Did Medieval Monks Actually Speak Latin?
1
JULIE BARRAU
O
ne of the most common distinctions made about medieval people wasthe classical distinction between
litterati
and
illiterati
, which found aquasi-equivalent in the more specifically medieval opposition of the
clerici
and the
laici
. It would be oversimplifying, but still roughly accurate,to
2
say that during the Early and High Middle Ages these terms separated thosewho knew how to read fromthose whodid not – this mostly meant how to readLatin. If there was going to be a place where it should have been straightfor-ward, and even required, to be
litteratus
between the tenth and the thirteenthcentury, it would surely be within the cloisters. Latin was sung, chanted andrecited during the many liturgical hours, and also read in the refectory, taughtin the monastic school, and copied and composed in the
scriptorium
. It was
3
I would like to thank Steven Vanderputten for his invitation to the conference in Ghent;
1
I am also very grateful to other speakers who were generous with their time to discuss my talk,amongthemGerdAlthoff,SusanBoynton,ElisabethvanHouts,DianeReillyand WimVerbaal;Susan Boynton and Elisabeth vanHoutsreadlaterdraftsand madesomemost needed commentsand criticisms. Finally, Joel Cabrita and Tim Gowers kindly edited drafts of this article – manythanks to them.See the seminal article of H. G
RUNDMANN
, “
Litteratus-Illiteratus
: Der Wandel einer
2
Bildungsnormvon Altertum zum Mittelalter”,
Archiv für Kulturgeschichte
40 (1958), pp. 1-65;on theindependentoriginsoftheseconceptualpairs, seeM. C
LANCHY
,
From MemorytoWritten Record, 1066-1307
(Cambridge, MA, 1979), pp. 177-191; for acriticaltakeon Grundmann, seeD.H. G
REEN
, “Orality and reading: The state of research in medieval studies”,
Speculum
65.2(1990), pp. 267-280, at pp. 274-276.For a general viewof tenth- and eleventh-centurymonastic houses as places of learning,
3
see P. R
ICHÉ
,
Ecoles et enseignement dans le Haut Moyen Age
, 2nd edn. (Paris, 1989), pp. 137-161.
Monastic Practices of Oral Communication (Western Europe, Eleventh-Thirteenth Centuries), ed. S. Vanderputten, Brepols, 2011, pp. 293-317
296
JULIE BARRAU
Latin, the title ofhis chapter was
Ne Teutonice loquamur
. There appears to be
14
little doubt which vernacular language the canons might have dared to speak amongst themselves. Nevertheless, monks were in a position to speak mostlyLatin if they wish-ed (and could). Linguistic practice must have varied both according to contextand over time, but it seems a reasonable starting point to suggest that, at leastwhen they were talking about important or elevated matters, they spoke Latinrather than any other language. That is certainly the view of certain scholars
15
who have studied the problem, amongthemThomas Haye in his recent synthe-sis on medieval spoken Latin. So should we assume that the numerous ex-
16
changes and conversations between monks that one finds inchronicles or hagi-ographies probably took place in some sort of Latin ratherthan in their mother tongue?The present studywill aimto present a somewhat different picture and willtentatively suggest that in many situations it was primarily the vernacular thatwas used. The actual linguistic practices of monks are very difficult to ascer-
17
tain, as the evidence is sketchy. I shall try to show here that learningLatin wasa long and difficult process, which suggests that not everybody would havetaken the trouble to master it. It will then be possible to put in context the fig-ure of the
monachus laicus
, the monk without Latin; this will lead to a pictureof heterogeneous communities where the vernacular may have been the
lingua franca
, even in formal occasions like chapter meetings. That will in turn
18
De Ordine canonicorum regularium
(ed.
PL
188, col. 1106).
14
Janet Burton, for instance, considers that a certain level of Latin was likely to have been
15
mastered by British monks (
Monastic and Religious Orders in Britain 1000-1300
(Cambridge,1994), p. 188).T.H
AYE
,
LateinischeOralität:GelehrteSpracheindermündlichenKommunikationdes
16
hohen und späten Mittelalters
(Berlin and NewYork, 2005). In his chapter on Latin in monasticeverydaylife(“
DasLateindesklösterlichenAlltags:AelfricBataundPaulSchneevogel
”,pp.68-77), he favours a picture of cloisters where Latin was a standard, if not universal, language:“Despite these limitations Latin probably played a central part in internal oral communication.In order to ensure that members of the order really did use Latin in their everyday activities, thisrequirement is repeatedlyimpressed upon them; in addition, the monasteryschool, in particular, providedintensivetraininginuseofthelanguage.Alongsidethestudyofwrittentexts,colloquialLatin, such as was necessary for standard situations of monastic life, was also practised”. (
Ibid.
, p. 71). The present study will partly qualify Professor Haye’s conclusions.Anyconclusion reached here could not and would not be presented as final; this studyis
17
very much part of a work in progress.In his extensive study of
conversi
, Wolfgang Teske estimates that they made up some
18
forty per cent of the brethren at Cluny during the abbacy of Odilo (W. T
ESKE
, “Laien,Laienmönche und Laienbrüder in der Abtei Cluny: Ein Beitrag zum ‘Konversen-Problem’ ”,
297
Did Medieval Monks Actually Speak Latin?
suggest that in the numerous texts where monks are described talking to eachother the language they used was more likely than not to be the vernacular;there were of course Latin speakers within monastic houses, but I will proposethat they were only an elite portion of each community, of a size that variedfrom one house to another. It is therefore important to keep in mind, when onetries to understand monastic culture by studying manuscripts from monasticlibraries, that those
codices
were used by only a fraction of the brethren.To start with, one can safely assume that for most of those who embracedmonastic life Latin was a foreign language when they entered the cloister. Thequestion of when Latin became solely a second language has been fiercelydebated by sociolinguists and historians for decades and is still not settled.
19
There is nevertheless a
terminus ad quem
on which everybody seems to agree: by the mid-tenth century, even in Italy there was a clear awareness that Latinand the vernacular had become two different languages. Therefore, during the
20
period that this paper is concerned with,Latinwas indeed a foreign language.
