Athenaeus 1.3A), it is clear that
by the end of the fifth century private libraries existed; Aristo-
phanes pokes fun at Euripides for drawing heavily on literary
sources in composing his tragedies (-^"0^943), and his own work,
being full of parody and allusion, must have depended to some
extent on a personal book collection.
There is no trace of any general library maintained at the public
expense at Athens, but it is likely that official copies of plays
performed at the leading festivals such as the Dionysia were kept at
the theatre or in the public record office. Pseudo-Plutarch {Lives of
the ten orators 841F) ascribes to the orator Lycurgus (c. 390-324 B.C.)
a proposal to keep official copies in this way, but the need would
probably have arisen earlier. We know that after the original
performance plays were revived from time to time. New copies of
the text must have been needed for the actors, and if they had been
obliged to obtain these by a process of transcription from private
copies it would be surprising that an almost complete range of plays
survived into the Hellenistic age.
The advance of education and science in the fourth century made
it only a matter of time before academic institutions with their own
6 Scribes and Scholars
signs: apart from the obelos wefindthe diple>, which indicated any
noteworthy point of language or content; the dotted diple
(TTpiCTiyfxvr}) > referred to a verse where Aristarchus differed in
his text from Zenodotus; the asteriskos -X- marked a verse incorrectly
repeated in another passage; the asteriskos in conjunction with the
obelos marked the interpolation of verses from another passage; and
finally the antisigma 3 marked passages in which the order of the
lines had been disturbed (Plates I and II).
It is natural that a complicated system of this kind, which had the
drawback that the reader wishing to discover a scholar's reasons for
placing a sign at any given point had to consult another book,
commended itself to scholarly readers only. No more than a tiny
proportion of the surviving papyri, about fifteen of more than six
hundred, display them. In the medieval manuscripts of the tenth
century and later they are usually omitted; but there is one famous
and important exception to this rule, the tenth-century Venetian
manuscript of the Iliad (Marc. gr. 454), which preserves a vast
collection of marginal scholia. As the commentary on an author was
written in the margins by this date, and not in a separate book,
there was less incentive to transcribe the signs; but fortunately the
scribe of the Venice manuscript was determined to copy what he
found in his exemplar without omission. Consequently the book
shows a great number of the conventional signs, and it is by far the
most complete and reliable source of our knowledge of this feature
of the work carried out by the Alexandrians. It does not, however,
always agree exactly in the use of the signs at the points where it
can be compared with a papyrus, and there are signs which are not
linked with a corresponding note in the scholia.
Though the Homer commentaries of Aristarchus and his col
leagues are lost, enough of them can be reconstructed from the
extant scholia, which are more copious than those on any other
Greek author, to allow us to form a good judgement of the scholarly
methods of the time. It is clear that many copies of the Homeric
text reached the Museum from widely different sources; the scholia
refer to texts coming from such places as Massilia, Sinope, and
Argos. These were sifted and evaluated by the scholars, but it is not
clear which text, if any, was taken to be the most authoritative. The
12 Scribes and Scholars
The great age of Alexandrian work occurred in the third and second
centuries; in the early part of the period the Museum was un-
rivalled. After a time, however, the rulers of Pergamum decided to
challenge this position by founding a library of their own. The
scheme is primarily associated with the name of king Eumenes II
(197-159 B.C.): vast buildings were erected, and excavation by
German archaeologists in the last century brought to light some
sections of the library. Much less is known of the Pergamene library
than of the Alexandrian. The librarians clearly undertook biblio-
graphical studies on a large scale, and literary men found it useful to
consult their work along with that of the Alexandrians (Athenaeus
8.336D, Dion. Hal., deDinarcho 1). But the Pergamene scholars are
not credited with editions of the classical authors and appear to
have confined themselves to short monographs on specific points,
sometimes directly in controversy with the Alexandrians. Their
interests were not exclusively literary; Polemon (c. 220-160 B.C.),
though he collected examples of parody, was first and foremost a
student of topography and inscriptions; these important topics of
historical scholarship had both remained outside the usual range of
studies undertaken in the Museum. The most famous name linked
with Pergamum is that of Crates (c. 200-c. 140 B.C.). He is known to
have worked on Homer; some of his proposals for emending the
text are preserved in the scholia, and he paid special attention to
geography in Homer, attempting to reconcile it with Stoic views on
the subject. He was also the first Greek to give lectures on literary
subjects in Rome (see p. 20).
The Stoics gave a good deal of attention to literature. To them an
important part of interpreting Homer was the application of
allegorical explanations, and one of their treatises on this, attributed
Antiquity *7
Although written records may have existed from very early times,
Latin literature did not begin until the third century B.C. Inspired by
Antiquity *9
The name Aelius is not in doubt, and his interest in Plautus and
the elucidation of archaic texts would naturally involve him in
scholarship of the Alexandrian type. Although Plautus is a far cry
from Homer, the nature of his text and the circumstances of its
transmission presented problems similar to those that had
exercised Hellenistic scholars and for which their critical methods
had an obvious relevance. Plautus' text needed to be standardized:
there was a mass of spurious plays, and the genuine ones contained
22 Scribes and Scholars
is clear that Cicero could depend on him to provide all the services
of a high-class publisher. Atticus would carefully revise a work for
him, criticize points of style or content, discuss the advisability of
publication or the suitability of a title, hold private readings of the
new book, send out complimentary copies, organize its distribution.
His standards of execution were of the highest and his name a
guarantee of quality.
From the exchange of letters between Cicero and Atticus we can
get a good idea of the casual and fluid nature of publication in the
ancient world. There was no copyright or royalty (hence the
importance of literary patronage) and private circulation could eas-
ily pass by degrees into full-scale publication; an author was able to
incorporate changes into a text he had already published by asking
his friends to alter their copies, but other copies would remain unal-
tered. Cicero drastically reshaped his Academica when Atticus was
in the process of having copies made and consoled him for the
effort wasted with the promise of a superior version. But copies of
the first draft were in existence; both 'editions' survived, and we
have a more substantial part of the first than of the second. Cicero
also protests that his Oratio in Clodium et Curionem, of which frag-
ments have survived in some scholia, was published without his
consent. In the Orator (29) he had incorrectly attributed some lines
of Aristophanes to Eupolis and asked Atticus to rectify the mistake
quickly in all copies (Att. I2.6a.i). In this case he succeeded in
correcting the tradition that has come down to us, but he was not so
lucky when in the Republic (2.8) he wished to alter Phliuntii (%s he
had wrongly called the inhabitants of Phlius) to Phliasii\ the sole
manuscript of the work that has survived still has Phliuntii, and it is
the modern editor who makes the change that Cicero requested.
The intellectual decline which had begun in the second century was
accelerated by the economic breakdown and political chaos of the
32 Scribes and Scholars
the book must have come from the early Christians; for while the
pagan codex was a rarity in the second century, the codex form was
already universal for biblical texts.
The advantages of the codex over the roll were many: it was
handier, more capacious, easier to consult, and it may have cost
rather less to produce. Reference was made still easier by number-
ing the pages, and the addition of a list of contents guarded against
forged interpolations and other interference with the text. These
were important considerations in the days when much of life
revolved around the authoritative texts of the Scriptures and the
Code. The importance of the codex for religion and law is obvious.
It had a relevance for literary texts too: a book which could hold the
contents of several rolls meant that a corpus of related texts, or
what was considered the best of an authors work, could be put
under one cover, and this was attractive to an age which was
inclined to trim its intellectual heritage to a manageable form.
The change from roll to codex involved the gradual but whole-
sale transference of ancient literature from one form to another.
This was the first major bottle-neck through which classical
literature had to pass. It must have been somewhat reduced in the
process, but the losses are not easily specified or assessed. There
was the danger that little-read works would not be transferred to
codex form, and in time their rolls would perish. A voluminous
author, if some of his rolls were not available at a critical moment,
might never recover his missing books.
Since some of the earliest surviving books of antiquity are parch-
ment codices of the fourth century, it may be appropriate to mention
at this point the separate question of the main scripts used in
Roman times for the production of books. These were Square
Capitals, Rustic Capitals, Uncial, Half-uncial. The only manuscripts
written throughout in Square Capitals are a few imposing manu-
scripts of Vergil; this script, modelled on the monumental style of
inscriptions, seems to have been introduced as a deliberate refine-
ment for de luxe copies of Rome's national poet. It is therefore
somewhat unfortunate that the standard and elegant capital book-
hand of antiquity, because of its comparatively less formal lines
when set beside this monumental script, should traditionally be
36 Scribes and Scholars
called 'Rustic Capital* (Plate IX), and this charming but rather mis-
leading name is now giving way to 'Canonized' or 'Classic Capital'
or plain 'Capital' script. The earliest specimens we can date are the
Gallus papyrus (Cairo, P. Qar Ibrim, c. 50-20 B.C.) and the fragment
of a poem on the battle of Actium (Naples, P. Here. 817), written
between the event it describes (31 B.C.) and the destruction of
Herculaneum (A.D. 79), where it was found. This hand continued in
much the same form down to the early sixth century; famous manu-
scripts in this script are the codex Bembinus of Terence (Vat. lat.
3226) and the great codices of Vergil, the Mediceus, Palatinus, and
Romanus. The other bookhands of the Roman period came into
being as the cursive forms of everyday writing were refined and
standardized by reference to the calligraphic bookhands. Whether
the dominant parent in the creation of Uncial was Rustic Capital, as
some think, or Cursive, this handsome rounded script emerged as a
fully developed hand in the fourth century and lasted until the
ninth. An early example is the Vatican palimpsest of the Derepuhlica
(Vat. lat. 5757 of the late 4th or early 5th cent., Plate X); one of the
finest is the fifth-century Puteanus of Livy's third decade (Paris lat.
5730, Plate XI). Further development from cursive, and in par-
ticular from the later, minuscule cursive, led to the creation of the
first minuscule bookhand, Half-uncial. There are a number of
classical texts written in this script, mainly papyri, but it was pre-
dominantly used for Christian books.
pagan writer and administrator, who made a moving plea for the
restitution of the Altar of Victory which had been removed from the
Curia. In 394 the leader of the last pagan resistance, Virius Nico-
machus Flavianus, was defeated by Theodosius and committed
suicide in the old tradition. At the centre of the pagan opposition in
the West were the Roman senators, who recaptured for a time the
spirit of their ancestors and rallied to the defence of their traditions
and heritage.
A vivid and sympathetic memorial to this movement is still
extant in Macrobius' Saturna/ia. The relevance of this learned
symposium lies in the setting and dramatispersonae. In the year 384,
on the occasion of the Saturnalia, a number of cultivated upper-
class Romans meet on successive days in the houses of Vettius
Agorius Praetextatus, Virius Nicomachus Flavianus, and Sym-
machus, and have much learned talk about religion, history, philo-
logy, and in particular their great pagan poet Vergil. Among those
present are other known opponents of Christianity. Servius is there
as a representative of professional scholarship, a little over-awed by
the company. We know that Praetextatus had died in 384,
Flavianus in 394; Macrobius has nostalgically recreated the great
pagan society of the past as a framework for his learned compilation
and we see its members, before their world had crumbled around
them, discussing the minutiae of Roman life and literature with the
sophisticated learning of the great Romans of the Republic.
Fortunately the triumph of Christianity did not remove the need
for readable texts of the pagan authors. Christians who were hostile
to pagan literature found themselves in an acute dilemma. It wras
clearly ill suited to be the basic stuff of Christian education. The
poets were polytheistic and the tales they told about their gods, and
particularly about the father of them all, were usually devoid of
edification or downright immoral; Roman rhetoric, though it could
be useful if employed in the right cause, encouraged glibness in
speech and argument out of keeping with simple piety; even the
philosophers, who had so much to offer to the Christian thinker,
also contained much that was inimical to religious faith and the
Christian way of life; the magnitude of the pagan achievement in all
spheres of human activity, of which its written as well as its material
3 Scribes and Scholars
the obligations of polite society and his own highly developed sense
of style made it difficult for him to turn over to the less sophist-
icated diet of Christian literature. The Roman educational system,
authors and gods and all, continued until the monastic and
episcopal schools were able to replace it with an education which,
however much it owed to the traditional system, was essentially
Christian in direction and purpose.
X. THE SUBSCRIPTIONS
means to this end was the demonstration that some of the import-
ant concepts of the new faith could be discussed in terms borrowed
from the classical philosophers, especially the Stoics and Plato. The
fusion of Greek and Christian thought in Justin and Clement
exemplifies this attitude.
Early church fathers of the highest authority were content that
Christians should read some pagan texts during their education.
When Saint Gregory Thaumaturgus attended Origen's school at
Caesarea in 233-8 he found his master encouraging pupils to read
classical literature, and especially the philosophers; only those
authors who denied the existence of a deity or a divine providence
were to be avoided (Migne, PG IO.IO88A, 1093A). It should be noted
that Origen's willingness to learn from pagan culture extended to
the realm of textual criticism. The interpretation of the Old
Testament had become a matter of controversy, since the Septua-
gint was at variance with some other early Greek versions, and diffi-
culty arose if precise interpretation of a passage was required.
Origen adapted the system of marginal signs used by the Alex-
andrian critics to the Old Testament; an obelus marked a passage
found in the Greek but not in the Hebrew, and an asterisk passages
in which the Hebrew agreed with translations other than the
Septuagint. In his Hexapla Origen went further and devised a
method of presenting the Hebrew text and the translations in
parallel columns. The successive columns were the original
Hebrew, the Hebrew in Greek letters, the Greek translations of
Aquila, Symmachus, the Septuagint, and Theodotion. The resulting
book, a cumbrous anticipation of a modern apparatus criticus, must
have been enormous, and no doubt partly for that reason it has not
come down to us in its original form, except that fragments of a five-
column version omitting the text in Hebrew characters survive as
the lower script in a Milan palimpsest (Ambros. S.P. 11.251, ohm O.
39 sup.).
The outlook of the fathers of the fourth century was no less
liberal. Saint Basil wrote a short treatise advising the young on the
best method of profiting from Greek literature, and Saint Gregory
of Nazianzus criticized the majority of Christians for their complete
rejection of pagan works, some of which he believed to be useful
50 Scribes and Scholars
book; yet copious marginal scholia are not commonly found until
the ninth century.
In t! lis connection it should perhaps be mentioned that Pro-
copius of Gaza (c. 460-c. 530) is supposed to have invented a form
of literature that bears some resemblance to scholia, namely the
catena, a running commentary on a book of the Bible which puts
together the opinions of several previous interpreters, normally
with verbatim quotations of their arguments. This invention
marked a new stage of biblical studies; but whether the catena is to
be regarded as a forerunner of classical scholia or as an imitation of
them is a question that has not yet been answered.
The last feature of this period which merits discussion is the
progressive narrowing of the range of literature normally read.
After the third century it becomes more and more uncommon to
find any educated man showing knowledge of the texts that have
not come down to the modern world. To explain this fact
Wilamowitz formulated the theory that in the second or third
century a school syllabus was selected by a prominent school-
master, and this became so influential that all schools adopted it.
With the general decline of culture and impoverishment of the
empire no texts outside this range were read and copied often
enough to be guaranteed survival. To take an example: seven plays
by Aeschylus and seven by Sophocles were selected, and because
of this no others have come down to us; nine or ten plays of
Euripides were chosen for reading in school, but in this case a
lucky chance led to the survival of a single manuscript containing a
number of other plays. Though the theory has much attraction
there is reason to think that it presents too schematic a view of the
history of texts. An initial objection is that there is no positive
evidence as to the identity of the schoolmaster in question. One
possible candidate would be Eugenius, who in the fifth century
wrote on the colometry of fifteen plays. If this figure is correct it
hints already at a selection of nine plays from Euripides and three
from each of the other dramatists; but the reduction of the set
books to the number of three is more probably a feature of the
revised curriculum of late Byzantine schools. However, when so
much remains unknown, it would be wrong to lay emphasis on our
54 Scribes and Scholars
have been used in the imperial archives from the middle of the
eleventh century.
The transliteration of old uncial books into the new script was
energetically undertaken by the scholars of the ninth century. It is
largely owing to their activity that Greek literature can still be read,
for the text of almost all authors depends ultimately on one or more
books written in minuscule script at this date or shortly after, from
which all later copies are derived; the quantity of literature that is
available to us from the papyri and the uncial manuscripts is only a
small proportion of the whole. In the process of transliteration
mistakes were sometimes made, especially by misreading letters
that were similar in the uncial script and therefore easily confused.
At many points in Greek texts there are errors common to all the
extant manuscripts which appear to derive from the same source,
and this source is usually taken to be a ninth-century copy. A
further assumption generally made is that one minuscule copy was
made from one uncial copy. The uncial book was then discarded,
and the minuscule book became the source of all further copies.
The theory has a certain a priori justification on two grounds, since
the task of transliteration from a script that was becoming less and
less familiar would not be willingly undertaken more often than was
absolutely necessary, and there is at least some likelihood that after
the destruction of the previous centuries many texts survived in one
copy only. But these arguments do not amount to proof, and there
are cases which can only be explained by more complicated hypo-
theses. In the tradition of Plato one manuscript (Vienna, supp. gr.
39) differs greatly from all others in its errors, and it is difficult to
believe it derived from the same ninth-century exemplar; it may
derive from the transliteration of a different uncial book, so that at
least two old books would seem to have survived the Dark Ages. A
confirmation of this is that when a Greek text has been translated
into an Oriental language at an early date, perhaps the fifth century,
the readings which are characteristic of the Oriental translation
may occur also in a small group of the Greek manuscripts. This is
true of the Armenian version of some of Plato's dialogues, the
Arabic version of Aristotle's Poetics, and the Syriac translation of
Saint Gregory of Nyssa's De virginitate. Another argument pointing
The Greek East 61
in the same direction can be drawn from the difficulty which arises
in the study of the manuscripts of some texts that were very
frequently read during the Middle Ages, such as the Euripidean
plays included in the school curriculum. Here the relation of the
manuscripts cannot be established precisely by the usual method,
since they do not fall into clearly defined groups that coincide
regularly in error. This situation presupposes that medieval scholars
and schoolmasters frequently compared their own copy of the text
with others and made alterations or added variant readings above
the line; this process is known as contamination or horizontal trans-
mission. In such cases it may be that more than one copy survived
the Dark Ages to be transcribed, so that two or more trans-
literations took place; alternatively, only one transliteration was
made but this copy was deposited in some central place where it
was consulted by interested readers and received as marginal addi-
tions the variant readings that had been found in other copies. It is
easy to imagine, though there is no external evidence for the
assumption, that such deposits of semi-official copies took place in
the library of the academy set up by Bardas. It is also possible that
similar copies existed in the patriarchal academy, for there is a
manuscript of Plato's Laws (Vat. gr. i) written in the early tenth
century with marginal variants added in the next century by a
scholar who refers to these additional readings as coming from 'the
patriarch's book*; unfortunately we cannot be sure whether this was
a private copy or part of the library in the seminary.
The Bardas school was founded under favourable conditions, and
was probably the centre of a lively group of scholars concerned to
recover and disseminate classical texts of many different kinds. Yet
it does not seem to have had the influence that might have been
expected, for there is very little reference to it at later dates. Its
professors are completely overshadowed by their contemporary
Photius (c. 8io-c. 893), a man of remarkable attainments who is
perhaps as important for his position in the Church and the affairs
of the government as for his devoted encouragement of learning.
Twice he held the patriarchal throne of Constantinople (858-67,
877-86), and in these years some of the negotiations which led to
schism between the Eastern and Roman churches took place; it is
62 Scribes and Scholars
that Photius summarizes many books that are now lost: that applies
for example to some twenty of the thirty-three historians he
discusses. Much can be learnt of the interests of a prominent
Byzantine figure of the time: in the secular texts historians are most
numerous, but among others there are orators, novelists, and
compilers of Atticist dictionaries. The latter are significant, for they
show the author's concern with stylistic considerations, which is
also shown by his frequent brief characterizations of the style of an
author; the desire to write and appreciate a good Attic style was
never far from the thoughts of Byzantine literati. The breadth of
Photius' interests is enormous. That a pious man and future
patriarch should bother to read the Greek novelists is surprising; he
enjoyed them linguistically, but could not bring himself to be
favourable to their contents. It is also notable that he read heretics
and anti-Christian writers; this is incidentally a strong argument
against the notion that the ecclesiastical authorities attempted to
impose a censorship. Philosophy is not well represented in the
Bibliotheca, but there is evidence of his knowledge in this field else-
where in his works. The most serious limitation of taste shown in
the book is the almost complete absence of poetry. One wonders
whether in this respect it is a true record of Photius* own reading.
