Obscuritas in Medieval and Humanist
Translation Theories
Reka Forrai
Despite the continuous attempts of medieval ists to d i spel it. the qualifier
saecula obscura still hangs over the Middle Ages like a dark cloud. Obscurity,
moreover, is a l i terary topos as weil as a historiographical one. Medieval
culture was often labeled obscure by poets and historians a l ike:
neither the Humanist Petrach, nor the enlightened Gibbon, for example,
thought very highly of it. But by calling the Latinity of the Hisperica
Famina or that of a labyrinthine scholastic argument obscure, we do
nothing more than admit that these texts are inaccessible to us.
Languages age and so do translations. Generation after generation,
words-regardless of whether they are used to write legal texts, philosophy,
poems or private letters-go from clear to blurred, transforming
reading into deciphering. All great works of Iiterature in the western
canon are re-translated by a lmost every generation since, after a while,
the language of the translation no Ionger clarifies, but obscures the
meaning of the original text. Translation is thus a particularly useful angle
from which to study obscurity, especially from the comparative perspective
of two historical periods like the Middle Ages and the
Renaissance. The Renaissance Greek-Latin (re)translation movements
justified and glorified themselves by condemning the medieval renderings
as obscuring their originals. Judging the medieval period according
to our own standards of clearness is a practice we have inherited from
the humanists. Many of our m isconceptions about the techniques of medieval
translation come from taking for granted the human ist critique of
them.
Rather than arguing that medieval translation practices were not abstruse,
I propose to investigate the different understandings of the term
obscuritas from Antiquity to the Renaissance. Obscurity was understood
in very d i fferent, sometimes Contradietory ways in ancient, medieval and
human ist translation theory. At first glance, two different interpretations
of obscurity are appa rent. Readers use it critica lly when obscurity is seen
1 58 RKA FORRAI
as darnage done to a clear text by an unskil led translator. Translators, on
the other hand, use it apologetically, attributing obscurity to the original
text itself. One’s first i mpression then is that readers blame the transla·
tors whenever they fail to understand a text, whereas translators blame
the text whenever they are unable to translate it clearly. But on closer
reading, much more is involved. ln the first case, the term is applied to a
fau lty translation. Here, obscuritas is an unfortunate new layer covering
the original text that has been produced by the shortcomings of the
translator’s craftsmanship and has to be removed i n order for the text to
be understood. l n the second case, it describes an i nherent characteristic
of the source-text. ln this instance, the obscure material usually strongly
resists the translator’s efforts. Obscurity belongs to the text’s nature: it is
intended to slow down and deepen the reading process. I will distinguish
between these two approaches by calling them rhetorical and
philosophical obscurity. Thus I argue for the existence of a positive d i ·
mension of obscurity i n the Middle Ages, which i s lost i n human ist rheto·
ric.
Already in classical Antiqu ity, Romans thought there was something
inherently obscure in the Greek language. For Lucretius (c. 99-55 BCE),
translating Greek philosophical ideas into Latin verses meant also a pu·
rification and simpl ification of an overly complicated system . ln this
process, the poverty of Latin is turned into an advantage:
Nor da I fail to understand that it is difficult to make clear the dark discoveries
of the Greeks in Latin verses, especially since we have often to employ new
words because of the poverty of the language and the novelty of the matters.1
The simplicity of Latin was seen to be in sharp cantrast with the sophis·
tication of Greek. The competitive Roman spirit translated this opposi·
tion into the antithetic pair of clearness versus obscurity, straight·
forwardness versus confusing intricacy.2
„Nec me animi fallit Graiorum obscura reperta I diff1cile in lustrare Latinis uersi·
bus esse. I multa nouis uerbis praesertim cum sit agendum I propter egestatem
linguae et rerum nouitatem“ (Lucretius, Oe rerum natura, I, 1 36-39, trans. W.H.D.
Rouse [Cambridge, Mass.: Harvard University Press, 1 959; reprint of the revised
third edition from 1937], 1 2-13).
Cf. Joseph Farrell, Latin Letters and Latin Cu/ture from Ancient to Modern Times
(Cambridge: Cambridge University Press, 2001) , 50-51 : “ l n his doxography Lu·
cretius systematically debunks the idea that Greek is superior to Latin as a me·
dium for poetry and phi losophy on every score: its supposedly greater beauty
and mell ifluous qualities. its lar9er vocatulary, the ease with which it forms com·
pounds, its capacity for subtle philosophical expression, all are revealed as traps
that Iead to obscurity. muddled thinki ng, silliness.“
MEDlEYALAND HUMANIST TRANSLATION THEORIES 159
l n his chapters on obscurity, Quintilian (born c. 35) opposed obscuritas
to perspicuitas, clarity 3 According to him, there are many ways to
create misunderstandings: excessively compl icated4 or excessively concise5
speech can be equally obscure, as are rhetorical figures when they
are used carelessly or excessively. He also cautioned rhetors against
those who value obscurity as a positive concept, confusing foggy formulation
with deepness of thought.G Ambiguitas is a synonym for obscuritas
Quintilian, lnstitutio oratoria B, I I, 1-1 1 , 1 2-21, trans. H. E. Butler (Cambridge,
Mass.: Harvard University Press, 1921 ), 196-208. On perspicuitas in translation
theory, see Frederick M. Rener, Language and Translation from Cicero to Tyler
(Amsterdam: Rodopi, 1989), 77-79.
„A greater source of obscurity is, however, to be found in the construction and
combination of words, and the ways i n wh1ch this may occur are still more numerous.
Therefore, a sentence should never be so long that it is impossible to
follow its drift, nor should its conclusion be unduly postponed by transposition
or an excessive use of hyperbaton. Still worse is the result when the order of the
words is confused as in the line: ln the midmost sea I Rocks are there by ltalians
altars ca/led‘ („Plus tarnen est obscuritatis in contextu et continuatione sermonis,
et plures modi. Quare nec sit tam longus ut eum prosequi non possit intentio, nec
traiectione vel ultra modum hyperbato finis eius differatur. Quibus adhuc peior
est mixtura uerborum, qualis in illo uersu: ’saxa uocant ltali med11s quae 1n fluctibus
aras'“: Quintilian, lnstitutio oratoria 8, II, 14, pp. 204-05).