21
FrühmittelalterlicheStudien
10(1976),pp.248-322and11(1977),pp.288-339,firstpart(1976),at p. 307). Whether that group was Latin-literate or not is therefore of some importance for understanding the linguistic dynamic of the community.For the avatarsofaquestion asked almost eightyyearsagobyFerdinandLot,seeF.L
OT
,
19
“Aquelleépoquea-t-oncessédeparlerlatin?”,
ArchivumLatinitatismediiaevi
6(1931),pp.97-159;D.N
ORBERG
,“Aquelleépoquea-t-oncessédeparlerlatinenGaule?”,
Annales:Économies,Sociétés, Civilisations
21 (1966), pp. 346-356; M. R
ICHTER
, “A quelle époque a-t-on cessé de parlerlatin en Gaule?Aproposd’une question malposée”,
Annales
ESC
38 (1983), pp.439-448.For recent and opposing views, see R. W
RIGHT
,
Late Latin and Early Romance in Spain and Carolingian France
(Liverpool, 1982), and M. B
ANNIARD
,
Viva voce: Communication écrite et communicationoraledu
IV
au
IX
siècleenOccidentlatin
(Paris,1992).Also
ID
.,“Languageand
e e
communication in Carolingian Europe”, in:
The New Cambridge Medieval History
, 2,
c. 700-c.900
, ed. R. M
C
K
ITTERICK
(Cambridge, 1995), pp. 695-708. See also the review of
Viva voce
byRogerW
RIGHT
in
Journal of MedievalLatin
3 (1993), pp.77-92.For amasterful presentation of Latin as a language spoken by both lay and clerical elites in the Carolingian world, see R.M
C
K
ITTERICK
,
The Carolingians and the Written World
(Cambridge, 1989), pp. 7-22. See alsoR. M
C
K
ITTERICK
, “Latin and Romance: An historian’s perspective”, in:
Latin and the Romance Languages in the Early Middle Ages
, ed. R. W
RIGHT
(London and New York, 1991), pp. 130-145.B
ANNIARD
,
Vivavoce
,pp.544-551,andW
RIGHT
,
LateLatinandEarlyRomance
,p.144.
20
Bernard Bischoff, in his important article on the teaching of foreign languages in the
21
Middle Ages,beginsbydismissingLatin asirrelevant to his topic. Hisreasonsarethe following:“LatinintheMiddleAgesdidnotbelongtoeitherofthetwonormalcategoriesofdeadlanguagesor living languages. Being the language of the Western Church which everylittle child admittedto an ecclesiastical school had to learn, it became for many centuries the general vehicle of spiritual culture and of practical record” (B. B
ISCHOFF
, “The study of foreign languages in theMiddle Ages”,
Speculum
36.2 (1961), pp. 209-224, atpp.209-210). I begto disagree: while one
298
JULIE BARRAU
It would be interesting to know what difference it made when a novice wasexposed to Latin if his mother tongue was a Romance language or a Germanicone. There is little available evidence to go beyond the obvious idea that Latinmust have looked and sounded much more familiar and less forbidding to ayoung Italian than to a Dane or an Anglo-Saxon. It is worth recalling here thefairly well-known story of Gunzo, an Italian who visited the abbey of Saint-Gall in 965 and was mocked by the German Benedictines for having made amistake in Latin. In this remarkable case, Gunzo’s blunder (he seems to haveused an accusative ending where an ablative was required), and the monks’reaction (at first a general laugh, later a short satirical poem apparently written by a young monk) were so painful for the Italian grammarian that heset out anelaborate defence of his Latin credentials. The Saint-Gall incident appears to
22
be what sparked the writing of his
Epistola ad Augienses
, his only extant work and a famous example of earlymedieval mastery of classical references. Gunzorefers to many distinguished authors to show that the use of cases can be moreflexible than the pedestrian classicist (here the “silly monk” and his master,whom he castigates throughout his text) might think. In an interesting exampleof linguistic self-awareness he also defends his lapse by drawing attention tothe similarity between Latin and his mother tongue. Gunzo’s mishap may
23
indicate that if Latin was easier to understand for speakers of Romance lan-guages, then it was paradoxically harder to speak properly because the close-ness of Latin to their vernacular was a constant source of confusion. In North-ern and Central Italy, where Gunzo is thought to have come from, specialistsnow think that vertical communication in Latin had been possible almost untilGunzo’s time (c. 900-950). Nevertheless, for the
litterati
born in Romance-speaking regions, the situation of
diglossia
may have made
lapsus linguae
quite easy, much more so than
lapsus calami
.
24
can readily agree that Latin cannot accurately be described as a dead or living language, it doesnot followthat it wasnot aforeign one.Whateverthe status of Latin in medieval societies, it hadtobelearnedasasecondlanguage,evenbyaseven-year-oldoblate–itwasnotanybody’smother tongue.Gunzo,
Epistola ad Augienses
, ed. K. M
ANITIUS
(Weimar, 1958:
MGH
Quellen zur
22
Geistesgeschichte des Mittelalters
2), pp. 22-23.
Ibid.
,p.27:“
Falso putavit S. Galli monachusmeremotumascientiagrammaticaeartis,
23
licet aliquando retarder usu nostrae vulgaris linguae, quae Latinitati vicina est
”.On diglossia and the coexistence of High (H) and Low (L) languages, H being
24
distinguished by its function, acquisition, prestige, literary heritage, and formalisation, see theclassicalarticlebyCharlesF
ERGUSON
, “Diglossia”,
Word
15 (1959), pp. 324-340. Seealso R.A.L
ODGE
,
French: from Dialect to Standard
(London and New York, 1993), pp. 87-95.
299
Did Medieval Monks Actually Speak Latin?
The teaching of Latin in the cloisters is still a shadowy subject, eventhough it has been much studied. A first difficulty is that the words used in
25
the sources to name the Latin language and its teaching often obfuscate theobject they describe.
Grammatica
and
litterae
are terms passed down by lateAntiquity authors, for whom they meant the scholarly knowledge of their mother tongue. The same words were used, usually without any added qualifi-cation, bymedieval writers who meant by them somethingaltogether different:the learning of a foreign language from scratch. But because they named this
26
grammatica
, it is easy to overlook that it meant the basic rudiments of thelanguage as well as more sophisticated stylistic studies. The dominance of thelatter in the manuscript tradition tends to overshadow the former, which thisarticle is mostly concerned with.