We know from his letters that he had read Aristophanes, Plutus and
Aeschylus, Prometheus Vinctus\ these and other school texts he might
well omit because they were set books already known to his
brother. But it looks as if other poetry did not interest him much,
and perhaps it had little appeal to intellectuals of his generation.
Another work deserving mention here is his Lexicon, the first
complete copy of which was discovered in 1959 in a remote
monastery in Macedonia. It is a typical work of its class, valuable for
its brief quotations of classical texts not now available. The purpose
was to amalgamate and revise various existing books of the same
kind; in the Bibtiotheca Photius remarks how useful such a book
would be. In his Atticism he was moderate and willingly admitted
words from poetic sources if they seemed the most expressive
means of conveying a notion. These quotations from the poets do
not imply a reading of the full text, but were probably drawn as such
from his sources. Besides this lexicon he was partly responsible for
64 Scribes and Scholars
With the death of Arethas some time in the thirties of the tenth
century a new period begins, in which eminent scholars and biblio-
philes are much more difficult to identify. Some stimulus to learning
was given by the activity of the erudite emperor Constantine VII
Porphyrogenitus (913-59). During a long period of enforced semi-
retirement he compiled various manuals of statecraft which survive
partially. These took the form of encyclopedic compilations based
on a very wide range of historical sources, and as such are of some
importance to classical scholars, since many of the texts do not
survive elsewhere. This great activity of Constantine was doubtless
66 Scribes and Scholars
Tzetzes are the latest Byzantines of whom we can say with certainty
that they could read more classical poetry than we can.
The reason for this lies in an event of the utmost importance
which Michael Choniates lived to see, the capture and sack of Con-
stantinople by the Fourth Crusade in 1204. Great damage was
done, and there is little doubt that libraries suffered severely. For
the historian of literature this sack of the city was a greater disaster
than the more famous one of 1453. In 1204 the rare texts men-
tioned in the previous paragraph were destroyed; at any rate there
is no trace of them when the seat of government was restored to
Constantinople in 1261 after the fall of the Latin kingdom. If the
events of 1204 had not taken place these texts might well have
found their way to the West through the agency of the numerous
Italian visitors and book collectors who went to Greece and
brought back manuscripts. By the time that the city fell into the
hands of the Turks little remained to be discovered by the
collectors; the only substantial and well-attested loss is recorded in
the statement of Constantine Lascaris that a complete copy of the
Universal History by Diodorus Siculus was destroyed by the Turks.
While the capital was occupied by the Franks and most of Greece
was parcelled out among Western barons, the Byzantine adminis-
tration dragged out a precarious existence at Nicaea, preserving the
empire's possessions in Asia Minor. Despite the drastic reduction in
the wealth and power of the empire this period of exile in Nicaea
was by no means one of the worst for literary studies. The emperors
John Vatatzes and Theodore Ducas Lascaris were concerned to
promote schools and libraries and eventually built up quite a tradi-
tion of secondary education. Little is known in detail, since very few
manuscripts can be identified as having been written in the Nicaean
empire, but it seems clear that poets and orators were studied, and
some of Theodore's own letters display cultivated and scholarly
attitudes. Other scholarly work was done by the monk Nicephorus
Blemmydes (c. 1197-c. 1272), who wrote on many topics, including
logic, physics, and geography, and made a journey to parts of the
old empire now under Latin control in search of books that could
not be found in Asia Minor. This is one of the few short periods in
which literary studies flourished outside the capital. It is also
The Greek East 73
The sixth century saw the final collapse of what remained of the
Roman empire in the West. In Italy the relatively enlightened rule
of Theodoric (493-526) was given distinction by the two most
notable figures of the transition period from the ancient to the
medieval world, Boethius and Cassiodorus; but it was followed by
the destruction of the Ostrogothic kingdom by the Byzantines and
a spectacular cultural decline. The provinces were to fare little
better. North Africa, now in Vandal hands, was soon to pass beyond
the pale of Western culture; some of its literary achievement, such
as the Latin Anthology, was transmitted in time to Europe and so to
posterity. Spain, prey to external attack and internal strife, was to
see a revival of Visigothic culture in the later sixth and early seventh
centuries, reaching a modest peak in Isidore of Seville, but it too
was to succumb in the early eighth century to the Moslem invaders.
Though traces of the older Roman culture lingered on among the
upper classes in Gaul, the Frankish Merovingian dynasty founded
by Clovis (481-511) was grotesquely ill-suited to foster any cultural
continuity.
The ravages of conquest and barbarism made the prospects for
cultural life extremely bleak, and within the narrowing world of
culture the place allotted to classical Latin literature was insecure.
Education and the care of books were rapidly passing into the
hands of the Church, and the Christians of this period had little
time for pagan literature. Decimated by the continued destruction
of war, faced by hostility or neglect at the hands of the new intel-
lectuals, the Latin classics seemed to have a slim chance of survival.
But the fundamental condition for their survival obtained: there
were still books. We do not know how much survived of the
8o Scribes and Scholars
9 Monasteries
0 100 2 0 0 Miles
liberal influences when the time was ripe, and reading could in any
case not be carried on without books.
While Italy had enjoyed its late Renaissance in the first half of the
sixth century, the blossoming of Visigothic culture in Spain did not
come until the late sixth and early seventh centuries. This revival
largely owes its place in the history of classical culture to the
achievement of its greatest writer, Isidore of Seville (c. 570-636).
Owing to the phenomenally rapid spread of his works throughout
Europe, an amazing accomplishment for the pre-Carolingian
period, Isidore was quickly established as one of the most influential
agents in the transmission and elucidation of ancient learning. His
Etymologies was at the same time the last product of the Roman
encyclopedic tradition and the starting-point for most medieval
compilations; its most frequently copied section, the first three
books covering the subjects of the trruium and quadrivium, must
have contributed enormously to the consolidation of the medieval
educational system. This systematically arranged encyclopedia,
packed with information and misinformation on every topic from
angels to the parts of a saddle, descends so often into false etymo-
logizing and the uncritical parade of absurd bric-a-brac that it can-
not be read without a smile. But Isidore wins one's respect, and
even affection, by his obvious appreciation of knowledge for its own
sake. Hostility to pagan literature is explicit in some of his public
pronouncements, and he was more at home in the neutral pages of
the scholiast and compiler than in the classical authors themselves,
whom with a few exceptions he quotes at second-hand; but his
curiosity knew no barriers and he took for granted the independent
value of profane culture. When he culls from the fathers of the
Church the scraps of classical poetry and pagan learning that they
contain and re-allocates them to their proper place in the tradi-
tional system of knowledge, this bishop is paradoxically recreating
in a resecularized form the basic structure of ancient learning.
However, the process that has preserved Latin literature for us
could not begin until there was a more sympathetic and more
positive attitude to classical authors than generally obtained on the
continent in the Dark Ages. Christians still lived in the shadow of
pagan literature; its achievement dwarfed their own, and its threat
The Latin West 85
to morals and doctrine was a real one. This was to change when
Latin culture was transplanted to a distant soil, where those eager
to learn the language of the Church could turn to antiquity without
any sense of inferiority or fear, since rivalry was out of the question
and men at large were protected from the dangers of ancient pagan-
ism by simple ignorance of the Latin language. But this spirit did
not percolate on any scale to the continent of Europe until the
Carolingian Revival in the late eighth century, and in the meantime
much of classical literature perished.
Although few ages are so dark that they are not penetrated by a
few shafts of light, the period from roughly 550 to 750 was one of
almost unrelieved gloom for the Latin classics on the continent;
they virtually ceased being copied. Among the mass of patristic,
biblical, and liturgical manuscripts that survive from this period
there are precious few texts of classical authors: from the sixth
century we have scraps of two Juvenal manuscripts, remnants of
one of the Elder and one of the Younger Pliny, but at least two of
these belong to the early part of the century; from the seventh
century we have a fragment of Lucan; from the early eighth century
nothing.
The fate that often overtook the handsome books of antiquity is
dismally illustrated by the surviving palimpsestsmanuscripts in
which the original texts have been washed off to make way for
works which at the time were in greater demand. Many texts that
had escaped destruction in the crumbling empire of the West
perished within the walls of the monastery; some of them may have
been too tattered when they arrived to be of practical use, and there
was no respect for rags, however venerable. The peak period for
this operation was the seventh and early eighth centuries, and
although palimpsests survive from many centres, the bulk of them
have come from the Irish foundations of Luxeuil and Bobbio. Texts
perished, not because pagan authors were under attack, but
because no one was interested in reading them, and parchment was
too precious to carry an obsolete text; Christian works, heretical or
superfluous, also went to the wall, while the ancient grammarians,
of particular interest to the Irish, often have the upper hand. But the
toll of classical authors was very heavy: amongst those palimpsested
86 Scribes and Scholars
we find Plautus and Terence, Cicero and Livy, the Elder and
Younger Pliny, Sallust and Seneca, Vergil and Ovid, Lucan, Juvenal
and Persius, Gellius and Fronto. Fronto survives in three pal-
impsests, fated always to be the underdog. Among the texts that
have survived solely in this mutilated form are some of outstanding
interest, such as the De republica of Cicero (Vat. lat. 5757, Plate X)
written in uncials of the fourth or fifth century and covered at
Bobbio in the seventh with Augustine on the Psalms, a fifth-century
copy of the Deamicitia and Devitapatris of Seneca (Vat. Pal. lat. 24)
which succumbed in the late sixth or early seventh century to the
Old Testament, and a fifth-century codex of Sallust's Histories
(Orleans 192 + Vat. Reg. lat. 1283B + Berlin lat. 40 364) which, in
France and probably at Fleury, was supplanted at the turn of the
seventh century by Jerome. Other important palimpsests are the
Ambrosian Plautus (Ambros. S.P. 9/13-20, ohm G. 82 sup.) and the
Verona Livy (Verona XL (38)), both of the fifth century.
The classical revival of the late eighth and early ninth centuries,
without doubt the most momentous and critical stage in the trans-
mission of the legacy of Rome, was played out against the back-
ground of a reconstituted empire which stretched from the Elbe to
the Ebro, from Calais to Rome, welded together for a time into a
political and spiritual whole by the commanding personality of an
emperor who added to his military and material resources the bless-
ing of Rome. Although the political achievement of Charlemagne
(768-814) crumbled in the hands of his successors, the cultural
movement which it fostered retained its impetus in the ninth
century and survived into the tenth.
The secular and ecclesiastical administration of a vast empire
called for a large number of trained priests and functionaries. As the
only common denominator in a heterogeneous realm and as the
repository of both the classical and the Christian heritage of an
earlier age, the Church was the obvious means of implementing the
educational program necessary to produce a trained executive. But
under the Merovingians the Church had fallen on evil days; some of
the priests were so ignorant of Latin that Boniface heard one carry-
ing out a baptism of dubious efficacy in nomine patria et filia et spiritus
sancti {Epist 68), and knowledge of antiquity had worn so thin that
the author of one sermon was under the unfortunate impression
that Venus was a man. Reform had begun under Pippin the Short;
but now the need was greater, and Charlemagne felt a strong
The Latin West 93
court to the monasteries and cathedrals, was the need for books;
these were produced on an unprecedented scale, in a flurry of activ-
ity which salvaged for us the greater part of Latin literature.
of Charlemagne about the year 790. The list includes Lucan, Statius'
Thebaid, Terence, Juvenal, Tibullus, Horace's Ars poettea, Claudian,
Martial, some of Cicero's speeches (the Verrines, Catilinarians, Pro
rege Deio/aro), and a collection of orations excerpted from_the Bella
anH Hi\fntv'/iP of Spltnqf Snrnn of the works in this impressive list
may have been ancient codices in capitals or uncials. The presence
of other books in the library may reasonably be inferred from other
evidence. Such rare works as Grattius' Cynegetka and Statius' S/'/vae
are quoted in the court poetry of the period, and Alcuin implies in a
letter to Charlemagne that a copy of the Elder Pliny would be to
hand. Paul the Deacon made his abridgement of Festus expressly as
a gift for Charlemagne's library, and we know that the Liber median-
a/is of Quintus Serenus was copied by imperial command. Some of
the books known to have been produced in the palace scriptorium
are remarkable both for the quality of their texts and for their
superb execution. Our best manuscripts of Lucretius and Vitruvius
(Leiden, Voss. Lat. F. 30; London, Harley 2767) were written there
about the year 800.
It is clear from the evidence that abbots and bishops who had the
right connections could enrich their libraries with copies taken
from the books in the palace library; and after Charlemagne's death,
although the details of the way in which this library was dispersed
are unknown, many of the books found their way to monastic
libraries. There is a remarkable correlation between the items in the
palace list and the works known to have been copied at Corbie
about the middle of the century: the unique Corbie manuscript
which contains the well-known collection of speeches and letters
taken from SallustCTVat. j a t . ^864])is the most striking example.
Another tell-tale item is the group of three Ciceronian speeches
which reappear in the important codex Holkhamicus, now in the
British Museum (Add. 47678): this was written at Tours in the early
years of the ninth centuryAlcuin was abbot of Saint Martin's at
Tours from 796 to 804and it can hardly be doubted that its parent
had been the copy in the palace library. Again, one of the most
famous manuscripts of Livy is the codex Puteanus of the third
decade (Paris lat. 5730, Plate XI), written in the fifth century in Italy
and the source of all the later manuscripts: this was copied at Tours
The Latin West 97
about the year 800 (its copy is Vat. Reg. lat. 762, Plate XIII) and
again at Corbie about the middle of the ninth century (Laur. 63.20),
a pattern which strongly suggests that the home of the Puteanus
was to be found in the palace. The importance of the palace
scriptorium seems to have continued under Charles' successor
Louis the Pious (814-40), for the manuscripts which have been
attributed to it during the period of his rule include such outstand-
ing books as the Bamberg manuscripts of Seneca's Letters (Class.
46) and Pliny's Natural History (Class. 42).
Books are naturally attracted to centres of power and influence,
like wealth and works of art and all that goes with a prosperous
cultural life. Some arrive as the perquisites of conquest, or as the
gifts that pour in unasked when the powerful have made their
wishes plain, some in response to the magnetic pull of an active and
dynamic cultural movement. Others were actively sought out by
those promoting the educational and cultural aims of the revival.
There was such a break in the copying of the classics in the Dark
Ages that many of the books that provided the exemplars from
which the Carolingian copies were made must have been ancient
codices, and this immediately raises a fundamental question: where
did all the books that have salvaged so much of what we have of
Latin literature come from? As far as we can tell from the evidence
available, the total contribution of Ireland and England, Spain and
Gaul, was small in comparison with what came from Italy itself,
from Rome and Campania and particularly, it would seem, from
Ravenna after its capture by the forces of Charlemagne. Nor did the
wholesale transference of classical texts to northern Europe exhaust
the deposits in Italy, for Italy continued, down to the end of the
Renaissance and beyond, to produce from time to time texts which,
as far as we can tell, had been unknown north of the Alps.
Gathering impetus with each decade, the copying of books went
on apace through the length and breadth of Charlemagne's empire.
Such ancient classical manuscripts as could be found, with their
imposing majuscule scripts, were transformed, often at speed, into
minuscule copies, and these in time begot further copies, branching
out into those complex patterns to which the theory of stemmatics
has reduced this fascinating process. The routes by which texts
98 Scribes and Scholars
fifth decade of Livy (Vienna lat. 15) and which had earlier been
circulating in the Low Countries; the main manuscript of Seneca's
De beneficiis and De dementia (Vat. Pal. lat. 1547), written in
northern Italy about the year 800; the codex Palatinus of Vergil
(Vat. Pal. lat. 1631), written in rustic capitals of the late fifth or early
sixth century; a famous palimpsest from Italy (Vat. Pal. lat. 24),
which had been made up from scraps of some of the oldest surviv-
ing books of antiquity, including codices of Seneca, Lucan, Fronto,
and Gellius.
The importance of the insular foundations of Fulda and Hersfeld
has already been mentioned. From these come the two manuscripts
of Ammianus Marcellinus from which the rest derive, and we owe
the survival of the Opera minora of Tacitus and the De grammaticis of
Suetonius to a manuscript written at either Hersfeld or Fulda and
preserved at the former (Iesi, Bibl. Balleani 8). Besides contributing
important manuscripts to the textual tradition of some authors,
such as the Younger Pliny, Aulus Gellius, Eutropius, and Nonius
Marcellus, Fulda played a dominant role in the history of other
texts: our only surviving medieval manuscript of Valerius Flaccus
(Vat. lat. 3277) was written there; of the two Carolingian manu-
scripts of Columella, one was written at Corbie (Leningrad, Class.
Lat. F. v. 1), the other at Fulda (Ambros. L. 85 sup.); the prime
source for the Historia Augusta (Val. Pal. lat. 899), which was itself
written in northern Italy, must have reached Fulda, for a direct copy
of it (Bamberg Class. 54) is of Fulda origin; books 1-6 of the Annals
of Tacitus have come down to us in a manuscript (Laur. 68.1)
written at Fulda and preserved at Corvey, a daughter-house of
Corbie; finally, to end on a lighter note, while one of the early
manuscripts of the cookery book of Apicius is a show-piece of the
script of Tours (LTrb. lat. 1146), the other (written in a mixture of
Anglo-Saxon minuscule and continental script) points almost
certainly to Fulda (New York, Acad. Med., MS. Safe).
Tours has already been mentioned as the source of some of our
earliest and finest Carolingian manuscripts; to these may be added
the oldest extant copy of Suetonius (Paris, lat. 6115). Tours itself
stood at the head of a chain of abbeys strung along the valleys of the
Loire and the YonneFleury, Ferrieres, Auxerrethat provided a
100 Scribes and Scholars
thick on the shelves of the libraries that their survival was no longer
in question: to this group we can assign Vergil and Horace (the
Satires and Epistles rather than the lyrics, which were less popular in
the Middle Ages), Lucan, Juvenal and Persiiis, Terence, the epics of
Statius, some of the rhetorical and philosophical works of Cicero
(the Letters and Speeches were still rare or unknown), the Catilina
and Jugurtha of Sallust, the Elder Pliny, Justinus, and Vitruvius. The
Elder Seneca and Valerius Maximus were available, as were Aulus
Gellius and the Letters of Seneca, but Gellius and Seneca both
circulated in two separate parts, of which one was much less
common than the other, and at this period complete copies were
rare or non-existent. Quintilian was less common than one would
expect (his place had been usurped by the AdHerennium and the De
inventione), and he too was incomplete; most of the manuscripts
were mutili, though a complete text was to be found in Germany in
the tenth century. Martial and Suetonius were not common, though
Einhard's Life of Charlemagne, thanks to a happy connection with
Fulda, is a brilliant adaptation of Suetonius' literary method and a
milestone in the development of secular biography. Plautus,
Lucretius, Livy, and the Younger Pliny were even scarcer, and the
great age of Ovid was still to come. Some authors existed in so few
copiessometimes only onethat their future was still precarious:
Cicero's Letters, Tacitus, Columella, Petronius, Apicius, Valerius
Flaccus, and Ammianus were all copied at this time, but not on a
scale that would ensure their survival through wars and acts of God
and the less dramatic but ever-present evils of mice and mould; it
was going to need another renovatio to make their position secure.
The few or unique copies of Tibullus and Catullus, Seneca's Traged-
ies and Statius' Sihae, were virtually in hibernation, while
Propertius, Seneca's Dialogues, Apuleius' Golden Ass, much of Taci-
tus, Manilius, Nepos, and Velleius Paterculus had shown no sign of
animation.
One cannot consider these facts without marvelling at the
slenderness of the thread on which the fate of the Latin classics
hung. In the case of many texts a single copy survived into the
Carolingian period, and often a battered one at that. When the
great period of the revival was over, some of the great works of
102 Scribes and Scholars
appetenda sapientia (EpisL i), he alone of the men of his age gives a
foretaste of the Renaissance. He was educated at Ferrieres and
completed his studies at Fulda under the greatest teacher of the
post-Alcuinian period, Hrabanus Maurus (780-856); he returned to
Ferrieres in 836 and was abbot from 842 until his death. His letters
are of great interest; despite his involvement in the affairs of the
world, his correspondence is dominated by his scholarly interests.