„Others are consumed with a passion for brevity and omit words which are actually
necessary to the sense, regarding it as a matter of complete indifference
whether their meaning is intelligible to others, so long as they know what they
mean themselves. For my own part, I regard as useless words which make such a
demand upon the ingenu ity of the hearer“ („Aiii breui tatem aemulati necessaria
quoque orationi subtrahunt uerba, et, uelut satis sit scire ipsos quid dicere uelint,
quantum ad alias pertineat nihili putant: at ego uitiosum sermonem dixerim
quem aud1tor suo ingenio intellegit“; Quintilian, lnstitutio oratoria 8, II, 1 9, pp.
206-07).
„Such express1ons are regarded as ingenious, daring and eloquent, simply because
of their ambiguity, and quite a number of persans have become infected by
the beliefthat a passage which requires a commentator must for that very reason
be a masterpiece of elegance. Nay, there is even a class of hearer who finds a Special
pleasure in such passages; for the fact that they can provide an answer to the
riddle fills them with an ecstasy of self-congratulation, as if they had not merely
heard the phrase, but invented it“ („l ngen1osa haec et fortia et ex ancipiti diserta
creduntur, peruasitque iam multos ista persuasio, ut id [iam] demum eleganter
atque exquisite dieturn putent quod interpretandum sit. Sed auditoribus etiam
nonnullis grata sunt haec, quae cum intellexerunt acumine suo delectantur, et
gaudent non quasi audierint sed quasi inuenerint“; Quintilian. lnstitutio oratoria
8, II, 21, pp. 208-09).
1 60 RKA FORRAI
in his rhetorical terminology 7
While a rhetorician should avoid obscurity, a philosopher can choose
to use it, if so he pleases. Late Antique philosophical commentaries often
claimed to clarify the thoughts of intentionally obscure or ambiguous
philosophers like the pre-Socratics or Aristotle. ln this case, the degree of
a text’s obscurity was considered a measure of the difficulty of its themes
and arguments. Calling a phi losopher obscure was not a critical judgment,
but an Observation about the Ievei of complexity of the work. This
obscurity could be caused by the compl icated suqject m atter, the philosopher’s
knotty argument, or the reader’s Ievei of understanding. According
to Cicero ( 1 06-43 BCE), phi losophical obscurity has two
acceptable sources: a philosopher may choose to write obscurely or his
subject-matter may require it.s
„Above alt, ambiguity must be avoided, and by ambiguity I mean not merely the
kind of which I have already spoken, where the sense is uncertain, as in the
clause Chremetem audivi percussisse Oemean, but also that form of ambigu ity
which, although it does not actually resu lt in obscuring the sense, falls into the
same verbal error as if a man should say visum a se hominem librum scribentem
(that he had seen a man writing a book). For although it is clear that the book
was being written by the man, the sentence ts badly put tagether and its author
has made it as ambiguous as he could. Again, some writers introduce a whole
host of useless words; for, in their eagerness to avoid ordinary methods of expression,
and allured by false tdeals of beauty they wrap up everything in a multitude
of words simply and solely because they are unwilling to make a d irect and
simple Statement of the facts: and then they link up and involve one of those
long-winded cl auses with others like it, and extend their periods to a lengths beyond
the compass of mortal breath“ („Vitanda in primis ambiguitas, non haec
solum, de cuius genere supra dieturn est, quae incertum intelleeturn facit, ut
‚Chremetem audiui percussisse Demean.‘ sed i l l a quoque, quae etiam si turbare
non potest sensum in idem tarnen uerborum uitium incidit, ut si quis dicat ‚uisum
a se hominem librum scribentem‘. Nam ettam si librum ab homtne scribi patet,
male tarnen composuerit, feceritque ambiguum quantum in ipso fuit. Est etiam i n
quibusdam turba inanium uerborum, qi. d u m communem loquendi morem reformidant,
ducti specie nitoris circumeunt omnia copiosa loquacitate, eo quod
dicere nolunt ipsa: deinde illam seriem cum alia simili iungentes miscentesque
ultra quam ullus spiritus du rare possit extendunt“: Quintilian, lnstitutio oratoria
8, I I, 16-17, pp. 204-07).
„Obscurity is excusable on two grounds: it may be del iberately adopted, as in the
case of Heraclitus, ‚The surname of the Obscure who bore,/So dark his phi losophic
Iore‘: or the obscurity may be due to the abstruseness of the subject and not
of the style – an instance of this is Plato’s Timaeus“ („Duobus modis sine reprehensione
fit, si aut de industria facias ut Heraclitus – cognomento qui OKOTEtv6<;
perhibetur quia de natura nimis obscure memoravit – aut cum rerum obscuritas
MEDIEVAL AND HUMANIST TRANSLATION THEORIES 1 6 1
Since the late Antique and medieval Greek-Latin translation canon
consisted mostly of philosophical and theological works, this concept of
philosophical obscurity was more prevalent than the rhetorical one. But
translators sti I I faced the question of what they should do with such obscure
passages? Should they leave them obscure or attempt to simpl ify
and clarify them? l n philosophical education this was the duty of the
commentator, but it was not clear whether translators were also commentators
or whether they should leave interpretation to someone eise.