27
Another problem is whether there were always schools in monastic houses,and what the schools taught. Some scholars have assumed that there was aschool in, ornear, everymonastic house. This might have been true in a weak
28
A survey of monastic Latin teaching can be found in R
ICHÉ
,
Ecoles et enseignement
,
25
pp.227-235;forratherpositiveassessmentsoftheteachingofLatininpre-ConquestEnglandanditsachievements,seeG.H.B
ROWN
,“ThedynamicsofliteracyinAnglo-SaxonEngland”,
BulletinoftheJohnRylandsUniversityLibraryofManchester
77.1(1995),pp.109-142,andM.L
APIDGE
,“Schools, learning and literature in tenth-centuryEngland”, in:
ID
.,
Anglo-Latin Literature 900-1066
(London, 1996), pp. 1-48. For less glowing conclusions, see D. B
ULLOUGH
, “Theeducational tradition in England from Alfred to Aelfric: Teaching
utriusque linguae
”, in:
Carolingian Renewal: Sources and Heritage
(Manchester, 1992), pp. 297-334. Much can also be found in A. G
ROTANS
,
Reading in MedievalSt. Gall
(Cambridge,2006), especiallypp. 71-79andpp.155-197 (ananalysisofthe
St. Gall Tractate
,a“‘howto’bookforteachingintermediateLatin reading”, probably written by Notker Labeo). For some lights about a later period, see J.M
URPHY
, “The teaching of Latin as a second language in the 12th century”,
Historiographia Linguistica
7 (1980), pp. 159-175, whose scope is broader than the monastic world.This is not to say that the same people were not aware of the difficulties inherent in
26
learning a foreign language. Peter Damian for instance explained that two Camaldolesi, sent off to evangelize king Boleslas’s Poland, needed seven painstaking years to master the local idiom:“
utpraedicarepostmodumpossent,Sclavonicamlinguamlaboriosediscerestuduerunt.Septimoveroanno, cum jam loquelamterrae plene cognoscerent, unum ad Romanam urbem monachummittunt,etpereumasummaesedisantistitepraedicandilicentiampetunt
”(
VitaBeatiRomualdi
,ed G. T
ABACCO
(Rome, 1957), p. 62).On
grammatica
and its teaching, see V. L
AW
,
Grammar and Grammarians in the Early
27
Middle Ages
(London, 1997). For a study that clearly distinguishes basic Latin teaching fromadvancedlinguisticandstylisticstudies,seeL.H
OLTZ
,“Lesnouvellestendancesdelapédagogiegrammaticaleau
X
siècle”,
MittellateinischesJahrbuch
24-25(1989-1990),pp.163-173.Seealso
e
M. I
RVINE
,
The Making of Textual Culture: Grammatica and Literary Theory, 350-1100
(Cam- bridge, 1994), pp. 334-364, 395-404.E. L
ESNE
,
Les écoles de la fin du
VIII
siècle à la fin du
XII
(Lille, 1940), p. 420. But, as
28
e e
300
JULIE BARRAU
sense – in each community there would have been monks literate enough toteach rudiments of Latin and singing. But teaching and learning Latin with thedidactic tools available in Western Europe between the tenth and the earlythirteenth centuries was likely to have been a time-consuming and gruelling process. For the most part, Latin classes seem to have been more about learn-
29
ing to read and compose a written text than about learning to speak. It is truethat during the Office a novice would become familiar with the rhythms andsounds of Latin before being informed of the meaning and syntactic articula-tion of the words. That may have helped when it came tolearningthe languagein a more structured and didactic way, but it would also have been perfectly possible to chant and even read a text with very little understanding of it. A
30
major difficultywas that for most of the period considered here the main gram-mar books used in the classroom were not written for students whose mother tongue was not Latin: Donatus’s
Ars minor
and its many abbreviations, sum-
31
notedbyPierreRiché,ittooktheconjunctionofmanyfactorstomakeamonasticschoolthriving.See P. R
ICHÉ
, “Les écoles de St. Gall des origines au milieu du
XI
siècle”, in:
Le rayonnement
e
spirituel et culturel de l’abbaye de St. Gall
, ed. C. H
EITZ
, W. V
OGLER
and F. H
EBER
-S
UFFRIN
(Paris, 2000), pp. 37-57, at p. 37.Adiscussionofmonasticschoolsisboundtofocusmainlyontheformation ofthe
oblati
,
29
the children who received all their education within the monastic enclosure; see M. L
AHAYE
-G
EUSEN
,
Das Opfer der Kinder: Ein Beitrag zur Liturgie-und Sozialgeschichte des Mönchtumsim Hohen Mittelalter
(Altenberge, 1991), especially pp. 241-257; on the Carolingian monasticschools and their “formation of an elite”, see M. D
E
J
ONG
,
In Samuel’s Image: Child Oblationin the Early Medieval West
(Leiden, New York and Cologne, 1996), pp. 232-245.
Conversi
of the pre-Cistercian type were characterized by their illiteracy; when oblation subsided, and thenwas forbidden, it is not clear how the recruits of the new type, young men with or without previous education (such as the Cistercian novices), would have acquired any more Latin thantheyhad when theyconverted to their new life.SeeG. C
ONSTABLE
, “‘Famuli’ and ‘Conversi’ atCluny: Note on Statute 24 of Peter the Venerable”,
Revue Bénédictine
83 (1973), pp. 326-350.This has been noted by many scholars. For instance Pierre Riché writes about earlier
30
times:“Itwasquitepossible to knowhowto read, to knowthe Psalterbyheart, and to knowhowto recite the office without really understanding Latin” (P. R
ICHÉ
,
Education and Culture in the Barbarian West, Sixth Through Eighth Centuries
, trans. J.J. C
ONTRENI
(Columbia, SC, 1976), p. 468); he reached similar conclusions in
ID
.,
Education et enseignement
, p. 235). Thesereferences are to an earlier period, but there is every reason to think that similar episodes couldhave happened in the tenth to twelfth centuries.H. G
NEUSS
, “The study of language in Anglo-Saxon England”,
Bulletin of the John
31
Rylands University Library of Manchester
72 (1990), pp. 3-32. Rare attempts to teach Latin asaforeignlanguageoriginated,unsurprisingly,inregionswithGermanicvernaculars.The‘Insular rudimentarygrammars’ were written in Latin, with the famous exception of Aelfric’s
Grammar
,an Old English adaptation of a Latin collection of
Excerptiones de Prisciano
, based on Priscian but also Donatus and Isidore. On Aelfricand thesuccessofhisGrammar,seeV. L
AW
, “Aelfric’s
Excerptionesde arte grammatica anglice
”,in:
EAD
.,
Grammar and Grammarians
, pp. 200-223.
301
Did Medieval Monks Actually Speak Latin?
maries and commentaries, and Priscian’s
Institutio de nomine pronomine et verbo
. As a consequence, for all their
auctoritas
theywere awkward tools for
32
their medieval users. Donatus, the most widespread of all available Latinschoolbooks, taught the eight parts of speech, and was universally revered, buthad been written for native speakers and was not in the least concerned withthe acquisition of vocabulary.Glosses and glossaries were probably the best place to turn to for a monk who wished to build up a command of Latin vocabulary:
no dictionaries in our sense were available before the twelfth century, and ... theinterlinear [glosses] had to play an important role in a programme to teach Latinvocabulary.
33
The situation partlychanged with the production of dictionaries and
derivatio-nes
from the late eleventh century but, as Richard Sharpe has pointed out, “theworks of Papias, Osbern, Hugutio and John of Genoa were not intended for readers in the first stages of learning Latin”. Glosses came in many forms,
34
marginal and interlinear, continuous and randomly scattered, copied at thesame time as their
lemmata
and added later on, Latin and vernacular, lexicaland syntactical. What lexical knowledge could be drawn fromthemtherefore
35
L. H
OLTZ
,
Donat et la tradition de l’enseignement grammatical. Etude sur l’”Ars
32
Donati” et sa diffusion (
IV
-
IX
siècle) et édition critique
(Paris, 1981). For a presentation of
e e
Donatus and the organising principles of his two
Artes
, see V. L
AW
,
History of Linguistics in Europe from Plato to 1600
(Cambridge, 2003), pp. 65-80. Donatus’s
Ars maior
and Priscian’s
Institutiones Grammaticae
were used at a more advanced level.H. G
NEUSS
, “
Anglicae linguae interpretatio
: Language contact, lexical borrowing and
33
glossinginAnglo-SaxonEngland”,
ProceedingsoftheBritishAcademy
82 (1993),pp.107-148,at p. 146. For a recent and stimulating analysis of medieval glossing and its meanings, see R.S
TANTON
, “Interpretation, pedagogy, and Anglo-Saxon glosses”, in:
ID
.,
The Culture of TranslationinAnglo-Saxon England
(Cambridge,2002),pp.9-54.ForarecentsynthesisonOldHighGermanglossesandtheirdidacticpurpose,seeJ.W
EST
,“IntoGerman:Thelanguageoftheearliest German literature”, in:
German Literature of the Early Middle Ages
, ed. B. M
URDOCH
(Rochester,NY,2004),pp.35-56,atpp.37-44.Forageneralapproachofearlymedievalglosses,seeM.I
RVINE
,
TheMakingofTextualCulture
,pp.371-393,andabouttheAnglo-Saxoncontext, pp. 426-449.R. S
HARPE
, “Vocabulary, word formation, lexicography”, in:
Medieval Latin: An
34
IntroductionandBibliographicalGuide
,ed.F.M
ANTELLO
andA.R
IGG
(Washington,DC,1996), pp. 93-105, atp. 97. On thoselexicographers see L.W. D
ALY
and B.A. D
ALY
, “Sometechniquesin mediaeval Latin lexicography”,
Speculum
39.2 (1964), pp. 229-239.For a typology of glosses, see P. L
ENDINARA
, “Anglo-Saxon glosses and glossaries: An
35
introduction”, repr. in:
EAD
.,
Anglo-Saxon Glosses and Glossaries
(Ashgate, 1999), pp. 1-26.