Anxious to increase the resources of the library at Ferrieres, which
had been modest enough in his student days, he writes far and wide
in his search for books, to Einhard (who had now left the court and
retired to Seligenstadt), to Tours, to York, to the Pope himself
However, he was not the only manuscript-hunter in the ninth
century: his distinction rests on the fact that he is avid to obtain
manuscripts of works which he already possesses, so that by colla-
tion he can correct and supplement his own text. He succeeded in
filling up some of the gaps in the incomplete Valerius Maximus that
had survived from antiquity by drawing on the rare epitome made
by Julius Paris in the fourth century. The following extract from a
letter written in 847 to a monk at Priim will illustrate his practice
{EpisL 69):
the Viking raids and other disorders of the ninth century and
recovery was slow. Two manuscripts written in Wales, one of the
late ninth century and containing Book I of Ovid's Ars amatoria
(Oxford, Bodl. Auct. F. 4.32, part IV), the other a copy of Martianus
Capella (Cambridge, Corpus Christi College 153) of similar date,
suggest that the Welsh, using continental exemplars, had initiated a
modest revival while their English neighbours were still in disarray.
But thanks to the monastic reforms of Dunstan and Ethelwold and
the establishment of contacts with such centres as Fleury and
Corbie, England began in the tenth century to import books from
the continent and with them the continental script. The most
important manuscript of Cicero's Aratea (Harley 647), written in
France in the Carolingian period, had arrived in England by the end
of the century and soon produced a whole brood of offspring on
English soil. Manuscripts ofJuvenal and Persius survive from tenth-
century England, including an attractive one in insular minuscule
(Cambridge, Trinity College O. 4.10), which must be among the last
of the classical manuscripts written in this script.
lines of Juvenal's sixth satire that are not to be found in any other
manuscript.
as well. When the classical heritage was absorbed into the systems
of contemporary thought, with its strong tendency to allegorize and
elaborate, it was bound to become distorted. It also suffered in other
ways. With so much else to occupy the mind, the wide reading of
ancient authors gave way to the more practical manuals, the
auctores to the artes, and the new grammars and rhetorics that came
into use were often scholastic in character. The classics still
remained a valuable quarry for moral anecdote and could provide a
curious age with information of all sorts; but form and style were no
longer part of the attraction, and matter could be more easily
assimilated when reduced to excerpts and exempb. At the same
time the writers of the twelfth and thirteenth centuries took their
place alongside the ancient authors; although they did not displace
them, the monopoly of the past had been broken.
For these reasons the century that witnessed the final triumph of
the Middle Ages in many fields is not a particularly enticing one for
the classical scholar. Manuscripts pour on to the market, but the
text of those authors who have been copied for generations is
getting more and more corrupt; the proportion of grain to chaff is
getting smaller, and the manuscripts themselves, with their heavy
Gothic appearance, are less alluring than those of previous cen-
turies. Despite all this, the classics survived the tide of scholasticism
and made significant advances where least expected. The heroes
of the period were the builders of the mighty philosophical and
theological systems, but amongst those intent on the organization
of knowledge were some who gave an important place to pagan
literature. Vincent of Beauvais, who died about 1264, is the most
monumental encyclopedist of the Middle Ages; his Speculum maius
was an attempt to put the whole sum of knowledge into one corpus.
Like so many others, he was anti-pagan in principle, but he saw the
value of profane texts and defends his use of them with a good con-
science. He draws heavily on classical authors; Ovid and Seneca far
outdistance the others, Vergil is eclipsed. A large part of his classical
quotations are taken from secondary sources, and the appearance of
fare authors like Tibullus is explained by his dependence on earlier
compilations, in particular theflorilegiumGaUicum.
About 1250 and within a few years of the publication of the
n6 Scribes and Scholars
of the greatest Roman poets had to wait a long time to come to life
aeain; and for his poems we are heavily indebted, as with so many
discoveries to come, to a new phenomenon, the wealthy private
book collector.
The Tragedies were not the only text of Seneca to achieve circula-
tion in northern Europe at this time. The Dialogues reached the
schools of Paris in the first half of the thirteenth century, having
worked their way north from Montecassino. They were known to
John of Garland as early as 1220 and fifty years later, somewhat
tardily but with tremendous excitement, their 'discovery' was
announced by Roger Bacon. Though this text too begins to
circulate again in northern France, among the first to make use of it
were Roger Bacon and John of Wales, both Franciscans and both as
much at home in Oxford as in Paris. They serve to draw attention to
the less spectacular but not inconsiderable contribution to the
promotion of classical studies already being made by the English
friars. Some of the English Franciscans actually compiled in the
thirteenth century a Registrum Hbrorum Angliae, a union catalogue of
books available in English libraries, a remarkable bibliographical
project in which some classical authors were included. John of
Wales's treatises, such as the Communiloquium and Compendi-
loquium, were full of references to the ancients and opened a wide
and flattering window on classical antiquity; they were intended not
only as aids for the teacher and preacher, but also as manuals for
polite conversation. Somewhat later Nicholas Trevet, a Dominican
but again mainly associated with Oxford and Paris, achieved such a
wide reputation for erudition and the exegesis of antique texts that
he received commissions from Italy to write commentaries on Livy
and Seneca's Tragedies. These prepared the way for the classicizing
group of friars who have been shown to have been active in
England in the early fourteenth century. This loosely knit group, of
which the most important are perhaps Thomas Waleys and Robert
Holcot, did much to popularize a knowledge of the ancient world
by introducing classical allusions to illustrate their biblical com-
mentaries and sermons and helping to create an audience with a
taste for ancient history and myth. With his classical scholarship,
seen at its best in his commentary on the first ten books of the De
n8 Scribes and Scholars
civitate dei, completed in 1332, his admiration for the ancients, and
his knowledge of rare texts, Thomas Waleys comes very close to
being a humanist and doubtless owes much of his special quality to
periods spent at Bologna and Avignon. He claims to have seen a
copy of Apuleius' Metamorphoses and can quote from Livy's rare
fourth decade, thanks to a book lent to him by the bishop of
Modena. The fondness for classical learning common to the group
might have developed into humanism had circumstances been
different; as it was, their lack of stylistic sophistication, their
medieval ways of thinking, their profession, and their lack of con-
tact with a leisured highbrow milieu prevented this from
happening; the movement took a different direction and faded out.
Thus more and more was added to the vast body of classical
books and learning which had been accumulating over the
centuries. Classical studies survived and advanced and were
successfully adapted to new tastes and conditions, but in a context
in which they were never really emancipated, could never really
catch fire. It was left to the humanists of the Renaissance, who drew
on this great medieval heritage with curiously little sense of debt, to
exploit what had been achieved in a new and vital way.
Under the Roman empire Italy had been to all intents and purposes
a bilingual country, but with the decline of the empire Greek fell out
of use except in the south of Italy and Sicily, where many towns
were by origin Greek colonies. Cassiodorus' monastery Vivarium
near Squillace is known to have had a collection of Greek books,
but there is no sign that they contributed in any tangible way to the
preservation of Greek. And in all the other parts of Western
Europe, where the language had never been so firmly established, if
indeed it was spoken at all, a knowledge of Greek became an attain-
ment of exceptional rarity throughout the Middle Ages. Even
diplomatic correspondence was sometimes delayed for lack of suit-
ably qualified translators and interpreters. Though the importance
of the language was often recognized, the history of Greek in
The Latin West 119
the Latin West is a series of brief episodes which never led to the
establishment of a lasting school.
The first of these episodes took place in England with the arrival
of the Greek-speaking missionaries Theodore (d. 690) and Hadrian
(A 710)- Interlinear glosses in about half a dozen manuscripts prove
that they taught some Greek in Canterbury, confirming the reports
in Bede's Ecclesiastical History.
The Carolingian revival of the ninth century created some inter-
est in Greek. A few bilingual biblical manuscripts survive, proved by
their handwriting to be products of the Latin world; they are
thought to come from the scriptorium at Saint Gall. In 827 the
Byzantine emperor sent a copy of Pseudo-Dionysius the Areopagite
to the French king (still preserved as Paris gr. 437), which served as
the basis for a translation of this highly popular forgery into Latin. A
few years later the Irishman John Scottus Eriugena used the manu-
script for his own translation of these works, and he also made some
translations from Gregory of Nyssa, Gregory of Nazianzus, and
Maximus the Confessor. But though some of his versions were
widely read he did not create a tradition of Greek learning, and few
other Greek texts were accessible for the time being except
Boethius1 versions of some of Aristotle's writings on logic and a
version of Plato's Timaeus made in the fourth century by Chalcidius.
In Rome a powerful figure at the papal court, Anastasius Biblio-
thecarius, added some historical and theological texts to the stock
of accessible Greek authors.
In the twelfth century the range of translations was increased
substantially. Much of the credit belongs to Burgundio of Pisa
(1110-93), wno n a d spent the years 1135-8 in Constantinople as an
interpreter and returned there later, taking the opportunity to
collect books. Copies of Galen which he used for his translations
can be identified by marginal annotations in his own hand. A more
obscure figure is James of Venice, a canon lawyer whose version of
Aristotle's Analytica posteriora was known to John of Salisbury in
11
59- Slightly better known are the inelegant and literal versions of
Plato, Euclid, and Ptolemy made in Sicily c. 1160 under the aegis of
Henricus Aristippus, archdeacon of Catania (d. 1162), who is said to
nave acquired some manuscripts sent as a gift by the Byzantine
120 Scribes and Scholars
THE RENAISSANCE
I. HUMANISM
and it was this command of the Latin tongue that enabled the
Renaissance man to impress his peers, denounce his enemies,
thunder in defence of creed or city. In turn this led to a more
sympathetic and comprehensive study of all aspects of ancient life
and to that feeling of identification, however illusory, with the men
and ideals of the ancient world which is the mark of neoclassicism.
This attempt to get closer to the classical spirit and to relive and
rethink the past in terms of the present completely transcends the
medieval approach to ancient letters. At last Latin literature was
emancipated from the role for which it had been so badly cast, that
of playing second-fiddle to religion; humanism was fundamentally
secular, and the thin but unbroken tradition of lay education in Italy
had doubtless contributed to this. The humanists were men of the
world, sometimes teachers of grammar or literature, more com-
monly notaries, papal secretaries, chancellors of cities. They were
usually book collectors, often on a large scale, and the growth of
private libraries and a commercial book trade helped to break the
long ecclesiastical monopoly of learning. At the same time the
movement quickly gained a foothold within the Church and soon
humanists were to be found in the highest positions of its hierarchy.
always as clear cut as one would wish, especially as Latin poets are
fond of echoing each other and the nature of poetic diction is such
that coincidence cannot always be ruled out. It has been claimed
that Lovato knew Lucretius, Catullus, the Odes of Horace, the
whole of Tibullus, Propertius, Seneca's Tragedies, Martial, the Silvae
of Statius, Valerius Flaccus, and Ovid's Ibis. Some of the items in
this list have succumbed to a more sceptical scrutiny: the evidence
for Lovato's knowledge of Catullus and Propertius has largely dis-
solved, though Catullus, soon to surface at Verona, was known to
later prehumanists, and the evidence for Lucretius looks dubious;
but precision is impossible and the acceptance that they may not
have known certain texts does not materially affect the importance
of Lovato and his circle. Other members of the group were
acquainted with a similar range of Latin poets, and among the prose
authors they studied evidence has emerged of a strong interest in
Cicero.
A clue to the source of some of Lovato's texts may lie in a manu-
script in the British Library (Add. 19906), which contains inter alia
Justinus' Epitome and a collection of Lovato's poems. At the end of
the Epitome the scribe has copied the subscription that he had
found in his exemplar, and this reveals that the manuscript he was
using was written in the monastery of Pomposa, in the Po delta, just
before 1100. It was one of the books produced when the library was
being enlarged at the instigation of abbot Hieronymus. Whether
or not Add. 19906 was copied by Lovato himself, about 1290, as has
been suggested, this manuscript establishes a link between Lovato
and Pomposa which can be supported by other evidence. Among
the classical texts known from the inventory to have been at
Pomposa as early as 1093 was a great rarity, the Tragedies of Seneca,
and this was probably none other than the extant eleventh-century
'codex Etruscus' of the Tragedies (Laur. 37.13 = E). Its parent may
well have come from Montecassino. Lovato's use of an E-type text
suggests that he had access to the Etruscus or something very close
to it. Thus Pomposa appears to have been one of the libraries on
which the prehumanists were able to draw, the Chapter Library at
Verona was clearly another, and Bobbio, which is known to have
had copies of Lucretius and Valerius Flaccus in the ninth century,
126 Scribes and Scholars
may ultimately explain their knowledge of other texts. But not all
the questions raised by their acquaintance with such a body of
poetry have been answered; when these texts were rediscovered
later, they came to light in France and Switzerland and Germany,
with the Alps between them and the Veneto. Padua is a somewhat
isolated and still obscure chapter in the story of the rediscovery of
antiquity.
Lovato has also left us a short note on the metre and prosody of
Senecan tragedy, remarkable in that it is derived, not from the
medieval manuals, but from an intelligent study of Seneca's own
practice. It was elaborated by his successors and is an indication of
the intense interest which the prehumanists took in Roman
tragedy. He also tried his hand at archaeology, and identified a
skeleton which some workmen had turned up as the remains of the
legendary founder of Padua, the Trojan Antenor, a gorgeous error.
From all this it is clear that something new had begun.
Something of a contrast is provided by another Paduan judge of
the same circle, Geremia da Montagnone (c. 1255-1321), who had
no literary ambitions and trod the well-beaten path of the didactic
florilegist: his Compendium moralium notabilium, probably put
together in the first decade of the fourteenth century, enjoyed a
wide circulation and was eventually printed in Venice in 1505.
Geremia is more typical of his period; but in some respects his com
pendium plants him firmly in the humanist group. His reading is
vast, his excerpts are systematically arranged, with chapter and
verse added, and he seems to be quoting at first hand from the
authors themselves; his notions of chronological sequence are not
bad for his time, and he makes a nice distinction (e.g. poeta and
versilogus) between classical and medieval writers. His quotations
from Catullus and Martial, from Horace's Odes and Ovid's Ibis,
together with his lavish use of Seneca's tragedies, show clearly the
influence of the local humanism.
Lovato's spiritual successor was his friend and fellow townsman
Albertino Mussato (1262-1329). A notary by profession, Mussato
achieved distinction in the worlds of politics, diplomacy, and liter
ature. Strongly influenced by Lovato, he read the same Latin poets
and delved more deeply into Senecan tragedy; he also wrote
The Renaissance 127
favourite books. The list is instructive, both for the works it contains
and their order of priority, and for those it does not contain; but it
must be remembered that the list belongs to the earlier period of
his life and that some of the books that he prized were among his
later discoveries. Cicero not unexpectedly heads the list, his 'moral'
works taking precedence. Next comes Seneca: the Letters have
pride of place; the Tragedies come later, and in a second and more
select list on the same page they are explicitly excluded from the
inner circle. The next main section is devoted to history, headed by
Valerius Maximus and Livy; there is a special category of exempla, in
which Macrobius and Gellius find their home. Poetry follows, with
Vergil, Lucan, Statius, Horace, Ovid, and Juvenal; Horace is
qualified praesertim in odis, a complete reversal of medieval taste. At
the end come the technical works, grammar, dialectic, and astro-
logy. Augustine is favoured with a list to himself, and with Boethius'
De consolatione Philosophiae he makes up the sum of the Christian
content. The only Greek work is Aristotle's Ethics (in Latin, of
course), and this disappears from the select list. Of the law, in which
Petrarch received his formal education at Bologna, nothing; the
writers of the Middle Ages are likewise rejected, made redundant
by direct contact with antiquity.
One of the first to fall under the influence of Petrarch's human-
ism was his younger contemporary Boccaccio (1313-75). Under the
patronage of its ruler, Robert of Anjou (1309-43), Naples had
emerged as an important intellectual centre quite early in the
century and it was here that Boccaccio spent his youth. His early
works, written in Italian, belong to the medieval tradition of
rhetoric and romance; it was largely his admiration for Petrarch,
whom in 1350 he got to know personally, that made him turn from
the vernacular to Latin, from literature to scholarship. As a scholar
he fell far behind Petrarch; he lacked the patience even to be good
at copying manuscripts. He was in the main a gatherer of facts
about ancient life and literature, and his encyclopedic treatises on
ancient biography, geography, and mythology, enjoyed a consider-
able vogue in the Renaissance and did much to promote the under-
standing of classical literature. He had a passionate interest in
poetry, and this led him along the lesser-known paths of Latin liter-
The Renaissance x
33
Cicero had left philosophy for a life of action and intrigue, it was his
blending of intellectual pursuits with a political career that roused
the admiration of Salutati and the later Renaissance.
copied from it before it was carried off to Italy by Poggio (Paris lat.
14749). This manuscript has now yielded up its secret and revealed
the hand of the French humanist Nicholas of Clamanges.
His next foray was in the summer of 1416, this time to Saint Gall
in company with three humanist friends, Bartolomeo da Monte-
pulciano, Gencio Rustici, and Zomino da Pistoia. The result was
three major discoveries: a complete Quintilian (previous humanists
had had to make do with mutili), Asconius' Commentary on five of
Cicero's speeches, and a manuscript containing four books (i-
iv.377) of the Argonautica of Valerius Flaccus. Poggio's manuscript
of Quintilian is of little value, but for Asconius we depend on three
copies which resulted from this trip, one made by Poggio, one (and
the best) by Zomino (Pistoia A. 37), and one derived from Barto-
lomeo's autograph (Laur. 54.5). The lost Sangallensis of the Argo-
nautica has to be reconstructed from a similar trio, one certainly in
Poggio's hand (Madrid, Bibl. Nac. 8514, ohm X. 81, which also con-
tains his Asconius); but for Valerius we have a complete and more
important ninth-century manuscript from Fulda (Vat. lat. 3277),
which was itself eventually taken to Italy.
Early in 1417, armed with official sanction, Poggio and Bar-
tolomeo made a highly organized expedition to Saint Gall and
other monasteries of the area: their finds included Lucretius, Silius
Italicus, and Manilius. The manuscripts which they found have
perished, but their legacy remains. The copy of Manilius which
Poggio had made is an important witness to the text (Madrid, Bibl.
Nac. 3678, ohm M. 31); his Lucretius fathered the whole race of
Italiy and all our manuscripts of Silius go back to the copies made as
a result of this expedition. At the same time Poggio acquired from
Fulda their famous manuscript of Ammianus (Vat. lat. 1873), which
he carried off to Italy; he also set eyes on their Apicius, and this too
was eventually taken to Rome, by Enoch of Ascoli in 1455 (New
York, Acad. Med. MS. Safe). Now or at some later date he also
obtained a manuscript of Columella (an author already known in
Italy), and his manuscript was probably the insular codex at Fulda,
which reached Italy in the fifteenth century (Ambros. L. 85 sup.).
In the summer of 1417 Poggio went on more extended travels in
France and Germany. He made two important discoveries. The first
138 Scribes and Scholars
was eight unknown speeches of Cicero: the Pro Caecina, Pro Roscio
comoedo, De lege agraria i-iii, Pro Rabirio perduellionis reoy In Pisonem,
Pro Rabirio Postumo. He found the Pro Caecina at Langres, the others
probably in Cologne cathedral. His autograph copy of these
speeches has now been discovered (Vat. lat. 11458) and so done
away with a tedious process of reconstruction. The second find was
one of the rarest of texts, the Sihae of Statius; our manuscripts of
these poems all descend from the copy that was made for him
(Madrid, Bibl. Nac. 3678, olim M. 31).
When the Council was over, Poggio spent some years in England,
where he found what he described as a particula Petronii, i.e. the
excerpta vulgaria\ it is from this manuscript that all the fifteenth-
century copies descend. On his way home in 1423 he found at
Cologne a second manuscript of Petronius containing the Cena Tri-
makhionis, and from the copy that he commissioned descends our
unique source for the complete Cena (Paris lat. 7989). It dis-
appeared from view while on loan to Niccolo Niccoli, fortunately to
reappear at Trogir in Dalmatia about 1650. The Propertius that he
sent to Niccoli in 1427 may have been another of the fruits of this
expedition.
Poggio's achievements in the field of discovery were prodigious;
his personal intervention in the history of many important texts was
decisive and influential. He also occupies a notable place in the
history of handwriting; though he may not have invented the new
humanistic script, as was once thought, he was one of its first and
finest exponents (Plate XVI). As time went by, and particularly from
the beginning of the thirteenth century, Caroline minuscule had
become more angular and thick and considerably less attractive,
developing into what is known as the Gothic hand. The humanistic
script was a deliberate return to an earlier form of Caroline minus-
cule; it appears to have been developed about 1400 in Florence,
with Salutati and his two proteges Niccolo Niccoli and Poggio all
playing a part. Niccoli seems to have developed the cursive form of
the new script. With the advent of printing the formal hand pro-
vided the model for the roman fount, the cursive for italic.