Rufinus of Aquileia (340/345-41 0), for example, chose to ernend Origen’s
( 1 84/185-253/254) work as weil as translate it. ln his prologue,
he affirmed that Origen’s On the Principles was in all respects d i fficult
and obscure, and that its subject-matter gave philosophers countless
troubles.9 This statement was followed by a brief description of his
methodology, in which he admitted that he rearranged Origen’s passages
as he had found it suitable, in order to clarify obscure ones-he claims.
however, that he did that using Origen’s own words from elsewhere.1o
non verbarum facit ut non Intell igatur oratio, qualis est in Timaeo Platonis“;
Cicero, Oe finibus II. V, 15, trans. H. Rackham [Cambridge, Mass.: Harvard University
Press, 1 91 4], 94-95). He then goes on to mention a third type of obscurity.
which has no explanation and 1s the fault of the writer. Cf. Jonathan Barnes,
„Metacommentary,“ in Oxford Studies of Ancient Phi/osophy 10 (1992): 267-8 1 .
See also Jaap Mansfeld, „l nsight by hi ndsight: Intentional Unclarity i n Presocratic
Proems,“ in Bulletin of the Institute of Classical Studies, 40 (1995): 225-32. For a
detailed discussion of the understanding of ambiguum and dubitabilis in medieval
philosophy see Orago Calma, „Du bon usage des grecs et des arabes. RMiexions
sur Ia censure de 1 277,“ in Christian Readings of Aristotle from the Middle
Ages to the Renaissance, ed. Luca Bianchi, Studia Artistarum 29 (Turnhout: Brepols,
201 1 ), 1 1 5-84.
„et praecipue istos, quos nunc exigis ut Interpreter, id est peri archon, quod uel
de princ1piis uel de principatibus dici potest, qui sunt re uera al1as et obscunssimi
et d ifficillimi. Oe rebus enim ibi talibus disputat, in quibus philosophi omni
sua aetate consumpta inuenire potuerunt nihil“; Tyrannii Rufini Opera, ed. M. Simonetti,
CCSL 20 (Turnhout: Brepols, 1961), 246. Cf. Marguerite Harl, „Origene et
les Interpretations patristiques grecques ce l“obscurite‘ biblique,“ Vigiliae Christianae
36, 4 (1982): 334-71.
10 „lf, however, spea king as he does to men of knowledge and discernment, he has
occasionally expressed h irnself obscurely in the effort to be brief, I have. to make
the passage clearer, added such remarks an the same subject as I have read in a
fuller form in his other books, bearing in mind the need for explanation. But I
have sa1d nothing of my own, simply giving back to him his own Statements
found in other places“ (Origen, On First Principles, trans. G. W. Butterworth [New
York: Harper and Row, 1966], lxiii); „Si qua sane uelut peritis iam et scientibus
loquens, dum breuiter transire uult, obscurius protulit, nos, ut manifestior fieret
162 REKA FORRAI
Rufinus’s method of dea l ing with the author’s obscurity i s thus an interventionist
one: Origen had supposed that his readers would be knowledgable,
but Rufinus did not and thus tried to make explicit whatever
was implicit in the original. Brevity here is a synonym for obscurity and
it was to be avoided because the danger of obscurity in a theological text
is that it can Iead to heretical interpretation. Rufinus also argued that if
knowledgable readers or scribes don’t ernend the text, then more obscur
i ties w i l l get generated for the readers.11 l n their debate on translating
Origen, Jerome {347-420) and Rufinus thus held opposite views about
the role of the translator: Jerome contested Rufinus’s tactic of combining
the two functions of translator and commentator.
The second most obscure Greek theologian after Origen is arguably
Pseudo-Dionysius the Areopagite. He was successfully translated into
Latin in the ninth century by John Scottus Eriugena. Eriugena chose a
different path from Rufinus. l n the preface to his translation of PseudoDionysi
us, he warned his eventual readers of the danger of finding his
version obscure, because he, as a faithful interpreter, had to leave the
text i mpenetrable.12 But he intended this more as a clarification than an
apology. ln his view, the obscurity was a lready there in the original, and
one way to try to understand it was to use the work of a commentator
like Maximus Confessor.13 Eriugena didn’t consider it his duty to make
locus, ea quae de ipsa re in aliis eius libris apertius legeramus adiecimus explanationi
studentes. Nihil tarnen nestrum diximus, sed licet in aliis locis dicta, sua
tarnen sibi reddidimus“ ( Tyrannii Rufini Opera, 246}.
11 „(everyone who shall either transcribe or read these books) shall ernend it and
make it distinct to the very Ietter, and shall not allow a manuscript to remain incorrect
or indistinct, lest the difficulty of ascertaining the meaning, if the manuscript
is not distinct. should increase the obscurities of the werk for those that
read it“ (Origen, On First Principles, lxiv); „et inemendatum uel non distinctum
codicem non habeat, ne sensuum difficultas, si distinctus codex non sit, maiores
obscuritates legentibus generet“ ( Tyrannii Rufini Opera, 246}.
12 „si obscuram minusque apertarn praedictae i nterpretationis seriem iudicaverit,
videat me interpretem huius operis esse, non expositorem“; E. Oummler, Ernst
Pereis and others, eds. MGH Epistolae 6 Karolini Aevi 4 (Berl i n : Weidmann, 1 902-
1925}, 1 59.
13 „Fortassis autem qualicunque apologia defensus, non tarn densas subierim caligines,
nisi viderem, praefatum beatissimum Maximum saepissime i n processu sui
operis obscurissimas sanctissimi theologi Oionysii Areopagitae sententias, cuius
symbolicos theologicosque sensus nuper Vobis similiter jubentibus transtuli,
introduxisse, mirabilique modo dilucidasse, in tantum, ut nulle modo dubitarem,
divinam clementlam, quae illuminat abscondita tenebrarum, sua ineffabili providentia
hoc d isposuisse, ut ea quidem, quae nobis maxime obstrusa in praedictis
MEDIEVALAND HUMANIST TRANSLATION THEORIES 163
the text maius apertarn (clearer) than it is, but delegated this task instead
to the expositor. the commentator on the work. ln the case of Rufinus,
these were overlapping functions, the translator having fu l l powers over
the author. But what seemed to be a possibility in Rufinus’s late Antiquity
was not even considered in the Middle Ages. Respect for the authority
of the theologian and fear of responsibility for the heretica l
accusations that might eventually result from combining interpretation
and translation reduced the translator’s freedom.