302
JULIE BARRAU
varied enormously; the occasional
interpretamentum
of a difficult word, oftena Latin synonym, could smooth the reading of the already advanced Latinist;the beginner would have required a more continuous support. When rearranged by themes or alphabetically, glosses became glossaries, which in turn becameimportant sources for lexicographers such as Papias. An intermediate stage can be found in the
glossae collectae
, or batch glosses, which were interlinear
interpretamenta
taken from their initial layout and listed, usually in the order they appeared in the initial manuscript, with their
lemmata
. There has been a
36
heated debate over the last three decades about whether glosses reflect actualclassroom practices. Here is not the place to continue the discussion, but,
37
whether glosses were used by masters and students in the classroom, or whether they were meant for private reading and study, they contributed to theacquisition of Latin. However, to gather from them a reasonably wide vocabu-lary would have been a long process. It is therefore doubtful that all youngmonks could have eventuallybecome sufficientlyfluent to read the Bible or theFathers or to sustain a conversation of any complexity. If there were no system-atic learningtools until the first dictionaries and lexicons, one can onlyassumethat reaching a moderately fluent level required a great deal of hard work on
Latinglosses,whosepedagogicalpurposeisoftenquiteclear,haveunfortunatelybeenmuchlessstudied than their vernacular cousins. See the
deploratio
of Gernot W
IELAND
, “Latin lemma – Latin gloss: The stepchild of glossologists”,
Mittellateinisches Jahrbuch
19 (1984), pp. 91-99.Since then some progress has been made towards the editing of Latin-Latin glossaries, mostrecentlythetwovolumesof
GlossaeBiblicae
,ed.P. V
ACIAGO
(Turnhout,2004:
CCCM
189A-B);these glosses, which include in the second volume the Leiden glossary and three versions of theReichenau glosses,aremuch moregrammaticalthanexegetical.Italsoseems,accordingtoTonyHunt, that lexical glosses and word-lists that translated Latin words into the vernacular were notin wide use in post-Conquest England until the thirteenth century (T. H
UNT
,
Teaching and Learning Latin in 13 C England
(Cambridge, 1991), pp. 4-18).
th
Foranexampleofbatchglosses,takenfromvarioussourcesthatP.Lendinaratraced,see:
36
P. L
ENDINARA
, “The glossaryin Oxford, Bodleian Library, Bodley 163", in:
EAD
.,
Anglo-SaxonGlosses and Glossaries
, pp. 329-355, at pp. 347-355.It is tempting to imagine these glosses read aloud by the master along with the text, but
37
glosses are not necessarily the direct reflection of teaching practices, as Michael Lapidge hasconvincingly argued; they could be copied at the same time as the text, and could be intendedmore for private study. See M. L
APIDGE
, “Thestudyof Latin textsin late Anglo-Saxon England,1, The evidence of Latin glosses”, in:
Latin and the Vernacular Languages in Early Medieval Britain
, ed. N. B
ROOKS
(Leicester,1982), pp. 99-140. But onecanfind verystrong arguments intheanswerofG. W
IELAND
, “The glossed manuscript: Classbookorlibrarybook?”,
Anglo-Saxon England
14 (1985), pp. 153-173. Twenty years into the debate, Gernot Wieland surveyed thequestion and its most prominent contenders, while still sticking to his guns, in
ID
., “Interpretingthe interpretation: The polysemy of the Latin gloss”,
The Journal of Medieval Latin
8 (1998), pp. 59-71.
303
Did Medieval Monks Actually Speak Latin?
the student’s part. Is it reasonable to assume that most monks wouldhave madesuch an effort?Acquisition of vocabulary was not the only hurdle, of course. When itcame to learning the rudiments of Latin grammar, it seems that the main exer-cise was ‘declension’. This exercise, modelled on Donatus’s
Ars minor
, con-sisted in giving all the particulars of the words encountered in a text and pro-ducing the declined forms of each word. It is described by John of Salisbury
38
as being applied to verse in twelfth-century Parisian schools. Its importance
39
in the monastic teaching of Latin is evident in a letter sent by Anselm, thenabbot of Bec, to the monk Maurice who was spending some time at Canter- bury. Maurice had started studyingunder a certain Arnulf, and Anselm vehe-
40
mentlyexhorted himto decline and decline again, even though this might seema rather elementary exercise. It must have been an efficient way to becomefluent enough to composea decent piece of Latin, but also time-consumingandtiresome for both student and teacher, as Anselm made clear in his letter:
you know that I have always found it very tedious to make children decline, so I amaware that you made a good deal less progress with me in that discipline than youshould have.
If Anselm of Bec could not bringhimself to make the effort, how manymasterswould actually have dedicated the necessary hours to this?In anycase, the proficiency monks attained would have been veryvariable.It must have depended heavily on individual masters and their commitment totheir task. We know of outstanding figures, and we know of masters listed in
On declension, see M
URPHY
, “The teaching of Latin”, pp. 168-172. On ‘parsing gram-
38
mars’, which were based on declension but applied it only to the words that they parse, see M.B
AYLESS
, “
Beatus quid est
and the study of grammar in late Anglo-Saxon England”,
Historio- graphia Linguistica
20 (1993), pp. 67-110. See also D. C
HAPMAN
, “
Anima quae pars
: A tenth-century parsing grammar”,
The Journal of Medieval Latin
12 (2002), pp. 181-204.John of Salisbury,
Metalogicon
,
I
. 24, ed. J.B. H
ALL
and K.S.B. K
EATS
-R
OHAN
39
(Turnhout, 1991:
CCCM
98) p. 51. John insists here and elsewhere on the importance of a goodcommand of
grammatica
before moving on to higher spheres of knowledge.Letter 64, in:
Opera omnia
, 3, p. 180: “
Audivi quoque quod ipse multum valeat in
40
declinatione, et tu scis quia molestum mihi semper fuerit pueris declinare, unde valde minusquam tibi expediret, scio te apud me in declinandi scientia profecisse. Hortor itaque et precor etutfiliocarissimopraecipio,quatenusquidquidabeo[Arnulf]legerisetquidquidaliudpoteris,diligentissime declinare studeas ... Quod si aliqua re obstante non potes ab eo legere: vel hoc stude ut librorum quos legisti, quoscumque potes et quibus horis potes, totos a principio usquead finem diligentissime, sicut supra monui, declines
”.