The bulk of Latin literature known to us had now been
recovered. The more important of the remaining discoveries may
The Renaissance *39
manuscript which had preserved it. The Cluny and Lodi manu-
scripts of Cicero, the Veronenses of Catullus and Pliny have
perished; only a few leaves survive of the Hersfeld Tacitus; the Cena
Trimalchwnis was nearly lost again for ever. Other manuscripts
unnamed and unsung survived into the Renaissance (their existence
can be inferred from their humanist progeny), but not beyond it.
Nor does the situation appear to have been much better in the six-
teenth century. Manuscripts were often treated with scant respect
by the printers to whom they had been entrusted and faced an
uncertain future when they had served their purpose. There were
some sad casualties. The Murbach codex of the Historia Augusta lent
to Erasmus survives as a few scraps in a binding (Nuremberg,
Fragm. Lat. 7); the fifth-century manuscript of Pliny's Letters, the
unique source for book X, now reduced to a fragment (Pierpont
Morgan Lib., M. 462), had triumphed over every hazard until
borrowed by Aldus from the abbey of Saint-Victor at Paris; the two
manuscripts from Worms and Speyer which Beatus Rhenanus and
Gelenius used for their edition of Livy disappeared from sight. But
these two scholars do not appear to have been as culpable as was
once thought. The Hersfeld manuscript of Ammianus Marcellinus
that Gelenius used for the Froben edition of 1533 survived until
later in the century, when it was dismembered not far from Hersfeld
to provide covers for account books (we still have six leaves: Kassel,
Philol. 20 27); and the Murbach codex of Velleius Paterculus used by
Rhenanus seems to have survived into the late eighteenth century,
being last heard of in a sale-room in 1786.
the great Livy volume which had been put together by Petrarch,
and his autograph notes can still be seen in its margins (Plate XV).
He dared to emend the Vulgate itself, and his notes and corrections
(1449), based on a study of the Greek original and early patristic
texts, were fully appreciated by Erasmus, who had them printed in
1505. He also found time to be a prolific translator from the Greek.
He firmly belongs to the first half of the century, and it is significant
that full recognition of his talents had to wait until his works were
put into print.
Politian was born at Montepulciano and educated at Florence.
He showed a precocious talent and was taken at an early age into
the household of Lorenzo de' Medici, who made him tutor to his
children and remained his lifelong friend and patron; by the age of
thirty he was a professor of such repute that he attracted scholars
from all over Europe to his lectures on Greek and Latin literature.
As well as being an influential teacher, he was the finest poet of his
time, both in Italian and in Latin; and as a scholar he at times tran-
scends his age and moves out of reach of any of his contemporaries.
Politian won his prominent position in the history of the classical
tradition both by his exact scholarship and by the way in which he
opened the eyes of his contemporaries to the full perspective of
ancient literature. Valla had recommended the study of Quintilian,
but his insistence on a classical norm in the writing of Latin had
tended to foster the predominant cult of Ciceronian Latin. Politian
firmly rejected Ciceronianism and chose to create an eclectic style
of his own which exploited the whole range of Latin: lnon exprimis
inquit aliquis 'Ciceronem'. Quid turn? Non enim sum Cicero, me tamen (ut
opinor) exprimo {Epist 8.16). In the same way he was the first to give
serious attention to the prose and poetry of the Silver Age.
Politian's great work of scholarship was his Miscellanea, a collec-
tion of studies of varying length on different points of scholarship.
The elegant and original format of this miscellany, which put as
much distance as possible between itself and the line by line com-
mentary on a specific text that had formerly been the fashion, was
well chosen to display the many sides of his learning. The first part
{Centuriaprima) was published in Florence in 1489; an autograph
draft of a second series of chapters came to light only recently. The
144 Scribes and Scholars
the sound and scholarly way in which Politian dealt with manu
script evidence:
Hunc librum de codice sanequam vetusto Angelus Politianus, Medicae
domus alumnus et Laurenti cliens, curavit exscribendum; dein ipse cum
exemplari contulit et certa fide emendavit, ita tamen ut ab illo mutaret nihil,
set et quae depravata inveniret relinqueret intacta, neque suum ausus est
unquam iudicium interponere. Quod si priores institutum servassent, minus
multo mendosos codices haberemus. Qui legis boni consule et vale. Floren-
tiae, anno MCCCCLXXXV, Decembri mense.
More fruitful than the contacts with the south of Italy were those
with Constantinople itself. The declining fortunes of the Greek
empire made it necessary to send frequent diplomatic missions
abroad to beg help against the Turkish invader; monarchs as far
distant as the king of England received these appeals. We have
already seen how a knowledge of Latin literature was made avail
able to the Byzantines through Maximus Planudes, who had served
on an embassy sent to Venice. It was almost exactly a century later
that another Byzantine diplomat, Manuel Chrysoloras, became the
first man to give regular lectures on Greek in Italy. He began in
Florence in 1397, which is therefore a date of fundamental import
ance in the cultural history of Europe, and continued his courses for
about three years before moving to Pavia for an equally brief stay.
He had several notable pupils including Guarino and Leonardo
Bruni. One important result of his teaching was that Latin trans
lations of Greek texts were prepared, and he insisted that the old
word-for-word style of translation should be abandoned, and that
attention should be given to ensuring some literary merit in the
version. An indication of his influence as an instructor is that his
textbook of Greek grammar entitled Erotemata gained a consider
able circulation, and eventually became the first Greek grammar to
be printed (in 1471); it was later used by such famous men as
Erasmus and Reuchlin.
During the fifteenth century the opportunities for an Italian to
learn Greek improved. A number of Byzantines came to live in Italy,
and after the defeat of their country in 1453 there was a stream of
i48 Scribes and Scholars
they had been forged by the Italians, the high quality of the Greek
was sufficient reply to the suggestion.
After this example of scholarly method used to refute the un-
scrupulous manipulation of texts we come to Bessarion's other
short work which shows his scholarship to advantage, and here too
the context is theological. After a reading from Saint John's Gospel
as part of the liturgy conducted in his house in Rome, lively con-
versation began as to the correct text of John 21: 22. The reading
had been performed from the Latin Vulgate, which erroneously
gave the word sic instead of si (the Greek has idv). Bessarion
pointed out in discussion that this was a simple case of a copyist's
error, involving only one letter. His audience was not completely
persuaded, and so he wrote a pamphlet to prove his point. Here
several important principles are enunciated, and the whole matter
is discussed with a common sense that seems natural to us but was
not welcome to the narrow-minded conservatism of men who
regarded every word of Saint Jerome's translation as sacred. Bes-
sarion states that the Greek text is the original and must have
precedence over the Latin translation, and is able to claim Augus-
tine's authority for this proposition. He also shows that early quo-
tations of the Greek text in Origen, Cyril, and Chrysostom all have
the same wording. Then he shows that the whole context of the
passage is not suited to the reading of the Vulgate. The work is of
great importance and anticipates the attitude of Erasmus in regard-
ing the Greek text of the New Testament as the only proper basis
for interpretation. It may owe something to Valla, who frequently
met Bessarion and had previously written but not published a tract
entitled Adnotationes in Novum Testamentum, in which he called into
question the Vulgate's accuracy.
To the Greek bishop who settled in Italy and whose scholarly
activity was devoted mainly to theology and philosophy Politian
(1454-94) offers a striking contrast. He is famous as a poet in his
vernacular language and in Latin, but was equally distinguished as a
scholar. Though primarily interested in ancient literature he had a
proper understanding of the subsidiary branches of knowledge,
such as epigraphy and numismatics, which make a contribution to
our general understanding of the ancient world. The combination
The Renaissance *53
While the new art of printing soon led to a spate of editions of Latin
classics from the seventies of the fifteenth century onwards, for
Greek texts the situation was quite different. Part of the reason for
this may have been the difficulty of designing a suitable fount of
type, in which the number of sorts would not be unreasonably
increased by the various combination of letters with accents and
breathings. Certainly some of the early printers, in a mistaken
desire to reproduce in print the appearance of contemporary Greek
script, devised founts of type that were expensive to operate with
and unsatisfactory in appearance. Even the famous Aldine founts,
which served for a very long time as the models for later typo-
graphers, are open to both these criticisms. But not all early printers
failed in this way; the type-face designed by the famous Frenchman
Nicholas Jenson, who worked in Venice, was an excellent piece of
work, and still better in some ways was that used for printing
passages of Greek in Politian's Miscellanea-, here the accents and
breathings were omitted and ligatures avoided, so that the text bore
no close resemblance to written script but was far more easily
legible. It is surprising that one of these simpler founts was not
immediately accepted as the standard.
More serious than the typographical difficulty was the lack of
The Renaissance *55
X. ERASMUS (C.I469-I536)
could muster, but they seem to have been an indifferent lot, with
one signal exception. He had access to readings from the Lorsch
manuscript of the Debeneficiis and De dementia (cf. p. 99), the arche-
type of the whole tradition. But he was inhibited by the critical
methods of his day: instead of basing his text of these works upon
this prime witness, he drew on it spasmodically to emend what he
had before him, and a great opportunity was lost.
5
The speed and vitality with which humanism had taken root and
flourished in Italy was unparalleled elsewhere. In France classicism
remained more traditionalist and made no such dramatic leap
despite its being open to Italian influence, particularly through
Avignon, from the early fourteenth century onwards. But the
strength and vitality of French medieval culture meant that French
humanism could absorb what it needed from Italy without being
too dependent on it and could strike out along its own path within
the broad lines of its own tradition. The sensitivity of French
scholars on this issue and the frequent signs of a reaction against
Italian scholarship reflect both their debt to Italian humanism and
the pride they took in the originality of their own achievement.
Pierre Bersuire (d. 1362) had been one of the first to benefit from
the cultural interaction fostered at Avignon and from personal
contact with Petrarch himself, who gave him his friendship and
help with his classical studies. His translation of Livy into French
was an important step in reinforcing the historian's new-found
popularity and his Qvidius moralisatus shows some Petrarchan influ-
ence; but his medieval ways of thinking were too strong for even a
Petrarch to change and he fell far short of being a humanist. But a
powerful group who thoroughly deserved the name had emerged in
France towards the end of the century, among them Jean de Mon-
treuil (1334-1418) and his intimate friend Nicholas of Clamanges
Scholarship since the Renaissance 171
(c. 1360-1437). Though they owed their wide familiarity with clas
sical authors, particularly Cicero, to contact with Italian humanists
and imported texts, their humanism was firmly rooted in the north
and they were well able to discover new texts on their own account.
Cluny in particular had proved a rich source. Not even a Poggio can
turn up a new text every time without being told where to look, and
the presence ofJean de Montreuil at the Council of Constance may
have had important side-effects. It can hardly have been a coincid
ence that Poggio found the Pro Caecina at Langres, where Nicholas
of Clamanges, the great connoisseur of Cicero's speeches, had been
canon and treasurer to the cathedral chapter. And although Poggio
claims credit for the discovery of the vetus Cluniacensis of the
speeches and indeed had it sent to Italy, the best and most con
scientious copy of the lost manuscripts is the one made before it
went to Italy, by Nicholas of Clamanges.
The apparently intermittent progress of French humanism was
strengthened by two events which took place in the second half of
the fifteenth century, the appearance of the first teachers of Greek
and the setting up of the first printing press in France. Earlier
attempts to organize Greek studies at Paris had proved abortive
and Gregorio Tifernate, who arrived in 1456, stayed only a few
years. George Hermonymus of Sparta, who came to France in 1476,
is best known for his failure to give much help as a teacher to either
Bude or Erasmus. But with the arrival ofJanus Lascaris in 1495 and
Girolamo Aleandro in 1508, Greek studies began to flourish and
became an important element in French humanism. The first
printers were German, the first book was a collection of the model
letters of the Italian humanist Gasparino Barzizza, but the pro
moter of the first press to operate in France was Guillaume Fichet,
master in theology and librarian of the Sorbonne, who in 1470
obtained authorization to set up a printing press in the College
itself. It made a decisively humanistic debut; it used the Roman
letter exclusively and its first publications were either straight Latin
texts, Sallust, Cicero, Juvenal, Terence, and the like, or works bear
ing on the cultivation of Latin style, such as Valla's Elegantiae and
Fichet's own Rhetorica. The first Greek book to be printed in France
appeared in 1507.
172 Scribes and Scholars
the field until Lachmann. One of the five manuscripts he used was
the ninth-century codex Quadratus (Leiden, Voss. Lat. Q. 94 = Q),
one of the two manuscripts on which the text is still based; it was
then at the monastery of Saint Bertin, near Saint Omer, and he had
access to a collation made for Turnebus. For Cicero's letters he used
a manuscript of outstanding merit which belonged to the Lyons
printer Jean de Tournes and was last heard of in 1580; for its read-
ings we are dependent on three French scholars of this period,
Lambinus, Turnebus, and Bosius.
The manuscript collectors of this age, often scholars and editors
themselves, made a signal contribution to classical studies. Con-
spicuous among them is Pierre Daniel (c. 1530-1603), a jurist of
Orleans, whose great coup was to succeed in buying manuscripts
from Fleury after its sack by the Huguenots in 1562. His collection,
now mainly at the Vatican or Berne, contained such important
relics of the scholarly heritage of that region as Lupus's copy of
Valerius Maximus (Berne 366). He also published editiones principes
of the Querolus (1564) and the longer version of Servius (1600), still
often referred to as Servius Danielis. Another was Pierre Pithou
(1539-96), who published the first editions of the Pervigilium Veneris
(1577) and the Fables of Phaedrus (1596), both based on ninth-
century manuscripts which remain prime witnesses to the text. His
use of good manuscripts enabled him to publish important editions
of Petronius, and he was the first to use the Lorsch manuscript for
the text of Juvenal and Persius (1585), the famous codex Pithoeanus,
now to be found, with many other of his manuscripts, at Mont-
pellier. Equally important was Jacques Bongars (c. 1554-1612),
whose enormous library, partly derived from the collections of
Daniel and Cujas and now at Berne, included such choice items as
the famous Irish manuscript of Horace (Berne 363) and our best
manuscript of Petronius (Berne 357). Indeed, the complicated
history of the text of Petronius in the latter half of the sixteenth
century epitomizes the activity of a group of French scholars of this
period, Pierre Daniel, the Pithou brothers, Bongars, Scaliger, and
the great professor of jurisprudence who had taught them all and
may have inspired this particular interest, Jacques Cujas. Its com-
plexity is also an indication of the difficulty of piecing together the
176 Sondes and Scholars
Scaliger was far from untouched by the religious troubles of his day,
but the way they bedevilled the scholarship of the sixteenth century is
more starkly illustrated in the case of his friend and younger contem-
porary Casaubon. Born in Geneva of refugee Protestant parents,
obliged to learn his Greek hiding in a cave in the French mountains,
unable to avoid being drawn into the wrangle because of his distinc-
tion as a scholar and forced to spend much of his time and talents on
arid polemic, this great French scholar finally found rest as a natural-
ized Englishman in Westminster Abbey. With him the French
scholarship of the period ended, as it had begun, on a chalcenteric
note. He was a man of vast industry and erudition, but had the rarer
gift of being able to use his learning as a commentator to illumine
rather than impress. He appears to have chosen to work on those
texts that offered the most scope to his wide knowledge, such as
Diogenes Laertius, Strabo, and Athenaeus. His choice of difficult and
often diffuse texts, with which most students of the classics have but a
passing acquaintance, means that his services are not always recog-
nized. For Casaubon is still with us. His Animadversiones on
Athenaeus formed the core of Schweighauser's commentary of 1801,
Strabo is still usually cited by reference to Casaubon's pages, his notes
on Persius loom large in Conington's commentary. Son-in-law to
Henri Estienne and for a time sub-librarian to de Thou at the royal
library, Casaubon was most at home in the world of books and manu-
scripts, able to find material for his own needs and to supply scholars
all over Europe. His use of manuscript material has not been properly
appraised, but he seems to have made no dramatic advances, except
that his second edition of Theophrastus' Characters (1599) added five
more characters (24-8) to those then known. Some of his most distin-
guished work was long buried in his unfinished commentary on
Aeschylus.
the errors of scribes and, though little that he says would come as
news to the great critics of his day, it is a gain to have certain valid
principles of emendation explicitly set out, even if the details need
refinement. Franz Modius (1556-97) is less noteworthy for his
scholarship, though he edited a number of Latin texts, than for his
insistence that conjecture alone is useless and even dangerous, that
there must be a proper balance between manuscript authority and
emendation, that recension is an essential preliminary to editing. In
this conviction, and also obliged by the political unrest of the
Netherlands to be on the move, he systematically explored the
manuscript collections of a wide area, extending from northern
France through the Low Countries to Fulda and Bamberg. His
activity is remarkable for its scale and his reports of manuscript
readings, found in his Novantiguae lectiones (1584), acquire great
value when the manuscripts themselves have been destroyed, as in
the case of the Cologne manuscript of Silius Italicus. The only other
first-hand report of the Cologne Silius is provided by his friend and
later enemy, Ludovicus Carrio (1547-95), who was similarly active
on a smaller scale. Jacob Cruquius worked almost exclusively on
Horace and owes his fame to his invention of the ghostly 'com-
mentator Cruquianus', now exorcized, and to his timely examina-
tion of four Horace manuscripts at the monastery of Saint Pierre au
Mont-Blandin, near Ghent, just before its destruction in 1566. One
of these was the very important, if controversial, Blandinius
vetustissimus, which assures the Bruges professor a fractional share
of the immortality that Horace so confidently forecast for himself.
It was a singular piece of good fortune for the new university of
Leiden that it should have attracted so soon after its inauguration
one of the most brilliant Latinists of the century. Justus Lipsius
(1547-1606) was a Catholic by upbringing and associated in his
early years with the university of Louvain, but his conversion to the
Protestant faith had opened the way for his being invited to the
chair of history at Leiden, which he held from 1579 to 1591, just as
his reconversion led to his return to Louvain in 1592, where he was
Professor of History at the university and of Latin at the Collegium
Trilingue. His achievement was based on a thorough knowledge of
the history and antiquities of Rome, reflected in his monographs
Scholarship since the Renaissance 181
good critic, led him astray in dealing with such authors as Horace,
and he earned notoriety by the amusing change he proposed to
make in the fable of the fox caught in the granary {Epistles 1.7.29).
Insisting that a fox will not eat grain Bentley proposed to read 'field-
mouse' (nitedula instead of vulpecula), quite oblivious of the con-
sideration that the author of the fable chose the animal as the
representative of cunning greed at the expense of the facts of
natural history. This insistence on logic, without consideration of
poetic and other forms of literary licence, mars Bentley's contribu-
tions to the emendation of leading authors that he edited, namely
Horace in 1711 and Terence in 1726, and the same is even more
true of his attempt to restore the works of Milton to what he
supposed to be their original state before a putative interpolator
imposed on the blind poet with a series of alterations of the text. On
the other hand, where hard facts were at a premium, as in the
astronomical poem by Manilius, Bentley's gifts were given a great
opportunity, and the opinion of experts is that he made contribu-
tions of the utmost brilliance to the interpretation of the hardest
passages of this very hard poem, the edition of which appeared in
1739 although the work for it had been done long before. It should
also be recorded that in dealing with Terence he displayed a
notable command of the principles of metre; in this field, however,
he acknowledged the importance of a sixteenth-century Italian
predecessor, Gabriele Faerno.
Bentley made many emendations in the text of other authors, of
which a high proportion have been accepted or seriously con-
sidered by subsequent editors. But two of his most valuable activ-
ities were projects that never came to fruition, editions of Homer
and the New Testament. As far as Homer is concerned, his most
notable discovery was that the metre of many lines could be
explained by postulating the existence of the letter digamma, a
notion which contributed as much as any other single discovery to
the understanding of this text.
Though Bentley is commonly thought of as a classical scholar
pure and simple because of his striking achievements in that field,
he was also of sufficient competence in theology to be appointed
Regius Professor of Divinity in 1717. Three years later he published
Scholarship since the Renaissance i87
century was told that the books could no longer be found. This
tantalizing state of affairs roused the curiosity of a local aristocrat
and antiquarian, the marquis Scipione Maffei (1675-1755). Besides
making a name for himself by writing the tragedy Merope, which
was a landmark in the revival of the Italian theatre, he found himself
involved in historical controversy in 1712, when he wrote a
pamphlet against the duke Francesco Farnese. Farnese had been
duped into purchasing the grandmastership of an order of Saint
John supposedly set up by the emperor Constantine. The pope and
the Austrian emperor swallowed the bait as well, and Farnese was
assigned for the use of his order the beautiful church of Santa Maria
della Steccata in Parma. Maffei demonstrated that the order must
be bogus, since all such orders were of medieval date, a fact which
did not save his book from being placed on the index of prohibited
literature.