Obscurity was thus valued and respected in theological d iscourse.
But what about other l iterary genres? The Neapolitan translation school
that flourished in the ninth and tenth centuries and specialized in hagiographic
texts despised obscurity deeply seeing in it a vice of translation.
Admittedly, the sources of this view arealso more problematic: condemnation
of the previous version was often part of thej ustification for a
new translation and thus cannot always be taken at Face value. However,
it is not by chance that these criticisms occurred mostly in the context of
translating hagiography, that is to say, a type of narrative, and not technical
writing.
One of the translators, Bonitus, complains both about the absurdity
and the obscurity of the earlier version of the Gesta Theodori.14 His colleague,
Guarimpotus, in his prologue to the Passio Blasii (BHL 1 380-
1 379), claimed that the other translation had lost the meaning and the
clarity of the original, truth had been replaced by falsity, clearness by obscurity,
and wise words had been turned i nto stupidity.1s He considered
it the duty of the translator to groom the text by reordering, cutting out
the superficial parts, adding what was missing and clarifying what was
unclear.16 The genre and the use of the texts required a certain Ievei of
stylistic attractiveness to faci l itate oral understanding. As Guari mpotus
beati Oionysii libris, aut vix pervia, sensusque nostros fugere videbantur,
aperi ret, sapientissimo praefato Maxime lucidissime explanante“; Maximi
Confessoris, Ambigua ad lohannem, iuxta Johannis Scotti Eriugenae latinam interpretationem,
ed. E. Jeanuneau, Corpus Christianorum Series Graeca 18, 3 (Turnhout:
Brepols, 1988), p. 3, I. 15-25.
14 „Tanta eas absurditate faminum, tantaque obscuritate sensuum replevere“; Bonitus,
„Vita Theodori,“ in AASS, February, vol. 2, 30-31.
lS „de virissimis falsa, de liquidissimis obscura ac de praeclaris reddire turpia“;
Guari mpoto, Passio 8/asii, in Paul Devos, „L’oeuvre de Guarimpetus hagiographe
napolitain,“ Analeeta 8ol/andiana 76 (1958): 1 57.
16 „inordinata componimus, superflua resecamus, quod deest adhibemus, quodque
obscurum est ad liquidum ducere curamus“; Guarimpoto. Passio 8/asii, 1 58.
1 64 RKA FORRAI
also argued, it is i mportant that a text that is intended to be read and I istened
to in the liturgy should avoid being ridiculed by the audience.17 For
these translators of hagiographic texts, there was no obscurity i n the
source-text. lt was caused, rather, by the translator’s m i scomprehension,
his inadequate skills, or his chosen methodology.
Word-to-word translation techniques, which became the standard
way to render treatises written in a techn ical language, be it phi losophical,
theological, legal or medical, seem inevitably to have produced obscurity.
The medieval Corpus Aristotelicum, which was built up during
the twelfth and thirteenth centuries with this method, so enraged Roger
Bacon that he could suggest only one remedy (which, luckily, he could
not carry out): to burn all the manuscripts.18
This type of criticism escalated with the arrival of Humanism. Let me
i l lustrate it with two incidents: first, a passage from Ambrogio Traversari
(1 386-1439), an ltalian humanist, who, in his preface to his translation
of John C l imacus’s Scala Paradisi (dedicated to Matteo G uidone, c. 1 4 1 9)
speaks i n the following terms about the previous translator of the work:
17
Meanwhile they will not in the least deny that h i s translation is extremely obscure.
What therefore is my cri me if what was translated obscurely I have tried
to render more clear or rather more Latin? Moreover, is it necessary to say
how erudite that translator was? They may cantend that he revealed hirnself to
be very learned in both languages. I, dissenting completely from them, will affirm
truthfully that i n neither was he fully adequate. For it wi ll be easily established
by anyone who has even a mediocre knowledge of the language that he
did not understand correctly most of the Greek. And whoever affirms that he
could have been erudite in Latin signifies with little doubt his own ignorance.
lf they will assert he was a holy man, easily and willingly I will agree. Because
he was a saint. however, it does not follow that he was erudite and capable of
translating. For holiness is one thing. erudition another. lndeed if he was a
saint, he ought not to have attempted wrat he could not execute properly, nor
to have approached this task which exceeded his power. For one causes injury
„absurdissima extitit Passio, ut non solum non intellegeretur, verum etiam ridiculum
legentibus et audientibus eius incompta denotaret obscuritas“; Guarimpoto,
Passio 8/asii, 1 5 8.
18 „Certus igitur sum quod melius esset latinis quod sapientia Aristotelis non esset
translata, quam tali obscuritate et perversitate tradita . . . et sie omnes qui aliquid
sciunt negl igunt perversam translationem Aristotelis, et querunt remedia sicut
possunt . . . si enim haberem potestatem super libros Aristotelis ego facerem omnes
cremari, quia non est nisi temporis amissie studere in i l l is, et causa erroris, et
multiplicatio ignorantiae ultra id quod valeat explicari“; Roger Bacon, „Compendium
studii phi losophiae,“ in Fratris Rogeri Baconi opera quaedam hactenus
inedita, ed. J.S. Brewer (London: Longmans, 1859), 469.
MEDIEVALAND HUMANIST TRANSLATION THEORIES 1 6 5
t o a learned man by rendering his utterance i n a n ignorant and rustic way.“19
The medieval translator referred to was Angelo Clareno {1 247-1 337), a
F ranciscan friar from Cingol i. Du ring the two long periods he had spent
in Greece-in the Corinthian bay ( 1 295-1297) and in eastern Thessaly
{ 1 2 98/9-1 304/5)-he translated a substantial amount of Greek spiritual
l iterature, including the Scala Paradisi of John C l imachus, a number of
writings of Basil the Great ( i ncluding the Rule, letters, and prologues to
several of his ascetic pieces), and a Ietter of Saint John Chrysostom to
Ciriacus. Accord ing to his hagiographer, he had acquired the language
through the Holy Spirit, while spending Christmas in a Greek monastery.