304
JULIE BARRAU
cartularies and named in chronicles, but it seems daring to conclude from thisanything about the form and quality of the teaching they delivered. It is clear from Anselm’s self-deprecating remarks that the basic training in
grammatica
took considerable effort on the masters’ part. It also tells us that while loftyintellectuals and busy abbots, who lived in memory as the pride of their breth-ren, may have been inspirational figures for the brightest recruits, they did notnecessarily contribute to the primary training of the young.
41
It is difficult to find out about the teaching of Latin in monasteries, buteven less is known about how monks were taught to speak Latin, as opposed toreading or chanting it. It is likely that the teaching in monastic schools was
42
mainlyaimed at preparingoblates and novices for liturgical services and teach-ing them to read, as is suggested by the provisions that were made for monksto borrow books from the library. That makes the Anglo-Saxon
colloquia
very precious, in that they were specifically intended for the teaching of spoken
43
Latin; theyused a fairly rudimentary syntax, and were probablymeant mainly
44
On Anselm’s narrow circle of pupils, see R. S
OUTHERN
,
Saint Anselm: A Portrait in a
41
Landscape
(Cambridge,1990), pp. 369-370, and 445: “Asfor the stimulation ofan audience, hefound it not in a classroom or in debate, but in colloquies with one or two like-minded youngmen”.SeealsoM.L
APIDGE
,“Aethelwold asscholarand teacher”,in:
ID
.,
Anglo-LatinLiterature900-1066
(London, 1993), pp. 183-211, at pp. 205-206, about Aethelwold’s role in Wulfstan of York’s education.Childrenweregivenincreasingliturgicaldutiesoverthetenthandeleventhcenturies,and
42
preparation for the officemust havetaken up most oftheir studying time. It was a priorityfor the boys to get their singing right, since otherwise theywere punished. On all this, see S. B
OYNTON
,“Trainingfor the liturgyasaformofmonasticeducation”, in:
MedievalMonasticEducation
,ed.C.M
UESSIG
andG.F
ERZOCO
(Leicester,LondonandNewYork,2000),pp.7-20.SusanBoyntonnevertheless emphasizes that school time and liturgical training should not necessarily seen asseparate, and competing, parts of the oblates’ lives; while learning by heart texts for the Office,children may have improved their understanding of Latin (
ibid.
, p. 11-14, and also
EAD
., “LatinglossesontheOfficehymnsineleventh-centurycontinentalhymnaries”,
TheJournalofMedieval Latin
11 (2001), pp. 1-26, particularlyatpp.4-8 on lexical, grammaticaland syntacticalglosses.On the increasing prerogatives of the
cantor
and the development of a double office of
cantor
/
armarius
, see M. F
ASSLER
, “The office of the cantor in early Western monastic rules andcustomaries: A preliminary investigation”,
Early Music History
5 (1985), pp. 29-51.
Aelfric’sColloquy
,ed.G.M.G
ARMONSWAY
(Exeter,1978);
Anglo-SaxonConversations:
43
The Colloquies of Aelfric Bata
, ed. S. G
WARA
with a translation by D. P
ORTER
(Woodbridge,1997). However down-to-earth, these colloquies must have been difficult enough for somestudents, since one of the four extant copies of Aelfric of Eynsham’s colloquy received acontinuousOldEnglishinterlineargloss.OnwhatAelfricBatarevealsofLatinteachingpractices,see D. P
ORTER
, “The Latin syllabus in Anglo-Saxon monastic schools”,
Neophilologus
78.3(1994), pp. 463-482.Seethe studentin Aelfric’s
Colloquy
who saysthatreadingandchantingarenotenough,
44
305
Did Medieval Monks Actually Speak Latin?
for lexical acquisition. But outside the British Isles, and after the Conquest,these kinds of dialogues did not continue. Following Donatus, many medievalgrammars up to the twelfth century were often inquestion/ answer form, espe-ciallythe ‘parsing grammars’, but it is not clear how much this had incommonwith the classroom practices.
45
Common sense suggests that among those who were taught Latin somewere less talented at it, or less interested, than others. The long training pro-duced
litterati
, in the medieval sense of the term: this did not imply muchabout their ability to speak Latin in daily life. Those who had the opportunityand the desire could probably eventually read Latin, understand the liturgyandin some cases compose a written text such as a letter, or copy one in the
scriptorium
. But this does not imply that they could speak, in the sense of beingable to produce entirely relevant sentences inthe different circumstancesof everyday life. Compared with written literacy, the ability to speak can beseen both as more elementary (snippets of everyday Latin must have been usedand understood by
illiterati
such as the
conversi
) and more complex (we allknow that understanding, reading and speaking a language are completelydifferent abilities). It has often been shown that reading and writing can be
46
completely dissociated skills; similarly, there are many reasons to think thatspeaking Latin did not necessarily follow an ability to read it, even less a ca- pacity to sing it. The constant liturgical hours and the readings in the refectory provided an intense background of Latin orality, with large numbers of Latinwords uttered and heard. Since this Latin was the praise of the Lord, it wasimportant that it was well pronounced. But orality is a misleading term here,
47
because that mass ofpronounced Latintells uslittle if anything about how wellthose who listened to it, or even chanted it, could come up with original sen-tences concerning their everyday life.
and he wants to learn to speak: “
vellem interim discere sermocinari latina lingua
” (
Colloquy
, p. 19).However, Martha Bayless thinks that it did (M. B
AYLESS
, “
Beatus quid est
”, pp. 78-79).
45
Itwasprobablytruethenasitisnowthatmutualintelligibilityisitselfamatterofdegree;
46
on the problematic delimitation of languages, see R. H
UDSON
,
Sociolinguistics
, 2nd edn. (Cam- bridge, 1996), pp. 34-38.This was a main concern of the Carolingian reformers; see R. W
RIGHT
, “Late Latin and
47
earlyRomance:Alcuin’s
DeOrthographia
andtheCouncilofTours(
AD
813)”,in:
Papersofthe Liverpool Latin Seminar
,
Third Volume
, ed. F. C
AIRNS
(Liverpool, 1981), pp. 343-359, at pp. 345-351. See also, with further thoughts on the consequences of the new pronunciation ondidactic practices,
ID
.,
Late Latin and Early Romance
, pp. 104-118. On the importance given toa proper pronunciation of the office, see B
OYNTON
, “Training for the liturgy”, pp. 10-11.