Maffei let it be known to the canon librarian of the cathedral in
Verona that he was very anxious to discover the fate of the manu-
scripts it had once possessed. One morning in 1712 the librarian
found them; they had been piled on top of a cupboard in order to
avoid damage from flooding, and then had been entirely forgotten.
The news was taken at once to Maffei's house, and he rushed over
to the cathedral in his night clothes and slippers. When he set eyes
on the books, a wonderful collection mostly of very early date, he
thought that he must be dreaming, but the dream proved to be a
reality and it was not long before he was studying the manuscripts
in his own home. The result of this study was a very important
theoretical improvement in the understanding of Latin book-
hands. Mabillon had divided them into five independent categories,
Gothic, Langobardic, Saxon, Merovingian, and Roman. But he had
said nothing about any possible relation between them. Maffei hit
on the fact that the explanation of the diversity of Latin scripts in
the early Middle Ages must be that in late antiquity there were
certain basic types, majuscule, minuscule, and cursive, and when the
Roman Empire broke apart variations of these scripts arose in-
dependently. It was this flash of insight which made palaeography a
subject with a clear theoretical basis. The only major advance sub-
sequently is the one associated with the name of Ludwig Traube
192 Scribes and Scholars
(a) Palimpsests
(b) Papyri
Until the end of the last century our knowledge of ancient texts
depended almost entirely on copies made during the Middle Ages,
whereas manuscripts dating back to the later centuries of the
ancient world formed only a tiny proportion of the total number
known. From the Renaissance onwards such discoveries as were
made of new texts, or more commonly, better manuscripts of texts
already known, usually consisted in the unearthing of neglected
196 Scribes and Scholars
and is right in at least two of the four places in which it differs from
the direct tradition.
VII. EPILOGUE
TEXTUAL CRITICISM
I. INTRODUCTORY
remained for Lachmann, in his edition of 1850, to apply his rules for
the mechanical application of the stemma and to give a classic
demonstration of the validity of the hypothetical archetype by
reconstructing its physical form and telling his astounded contem-
poraries how many pages it had, and how many lines to the page.
also be given in which authors' revisions are visible but have not
affected the stemma so seriously. At Martial 10.48.23 it looks as if
the name of a charioteer stood in the first edition of the book, but
for the second edition, prepared after his death, the no longer
topical name was replaced by a word indicating the team to which
he belonged. A possible explanation of the incoherence of a scene
in Aristophanes' Frogs (1437-53) is that some of the lines come
from a revised version of the play. Similarly Galen (15.624) attrib-
utes the confused state of one of Hippocrates' works to marginal
additions and revisions by the author.
But Quintilian (9.3.8; though his MSS. too are corrupt) evidently
read qui at the beginning of the relative clause, which led to the
necessary conjecture:
And it is not only ancient authors whose quotations hint at the right
text. Some good manuscripts that are now lost survived until late in
the Middle Ages and were consulted by scholars of that period. A
medieval source of secondary tradition which is often important is
the Suda lexicon. Two examples may be given of good readings that
it offers in quotations of Aristophanes: {a) Knights 254, where it has
<f>vyevt whereas the manuscripts have the unmetrical ecfrvyt. (b)
Clouds 215, where the manuscripts give TT&VV (fipovri^re, whereas
the Suda has ^Ta^povri^Tt, an amusing new coinage, probably
correct (the Ravenna scholiast also apparently had this reading).
It should not be supposed, however, that secondary tradition is
an unfailing source of correct readings. Ancient and medieval
writers were even less inclined than their modern counterparts to
follow the advice of Dr. Routh, president of Magdalen College,
Oxford, from 1791 to 1854, who when asked by Dean Burgon
whether after a long career he had any advice to offer young
scholars replied 'Always verify your references'. Quotations were
usually made from memory and there were good reasons for this
220 Scribes and Scholars
practice. Ancient books of the roll type did not facilitate quick
reference, and when the codex form was adopted readers still
could not count regularly on such aids as page numbering,
chapter division, and line numbering, which are part of every
modern text. These considerations must be borne in mind in cases
where the manuscripts of an author offer one reading, and there
is an acceptable but not absolutely certain alternative in the
secondary tradition. Either reading may be what the author wrote.
But in the majority of cases editors will probably be right to
follow the primary tradition; the divergent reading of a quotation
is likely to be right only if the purpose of the quotation was to
emphasize or illustrate the divergent words or phrases, whereas if
the divergence is incidental to the quotation it is probably due
merely to lapse of memory.
The difficulty of enunciating firm principles may be shown by a
few examples:
{a) Aristophanes, Acharnians 23: the MSS. have ovS'ol -npvTavtic
rJKovciv, d\y aatpiav \ TJKOVTC KT\. The Suda lexicon quotes the
line with the reading dcopia, in the article which is devoted to this
word. In principle there is some reason to hope that this might be
an accurate quotation, but the dative is less idiomatic than the
accusative in this type of adverb and for that reason is rejected by
the editors.
(b) In the same play at 391-2 the MSS. offer:
These lines are quoted in the Suda article on Sisyphus, but with two
differences: in the first line iVa is replaced by dAAd, and in the
second elcSi^erat by 777)0 c S ef era 1. Since the quotation is not
intended to illustrate either of these words, there is not much
reason to trust in its accuracy. Yet on other grounds dAAd may well
be right, and npocSe^rai is as likely to be right as the rather
puzzling cicBigcTai.
(c) Lucretius 3.72 is given in the MSS. as:
The grammarian Charisius (p. 265, Barwick 2 ) quotes the line with
the variant longiter, and says that it was his purpose to exemplify the
form by this quotation. For this reason editors usually, though not
invariably, adopt the reading.
such cases the more difficult reading may be more difficult because
it is wrong.
VIII. CORRUPTIONS
view. Typical causes of error within this class are (i) the lack of divi-
sion between words in many manuscripts, (ii) a close similarity of
certain letters in a script which results in their being confused, (Hi)
the misreading of an abbreviation; apart from ordinary signs repres-
enting syllables or common short words there was a special method
of abbreviation for certain key terms of Christian theology; these
are known as nomina sacra, and are frequent in both Greek and
Latin texts. In both languages abbreviations are so numerous and
complex that the study of them forms a subject in itself (Plate V).
(iv) Since numerals were represented by letters in both languages
they were often incorrectly transmitted, a fact which is a serious
hindrance to students of economic and military history. Perhaps
one may add here (v) the confusion of two words of similar shape or
spelling even when there is no immediate cause of confusion in the
forms of the individual letters.
8e IJaXXdc Sophianus.
(ii) The following letters are commonly confused in Latin scripts:
In capitals: ILT EF PT PF PC BR HN OQ COG and such combina-
tions as M NI.
In uncials: ILT FPR CEOGU and such combinations as U CI. Char-
acteristic of uncial, as opposed to capital, is the confusion of EU
(now rounded in form) with the group COG.
In minuscule: au oecldnu sfct and various letters or combinations of
letters made of one or more downstrokes (minims), e.g. the letters
in minimum. The confusion of prrn ns is characteristic of insular
script; in Visigothic the peculiar /, and in Beneventan both the t
and the a cause difficulty (sec Plate XIV).
1. Seneca, Epist. 81.25.
Manifestum etiam contuenti discrimen est.
coniventi codd. recc.
Scribes and Scholars
2. Lucretius ii.497.
quare non est ut credere possis
esse infinitis distantia femina formis.
semina O c
Groups of letters liable to be confused with each other in Greek
script are:
In uncials: AAA E80C1CK IT.
In minuscule: /?K/X p.v a tv.
1. Aristotle, Poetics 1462 b 3.
AfyCU 8'OLQV LTIC TOV 0181TTOVV Btil) TQV CO^OKXCOVC V 7TCIV OCOIC
rj iSCac.
rj iSiac MSS.: -q IXtdc a humanist corrector (confusion of A and A).
2. Julian, EpisL 23.
80c fioC TI Kara TOVC fieXiKrdc i7Tiv prjropac.
fieXiKT&c Aldine edition: /xeA^rdc VL: pLeXirovc N: fieXrlcTovc Jackson.
fi and fi are very similar in minuscule; here the corruption is aided by
the uncial confusion of IC and K.
ovv may well conceal the nomen sacrum ovpavov (ovvov), which has
been further corrupted.
does to the metre; but sometimes the word used as a gloss had
the same metrical value as the word in the text and replaced it
without impairing the metre, and examples of this process are not
easy to detect. The detection of (iii) glosses in a prose text is often
of the greatest difficulty. Many passages contain explanatory
phrases which are not strictly required for the sense but offer no
offence to grammar or syntax. These phrases present problems
which may remain insoluble. Two texts which scholars have
recently discussed in detail in the light of these questions are
Petronius* Satyricon and Cicero's Tusculan Disputations, (iv) A rare
but interesting corruption is the addition to a text of a parallel
passage originally written in the margin of a book by a learned
reader. This may happen in verse or prose. Cases are known from
Greek tragedy, and Galen (i7(i).634) noticed that it had hap-
pened in one of the Hippocratic treatises.
(iv) 1. In the margin of the Medicean Aeschylus at Persae 253 the line
Sophocles Antigone 277 has been written. In copies of M this line has
been incorporated into the Aeschylean text.
2. Vergil, Aen. ii.76.
ille haec deposita tandem formidine fatur.
Sinon is about to explain himself. This line is omitted in P and added at
the foot of the page in M by a later hand. It is a doublet of iii.612, where
it is in place, and has been added here because of the similarity between
this passage and that in book iii.
1. Juvenal, viii.148.
Ipse rotam adstringit sufflamine mulio consul.
In some manuscripts this has been, not unnaturally, corrupted to sufflamine
multo. At this point deliberate interpolation steps in and restores the metre:
other manuscripts read multo sufflamine.
the manuscripts which transmit any one of them. The text of one
version, perhaps descended from an earlier or better source, may
allow the editor to correct that of another, but it is not always easy
to decide whether a version is wrong or simply different. And which
does he choose to edit? In some cases he will print the original or
dominant version, and then add additional or divergent passages at
the foot of the page. If typography and format permit, he may
produce a synoptic edition in which the different recensions are
printed alongside each other in parallel columns. Or it may be
advisable to edit each version separately. The choice will depend on
the merits and conditions of each particular text.
XI. CONCLUSION
CHAPTER I
I. Ancient books
(which is not certain). It has been suggested that P. Oxy. 3672 of Plato, Laws
vi will have been about 12 metres long.
The use and gradual refinement of punctuation are still debated; apart
from Turner's book cited above see R. Pfeiffer, History of classical scholarships
Oxford 1968, pp. 178-81; another useful survey is given by M. Geymonat in
the Enciclopedia Virgiliana, s.v. interpunzione, vol. 2 pp. 998-1000. H.-I.
Marrou, Histoire de Fducation dans Tantiquit, Paris 19656, p. 602 n. 30,
suggested that books with full punctuation never passed into general use
but were confined to teachers and pupils (his remark applies particularly to
books in the Roman period). An interesting example of a conjecture that
finds part of its justification in the absence of punctuation from ancient texts
is given by H. Lloyd-Jones, The justice ofZeus, Berkeley 19832, p. 213 n. 23.
Horace, Epistles 1.15 presents a problem: the syntax is so complicated that
the modern reader is bound to ask how his ancient counterpart could find
his way through the text.
books, down to the end of the first century A.D., were provided with more in
the way of punctuation and aids to the reader than was generally customary
in antiquity was propounded by R. P. Oliver, TAPA 82 (1951), 241-2, and
further developed by E. O. Wingo, Latin punctuation in the classical age, The
Hague 1972. A more sceptical view is taken by G. B. Townend, who dis-
cusses the difficulties of punctuating Latin texts, with particular reference to
hexameter poetry and Vergil: CQ 19 (1969), 330-44, Proceedings ofthe Virgil
Society, 9 (1969-70), 76-86. See too W. Mller, Rhetorische und syntaktische
Interpunktion (Diss. Tbingen 1964), and the literature cited in section I
above. Other aspects of Roman books and their production are discussed
below.
There is no full and comprehensive account of scholarship during the
Roman period. The primary material can be found in G. Funaioli, Gram-
maticae Romanaefragmenta, vol. 1, Leipzig 1907, and its continuation, A. Maz-
zarino, Grammaticae Romanaefragmentaaetatis Caesareae, vol. 1, Turin 1955.
Brief accounts are given by A. Gudeman, Grundriss der Geschichte der
klassischen Philologie, Leipzig 19092 (reprinted Darmstadt 1967); F. Leo,
Geschichte der rmischen Literatur, vol. 1, Berlin 1913, pp. 355-68; G. Funaioli,
Studi di letteratura antica, voi. 1, Bologna 1946, 2o6ff. Much can also be
derived from Leo's Plautinische Forschungen, Berlin 19122, and G. Pasquali's
Storia della tradizione e critica del testo, Florence 19522. The aspects of scholar-
ship in the Roman period that most concern us here have been examined by
J. E. G. Zetzel in Latin textual criticism in antiquity, New York 1981, which
deals with a number of topics treated in this and the following sections; for
important additions and qualifications, see the reviews by H. D.Jocelyn,
Gnomon, 55 (1983), 307-11, and M. D. Reeve, CPh 80 (1985), 85-92.
The history of the text of Plautus in antiquity throws some much needed
light on the beginnings of Roman scholarship, murky though that story
sometimes is: cf. W. M. Lindsay, The ancient editions ofPlautus, Oxford 1904;
F. Leo, Plautinische Forschungen, pp. 1-62, Pasquali, Storia della tradizione,
pp. 331-54. The history of the text of Terence in antiquity has been the
subject of recent studies, where reference to earlier work may be found: J. N.
Grant, Studies in the textual tradition of Terence (Phoenix, Suppl. Vol. 20),
Toronto 1986, pp. 1-96; H. D. Jocelyn, 'Two questionable areas of the
directly transmitted text of Terence's Adelphoe (w. 115-19; 584-6)', in
S. Prete (ed.), Protrepticon. Studi classicied umanistici in onore di G. Tarugi, Milan
1989, pp. 45-54. On the composite nature of the prologue to the Poenulus,
see Jocelyn, ' Imperator histricus\ YCS 21 (1969), 97-123; for the date of the
alternative ending to the Andria of Terence, O. Skutsch, 'Der zweite Schluss
der Andria\ RhM 100 (1957), 53-68. There is a good account of the history
Notes to Chapter i 249
of the text of the dramatic poets in ancient times, with particular reference
to Ennius, in Jocelyn, The tragedies ofEnnius, Cambridge 1967, pp. 47-57.
Recent critics have attributed the disorder of the De agri cultura to Cato
himself: for a rsum of the question, see A. E. Astin, Cato the censor, Oxford
1978 PP- ^9I~203'
On the Anecdoton and the critical activity of Aelius Stilo and his circle, see
S. F. Bonner, 'Anecdoton Parisinum', Hermes, 88 (i960), 354-60. The Anec
doton and the whole subject of critical notae and their use have been re
examined in a series of articles by H. D. Jocelyn, 'The annotations of
M. Valerius Probus', CQ 34 (1984), 464-72, 35 (1985), 149-61, 466-74. The
composite nature of the material in the Anecdoton rules out derivation from
a single source, such as Suetonius' De notis.
Attention focuses on Atticus because of the wealth of information in
Cicero's letters, in contrast to our ignorance of the book trade at Rome
before his time. His precise role has been discussed by R. Sommer, Hermes,
61 (1926), 389-422; see further R. Feger, RE> Suppl. 8 (1956), 517-20,
K. Bchner in Geschichte der Textberlieferung der antiken und mittelalterlichen
Literatur, vol. 1, Zrich 1961, 328. The evidence is conveniently presented
(and interpreted with predictable bias) by J. Carcopino, Les Secrets de la
correspondence de Cicron, vol. 2, Paris 1947, pp. 305-29.
A very lucid account of the process of 'publication' in Roman times is
given by Raymond J. Starr, 'The circulation of literary texts in the Roman
World', CQ 37 (1987), 213-23. See also B. A. van Groningen, EKAOEIZ,
Mnent Sen 4, 16 (1963), 1-17; Kenney, op. cit., 19.
fanjniage: The grammarian and society in late antiquity, Berkeley and Los
Angeles 1987. On aspects of their philological work, we have Zetzel, I^atin
textual criticism, and Timpanaro, Per la storia. For a general study of
Donatus, L. Holtz, Donai et la tradition de renseignement grammatical tude sur
[Ars Donati et sa diffusion (IVC-W sicles) et dition critique, Paris 1981. There
is an excellent brief account of the commentators on Horace in R. G. M.
Nisbet and M Hubbard, A Commentary on Horace: Odes, Book I, Oxford 1970,
pp. xlvii-li.
X. The subscriptions
The fundamental work on the subscriptions is still that of Otto Jahn, 'ber
die Subscriptionen in den Handschriften rmischer Classiker', Berichte ber
Notes to Chapters i - 2 2
53
die Verhandlungen der Schsischen Geseilschaft der Wissenschaften, Phil.-hist.
Classe 3 (1851), 327-72. J. E. G. Zetzel has provided a useful list of the
subscriptions in manuscripts of secular authors in Latin textual criticism,
pp. 209-31. To the literature on subscriptions already cited in VI above
should be added Zetzel, ' T h e subscriptions in the manuscripts of Livy and
Fronto and the meaning of emendatici, CPh 75 (1980), 38-59; O. Pecere,
'Esemplari con subscriptiones e tradizione dei testi latini, l'Apuleio Laur. 68,
2 in C. Questa and R. Raffaelli (edd.), / / libro e il testo, Urbino 1984,
pp. 111-37; 'La tradizione dei testi latini tra IV e V secolo attraverso i libri
sottoscritti', in A. Giardina (ed.), Tradizione dei classici trasformazioni della
cultura (Societ Romana e Impero Tardoantico, 4), Rome-Bari 1986,
pp. 19-81, 210-46. The last contains ample bibliography of the whole
subject.
For the imperial fora as intellectual centres, see H.-I. Marrou, Mlanges
^archologie et dhistoire de TEcole franaise de Rome, 49 (1932), 94-110,
reprinted and enlarged in Patristique et humanisme: Mlanges, Paris 1976,
pp. 65-80.
CHAPTER 2
A chapter in the history of annotation, London 1905, pp. 47-57; see also the
note infra on Erasmus.
plates have been proposed for the composition of the Bibliotheca, and there is
j . a r e e ment about the method of composition. The passage in which he de-
scribes the meetings of friends at his house is translated and discussed in
P Lemerle's valuable study, Le premier humanisme byzantin, Paris 1971,
D JQ7-8 (pp. 229-30 in the English tr., Byzantine humanism: thefirstphase,
Canberra 1986).
CHAPTER 3
culture dans I* Occident barbare, VIe-VIII* sicles, Paris 19723 (translated into
English by J.J. Contreni, Education and culture in the barbarian West, sixth
through eighth centuries, Columbia, S.C., 1976); B. BischofT, 'Scriptoria e
manoscritti mediatori di civilit dal sesto secolo alla riforma di Carlo
Magno', Settimane di studio del Centro italiano di studi sui alto medioevo, 11
(Spoleto 1963), 479-504, reprinted in his Mittelalterliche Studien, vol. 2, Stutt-
gart 1967, pp. 312-27; also the works earlier cited by Laistner and Marrou.
The fundamental work is E. A. Lowe, Codices Latini Antiquiores, vols. 1-12,
Oxford 1934-71, with Addenda by B. Bischoff and V. Brown in MS 47
(1985), 317-66. This contains facsimiles and descriptions of all Latin manu-
scripts prior to the ninth century.
On the falling off of classical culture during the Dark Ages, see L. D.
Reynolds (ed.), Texts and transmission: a survey of the Latin classics, Oxford 1983,
pp. xiv-xvii.
The passages of Seneca cited by Johannes Lydus may be found in the
edition of the Naturales quaestiones by A. Gercke, Leipzig 1907, pp. 157-9,
and those of Petronius quoted by Fulgentius in the edition of the Satyricon
by K. Mller, Munich 1961, pp. 185-94; see too V. Ciaffi, Fulgenzio e Petronio,
Turin 1963. Martin of Braga's Formula vitae honestae, dedicated to the Suevic
king Mir and written between 570 and 579, is an adaptation of a lost work of
Seneca, probably the De officiis (cf. E. Bickel, RAM 60 (1905), 505-51).
Unlike his De ira, a carefully constructed mosaic of borrowings from
Seneca's treatise ofthat name, which survives in only one medieval manu-
script (Escoriai M.rii.3, of the tenth century), the Formula was an extremely
popular work in the Middle Ages and later. Often entitled De quattuor
virtutibus cardinalibus, it was commonly attributed, with perspicacity if not a
sense of poetic justice, to Seneca. For further information see C. W. Barlow,
Martini episcopi Bracarensis opera omnia, New Haven, Conn. 1950.