Another indignant voice was that of Leonardo Bruni (c. 1 370-1444 ) .
Encountering the earlier version o f Aristotle’s Nicomachean Ethics sent
h i m into fits of rage and contempt. ln 141 7 he attempted to replace this
earlier, medieval version with a fresh one by hi mself.20 ln the preface to
his translation, he called the medieval version of the Aristotelian text
more barbarian than Latin, immature, ignorant, absurd and awkward,
and the translator half-Latin and half-Greek, incompetent in both languages,
an author of a work that is altogether unworthy of Aristotle and
of the Latin language, perverted, full of twisted words, obscure concepts
19 „Praeterea traductionem illam esse obscurissimam ne ipsi quidem negabunt.
Quod ergo crimen meum est, si quod ille obscurius transtulit, apertius ipse, et a l i quante
etiam latinius convertere conatus sum? Porro quam fuerit i l l e lnterpres
eruditus quid adtinet d i cere? Contendant isti peritissimum i l lum in utraque lingua
exstitisse: ego ab i l l i s Ionge dissentiens, i n neutra illum satis plenum fuisse
veraciter adseverabo. Nam graeco pleraque non recte intellexisse cuilibet eius
l i nguae vel mediocriter perito facile constabit: et latine erudite posuisse, qui adfirmat
sese imperitissimum esse haud obscure significat. Sanctissimum i l l u m
fuisse virum s i adseverant; facile, a c perlubenter consentiam: non tarnen, quia
sanctus fuerit. eruditum etiam fuisse sequitur, atque idoneum ad transferendum.
Aliud enim sanctitas est, atque aliud eruditio. lmo vero si sanctus fuit; ne id
quidem tentare debuit, quod commode im::>lere non passet, neque id onus subire,
quod virium suarum excederet modum. Facit enim iniuriam doct1ssimo viro, qui
illum i mperite, ac rustice loquentem reddit;“ Ambrosii Traversarii Generalis
Camaldulensium latinae epistolae, ed. Laurentius Mehus, 2 vols, (Fiorentiae: ex
Typograph10 Caesareo, 1 759; reprint Bologna: Forni. 1 968), vol. 2, col. 962 (book
23, Ietter 7); trans. Charles L. Stinger, Humanism and the Church Fathers. Ambrogio
Traversari {1386-1439] and Christian Antiquity m the lta/ian Renaissance (AIbany:
State University of New York Press, – 977 ) , 1 1 1 .
20 Cf. Hanna-Barbara Gerl, Philosophie und Philologie. Leonardo Brunis Ubertragung
der Nikomachischen Ethik in Ihren Philosophischen Prämisen (Munchen: Wilhelm
Fink, 1981 ) . For related writings of Bruni, cf. Paolo Viti, ed., Leonardo Bruni, Sul/a
perfetta traduzione (Napoli: Liguori, 2004).
1 66 RKA FORRAI
and a shaky doctrine.21 The identity of the medieval translator(s) was
unknown to Bruni at the time but he used a version that had been translated
in the mid-thi rteenth century by Robert Grosseteste, bishop of Lincoln,
and revised by W i l l i am of Moerbeke, the famous medieval translator
of the entire Corpus Aristotelicum.
Moerbeke, Grosseteste and Angelo Clareno had no chance to defend
themselves from these accusations of obscurity. Their cause was taken
up, however, by Alonso of Carthagena (1 384-1456), a converted Jew
from Spain, bishop of Burgos, famous church politician, canon lawyer,
and learned human ist, who wrote a l i ttle treatise against Bruni’s accusations.
22 From Alonso’s defense, it is clear that humanist and medieval
translation theories operated in two entirely d ifferent conceptual
worlds, and thus must be j udged according to their own criteria, rather
than each other’s and our own.
These medieval translators practiced the most widespread method
among the medieval guild of translators, that is to say, the so called verbum
e verbo method. We would now call it Iitera! translation in English,
that is, a word-for-word faithful following of the original. This translation
practice conceives of the sentence as a chain, where only two elements
have semantic value: the chain itself and the l inks, or words, of which it
is composed, which are defined by their meaning and their position in
the chain, and not, for example, by their relation to other links i n the
chain.
Why did medieval translators have such a notorious predi lection for
Iitera! translation? How, if at all, can such a practice be explained? This
2l „0 ferreum hominem! Hoceine est interpretari? . . . Ego igitur infinitis paene
huiusmodi erroribus permotus, cum haec indigna Aristotele, i ndignaque nobis ac
lingua nestra arbitrarer, cum suauitatem herum li brorum, quae Graeco sermone
maxima est, in asperitatem conversam, nemina intorta, res obscuratas,
dectrinam labefactatarn viderem, Iaberem suscepi novae traductienis, in qua, ut
cetera emittam, id assecutum me pute, ut hes libros nunc primum Latines
feceri m, cum antea non essent“; A. Birkenmajer, „Der Streit des Alonso von
Cartagena mit Leenarde Bruni Aretino,“ in Vermischte Untersuchungen zur
Geschichte der mittelalterlichen Philosophie, ed. Clemens Baeumker {Munster:
Verlag der Aschenderffschen Verlagsbucrhandlung, 1922), 1 59.
22 See Gonzalez Rolan, A. Moreno Hernandez, P. Saquiero Suarez-Somonte, Human·
ismo y Teorfa de Ia Traducciön en Espana e ltalia en Ia primera mitad del siglo XV.