306
JULIE BARRAU
If one accepts that it must have been a major undertaking to attain a levelof Latin that made it possible to understand and produce complex sentences, itthen becomes pertinent that texts that indicate explicitly what language wasspoken between monks are rare. In the huge bulk of sources that describe lifein monasteries directly or obliquely, a remarkably small proportion make anymention of which language is spoken. In a lay context it was clear why lan-guages used between characters were not specified – people would use thelocal vernacular. But such a ‘default language’ does not so obviously exist ina monastic environment. Although it might have been the impression clericalauthors wanted to convey, whether consciously or unconsciously, Latin wasnot the vernacular of the cloisters.I would like to approach the question froma complementary point of view.I shall focus here on a few texts in which it is explicitly said that Latin
was
used, and try to understand what that tells us about the actual linguistic situa-tion in monasteries. Far from being shown buried deep in silence, monks doindeed speak in these sources, be they collections of
exempla
, chronicles or
Vitae
, to name just the main genres, but the language in which they speak israrely mentioned. The working hypothesis of this article is quite straightfor-ward: if it is specified that the language used was Latin, then the expectedlanguage would have been the vernacular. Such evidence might suggest notonly that the daily language of the professed choir monks could have been their mother tongue, but that this does not seem to have been particularly surprisingor problematic to the authors who described it.Benedict of Peterborough reports an interesting exchange in his collectionof the miracles of Thomas Becket. Benedict tells his readers that a short time
48
after the murder he met the ghost ofthe archbishop, and dared to speak to him.He specifies that he addressed Thomas in French and that the ghost replied inLatin. This linguistic asymmetry can be interpreted in various ways. Becket’sreply in Latin could be a sign of his sanctity and his proximity to the divinesphere, or a way for his hagiographer to reply to the malevolent gossips whoaccused the former chancellor of being a poor Latinist while he was alive.Perhaps the simplest explanation, though, is that for a monk of CanterburyFrench was the default language in which to address a fellow cleric, be he a
“
Etaccedenscominus,petitaetacceptabenedictione,subjeci:‘Ne,quaeso,domineiras-
48
carissiteinterrogavero’.Etille‘Loquere’.‘Domine,inquam,nonnemortuuses?’AtilleGalliceinterroganti respondit sermone Latino: ‘Mortuus fui, sed surrexi’
“ (Benedict of Peterborough,
Miracula S. Thomae
, ed. in:
Materials for the History of Archbishop Thomas Becket
, ed. J.C.R
OBERTSON
, 7 vols. (London, 1875-1885:
Rolls Series
67.1-7), 2, p. 27).
307
Did Medieval Monks Actually Speak Latin?
dead prelate or not. Peter the Venerable, in his
De miraculis
, recounts the storyof Turquillus, a monk of Cluny to whom he had entrusted the nunnery of Marcigny. The abbot of Clunywas veryfond of that house, and probablychosethe man he appointed with great care. On his deathbed Turquillus had a vi-
49
sion (of “the Lord and his sweet entourage”) which he described to the brothersgathered around him – and he did so, Peter tells us, in Latin. As it is told to
50
us, this story suggests that a question was put to him in French and that hereplied in Latin. But why did Peter the Venerable specify this? And why washe so insistent? “He did not say it in the vernacular, but as I have written it, inLatin”. Turquillus was a monk, a fully professed monk, and Peter does not present him as particularly uneducated or illiterate. Rather, it seems to be im- plicit that the question was put to him in French and that the abbot of Clunydidnot seektohide or excuse this use of the vernacular. It was the use of Latin thatwas exceptional.It is worth questioning whether Latin was the language generally spoken, both because the vernacular mayhave been a more obvious choice and becauseit is likelythat a number of monks were simply not able to read or speak Latin:the
monachi laici
, the illiterate monks. Caution is necessary when sources
51
describe illiterate clerics, since to criticize somebody’s Latin was to criticizemore generally his competence as a man of the church. When Anselm of Bec
52
wrote to Paul of Caen to congratulate him on his appointment as abbot of St.Albans just after the Conquest, telling him that he would not be able to com-municate with his brethren, and should teach them
exemplo
if not
verbo
, he
Peter’s mother was cellerar at Marcigny for some twenty years; it was also the house
49
where, at the same time, Adela of Blois, the daughter of William the Conqueror and mother of King Stephen, retired. Marcigny was praised by the abbot of Cluny for its strict enclosure. OnPeter’s friendship with Turquillus, see E.M. W
ISCHERMANN
,
Marcigny-sur-Loire: Gründungs-und Frühgeschichte des ersten Cluniacenserinnenpriorates (1055-1150)
(München, 1986), pp. 110, 121-122.“
Cumque illum in cinere et cilicio positum, fratres undique circumstarent, repente ab
50
eis spiritu raptus obmutuit. Qua de causa frater quidam ex his qui astabant, Unfredus nomine,honestae vir conversationis, ausu familiaritatis qua prae caeteris eisemperadhaeserat, ad eum se inclinavit, et si aliquid spiritualium rerum vidisset, quaesivit.
Ad quem ille his verbis:
‘Vidi,
ait,
Dominum, et dulce consortium ejus’.
Hoc cum non vulgaribus, sed ita, ut expressi, verbisLatinis dixisset,
parvo interveniente temporis spatio, defunctus, est
” (Peter the Venerable,
Demiraculis libri duo
, ed. D. B
OUTHILLIER
(Turnhout, 1988:
CCCM
83), pp. 63-64).On the Cluniac
conversi
, see T
ESKE
, “Laien, Laienmönche und Laienbrüder”.
51
On this frequently used technique of disparaging fellow clerics, see H
AYE
,
Lateinische
52
Oralität
, pp. 9-12.
308
JULIE BARRAU
assumedthat the English monks could not be addressed in Latin. It is difficult
53
to tell whetherhe knew that they were not Latinate, or whether he assumed outof mere prejudice that those “Barbarians” probably would not be.Gerald of Wales is an interesting source to turn to here, as long as one bears in mind that he often had an agenda: sometimes self-glorification, some-times a bitter portrayal of the church of his day. Nevertheless, he has one
54
useful quality, which is that in many of his works he is clearly very aware of the multilingual nature of the British Isles in the late twelfth century. In the
Gemma ecclesiastica
, in a chapter on illiterate clerics, he tells two anecdotesabout abbots with no Latin. One is anonymous, which makes the story’s au-thenticity rather dubious, but the other abbot is named: he is Robert, abbot of Malmesbury, one of the most important abbeys in England, from 1159 to1176. Robert was accused of
illiteratura
by his monks ... and it seems that
55
they were right. The two papal legates asked him to translate into French averse from the Acts of the Apostles, “
Factus est repente de coelo sonus
” (Acts2.2): “Suddenly a noise came from the sky”. Robert seems to have been awareof the gaps in his knowledge (“
stetit et hesitavit
”), but he had a try, placing histrust in the similarity between
repente
and the French verb
se repentir
. Unfor-tunately, the latter word comes from the late Latin verb
repoenitere
, which hasnothing to do with the classical adverb
repente
. A similar confusion betweenFrench and Latin appears in the other anecdote.Gerald of Wales’ abbots did not just have a rather rudimentary commandof Latin: they were genuine
illiterati
. It is clear that Robert, who could nottranslate a very simple phrase, must have spent the greater part of his life par-ticipating in services without understanding a single word; he would clearlyhave had to speak French to his monks. Even though his
illiteratura
was obvi-
“
Quamvis enim barbaris vestra praelata sit sanctitas, quos verbis docere propter
53
linguarum diversitatem non potestis, non tamen omnino vos apud districtum Judicem excusarevaletis, si alios lucrari Deo negligitis. Quod enim illis sermone non potestis dicere, vita potestisostendere
” (Letter 80,
Opera omnia
, 3, p. 203). On St. Albans before and immediately after theConquest, see R.M. T
HOMSON
,
Manuscripts from St. Albans Abbey, 1066-1235
, 2 vols.(Woodbridge, 1982), 1, pp. 8-14.On Gerald’s criticism of monks, see Knowles,
The Monastic Order
, pp. 664-674.