For bibliographical material on Cassiodorus see the important article by
A. D. Momigliano, 'Cassiodorus and the Italian culture of his time', PEA 41
(1955), 207-45, reprinted (with a more select bibliography) in his Secondo
contributo alla storia degli studi classici, Rome 1960, pp. 219-29, and Studies in
Historiography, London 1966 (paperback 1969), pp. 181-210. J.J. O'Donnell,
Cassiodorus, Berkeley 1979, stresses the limited aims that Cassiodorus had in
founding Vivarium and seeks to minimize his influence (see the review by
Averil Cameron, 'Cassiodorus Deflated', JRS 71 (1981), 183-6).
The theory that the oldest Bobbio manuscripts came from Vivarium,
advanced by R. Beer in 1911, has crumbled under the attacks of numerous
scholars, in particular that of P. Courcelle, Les Lettres grecques, 357-88. The
identifications proposed by Courcelle are examined by H. Bloch in his
Notes to Chapter 3 259
two valuable articles on the scholarly activity of the Irish by B. BischofF, 'II
monachesimo irlandese nei suoi rapporti col continente', Settimane, 4
(Spoleto 1957), 121-38, and 'Wendepunkte in der Geschichte der latein-
ischen Exegese im Frhmittelalter', Sacris Erudiri, 6 (1954), 189-279; both
are reprinted in Mittelalterliche Studien, vol. 1, pp. 195-205, 205-73. For a
handsome introduction to early Irish culture, see L. Bieler, Irland, Wegbereiter
des Mittelalters, lten etc. 1961 (English edition, Ireland, harbinger of the
Middle Ages, Oxford-London 1963).
There is a survey of the whole question of classical learning in the British
Isles in T.J. Brown, 'An historical introduction to the use of classical Latin
authors in the British Isles from the fifth to the eleventh century', in La
cultura antica nel occidente latino dal VII alF XI secolo, Settimane, 22 (Spoleto
1975), voi. 1, pp. 237-99.
There is now firmer evidence for classical texts in Anglo-Saxon England.
J. D. A. Ogilvy, Books known to the English (Medieval Academy of America
Publications, 76), Cambridge, Mass., 1967, still useful but often uncritical,
has been overtaken in part by H. Gneuss, 'A preliminary list of manuscripts
written or owned in England up to 1100', Anglo-Saxon England'9 (1981), 1-
60; M. Lapidge, 'Surviving Booklists from Anglo-Saxon England', in
M. Lapidge and H. Gneuss (edd.), Learning and literature in Anglo-Saxon Eng-
land' studies presented to Peter Ctemoes on the occasion of his sixty-fifth birthday,
Cambridge 1985, pp. 33-89. Bede's classical sources are thoroughly studied
by M. L. W. Laistner, 'Bede as a classical and a patristic scholar', Tr. Royal
Hist Soc, 4th ser. 16 (1933), 69-94, and 'The library of the Venerable Bede',
in Bede: his life, times, and writings, ed. A. H. Thompson, Oxford 1935, pp. 237-
66; both are reprinted in his collected essays, The intellectual heritage of the
early Middle Ages, ed. Chester G. Starr, Ithaca, N.Y., 1957, pp. 93-116, 117-
49. On Bede see also R. W. Southern, Medieval humanism and other studies,
Oxford 1970, pp. 1-8.
Alcuin's lines on the contents of the library at York are to be found in his
Versus de Sanctis Euboricensis Ecclesiae, 1535fr., ed. E. Dmmler, MGH, Poetae
latiniaeviCarolini, vol. 1, Berlin 1880-1, pp. 203-4, and now re-edited with a
commentary by P.J. Godman, The bishops, kings, and saints of York (Oxford
Medieval Texts), Oxford 1982.
One of the difficulties of assessing the part played by the Irish and the
English in the transmission of classical texts is the imprecise nature of the
term 'insular tradition'. This may be postulated in a variety of cir
cumstances: when one or more manuscripts of a text were actually written
in Britain, or written in insular script on the continent, or associated with
some Irish or Anglo-Saxon foundation, or showing traces, more or less
conjectural, that a lost exemplar belonged to one of these categories.
The route which the flow of cultural life followed from Italy to Britain and
then back again to the continent is very much the romantische Strasse for the
transmission of texts and it can be dramatically documented for some
biblical traditions, as in the case of the Fulda and Echternach Gospels
(Fulda, Bonifat. 3; Paris lat. 9389) and, still more clearly, the codex
Amiatinus (Laur. Amiat. 1 ), where the part played by England in the story is
beyond doubt. This great bible was written at Wearmouth or Jarrow as part
of Ceolfrid's project to produce three complete bibles, or 'pandects', and was
almost certainly used by Bede himself, but much of its decoration was
modelled on that of the lost codex grandior of Cassiodorus, written at
Vivarium and brought from Rome to Northumbria by Ceolfrid as one of the
fruits of his journey with Benedict Biscop in 678; he was taking the
Amiatinus to Rome, as a present for the pope, when he died at Langres in
716 (cf. R. L. S. Bruce Mitford, Journal of the British Archaeological Association,
3rd ser. 32 (1969), 1-25; J. W. Halporn, 'Pandectes, Pandecta, and the
Cassiodorian commentary on the Psalms', Revue bndictine, 90 (1980), 290-
300, especially 297fr.).
Unfortunately, it is not easy to substantiate such romantic journeys for
classical texts and some of the hastily posited insular traditions once fash
ionable have evaporated, such as the insular pre-archetype of Lucretius,
an author who does not appear to have reached England before the
fifteenth century (cf. F. Brunhlzl, Hermes, 90 (1962), 97-104; V.Brown,
HSCP 72 (1967), 301-10). T.J. Brown (op. cit., 281-9) examines the sig
nificance of insular symptoms in classical texts.
22 Notes to Chapter $
For the key to the Palace Library see B. BischofF, 'Die Hofbibliothek Karls
des Grossen', in Karl der Grosse, pp. 42-62 { Mitt Stud vol. 3, pp. 149-69);
he has also edited a facsimile of the manuscript containing the book-list,
Sanwielhandschrifi Diez. BSant 66. Grammatici latini et catalogus librorum, Graz
*973- I* *s noteworthy that the compiler of the book-list, an Italian, was
interested in jotting down only the classical books in the collection. For the
attribution of Lucretius and Vitruvius to the Palace School, see IMU 15
(1972), 38 n. 3, Mitt Stud. vol. 3, p. 282.
On the libraries of Charlemagne's successors, Louis the Pious and
Charles the Bald, see B. Bischoff, 'Die Hofbibliothek unter Ludwig dem
Frommen', in J.J. G. Alexander and M. T. Gibson (edd.), Medieval learning
and literature: essays presented to Richard William Hunt, Oxford 1976, pp. 3-22
(~Mitt Stud. vol. 3, pp. 170-86); R. McKitterick, 'Charles the Bald (823-
877) and his library', The English Historical Review, 95 (1980), 28-47. Among
the books one knows to have been offered to Charles the Bald or written for
him was a manuscript of Vegetius which Freculphus, bishop of Lisieux, had
pecially corrected and prepared for him. A fine copy of Apicius (Vat. Urb.
lat. 1146, written at Tours) is thought to have been another present.
The sources on which the Carolingians drew for their books are dis
cussed, with special reference to the Hoibliothek, by Bischof!, 'Das bene-
djktinische Mnchtum und die berlieferung der klassischen Literatur',
Studien und Mitteilungen zur Geschichte des Benediktiner-Ordens und seiner
^ ^ . 92 (1981), 165-90, particularly 170ft'.; see too L. D. Reynolds (ed.),
Vftrandtransmission: a survey of the Latin classics, Oxford 1983, pp. xvii-xxiv.
'.' For Hadoard and the Corbie manuscripts, see B. Bischoff, 'Hadoardus
;pW the manuscripts of classical authors from Corbie', in S. Prete (ed.),
fJtdaskaliae:studies in honor of^Anselm M. Aibareda, New York 1961, pp. 41-57,
3jprinted (in German) in Mitt Stud vol. i, pp. 49-63.
264 Notes to Chapter 3
last two articles provide an excellent study of John's use of his classical
sources. On William of Malmesbury: H. Farmer, ' William of Malmesbury's
life and works', Journal of Ecclesiastical History, 13 (1962), 39-54, and a series of
articles by R. M. Thomson, notably 'The reading of William of Malmesbury',
Revue bndictine, 85 (1975), 362-402, 86 (1976), 327-35, 89 (1979), 313-24,
and 'The "scriptorium" of William of Malmesbury', in M. B. Parker and
A. G. Watson, Medieval scribes, manuscripts and libraries: essayspresented to N. R
Ker, London 1978, pp. 117-42. William's knowledge of Cicero is particularly
impressive, but difficult to estimate precisely, because the corpus of works in
the Cambridge manuscript (copied at Cologne in 1444) seems to contain
later, continental accretions: see Thomson (1975), 372-7, (1976), 330,
(1979), 316.
The Florilegium Gallicum has been discussed in relation to the classical
texts it contains by B. L. Ullman in a series of articles in Classical Philology,
23-7 (1928-32); for a summary see the last, pp. 1-42. Also A. Gagner,
Florilegium Gallicum, Lund 1936, and the partial edition by J. Hamacher,
Florilegium Gallicum: Prolegomena und Edition der Excerpte von Petron bis Cicero,
De oratore (Lateinische Sprache und Literatur des Mittelalters, 5), Frankfurt
1975; R. Burton, Classical poets in the Florilegium Gallicum, Frankfurt 1983. It
was used by the author of another important florilegium, the Moralium
dogma philosophorum, ed. J. Holmberg, Uppsala 1929, and in the thirteenth
century by Vincent of Beau vais (cf. p. 115).
B. Munk Olsen has provided a complete catalogue of florilegia containing
excerpts from classical authors up to the year 1200, amounting to more than
seventy different compilations: 'Les Classiques latins dans les florilges
mdivaux antrieurs au XIIIe sicle', RHTg (1979), 47-121,10(1980), 115-
72. He further analyses their form and purpose in 'Les Florilges d'auteurs
classiques', Les Genres littraires dans les sources thologiques et philosophiques
mdivales. Dfinition, critique et exploitation. Actes du Colloque international de
Louvain-la-Neuve 25-27 mai 1981 (Publications de l'Institut d'Etudes Mdi
vales, 2nd series: Textes tudes, congrs, vol. 5), Louvain-la-Neuve 1982,
pp. 151-64.
For an excellent analysis of the way in which a Latin classical author can
be used and adapted for Christian purposes in this period, see J. M.
Dchanet's study of Guillaume de Saint-Thierry, ' Seneca Noster. Des lettres
Lucilius la lettre aux frres du Mont-Dieu', Mlanges Joseph de Ghellinck
vol. 2, Gembloux 1951, pp. 753-66. The popularity of Ovid in the twelfth
century, less of an aetas Ovidiana than had been commonly supposed, is
studied by B. Munk Olsen, 'Ovide au Moyen ge (du XI e au XIIe sicle)', in
G. Cavallo (ed.), Le Strade del testo, Bari 1987, pp. 67-96.
Notes to Chapter 3 269
CHAPTER 4
1. Humanism
The first important discussions of the origin of the term 'humanist'
appeared almost simultaneously: P. O. Kristeller, 'Humanism and schol-
asticism in the Italian Renaissance', Byzantion, 17 (1944-5), 346-74, and
A. Campana, 'The origin of the word "humanist'", JWI9 (1946), 60-73. For
recent bibliography and further discussion, see R. Avesani, 'La professione
dell' "umanista" nel cinquecento', IMU13 (1970), 205-323.
The wider problem of the origin of Italian humanism and its place in the
Notes to Chapter 4 271
ntext of the Renaissance is beyond the competence of this book, but some
f the more general studies which have influenced the shape of this chapter
v be conveniently mentioned here: P. O. Kristeller, various essays and
lectures collected together in Studies in Renaissance thought and tetters, Rome
oc6, and Renaissance thought, vols. 1-2, New York 1961-5; Kenneth M.
Cgtton, 'The Byzantine background to the Italian Renaissance', Proc Amer.
Philosoph. Soc 100 (1956), 1-76; F.Simone, l Rinascimento francese, Turin
10652 (updated English version by H. Gaston Hall, The French Renaissance,
London 1969); Beryl Smalley, Englishfriars,pp. 280-98; B. L. Ullman, Studies
fa the Italian Renaissance, Rome 1955^. Weiss, The dawn ofhumanisn in Italy,
London 1947, The spread ofItalian humanism, London 1964, and The Renais-
ffice discovery of classical antiquity, Oxford 1969.
Though in many respects out of date, the fundamental works on the
ffcdiscovery of classical texts remain those of R. Sabbadini: Le scoperte dei
codia latini e greci n1 secoli xiv e xv, 2 vols., Florence 1905-14, reprinted
with author's additions and corrections and an appreciation by E. Garin,
Florence 1967; Storia e critica di testi latini, Catania 1914, of which a second
edition has now been produced (Medioevo e Umanesimo, no. 11, Padua
1971 ) with new indexes and a full bibliography of Sabbadini's works.
For a concise account of the development of both dictamen and Renais-
sance rhetoric, see Kristeller, Renaissance thought and classical antiquity (ed.
M. Mooney), New York 1979, pp. 228ff. A. Grafton and L. Jardine, in their
book From humanism to the humanities, London 1986, offer a stimulating
examination of both the methods and the merits of humanist education.
For the history of humanistic script, together with biographical data on
some of the more prominent humanists and samples of their handwriting,
see B. L. Ullman, The origin and development ofhumanistic script, Rome i960;
A. C. de la Mare, The Handwriting of Italian Humanists, vol. 1, part 1, Oxford
1973. Other works are cited where relevant.
IX. The first Greekprinted texts: Aldus Manutius and Marcus Musurus
X. Erasmus
1
Erasmus classical scholarship is described in R. Pfeiffer, History of classical
scholarship 1300-1850, Oxford 1976, pp. 71-81. Pfeiffer cites a revealing
aphorism: incorrect punctuation, a tiny detail in itself, is enough to give rise
to heresy (tantula resgignithaereticum sensum). Although this notion is found
in Photius, there is no sign that Erasmus knew his writings. P. S. Allen, PRA
11 (1924), 349-68, argues that Erasmus' chief services to learning were in
editing patristic texts, mainly of Latin Fathers (his attempts to assemble
material for an edition of Chrysostom never made enough progress for
printing to begin). One important facet of Erasmus' activity is dealt with by
E. Rummel, Erasmus as a translator of the classics, Toronto 1985.
For the preparation of the Alcal Bible and Erasmus' New Testament see
the summary in B. M. Metzger, The Text of the New Testament, Oxford 19682,
pp. 96-103. Erasmus' use of manuscripts for his edition and its subsequent
revised impressions has to be worked out from various sources, including
his letters and passing remarks in his commentaries on the New Testament
(the idea of a systematic exposition of the manuscripts used for an edition is
relatively modern). The facts stated in the text depend on P. S. Allen's intro
duction to letter 373 in vol. 2 of Opusepistolarum Desiderii Erasmi Roterodami,
pxford 1906-54, pp. 164-6. Allen's account seems to be reliable, with the
jpasible exception of his statement about the Leicester codex. With regard
fp the Vatican codex B (Vat. gr. 1209), Erasmus had learned of its existence
jpl 1521, and when he was reminded of its importance some years later by
W*e Spanish humanist and theologian Seplveda he failed to respond as he
flfeould have done. In his reply to Seplveda he suggested that a Greek
*panuscript which supported some readings of the Vulgate had probably
28o Notes to Chapter 4
been tampered with, not realizing that the great age of B made this
relatively implausible; and he advanced the exaggerated but not entirely
unreasonable proposition that the only way to be sure of recovering the
original Greek was to go back to the text as cited by patristic authorities of
the third, fourth, and fifth centuries (letter 2905, written in 1534). An excel-
lent new study of Erasmus is J. H. Bentley, Humanists and Holy Writ,
Princeton 1983; see pp. 153-4, 158 on the maxim diffkiior lectio potior. (On
this point it is worth recording here a fact kindly communicated to us by Dr
R.J. Durling: Galen is very close to this concept. He can be found expressing
a preference for old words which are more difficult of explanation ( Corpus
medicorum graecorum 5.10.2.2, p. 178, 17-18) and understands that they
would have been changed into something easier if alteration of the text had
occurred (ibid., 121.17-18).)
With regard to Erasmus' intentions in editing the New Testament, H.J.
de Jonge, 'Novum Testamentum a nobis versum: the essence of Erasmus'
edition of the New Testament', JTS 35 (1984), 394-413, argues that this
work is not to be treated as an edition of the Greek text. But in defence of
the conventional view one may say: (1) Erasmus collated MSS. in both
languages, so there must be a sense in which he is trying to establish the
precise nature of the graeca Veritas, and that procedure, however incomplete
by our standards, was not slapdash by the standards of the time and must be
thought of as putting him into the category of editor. (2) It is obvious why
he puts emphasis on the Latin rather than the Greek: he wanted to be read
and knew that only a tiny fraction of the educated class had enough Greek
to follow the original (de Jonge realizes this, pp. 401, 406, without drawing
the necessary conclusion). (3) There could have been a prudential reason for
Erasmus' presentation of his work as essentially concerned with the Latin, if
there were potential critics of his and Valla's idea of concentrating on the
graeca Veritas.
The story of Erasmus and the comma Johanneum may have been slightly
embroidered; see H.J. de Jonge, Ephemerides theoogicae Lovanienses 56
(1980), 381-9. But it is clear that Erasmus exposed himself to opponents
acting in bad faith.
A good way of approaching the Adagia is to read M. M. Phillips, The
'Adages* of Erasmus, Cambridge 1964; on pp. 65-9 there is an account of the
polemic arising from Erasmus' stay in Aldus' house. His stay is also
described by D.J. Geanakoplos, Greek Scholars in Venice] see especially
pp. 273-5 for the question of the so-called Erasmian pronunciation of
Greek.
For further orientation and bibliography, P. Petitmengin, 'Comment
Notes to Chapters 4-5 281
CHAPTER 5
IV. Richard Bentley (166 2-174 2): classical and theological studies
demie tradition over the centuries. See further P. Peeters, I'/uvredes Bol'-
faidistes, 2nd ed., Brussels 1961.
Traube and Knowles give all the essential guidance for further reading
about the Maurists and Maffei. But on the latter one may also refer to an
essay by A- Momigliano, 'Mabillon's Italian disciples', in Terzo contributo alla
storia degli studi classici e del mondo antico, Rome 1966, pp. 135-52. Maffei's
letters, in addition to the works mentioned by Traube, contain some state-
ments about palaeography; see nos. 158 and 160 in the Epistolario, ed.
C.Garibotto, Milan 1955, pp. 199-201, 203-4. His palaeographical insight is
perhaps partially anticipated in Janus LascanV letter of dedication prefaced
to his edition of the Greek Anthology.
Perhaps the earliest catalogue of a manuscript collection is that published
as a memorial to the learned Spanish bishop Antonio Agustin, Antonii
Augustini Tarraconensium antistitis bibliothecae graecae anacephaleosis, Tarragona
1586. A few years later appeared D. Hoeschel's Catalogusgraecorum codicum
qui sunt in bibliotheca Reipublicae Augustanae Vindelicae, Augsburg 1595; this is
a scholarly publication with indications in the margins of the extent to
which the texts have been published. Also valuable, if somewhat less
detailed, was Thomas James' Ecloga Oxonio-Qmtabrigiensis of 1600, giving an
account of the collections of the two English universities.
(I). In addition Sir Harold Bell, EgyptfromAlexander the Great to the Arab con-
quest, Oxford 1948, provides an excellent introduction from a cultural and
historical point of view.
The codex containing Menander's Dyscolus has given us in addition a sub-
stantial proportion of his Aspis and Samia; the fragments of the latter over-
lap to some extent with those already known, with the result that in these
passages we possess two uncommonly early witnesses to the text of a clas-
sical author.