Ediciön y Estudio de Ia Controversia Alphonsiana: Alonso de Cartagena vs. L. Bruni
y P. Candida Decembrio, (Madrid: Edicienes clasicas, 2000) and Marfa Morras. „EI
debate entre Leenarde Bru ni y Alonso de Cartagena: las razones d e una polemica,“
Quaderns. Revista de traducciö 7 (2002): 33-5 7.
ME DIEVALAND HUMANIST TRA SLATION THEORIES 1 6 7
question has troubled specialists o f medieval translation theory a n d
practice for a long time,23 a n d although it has not been completely a n swered,
major misconceptions have already been removed: the Iack o f a
good knowledge of Greek or Latin, for example, is no Ionger considered a
sufficient explanation for the phenomenon. The l iteral method was not
chosen because of one’s l i m ited capacities (even if, from certa i n viewpoints,
these capacities, or rather the tools for developing them, were
quite l i mited). Translators were often good rhetoricians when it came to
their own prose and the commentaries they often provided on the texts
they translated show that they grasped the texts‘ meani ngs perfectly
even if they did have problems with Greek and thus rendered it somewhat
awkward ly. Word-for-word translation was not therefore a primitive
form of interpretation, not the result of a handicap, but a choice.
Translators were conscious of the impossibility of creating a perfect
translation.
A translation was not expected to be explicit. or clear, or, horribile
dictu, beautiful. Texts were to be concise and terminologically coherent,
and should not attempt to interpret themselves, so to speak. For this,
there were commentaries. A text had to be deciphered, and if you manage,
promises Alonso, what seemed so repulsive at the beginning, w i l l
actually become beautiful and not a syllable w i l l be in the wrang place.24
ln the effort of reading, one had to distinguish between text and commentary.
25 This too is a very medieval concept, rooted in late antique
educational practice. Hellenistic phi losophical and l iterary exegesis, at
least the way it was practiced i n schools, was based on I itera I exposition
followed by a paraphrase type of commentary. Thus I itera I translation is
the interlingual application of an originally intra-lingual textual transformation,
which in turn was a school-technique of textual exegesis. Jerome
hirnself says that it is the commentator’s role to make plain what the a u thor
expressed obscurely.26
23 Cf. Paolo Chiesa, „Ad verbum or ad sensum? Model l i e coscienza metodologica
della traduzione tra tarda antichitt e alto medioevo.“ Medioeva e Rinascimento 1
(1987): 1-51.
2• „sed cum studiosi ingenio vel glossarum auxilio quod conceperit pandere cogitur,
stc eius dulce fulget eloquium, ut eius maiestatem mtran cogamur et nedum
verbum aliquod, sed nec syllabam deficere arbitremur, quae obmissa videbantur,
ex industria sie conscripta cernentes“; Bi rkenmajer, Der Streit, 1 67 .
2 5 „textuum a c glossarum non debet similis esse locutio“; Birkenmajer, Der Streit,
1 67.
26 „Commentarii quid operts habent? Alteritts dicta edtsserunt, quae obscurae
168 REKA FORRAI
Medieval translations are strongly dependent on this concept of
reading. Texts were expected to be obscure and had to be unlocked; they
did not unlock themselves. And this in turn brings us to the issue of
meaning. Texts did not explain themselves because their explanation d i d
not l i e within, but beyond the shell o f letters, words, a n d language i n
general. Bruni repeatedly defines translation as a rendering from one
language to another.27 Alonso, on the other hand, explicitly says that he
does not know Greek and does r.ot even care about it. F or one has to understand
not what Aristotle wrote down in Greek, but what he thought,
what he must have meant.2B For this, one does not need to use Greek
texts, but simply sound reasoning, as Greek texts might be faulty themselves,
not presenting very clearly what Aristotle should have had in his
mind. Also, chances are that Aristotle might have meant somethi n g more
reasonable than what he actually said.29
Accord ing to this reasoning, if someone finds in a Greek text that 2
plus 2 equals 5, one should translate 2 plus 2 equals 4, as there are obvious
extralingual elements which support the verity of the second version,
and refute the logic of the first. ln philosophy, this ultimate external
reference point is reason. Different idioms follow and express the same
reason; that is why, Alonso argues, there is no need for him to know
Greek in order to critically assess the translation. ln theology, this reason
is God, or his revelation. The external pressure of orthodoxy upon
translators p layed a huge role in shaping translation techniques. Texts
were supposed to be fa ithful not to the I iterary category of what could be
today called the author’s i ntention, but rather to the religious system of
which they were part.
Beyond the verbum and sensum. there was a category called veritas
that is perhaps much closer than sensum to what we would term „meanscripta
sunt, plano sermone mantfestant“; Hieronymus, Apologia Adversus Libros
Rufini, I, 1 6 in P Lardet, ed .. Apologie Contre Rufin, SC 303 (Paris du Cerf. 1983).
44. 27 „I nterpretatio autem recta, si graeco respondet, vitiosa, se non respondet. ltaque
omnis interpretatio contentio unius l i nguae ad alteram est“: Birkenmajer, Der
Streit, 189. 28 „Non ergo an in Graeco sie scripturn est, sed an stc scribi potuit, ut translator
noster edixit illis in locis, ubi dire reprehensus est, inquiramus“: Birkenmajer, Der
Streit, 166.
29 „Cum igitur Aristoteles ipse non rationem ab auctoritate, sed auctoritatem a
ratione consecutus est, quicquid rationi consonant, haec Aristoteles dixisse
putandus est et Graece arbitremur scripturn fuisse. quicquid Latinis verbis translatio
nostra sapienter depromit“: Birkenmajer, Der Streit, 1 66.