54
“
Item exemplum de abbate Malmesburiensi Roberto, qui nostris diebus accusatus est a
55
monachis suis super illiteratura, et a judicibus a papa Alexandro
III
delegatis, Bartholomeo scilicet Exoniensi et Roger Wigorniensi, ex hoc probatus:
Factus est de coelo sonus
etc. Cumcaetera Gallice interpretatus esset, veniens ad
repente,
stetit et hesitavit, tandem vero dixit:‘Repente,
il se repenti’ “ (
Gemma ecclesiastica
,
Giraldi Cambrensis opera
, ed. J.S. B
REWER
, 8vols. (London, 1861-1891:
Rolls Series
21.1-8), 2, p. 346).
309
Did Medieval Monks Actually Speak Latin?
ous, he was not on that account regarded as unfit to remain an abbot. If Geraldis to be believed, his qualities as an administrator were considered more impor-tant. If one could be an abbot with no Latin, then it is highly likely that a monk could get away without it, and that a monastery in which communication took place in the vernacular was not exceptional. This is confirmed by sources morecandid than Gerald.Geoffrey, the abbot of the Trinity of Vendôme, tookupthe case of a monk called Savaric, who was wrongly accused by his abbot of serious disobedi-ence. In the first of two letters that he addressed to Bishop Renaud of Angers
56
on the subject, Geoffrey specified that Savaric had had to defend himself inFrench rather than Latin because he was
laicus
, that is to say, illiterate. Geof-frey does not seem to have been particularly apologetic about that deficiencyin his protégé, or to have thought less of him, since he calls Savaric “
domnusSavaricus
”. It is just a simple fact that explains whyhe had to intercede, since
57
Savaric could not write to the bishop. In the Cistercian
Exordium magnum
there are several stories praising lay brothers (
conversi
), and in one of them a
conversus
is miraculously made literate on his deathbed. But illiteracy was
58
not the monopoly of the lay brothers: Walter, a professed monk, whose devo-tion was “not inferior to that of the scholars and the learned”, was rewarded bya dream that taught him a mass, which he could thereafter sing along with hisliterate brothers. Conrad of Eberbach plainly writes that before the miracu-
59
lous dream the monk “since he was illiterate did not understand a word of themass” (“
eandem missam, quam antea penitus utpote litteras nesciens ignora-bat
”). There is no indication in the
exemplum
that Walter’s ignorance of Latinwas particularly remarkable.At the beginning of the twelfth century Pope Calixtus
II
addressed a letter to the abbot and monks of Schaffhausen, a Swiss Benedictine abbey, in whichhe offered his direct protection to the house. At the end of his letter the popeasked the abbot to send him one of his monks, as he had already asked in ear-
GeoffreyofVendôme,
Oeuvres
,ed.G.G
IORDANENGO
(Paris,1996),Letter115,pp.220-
56
222: “
Ad cujus objecta monachus, quia laicus est, non Latina, quam non didicit, lingua, sed maternarespondet,etdicitinhoctantummodoseesseculpabilem,quodabbati,utseabsentaret, praecipienti oboedivit
”.
Ibid.
,Letter116,p.224:“
Mandastis,dilectissimedomine,perfratremnostrumRotbertum
57
ut domnus Salvaricus ad abbatem Lambertum revertatur
”.Conrad ofEberbach,
Exordiummagnumcisterciense
,
IV
.17,ed. B. G
RIESSER
(Turnhout,
58
1994:
CCCM
138), pp. 261-263.
Ibid.
,
IV
.15, pp. 258-259. There is another glorification of a
monachus laicus
at
IV
.12,
59
pp. 253-255.
310
JULIE BARRAU
lier letters, and specified that the brother should be able to speak German andLatin. Calixtus would obviously himself speak Latin, and the same would
60
have been true of the officials of the papal court. If the pope requested this, itwas very probably not obvious to him (or to his secretary) that a monk fromSchaffhausen, whose native language was most likely to be German, wouldalso speak Latin. And this was the case even for a monk sufficiently importantin the monasteryto be sent to the pontifical court – this calls into question eventhe idea that in all communities there would be a properly Latinate elite. Theabbot AdamofSaint-Trond, at the end of the thirteenth century, is presented inthe chronicle of his abbey as an ‘abbot with no Latin’, even though in other respects he is presented very positively. We are told of a visit from the Do-
61
minican Nicholas Boccasini, the future pope Benedict
XI
. When he was aboutto join the abbot’s table, an ‘honest man’ suggested that he should speak French. The guest followed his advice and spent an excellent evening. Sometime later, he had become pope and a delegation from Saint-Trond came to present a request. The pope asked if they were the monks of the “abbot with noLatin”, of whom he had such pleasant memories.When theysaid yes, he imme-diatelygranted their request. It is strikingthat the chronicler is not embarrassedto recount this storyand that the linguistic deficiencies of the abbot even seemto have worked to the abbey’s advantage.
Monachi laici
are likely to have been a normal group of any monasticcommunity, of varying importance from one house to another and from onecentury from another. The
conversi
of what Giles Constable calls the ‘old
62
“
Rogamus etiam, sicut aliis jam litteris rogavimus, ut nobis unum ex vestris fratribus
60
dirigatis, qui et Teutonicam linguam noverit et Latinam
” (ed.
PL
163, col. 1156).
Gesta abbatum Trudonensium, Continuatio Tertia: De Gestis Ade abbatis
(
MGH SS
10,
61
p.411):“
Cuiadaulamaccedentiaquodamhonestovirosuggestumest,quodGallicalingua,non Latina, cum abbate haberet colloquia. Quod ille gratanter intelligens, dulcia familiaritatisobsequia promittens, jocunde delectati sunt, status sui qualitatem alter alteri mutua collationeverborum depromentes ... Que cum ad noticiam pape relate essent, vocatis ad se nunciis illis, papa quesivit, si causa esset abbatis sine Latino, qui nuper ei in claustro sancti Trudonis tam jocundum prandium exhibuit. Illis id affirmantibus, papa indilate jussit in omnibus petitionemillorum ad effectum produci
”.On the variety of situations, and the difficulty of drawing general conclusions from
62
particular situations, see J. D
UBOIS
, “L’institution des convers au
XII
siècle: Forme de vie
e
monastique propre aux laïcs”, in:
I laici nella “Societas christiana” dei secoli
XI
e
XII
: Atti dellaterza Settimana internazionale di studio, Mendola, 21-27 agosto 1965
(Milan, 1968), pp. 183-261. The most extensive study of
conversi
is probably that of W. T
ESKE
, “Laien, LaienmöncheundLaienbrüder”,firstpart(1976),particularlyfromp.278onwards,andsecondpart(1977).Of particular interest are the lists of converts compiled from eleventh- and twelfth-century Cluniac
311
Did Medieval Monks Actually Speak Latin?
type’, as opposed to the Cistercian lay brothers, were real professed monks.