Literary papyri, including the very small number of Latin texts among
them, are listed with a bibliography by R. A. Pack, The Greek and Latin literary
textsfromGreco-Roman Egypt, Ann Arbor, Mich., 1965. For statistics showing
how the main authors are represented at various dates see W. H. Willis,
GRBS 9 (1968), 205-41. Latin papyri are surveyed in R. Cavenaille, Corpus
papyrorum latinorum, Wiesbaden 1958. An important recent addition is the
scrap of Gallus, published by R. D. Anderson, R. G. M. Nisbet, P.J. Parsons,
JRS 69 (1979), 125-55. The fragment of Livy has been edited with full
description and commentary by B. Bravo and M. Griffin, 4Un frammento del
libro XI di Tito Livio?', Athenaeum 66 (1988), 447-521. It is now in the
Coptic Museum in Cairo (inv. N15/86). Another useful reference work is
J. van Haelst, Catalogue despapyrus littrairesjuifs et chrtiens, Paris 1976.
(c) Other manuscript discoveries. A history of the Homeric Question is
given by Adam Parry in the introduction to the collected papers of his father
Milman Parry, The making ofHomeric verse, Oxford 1971, pp. xiii-xv.
Matthaei's find is discussed by O. von Gebhardt, ZentralblattfirBiblio-
thekswesen 15 (1898), 442-58.
Leopardi's disappointing experience is recounted by S. Timpanaro in Dif-
ferenze, 9 (Studi in memoria di Carlo Ascheri), Urbino 1970, pp. 357-79.
Heiberg's find of Archimedes was announced in Hermes, 42 (1907), 235fr.
On Juvenal VI and Saint Cyprian see J. G. Griffith, Hermes, 91 (1963), 104-
14. For the new letter of Saint Cyprian see also M. Bvenot, The tradition of
manuscripts: a study in the transmission of St Cyprian s treatises, Oxford 1961. The
Epigrammata Bobiensia were edited by A. Campana and F. Munari, Rome
1955. For the new lines of Rutilius Namatianus recovered from Turin F.iv
25, see M. Ferrari, 'Spigolature bobbiesi, I: In margine ai Codices Latini
Antiuiores, II: Frammenti ignoti di Rutilio Namaziano', IMU 16 (1973), 1-
41. The newly discovered letters of Augustine have been edited by J. Divjak,
Sancii Aurelii Augustini Opera Epistulae ex duobus codicibus nuper in lucem
prolatae (CSEL 88), Vienna 1981. The two manuscripts containing the
letters are Paris lat. 16861 and Marseilles, Bibl. Mun. 209, of the twelfth and
fifteenth centuries respectively.
Notes to Chapter 5 287
f
I " f /d ) Epigrophic texts. Roberto Weiss, in The Renaissance discovery of classical
UMtouityy Oxford 1969, traces the beginnings of an interest in the tangible
ujnains of antiquity. For the impact of inscriptions on literary texts, see the
action 'Rapports avec la littrature' of the 'Bulletin pigraphique1 of J. and
& Robert in the Revue des tudes grecques.
if The Res gestae of Augustus, sometimes referred to as the Monumentum
hflcyranum, has been frequently edited and it will suffice to mention the edi-
gjons ofJ. Gag, Paris 1935, and P. A. Brunt-J. M. Moore, Oxford 1967. The
laudatio Turiae has been edited with a translation and commentary by
lit. Durry, logefimbredune matrone romaine, Paris 1950. The Lyons tablet was
(fct published by G. Paradin in his De antiquo statu Burgundiae, Lyons 1542.
vpor the monument of Antiochus of Commagene and its importance for
t^e history of ancient prose style, see K. Humann and O. Puchstein, Reisen in
f&inasien und Nordsyrien, Berlin 1890, and E. Norden, Die antike Kunstprosa,
pia ed., Leipzig-Berlin 1909, vol. 1, pp. 140fr.
.1. Diogenes of Oenoanda has been recently re-edited and studied by C. W.
Chilton, Diogenis Oenoandensisfragmenta,Leipzig 1967; Diogenes of Oenoanda:
the Fragments, Oxford 1971. There is an important article on the inscription
by J. Irigoin, Studifilologicie storici in onore di Vittorio De Falco, Naples 1971,
pp. 477-85; and for recent progress in the finding of new fragments, see
M. F. Smith, Anatolian Studies 29 (1979), 68-89 (with, on pp. 87-8, a biblio-
graphy of recent finds).
For the early Christian hymn, see P. Maas, Kleine Schriften, Munich 1973,
p. 315. The statue of Socrates and another epigraphic testimony to the text
of Plato are discussed by A. Carlini, Studi sulla tradizione antica e medioevale del
Fedone^ Rome 1972, p. 74.
Pompeian graffiti have been collected and edited by E. Diehl, Pompeianische
Wandinschriften und Verwandtes, 2nd ed. (Kleine Texte fur Vorlesungen und
bungen 56), Berlin 1930. For the corpus of epigraphic poetry, see the rel-
evant parts of the Anthologialatina edited by F. Biicheler and E. Lommatzsch,
Vols. 1-3, Leipzig 19302, 1897, 1926; E. Engstrm, Carmina latina epigraphica,
Gothenburg 1911. The occurrence of armavirumque on ancient walls is docu-
mented by R. P. Hoogma, Der Einuss Vergib auf die Carmina Latina Epi-
pvphica, Amsterdam 1959, pp. 222f. The text of Propertius 3.i6.i3f is
facussed by M. E. Hubbard in 'Propertiana', CQN.S. 18 (1968), 3i8f.
VII. Epilogue
The role of the classics and the advances of scholarship in the period which
saw the growth of modern Europe constitute an exceptionally complex
288 Notes to Chapters 5-6
theme, which has not yet been the subject of a comprehensive and satisfying
study. Some guidance may be had from Wilamowitz's famous brief sketch of
the history of scholarship, Geschichte der Philologie, Berlin 1921, revised 1927,
translated into English as History of classical scholarship by A.Harris, ed.
H. Lloyd-Joncs, London 1982, and from R. Pfeiffer, History of classical scholar-
ship 1300-1650, Oxford 1976. There are also contributions in the two
volumes edited by R. R. Bolgar, Classical influences on European culture AT).
1500-17'oo, and Classical influences on Western thought, Cambridge 1976, 1979.
A challenging hypothesis about changes in educational practice is put
forward by A. Grafton and L. Jardine, From humanism to the humanities\
London 1986. Readers with an interest in an important period of English
scholarship should consult M. L. Clarke, Greekstudies in England'iy'00-1830,
Cambridge n.d. (1945). Further information is provided by C. O. Brink,
English classical scholarship: historical reflections on Bentley Porson and Housman,
Cambridge 1986; this study is discussed at length by H. D. Jocelyn, Philolog\'
and education, Liverpool 1988. Wolf's Prolegomena have been translated into
English with notes by A. Grafton, G. W. Most, and J. E. G. Zetzel, Princeton
1985.
CHAPTER 6
Our account of stemmatic theory and the history' of its evolution depends
on P. Maas, Textual criticism, Oxford 1958, and S.Timpanaro, La genesi del
metodo del Lachmann, 3rd edition, Padua 1981. Timpanaro's second and third
appendices are also important explorations of areas of stemmatic theory.
Maas' exposition is so brief as to verge on the obscure, and some of the finer
points of stemmatic theory require a full statement in order to make latent
assumptions explicit. Timpanaro, Maia, 23 (1970), 289, pointed to one such
assumption in the fourth of the inferences from our hypothetical stemma on
p. 211, where we have added a parenthesis to meet the case: the agreement
of one of the MSS. XYZ with indicates the reading of a, providedthat they
readings of the other two of the MSS. XYZ disagree with each other; if they
agree, as can happen, the tradition has been affected by contamination or
emendation. L. Canfora, Belfagor, 23 (1968), 361-4, has directed attention to
some other obscurities in Maas's presentation of the theory.
Limitations of the stemmatic method were emphasized by G. Pasquali,
Storia della tradizione e critica del testo, 2nd ed., Florence 1952, and they have
been urged more recently, but perhaps too passionately, by R. D. Dawe. The
collation and investigation of manuscripts of Aeschylus, Cambridge 1964. We have
Notes to Chapter 6 289
tried to make it clear in our text that controversy on this subject is largely
displaced. Maas knew as well as anyone else that there is no simple answer
to the problems of a contaminated tradition, but some critics have failed to
notice his explicit statement on the matter. Others, perhaps unduly
impressed by the wealth of examples in Pasquali's rather discursive but
deservedly famous book, most of which is devoted to unusual traditions,
have assumed that contamination is the rule rather than the exception, and
that consequently Maas's theory is of no practical use. We doubt whether
Pasquali could have wished to create this impression, and it must be
stressed that in many traditions the amount of contamination that has taken
place is not sufficient to prevent the useful application of stemmatic theory.
It may be worth adding here that an interesting eliminatio codicum has
recently been performed in the stemma of Aristotle's RJietoric, where the
tradition is not entirely free from contamination; see R. Kassel, Der Text der
Aristotelischen Rhetorik, Berlin 1971, pp. .54-5.
An interesting and extremely complicated discussion has been conducted
by scholars concerned with classical and medieval texts about the relative
frequency of various types of stemma. It was begun by Joseph Bdier, who
observed that an extraordinarily high proportion of the stemmata recon-
structed by editors of medieval texts have two main branches, in other
words they entail the view that two copies, and no more than two, were
made from the archetype. Bdier believed that this observation would be
found to apply to editions of classical authors. He thought that scholars
reached such conclusions because they allowed themselves to be affected by
subjective considerations such as the desire to see all questions of textual
variation in terms of a dichotomy between truth and error, and he eventu-
ally gave up hope of establishing stemmata, preferring to edit on the basis of
one manuscript. This is not acceptable as a general principle, however use-
ful or indeed necessary it may be as a procedure in dealing with certain
works of medieval literature. Later contributors to the discussion have
argued at great length about the possible statistical justification for a pre-
dominance of two-branch stemmata. Recently it has been emphasized that
proper allowance must be made for cultural conditions during the period of
transmission: is it for instance likely that many medieval books were
damaged or destroyed before more than two copies could be made from
them, or that some books were placed on deposit in central libraries avail-
able to students so that a large number of copies were taken? The fact is that
some evidence can be adduced for each of these opposing hypotheses, but
we do not have much information. A further difficulty is that the ability of
bribes to make emendations of obvious errors in the texts, either by
20 Notes to Chapter 6
Mote: the number of collections, both public and private, which contain Greek and
Latin manuscripts of direct importance for the subjects treated in this book, is very
considerable. Guides to printed descriptions or handwritten catalogues are given for
Greek by M. Richard, Rpertoire desbibliothques etdescatalogues desmanuscritsgrecs; 2nd
ed., Paris 1958, with a Supplement, Paris 1965, and for Latin by P. O. Kristeller, Latin
nmnuscript books before \ 600,3rd ed., New York 1965. The history of manuscripts since
their discovery in the Renaissance is in some cases very complicated. The formation
of some of the major libraries has been made the subject of specialized monographs
which are outside the scope of the present book. It would be useful if students were
able to refer to a short account of the movements of manuscripts from the Renais
sance to the present day, which would explain the names and present location of the
various collections and would incidentally cast an interesting light on a section of
European cultural history. At the moment there does not seem to be a study which
precisely fills this need, but the chapter on the 'Nomenclature of manuscripts' in
F. W. Hall, A companion to classical texts, is still valuable and can now be supplemented
by W. Fitzgerald, 'Ocelli nominum. Names and shelfmarks of famous familiar manu
scripts', MS 45 (1983), 214-97, 48 (1986), 397-421; for a selective treatment of the
subject see M. R. James, The wanderings and homes of manuscripts (Helps for Students of
History no. 17), London 1919, G. Laurion, 'Les Principales Collections de manuscrits
grecs', Phoenix, 15 (1961), 1-13.
A: MANUSCRIPTS
ATHOS
Almost all the monasteries on Mount Athos have a number of manuscripts; some of the
collections are extremely large and have a nucleus of books acquired in the Middle
Ages.
Lavra 184: 67
Vatopedi 747: 67
BAMBERG, Staatsbibliothek
Class. 31:108
35-108
35a: 108
42: 97, 108
44: 108
46: 97, 108
54:99
BERNE, Burgerbibliothek
T h e important collection of Jacques Bongars, which included manuscripts formerly in
the possession of Daniel and Cujas, was presented to Berne.
357: 175
363: 175
366: 105, 175
CAMBRIDGE, Peterhouse
169: 269
ERLANGEN, Universittsbibliothek
380: 107
FULDA, Landesbibliothek
Bonifatianus 3: 261
HEIDELBERG, Universittsbibliothek
The Heidelberg collection, originally very large, was once owned by the Elector
Palatine; hence the designation Palatini. After the capture of Heidelberg in 1623 the
library was given by Maximilian I, duke of Bavaria, to Pope Gregory XV in return for
financial assistance, and most of the manuscripts are still in the Vatican Library. In
1797 Napoleon removed some of the books to Paris (a fate suffered by many other
Italian libraries), and when the books were restored in 1815 the University of Heidel-
berg, aided by the king of Prussia, persuaded Pope Pius VII to let a few volumes
return to their original home.
Pal(atinus) gr. 23: 66, 182
KASSEL, Landesbibliothek
Philol. 2 27: 140
JERUSALEM, Patriarchate
36: 195
LEN, Catedral
Fragm. 3: 263
B: PAPYRI
Alter publication many papyri have been distributed to libraries or other institu-
tions which bore part of the cost of the excavations in Egypt. In such cases they
usually receive a shelf-mark in the library in question, in addition to retaining their
serial number in the original publication.
P. Berol. 5006: 50 P. Lit. Lond. 108: 196
SSM: 54 121:34
9722: 54 P. Oxy. 225: 244
9780: 10, 18 852: 4, 196
9875:4, 197 1174: 196
2192:10
P. Bodmer 4: 4, 196, 217 2258: 54
P. Cairo inv. 43227: 196 2536: 45
Qar Ibrim: 36, 247, 286 3672: 245
P. Colon, inv. 4780: 196 P. Ptrie I 5-8: 244
P. Here. 817:36 P. Rylands 26: 45
1012: 18 457: 196
1423: 244 P. Sorbonne 72: 4, 196
1471: 244 2272: 4, 196
1497: 244 2273: 4. 196
P. Lille 763+73: 14 2328: 197
76d: 246 Oxford, Bodleian Library,
P. Lit. Lond. 46: 196 MS. gr. class, a. i(P): Plate 1
70: 197
GENERAL INDEX
Campano, G. A., 277 38, 40, 81, 86, 89, 92, 96, 98, i o i , 103,
Campesani, Benvenuto, 127, 272 104, 107, 108, 109, i n , 112-13, 113-
Cangrande della Scala, 127 14, 125, 127, 131, 135-6, 138, 139,
Canter, Wilhelm. 179-80. 283 140, 143, 144, 171, 173, 174, 175, 181,
Canterbury, 87, 88, 110, 119 193, 194, 208, 210, 213, 215, 218, 225,
Capitals, Rustie, 35-6; Fiate IX; Square, 228, 250, 268, 275, 291; Academica, 24,
35; letters confused in, 223 103; AdAtticum, 127, 131, 135-6, 144,
Cappelli, Pasquino, 135 175, 225; Ad Brutum, 139; Ad
Cardona.J. A. de, 291 familires, 114, 135, 144, 210, 213;
Caroline minuscule, 94-5, 138, 262, Aratea, 104, 109; Brutus, 139; De
Plate XIII amicitia, 103, i n ; Dedivinatione, 103;
Carolingian libraries, 95 ff. De fato, 103; De finibus, 108; De
Carolingian revival, 85,92 ff.; scholarship. inventione, 82, 83, 89, 101; De lege
102 ff agraria, 31, 40, 138, 250; De legibus,
Carrio, Ludovicus, 180 103, 173; De natura deorum, 103; De
Casaubon, Isaac, 165. 176-7, 282 officia, 38, 103; De oratore, 103, 104,
Cassiodorus, 79, 82, 118, 258, 261 107, 139; De republica, 24, 36, 86, 107,
Catalans, 73 112, 194, Plate X; Desenectute, 92, 103;
Catena, 53 Hortensius, 112; In Catilinam 96, Plate
Cato, 19, 29, 30, 31, 145, 249; De agri XVI; In Clodium et Curionem, 24; In
cultura, 19, 145, 249; Speeches, 19 Pisonem, 103, 138; In Verrem, 96;
Catullus, l o i , 102, 107, 112, 125, 126, Orator, 24, 38, 139; Paradoxa, 103;
127, 135, 140, 141, 144, 176, 210, 230, Philippics, 103; Pro Archia, 108, 131;
266, 271, 272, 274 Pro Caecina, 138, 171; Pro Caelio, 136;
Celsus, 107, 145 Pro Cluentio, 131, 133, 136; Pro Fiacco,
Celsus, Titus Julius, 25 103; Pro bonteio, 103; Pro Milone, 136;
Cencio Rustici, 137 Pro Murena, 136; Pro Rabirio perd.,
Ceotfrid, abbot, 88, 261 138; Pro Rabirio Postumo, 138; Pro rege
Chalcidius, 119 Deiotaro, 96; Pro Roscio comoedo, 138;
Chalcondyles, Demetrius, 51 Pro Roscio Amerino, 136; Timaeus, 103;
Charisius, 221 Tusculan Disputations, 92, 103, 228,
Charlemagne, 92 ff, 96, 97, 98, 102, 106, 291
236, 262 Ciceronianism, 143,164-5,173 181, 281
Charles Martel, 90 Clark, A. C , 136
Charles the Bald, 103, 263 Claudian, 96, 108, 183, 266
Chartres, n o , 113, 130 Claudius, 200
Chi, 15 Clement VIII, Pope, 169
Choeroboscus, 54 Clement of Alexandria, 49, 168
Choniates, Michael, 71 'Closed' tradition, 211
Choricius, 50 Cluny, 136, 171, 274
Christians, attitude to pagan culture, Codex. 34-5, 42, 251
36ff., 48-51, 63, 79, 84-5, 231, 252, Codicology, 257
254 Collge de France, 172, 282
Christina, Queen of Sweden, 184 Collge des Lecteurs Royaux, 172
Chrysococces, George, 149 Collegium Trilingue, 178, 180
Chrysoloras, Manuel, 135, 147, 155, 277 Cologne, 108, 138, 180
Chrysostom, Saint John, 68, 152, 169, Colometry, 4-5, 15, 53
Colonna, Landolfo, 130
Cicero, 19, 22, 23-4, 26, 27, 30, 31, 36, Columba, 87
3o6 General Index
Otranto, 73, 149 134, 135-6, 143, 146, 168, 170, 194,
Otto I, 107 272-3, Plate XV
Otto III, 107, 130, 266 Petronius, 81, 100, i o i , 102, 105, 113,
Ottoman dynasty, 107 138, 140, 175, 199, 204, 223, 228, 258,
Ovid, 25, 74,81,86, 89,98,100,101,109, 282, 291
m , 112, 115, 125, 126, 132, 133,145, Petronius, Prosper, 197
170, 183, 184, 218, 266, 268; Amores, Phaedrus, 175
98; Ars amatoria, 100, 109, 266; Hero- Phalaris, letters of, 185
ides, 74, 98; Ibis, 125, 126, 133; Meta Philo of Byzantium, 56
morphoses, 74, 100; Remdia amoris, Philo (Judaeus), 58, 173
i n ; Tristia, 145 Philochorus, 18
Oxyrhynchus, 177 Philodemus, 196
Philostratus, 157
Padua, 124 fr., 130, 136, 156 Photius, 47, 48, 54, 61-4, 66, 68, 167,
Palaeography, development of, 189-92, 256-7, 279
284-5 Phrynichus, 46-8
Palaeologi, Palaeologan Renaissance, Physiologus, 256
68-9, 73flf.,217 Pichena, Curzio, 181
Palimpsests, 85-6, 112, 192-5, 259, 285 Pico della Mirandola, 277
Palladius, 235 Pietro d'Abano, 277
Panathenaea, 1 Pilato, Leonzio, 146-7
Pandects, Florentine (Codex Pisanus), Pindar, 9, 15, 45, 70, i68, 225
147 Pippin II, 89
Panegirici latini, 139 Pippin the Short, 92
Panormita, see Beccadelli, Antonio Pirmin, 90
van Papenbroek, Daniel, 189, 284 Ptsistratus, 1, 5
Paper, 59 Pithou, Franois, 175
Papyri, 46, 59, 195-7, 244, 285-6 Pithou, Pierre, 175
Papyrus, 2-4, 18, 34. 59, 244 Plantin, Christopher, 178-9, 181, 283
Parchment, 3, 34-5, 59, 75; parchment Planudes, Maximus, 66, 73-5, 234, Piate
notebooks, 34 VI
Paris, n o , 113, 117, 129, 165-6 Plato, 2, 4, 7, io, 50, 51, 56, 58, 6o, 61,
Parthenon, 71 64-5, 68, 119-20, 148, 149, 151, 155,
Passow, F., 204 201, 217, 256, 269, 277, 287, Piate III
Paul the Deacon, 22, 93, 96, 235-6 Plautus, 19-20, 21-2, 28, 29, 81, 85, 86,
Pausanias (lexicographer), 46 101, 107, 112, 139, 174, 181, 194, 248,
Pausanias (author of the Periegesis), 65 274, 282; Palatine recension (P), 20,
Pavia, 141, 147 23, 107; Casina, 19; Epidicus, 22-3;
Pehlevi, 256 Miles, 20, 23; Nervolaria, 22; Poenulus,
Pelagonius, 145 19, 248; Pseudolus, 20; Vidularia, 194
Peregrinano Egeriae, 55 Plethon, George Gemistus, 149, 157
Pergamum, 3, 16, 44, 246 Pliny the Elder, 30, 32, 81, 85, 86, 89, 91,
Persian, 57 96, 101, 108, 113, 127, 145, 162, 172,
Persius, 28, 42, 81, 86, 89, 98, 101, 109, 183, Piate XII
112, 175, 177, 218 Pliny the Younger, 25, 81, 86, 98, 99,
Pervigilium Veneris, 175 101, 127, 140, 218
Petau, Paul, 184 Plotinus, 167
Peter of Pisa, 93 Plutarch, 68, 74, 147, 153, 167, 172, 230,
Petrarch, 106, 108, 123, 128-32, 133, 232
312 General Index
Poelman, Theodore, 179 Quintilian, 25, 27, 30, 82, 83, 100, 101.