MEDIEVALAND HUMANIST TRANSLATION THEORIE$ 1 69
ing“ today. Nevertheless, the task of the translator was not to grasp and
to express this truth, but only to present a version of the text that would
al low the reader to reach its veritas by himself. A translator was supposed
only to make this veritas accessible, rather than express it, since
translation was not supposed to interpret in the sense of deciding on a
meaning. On the Ievei of terminology, perhaps this can be caught in the
distinction between interpretare and intellegere, the first being the task
of the translator, the secend the task of the aud ience, that is to say, the
reader or commentator. lt is a long these l ines that Boethius distinguishes
between his translation and his commentary on the Isagoge of Porphyrus:
he was after uncorrupted truth, not beauty of the style, when he
translated the work. While he claims he is gui lty of translating faithfully,
he seems to think this is unavoidable, and should be remedied later via
commentaries.JO
That is one of the reasons why the contamination of Latin with Greek
and other foreign expressions does not seem as scandalous to medieval
scholars as it does to Bruni, who was obviously brought up on Quintilian’s
notion of various lexical obscurities to be avoided. Every thought,
every concept was thought to have a perfect expression, or rather, a
most concise and more precise expression, which needed to be found
regardless of the language. A concise foreign word was considered superior
to a loose circumscription in Latin, said Alonso.J1
Accordi n g to Alonso words are like hostages taken in wartime from
the enemyn And in the war of scientific d iscussion, one needed to be
rigorous and accurate, and not to complicate what is simple. He argued
that one needs to examine the semantic field of the Latin words, rather
than looking for superficial equivalence with Greek, as the Latin term
should refer back to the essence of the philosophical discourse, rather
JO „Secundus hic arreptae expositionis Iabor nostrae seriem translationis expediet.
in qua quidem uereor ne subierim fidi interpretis culpam, cum uerbum uerbo
espressum comparatumque reddiderim. Cuius incepti ratio est quod in his
scriptis in quibus rerum cognitio quaeritur, non luculentae orationis lepos sed
incorrupta ueritas exprimenda est“; Anicii Manlii Severini Boethii in Isagogen Porphyrii
commenta, ed. S. Brandt, CSEL 48 (Vienna: Tempsky and Leipzig: Freytag,
1906). XVI I I .
3l „Nonne melius fuit i l l a, ut iacebant, dimittere, ut sub nostris refulis declinata inter
Latina haberemus, significatione earum per descriptiones et sequentia plene
percepta – quam circumlocutionibus totam scripturae seriem perturbare?“;
Birkenmajer, Der Streit, 1 69.
32 „quasi ab hostibus capta a l ienas voces et nomina“; Birkenmajer, Der Streit, 1 68.
1 7 0 RtKA FORRAI
than to the way it was expressed i n Greek.33
There was, therefore, a crucial difference in the attitude of medieval
and human ist translators towards obscur i ty. For the former, it was a
philosophical, theological concept, an admirable qua l i ty of dense and
concise texts, which could also act as a filter and defend the text from i n ept
readers. Unlocking obscure passages was t h e r o l e of t h e commentator
rather than the translator. As the Neapolitan hagiographic
translations testify, however, not all obscurity was tolerated: narrative
texts, especially those used in l i turgy, were to be polished in order to facil
itate their immediate grasp by the audience.
Human ists, on the other hand, operated with the rhetorical concept
of obscuritas. Criticism based on such a concept would, however, have
been meaningless to medieval translators of phi losophical works: they
would never have dreamed of trying to find and restore elegantia to the
Aristotelian corpus34-neither would we, for that matter. For humanists,
obscurity was a rhetorical vice to be avoided, in cantrast with clarity and
elegance. Thei rs was a pur ist approach that resented the usage of Greek
neologisms or of any technical vocabulary in fact. During the late Renaissance,
this conflict over translation methodologies became part of a larger
debate between scholastics and humanists, philosophy and
rhetoric.35 As a result, obscurity lost the positive connotations of its
33 „Quisquis tamen i l l e fuerit, obscuritate arguendus non est, cum in omnibus fere
scientiis textuum conditores brevitati studuerunt. Nam sicut alia principem, alia
oratorem decet oratio et al iter iudicem, aliter advocatum congruit loqui, sie textuum
ac glossarum non debet similis esse locutio: nam breviter textus nos docet,
glossule vero qu id textus senserit aperire solent; quod tam in l iberalibus artibus,
quam in naturalibus scientiis ac iurium doctri nis saepe repertum est, ut, his
saepe solis verbis plerumque contenti sint, quibus conceptus sensus includi vix
valuit, adeo quod plerique rudimenta artium amore brevitatis adinuenta duxuerunt.
Non ergo translationis incusandus est, qui recte i ntellectus breuibus
uniuersa conclusit. Procul dubio enim in pr imis armis quodammodo translatio
haec defendere se uidetur et uiolentiam legentis uiri liter propulsare; sed cum
studiosi ingenio uel glossarum auxilio quod conceperit pandere cogitur, sie eius
dulce fulget eloquium, ut eius maiestatem mirari cogamur et nedum uerbum aliquod,
sed nec syllabam deficere arbitremur, quae obmissa uidebantur, ex industria
sie conscri pta cernentes“; Birkenmajer, Der Streit. 1 67.
34 The Humanists adhered to Cicero’s statement about Aristotle’s „pouri ng forth a
golden stream of eloquence“ („flumen orationis aureum fundens Aristoteles“;
Cicero, Academicorum Priorum Liber I I , 38, in Cicero, Oe natura deorum. Academica,
trans. H. Rackham [Cambridge, Mass.: Harvard University Press, 1 933]. 620-
2 1 ) .
35 Cf. Erika Rummel, Humanist-Scholastic Oebare in the Renaissance and Reformation
MEDlEYALAND HUMANIST TRANSLATION THEORIE$ 1 7 1
semantic field. Medieval translation theory and practice, however, remind
us that obscurity is inherent in human d iscourse: i n herent in language,
i nherent in philosophy, inherent in theology. i n herent in translation.
lt is a manifold and powerful presence that requires a manifold
methodology that is genre and audience dependent.