Poggio, 3 1 , 129, T34, 136-8, 139, 142, T12, 113, 137, 143, 219, 226, 272, 276
145, 150, 164, 171, Plate XVI Quintus Smyrnaeus, 149
Polemon, 16
Politian, 141, 143-6, 148, 152-4, 174,
185, 210, 213, 276, 277 Ratherius, 106-7, 127 266
Pollio, C. Asinius, 23 Ravenna, 41, 81, 97, 106, 129
Pollux, 46, 157 Reagents, use of, 194-5
Polybius, 66, 168 recentiores, non dtriores, 218, 291
Polycrates, 5 Registrum librorum Angliae, T 17, 269
Pompeius Trogus, 32, 89 Reichenau, 90, 98, 100
Pomposa, 125 Remigius of Auxerre, 105
Porphyrio, 32, 251 Remmius Palaemon, 27
Porson, Richard, 205 Res gestae Divi Augusti, 200. 287
Praetexlatus, Vettius Agorius, 37, 39, 42 Reuchlin, Johann, 147
Prehumanists, 124fr., 130, 136, 271-2, Rhazes, 256
Rhenanus, Beatus, 139, 140
277
Rhetoric, 45, 52, 58, 78
Prapea, 133
Rhodes, 17, 20, 44
Printing, 141-2, 144, 154fr, 171, 178-9.
Richard of Fournival, 116-17, 129, 269
2o8, 278, 282
Ritschi, F. W., 210
Priscian, 33, 103
Robert of Anjou, 132
Probus, M. Valerius, 21, 27-9, 31, 250
Robert of Cricklade, 113
Proclus, 120
Robortello, Francesco, 167, 281. 283
Procopius (of Gaza), 50, 53
Prohaeresius, 50 Roll, 2-4, 18, 34-5, 244
Pronunciation of Greek, 159 Rome, 81, 83, 88, 97, 141, 275
Propertius, 101, 102, 112, 113, 116-17, Royal Society, 203-4
125, 129, 138, 145, 181, 201-2, 271, Rusticius Helpidius Domnulus, 106, 129
Rutilius Namatianus, Claudius, 199, 286
273. 2**7
Prdentius, 41, 183, 265
Psellus, Michael, 67-9, 257 Sabinus, Flavius Julius Tryphonianus, 42
Pseudo-Apuleius, 83 Saint Denis, abbey of, 100
Pseudo-Callisthenes, 236, 256 Saint Gall, 87, 98, 100, 119, 137, 269
Pseudo-Dionysius the Arcopagite, 119- Saint John Lateran, 83
20 Saint Omer, abbey of Saint Bertin, 175
Pseudo-Ovid, 139, 263 Saint Victor, abbey of, 140
Pseudo-Plutarch, 5 Salerno, n o
Pseudo-Quintilian, 108 Sallust, 26, 27, 28, 29, 86, 96, 98, 101,
Pseudo-Seneca, 142 112, 127, 171, 218
Ptolemy I, 6 Sallustius, 40
Ptolemy II Philadelphus, 6 Salmasius (Claude de Saumaise), 182,
Ptolemy Euergetes II, 17 184
Ptolemy (astronomer and geographer), Salutati, Coluccio, 134-6, 136, 138, 144,
75,110, 119-20,148,162 273-4. 274
Punctuation, 4, 9, 15, 245, 247-8, 253 Samarkand, 59
Sannazaro, Iacopo, 139
Quadrigarius, Claudius, 30 Sappho, 50, 54
quadrfvium, 33, 75, 84 saut du mme au mme, 226
Quero/us, 175 Savile, Sir Henry, 169
General Index 3*3
Scaliger. Joseph Justus, 173, 174. 175, Sophocles, 53, 66, 70, 156, 173, 174,
176-7, 182, 210, 282 195-6, 227, 229
Scaliger, Julius Caesar, 173, 282 Sorbonne, college of, 166,171; library of,
Scholastic age, 114 fr. 129
Scholia, 10-18, 52, 76, 188. 254 Sosii, 25
Scholia Bobiensia, 194 Speyer, cathedral of, 130, 140
Schoppe, Caspar, 283 Statilius Maximus, 31, 40
Science, development of" Greek, 5-6 Statius, 89, 92, 96, 98, 101, 112, 125,132,
Scipionic circle, 19 138, 145, 263, 276
Second Sophistic age, 47, 185 Stemmatic method, 207II, 288-9
Secondary- tradition, 219-21 Stcphanus, sec Estienne
Sedulius, 41 Stephanus of Byzantium, 157
Sedulius Scottus, 87, 102-3 Stesichorus, 14, 246
semeia, see Signs, critical stigme, 10
Seneca the Elder, 98, 101, 183 Stilo, see Aelius
Seneca the Younger, 25, 26, 8 i , 86, 97, Stobaeus,John, 179, 185
99, 100, 101, 108, 109, i n , 112, 113, Stoics, Stoicism, 16, 45, 49, 181
114, 115, 116, 117, 125, 126, 128, 131, Stoudios monaster}', 59
132, 135, 145, 162-3, 165, 181, 182, Strabo, 6, 44, 148, 177
183, 193, 218, 223, 224, 225, 227, 228- Studemund, W., 193, 194
9, 258, 265, 267, 268, 269, 272, 281; Sturmi, 90
Apocolocyntosis, 112; De amicitia, 86. Surez, J. M , 200
193; Debenefciis, 99, 163; De dementia, Subscriptions, 3 1 , 39-43, 145-6, 252-3,
99, 163; De forma mundi, 83; De vita Plate IX
pains, 86, 193; Dialogues, 101, 109, Suda, 66, 219-20, 238
117, 227, 267, 269; Letters, j^j, 101, Suetonius, 20, 28, 81, 99, 101, 105, 112,
103, 108, 112, 132, 223, 224, 225, 228- 113. 127, 162, 276
9, 230, 231, 265, 268; Naturai Ques- Suidas, see Suda
tions, 81, 100, 114, 258, 281; Tragedies, Symmachi, 40, 41, 81
101, 116, 117, 125, 126, 128, 132, 135, Symmachus, Aurelius Memmius, 40, 83
145, 181, 182, 183, 218, 269, 272 Symmachus, Q. Aurelius, 3 6 - 7 , 4 0 , 4 1 , 42
Sens, abbey of Sainte Colombe, 174 Symmachus (translator of Old Testa-
Septuagint, 49 ment), 49
Seplveda, Juan Cins de, 279 Syntax, study of, 45
Serenus, Quintus, 96 Syriac, 17, 55-7, 188, 255
Servius, 27, 28, 32-3, 37, 91, 175
Servius Claudius, 21, 22, 23 Tabourot, Etienne, 174
Sextus Empiricus, 167 Tacitus, 32, 99, 101, 102, 109, 112, 133,
Sicily, 73, 118-20, 146 139, 140, 165, 181, 183, 200, 218, 234,
Sigla, 105, 179, 238 236, 283, 291, Plate XIV
Signs, critical, 11-14, 21, 28, 49, notes to Tacitus, the emperor, 32
Plate II Tarasius, 62
Silius Italicus, 100, 137, 180 Tassin, Dom Rene, 193
Simon, Richard, 188, 284 Terence, 20, 26, 28, 29, 33, 36, 81, 86, 96,
Sinope, : i 98, t o i , 112, 144, 162, 168. 171, 186,
Sirmond, Jacques, 200 248, 262-3; Adeiphi, 29; Andria, 20,
Sixtus V, Pope, 169 248
Socrates, 201 Teubner, B. G., 179, 204
Solinus, 32 Teucer of Babylon, 256
3H General Index
Textual criticism, early manuals of, 167, Turks, 47, 72-3, 147
179-80, 283 Turnebus, Adrianus, 173-4, 175, 282
texttis receptus, 208-9 Type, founts of, Greek, 154, 156; italic,
Theagenes of Rhegium, 9 138; roman, 138
Theocritus, 45, 48, 153, 155, 156, 158 Typographia Vaticana, 169
Theodegius, 58 Tzetzes, John, 71
Theodore the geometrician, 58
Theodore Ducas Lascaris, 72 Uncial script, Greek, 59-60; Latin, 35,
Theodore of Tarsus, 87-8, 119 36, 88, 94, Plates X, XI
Theodoric, 79 Universities, ancient, 44, 51, 54;
Theodosian code, 52 medieval, 58, 61, 67, n o
Theodosius I, 37 Uspensky Gospels, 59
Theodosius II, 54 utrum in aiterum, 221
Theodotion, 49
Theodulfus of Orleans, 93, 105, 265 Valerius Flaccus, 92, 99, TOO, 10I, I O 8 ,
Theognis, 66 113, 125, 137,144. 145. 183
Theognostus, 54 Valerius Maximus, i o i , 103, 104, 105,
Theon, 44 113, 129, 1 3 2 , 1 7 5 , 2 7 4
Theophrastus, 7, 55, 56, 148, 156, 173, Valla, Lorenzo, 131, 141, 142-3, 150,
177. 197 152. 153, 164, 171, 275-6, Plate XIV
Theopompus, 18 Vargunteius, Q 21
Thersagoras, 10 Varro, 3, 19, 20, 21, 22, 23, 109, 127,133,
Thessalonica, 69, 73 145
Thomas Magister, 47 Vegetius, 89, 98, 103, 112, 263
de Thou, Jacques Auguste, 168, 177 Velleius Paterculus, 100, 101, 139, 140,
Thucydides, 4, 18, 47, 68, 148, 153, 217, 2
75
"5 Vellum, see Parchment
Tibullus, 96, 101,108,112,113,115,116, Venice, 73, 148, 150
" 5 > 135 Vercelli, chapter library of, 135
Timon of Phlius, 6 Vergil, 21, 25, 26-7, 28, 2 9 , 3 0 , 3 3 , 36, 37,
Timotheus, 4, 197 39, 41, 81, 86, 89, 99, 100, 101, i n ,
Tiro, 30, 31 112, 115, 131, 132, 144, 145, 183, 201.
Toledo, n o , 120 219, 229, 249-50, 263, 276, 287, Plate
Tours, 96, 98, 99, 104, 105, 145, 263, IX; Appendix Vergihana, 133
notes to Plates XI, XIII Verona, 81, 107, 125, 127, 131, 140.
Trajan, 25 190-1
Transliteration, 60-1 Verrius Flaccus, 22, 25, 27, 32, 235-6
Transposition, 229-30 Vesalius, 204
Trapezuntios, George, 150 Vespasiano da Bisticci, 141
Traube, Ludwig, 191-2 Vettius Valens, 256
Traversari, Ambrogio, 277 Vettori, Pier, 166, 168, 174, 281
Trebizond, 150 vtus C/uniacensis, 136, 140, 171
Trevet, Nicholas, 117, 128 Vibius Sequester, 235
Trent, Council of, 165 Vicenza, 127
Triclinius, Demetrius, 75-8, 174, 231-2, Victorianus, Tascius, 40, 42
234, notes to Plates VI, VII, Vili Villoison, Jean Baptiste Gaspard d'Ansse
trivium, 33, 75, 84 de, 198
Tryphon (bookseller), 25 Vincent of Beau vai s, 115, 268
Tryphon (grammarian), 44 Vndolanda tablets, 247
Generai Index 3*5
I. Oxford, Bodleian Library, MS. gr. class a.i(P). 2nd cent. The papyrus
known as the Hawara Iliad. The fibrous nature of the material is clearly
visible.
II. Venice, Biblioteca Marciana, MS. gr. 454, fol. 41 r . 10th cent. This famous
book from the collection of Cardinal Bessarion is generally known as
Venetus A of the Iliad.
Plates I and II illustrate the relation between the Alexandrian critical
signs and the commentary. They both show the same passage of the Iliad(ii.
856fr.); the papyrus has the critical signs in the margin but no scholia, the
manuscript has both. It is not surprising to find that the signs are not quite
identical. A dipl seems to be the correct sign at 856, but the papyrus appar-
ently has the dipl'periestigmen\ the scholia have simply a geographical note
about the Alizones and add that there was another Hodios in the Greek
camp, but there is no indication of a difference between Aristarchus and
Zenodotus here. At 858 the papyrus has a dipl, and the manuscript a note
that the name Chromis is elsewhere given as Chromios. 859-61 are obel-
ized in the papyrus, 860-1 in the manuscript, and the ground given in the
latter is that in the battle by the river the death of Chromis is not related,
whereas Homer is careful to record the death of any commander of a con-
tingent; this is a good example of an argument that does not satisfy the
modern reader. At 863 both books have a dipl, and the manuscript gives a
geographical note on the use of the name Phrygia.
III. Oxford, Bodleian Library, MS. E. D. Clarke 39, fol. 113'. A.D. 895. Plato;
the plate shows the opening of the Sophist. The text was written for Arethas
by the scribe known as John the Calligrapher, who also prepared for
Arethas a copy of Aelius Aristides (MSS. Laur. 60.3 and Paris gr. 2951). The
marginal scholia are in Arethas' own hand; the first note begins: av-nq 77
Xaia ov\ toe rivzc vireXaov rrjc '/am'ac icnv dXX nfjc 'IraXiac, tin e
CTpdwvi TTeiQecdai r yOjyp<f>aj.
IV. Oxford, Bodleian Library, MS. Auct. V. 1.51, fol. 94r. Late 10th cent.
Notes on the Odyssey. The plate shows the outline of the story of Book XI
(the descent to the underworld), followed by the beginning of the
vocabulary list for that book. Such aids were necessary for readers in the
Middle Ages and their existence throws light on the school curriculum. This
3i8 Notes to the Plates
MS. is from the collection of Giovanni Aurispa and later belonged to the
monastery of San Marco in Florence. The opening clause reads: d-nayyeXXei
nwc Kara rr\c KipKrjc ivroXc Xafiv tc 'Atov Ka-rqXQtv.
V. Oxford, Bodleian Library, MS. Auct. T. 4.13, fol. i32 r . n t h cent. The
archetype of the works of Epictetus. It is thought that Arethas possessed a
manuscript of this text, of which the Oxford MS. is perhaps a direct copy.
The script displays a certain number of abbreviations. The opening words
read: Xevdepoc criv >v wc ovXtrai, v oV vayt<cai CTIV OVT
KioXvcai tdcacOai.
VII. Oxford, Bodleian Library, MS. Holkham gr. 88, foi. 207". 15th cent.
Aristophanes. This copy shows the text and scholia in the recension
prepared about a century earlier by Demetrius Triclinius; it is the only
known manuscript source for his scholia to four of the plays. Note in this
plate Triclinius' scholium on metre at the foot of the page (beginning rj
c0cic roi) -napvroc SpfjLCLToc) and his misleading title to the main body
of scholia, 'old scholia by Aristophanes the grammarian'.
IX. Florence, Biblioteca Medicea-Laurenziana, MS. Laur. 39.1, fol. 8r. This
is the codex Mediceus of Vergil, written in Italy in the 5th century. It was
corrected at Rome by Asterius, consul in 494, and later found its way to
Notes to the Plates 319
Bobbio. The script is rustic capital. The plate shows the end of the Eclogues
(10.61-77). Abbreviations are few, here restricted to B- (bus) and Q*
(cue). Among the corrections, apparently by Asterius himself, we may note
in line 62 the change from DRUSUM to RURSUS and NABIS to NOBIS,
and in line 70 HAES corrected to HAEC. In line 63 the reading RURSUSM
suggests a duplex lectio at an earlier stage in the tradition (RURSUS/M). In
the space left vacant at the end of the Eclogues Asterius has added a sub
scription recording his work on the manuscript, rounded off with a poem in
elegiacs. The subscription is written in a small hand and is somewhat
defaced; the first part reads (the abbreviations are expanded and enclosed
within brackets): Turcius Rufius Apronianus Asterius v(ir) c(larissimus) et
inl(ustris), ex comit domest(icorum) protect(orum), ex com(ite) priv(atarum)
largit( ionum), expraef\ ecto) urbi, patricius et consul ordin ( anus) legi et distinexi
codicemfratris Machariiv ( in) c( larissimi) non meifiduciaset eius cuisiet ad omnia
sum devotus arbitrio XI Kai Mai(as) Romae.
X. Biblioteca Apostolica Vaticana, MS. Vat. lat. 5757, fol. 171'. The
famous palimpsest of Cicero's De republica. Originally a de luxe edition of
Cicero, it was reused in the 7th century at Bobbio to copy a text of Au
gustine on the Psalms. The primary script is a bold uncial of the late 4th or
early 5th century, the secondary script is a small uncial of the 7th century.
Here we have part of De republica 2.33. The lower text reads: ENIMSERPIT/
SED VOLAT IN/OPTIMUM STA/TUM INSTITU/TO TUO SERMO/NE
XI. Paris, Bibliothque Nationale, MS. lat. 5730, fol. 77v. This uncial manu
script of Livy's third decade was written in Italy in the first half of the 5th
century and is the parent of all the complete extant manuscripts of this
decade. A direct copy of it, written at Tours about the year 800, is shown on
Plate XIII. For the history of this manuscript in the Carolingian Renaissance
see pp. 96-7. The parchment is so fine and thin that in places, as here, the
writing shows through from the other side of the leaf. Both this plate and
Plate XIII show the beginning of Book XXIII; in line 11 a second hand has
'corrected' MOPSIORUM to COMPSINORUM, and this further change in
an already corrupt passage has established itself in the text of the copy
(Plate XIII).
XII. Leiden, Bibliotheek der Rijksuniversiteit, MS. Voss. Lat. F. 4, fol. 20v.
This beautiful manuscript of the Elder Pliny, written in Anglo-Saxon
320 Notes to the Plates
XIII. Biblioteca Apostolica Vaticana, MS. Vat. Reg. lat. 762, fol. 32".
Caroline minuscule. This manuscript of Livy's third decade was written at
Tours about the year 800 and copied directly from the 5th-century uncial
manuscript shown on Plate XI. Both plates show approximately the same
passage from the beginning of Book XXIII. The survival of both ancient
exemplar and minuscule copy enable one to examine the mistakes which
arise when a medieval scribe copies an ancient book, and the errors made
in this transcription have been collected and studied (F. W. Shipley,
Certain sources of corruption in Latin manuscripts, New York 1904). The words
are divided for the most part, there is little punctuation, and abbreviations
are few, e.g. q\ (que), b\ (bus), p (prae). Some cursive elements survive
from earlier scripts, the open a which appears along with the other forms
of the letter and the ligatures of et, ri, st. The open a disappeared in time,
and the majuscule TV, here used alongside the minuscule form, later had a
more restricted use.
XV. London, British Museum, MS. Harley 2493, fol. IOI V . The history of
this manuscript of Livy, written about the year 1200 and later in the
possession of both Petrarch and Valla, is told on pp. 130-1. The plate
shows one of the passages of Livy (21.46.3) used by Valla to discredit the
scholarship of his rivals Panormita and Facio (see pp. 142-3). The manu
script offers the corrupt ex quo propinquo. Valla points out that, while his
rivals had failed to see anything wrong with the transmitted text, Petrarch
had long since altered ex quo to ex loco, and Petrarch's correction can still
be seen in the text. In the margin Valla has written his own conjecture
exque, an emendation accepted by modern editors. One may note in pass
ing the increased use of abbreviation at this period.
Notes to the Plates 321