(Cambridge, MA: Harvard Un iversity Press, 1995), 1 1 : „The battle l i nes were
drawn and the stereotypes establ ished: all scholastic theologians were obscurantists
who had never read classical authors, wrote atrocious Latin, and were in·
terested only i n esoteric quibbles, while all humanists were grammarians and
wordspinners, i nterested in form rather than substance, pseudo-Christians
whose brains had been addled by reading pagan literature.“
Obscurity in Medieval Texts
MEDIUM AEVUM QUOTIDIANUM
SONDERBAND XXX
Obscurity in Medieval Texts
edited by
Lucie Dolezalova, Jeff Rider,
and Alessandro Zironi
Krems 2013
Reviewed by
Tamas Visi
and Myriam White-Le Goff
Cover designed by Petr Dolezal with the use of a photo of the interior of
the Church of the Holy Sepulchre in Jerusalem (photo Lucie Dolezalova)
GEDRUCKT MIT UNTERSTÜTZUNG
DER
CHARLES UNIVERSITY RESEARCH DEVELOPMENT PROGRAMS
„UNIVERSITY CENTRE FOR THE STUDY OF ÄNCIENT AND MEDIEVAL
INTELLECTUAL TRADITIONS“
UND
„PHENOMENOLOGY AND SEMIOTICS“ (PRVOUK 1 8)
80TH AT THE FACULTY O F HUMANITIES, CHARLES UNIVERSITY IN PRAGUE
UNDDER
CZECH SCIENCE FOUNDATION
WITHIN THE RESEARCH PROJECT
„INTERPRETING AND APPROPRIATING ÜBSCURITY
IN MEDIEVAL MANUSCRIPT CULTURE“
(GACR P405/1 0/Pl 1 2)
A l l e Rechte vorbehalten
-ISBN 978-3-901094-32-13′.3
Herausgeber: Medium Aevum Quotidianum. Gesellschaft zur Erforschung der materiellen
Kultur des Mittelalters. Körnermarkt 1 3. 3500 Krems, Österreich. Fur den
Inhalt verantwortlich zeichnet die Autorin, ohne deren ausdruckliehe Zustimmung
jeglicher Nachdruck, auch in Auszügen, nicht gestattet ist. Druck: KOPITU Ges. m. b.
.• iedner Hauptstraße 8-10, 1 050 Wien, Österreich.
,s \i !.Ut ‚o ,… ….
Acknowledgements
List of Figures
T able of Contents
Textual Obscurity in the Middle Ages (lntroduction)
Lucie Dole2alov. Jeff Rider. and Alessandro Zironi
„Ciarifications“ of Obscurity:
Conditions for Proclus’s Allegorical Reading of Plato’s Parmenides 1 5
Florin George Cäl ian
Lucifica nigris tune nuntio regna figuris. Po!!tique textuelle de I‘ obscuritas
dans I es recueils d‘!!nigmes latines du Haut moyen Age (V He-V I I I • s.) 3 2
Christiane Veyrard-Cosme
The Enigmatic Style in Twelfth-Century French Literature 49
Jeff Rider
Mise en abyme in Marie de France’s „Laüstic“ 63
Susan Small
Perturbations of the Soul: Alexander of Ashby and Aegidius of Paris an
Understanding Biblical Obscuritas 75
Greti Dinkova-Bruun
Versus obscuri nella poesia didascalica grammatocale del X I I I sec. 87
Carla Piccone
Disclosing Secrets: Vorgil on Middle High German Poems 1 1 0
Alessandro Zironi
Obscuritas tegum: Traditional Law. Learned Jurisprudence, and Territorial
Legislation (The Example of Sachsenspiegel and fus Municipale Maideburgense) 124
Hiram Kümper
Ta Be Born (Aga in) from God:
Scriptural Obscurity as a Theological Way Out for Cornelius Agrippa 1 45
Noel Putnik
Obscuritas in Medoeval and Humanist Translation Theories 157
R!!ka Forrai
The Darkness Within:
First-person Speakers and the Unrepresentable 1 72
Päivi M. Mehtonen
Contributors 1 90
Index nominum 1 94
Index rerum 197
Acknowledgements
This volume grew out of a conference held in Prague in October 6-8. 201 1 .
The conference and the book were supported by a post-doctoral research
grant from the Grant Agency of the Czech Republic, “ l nterpreting and
Appropriating Obscurity i n Medieval Manuscript Culture“ no. P405/1 0/
P1 1 2 undertaken at the Faculty of Arts at the Charles University in Prague,
by The Ministry of Education, Youth and Sports through l nstitutional
Support for Longterm Development of Research Organizations to the
Faculty of Humanities of the same university (PRVOUK 1 8 and UNCE
204002), and by the European Research Council under the European
Community’s Seventh Framewerk Programme ( FPJ/2007-2013) I ERC
grant agreement No. 263672. We are much grateful to these i nstitutions.
Further thanks goes to the individual contributors to this volume who have
been very quick and patient during the process, as weil as to Petr Dolezal
for the cover design and Adela Novakova for the index.
List of Figures
Figure 1 : Scene from one of the Saxon Mirror’s codices picturati (Wolfenbuttel, HerzogAugust-
Bibliothek, Cod. Guelf. 3.1 . Aug. 2°, fol. 34r).
Figure 2: Index for a manuscript of the Richtsteig Landrecht (Göttweig, Sti ftsbibliothek,
Cod. 364rot, fol. 526r).
Figure 3: Printed text of a Saxon Mirror with Gloss (Christi an Zobel, Leipzig, 1 569).
Figure 4: A remissorium from a Saxon Mirror edited tn 1536 by Chistoph Zobel (Leipzig).
Figure 5: Editorial report for a Saxon Mirror pri nted in 1545 by Nikolaus Wolrab
(Leipzig).
Figure 6: Sebastian Stelbagius, Epitome sive summa universae doctrinae iusticiae legalis
(Bautzen, 1 564 ).
Figures 7 and 8: Melchior Kling, Das Gantze Sechsisch Landrecht mit Text und Gloß in eine
richtige Ordnung gebracht (Leipzig 1 572).