63
At least till the mid-twelfth century they were full members of the community(except among Cistercians and Carthusians), though their somewhat junior status was marked in the liturgy; they were often defined by their inability to
64
sing. It is worth examining what their total or partial Latin illiteracy implied
65
for the linguistic functioning of their communities, since they would have hadto be addressed in the vernacular, or maybe in a sort of Latinate pidgin. Theyattended the chapter meetings: could that mean that those were not held inLatin? Was the Rule read aloud in Latin, or in the vernacular? When Peter theVenerable read aloud to his monks a letter from the pope, he probably impro-vised a vernacular version since he wrote that he read it for the benefit of bothliterates and illiterates. It is relevant for this study that the decline of oblation
66
from the eleventh century led monasteries to receive ever-increasing numbersof youths and young men past school age. Some of them entered their new lifealready literate, but there was no strict requirement of literacyin a new recruit,at least till the earlythirteenth century. As for those whobecame monks while
67
sources, in the second part, pp. 320-336, with laymen and clerics clearly identified.C
ONSTABLE
,“‘Famuli’and‘Conversi’atCluny”,pp.334-340.Hedefinesthe“oldtype”
63
conversi
as “monks who had entered the monastic life as adults and were illiterate and in nodegree of clerical orders” (pp. 334-335) For Giles Constable the statute of Peter the Venerableis not a major turning point towards a status of Cluniac conversi more similar to that of theCistercianlaybrothers;suchassimilationisnotreallydocumenteduntilthebeginningofthenextcentury (
ibid.
, pp. 348-350). On
conversi
at Clunyand their evolution, see also D. I
OGNA
-P
RAT
,
Ordonner et exclure: Cluny et la société chrétienne face à l’hérésie, au judaïsme et à l’islam,1000-1150
(Paris, 1998), pp. 51-52.On their junior status, as reflected in the liturgy, see T
ESKE
, “Laien, Laienmönchen and
64
Laienbrüder”,<firstorsecondpart?>pp.293-306, and C
ONSTABLE
, “‘Famuli’ and ‘Conversi’”, pp. 336-340.In lateeleventh-centuryCluny, afteramonkwasprofessedtheabbot decided whetherhe
65
would be a
cantor
or a
conversus
, to whom practical tasks would be given during the office: “
et domnusabbassecundumgratiam,quaesingulisestdata,jubeteosdecaeteroinecclesiaservire, scilicet, ut legant, et cantent, qui hoc sciunt; et qui aliud non valent, candelabra et thuribula portent
” (Ulrich of Cluny,
Epistola nuncupatoria
<check>, ed. in:
PL
149, col. 714).Peterreadtheletter“
tamlitteratisquamilliteratisquosconversosvocamus
”(Letter112,
66
The Letters of Peter the Venerable
, ed. G. C
ONSTABLE
, 2 vols. (Cambridge, MA, 1967), 1, p. 300).One had to meet various requirements to be accepted into a Cistercian monastery, but
67
literacy was not one of them in the twelfth century. There is only one twelfth-century Cistercianstatute that has anything to do with education; it is a reference to reading times for novices andmonks. See
Statuta capitulorum generalium Ordinis Cisterciensium
, ed. J.M. C
ANIVEZ
, 8 vols.(Louvain, 1933-1941), 1, p. 31: “
Nullus puerorum doceatur litteras intra monasterium vel inlocismonasterii,nisisitmonachus,velreceptusinprobationenovitius.Quibustemporelectionis
312
JULIE BARRAU
being doubly illiterate, unable to read and without Latin, it is unlikely that oneyear of novitiate could improve their situation much. The differences in intel-lectual and linguistic abilities that must have always existed are likely to havedeepened; not necessarily only between
nutriti
and
conversi
, but also, among
68
the new adult brothers, between those who could pick up the sacred tonguerelatively easily or knew it already, and those for whom the liturgy and thewritten word would remain forever quite opaque. Monks were not supermen.Vernacular languages may therefore be the principal but barely acknowl-edged linguistic vehicle of everyday life, because a significant part of the brethren could speak nothing else. But this is not all. We know of monks andabbots using the vernacular without fear of reprimand, but we also know of times where the vernacular was used for prestigious activities suchas scripturalcommentary. For instance, the dedicatory letter that Abbo of Fleury puts at the beginning of his
Life of Saint Edmund
informs us that the archbishop of Can-terbury, Dunstan, used to comment on the divine word to his monks in Latin, but also in English. Williamof Malmesbury, in his
Life of Wulfstan
, presents
69
the bishop of Worcester commenting in his mother tongue on the text that wasread in the refectory. Another example of Latin and vernacular being used
70
interchangeably by a literate dignitary is given by the Chronicle of Battle Ab- bey; it says that the abbot Odo (1175-1200) used Latin and French equally for his scriptural lessons. It is also interesting that Aelred of Rielvaux’s biographer decided to show him saying on his deathbed “
for crist luve
” because that wassofter on the ear: “
dulcius quodammodo auditur
”. In that Cistercian context
71
discere licet
” (year 1134). Two thirteenth-century statutes do call for
litteratura
, and their mainmotivationisthereputationoftheOrder;see
Statuta
,ed.J.M.C
ANIVEZ
(Louvain,1934),2,p.93-94 (year 1231), and
ibid.
, 3, p. 115 (year 1273).See the “
Contentio inter conversos et nutritos
” reported in an Anselmian text. See
68
Memorials of St. Anselm
, ed. R.W. S
OUTHERN
and F.S. S
CHMITT
(London, 1969), pp. 68-69.AbboofFleury,
PassioSanctiEadmundi
,
Praefatio
,ed.in:
ThreeLivesofEnglishSaints
,
69
ed.M.W
INTERBOTTOM
(Toronto,1972),pp.67-68(Abbo’svisittoDunstanprobablytookplacein 985): “
Audierant enim quod jam pluribus ignoratam, a nemine scriptam, tua sanctitas exantiquitatismemoria collectam, historialitermepraesente retulisset domino Rosensis Ecclesiaeepiscopo,etabbatimonasteriiquoddiciturMealmesbyry,acaliiscircumassistentibus,sicuttuusmos est, fratribus quos pabulo divini verbi Latina et patria lingua pascere non desinis
”.This story about Wulfstan, as well as the following two about Ailred and Odo, is found
70
in M. R
ICHTER
,
Sprache und Gesellschaft im Mittelalter
, pp. 91-93.
The Life of Ailred of Rievaulx by Walter Daniel
, ed. M. P
OWICKE
(Oxford, 1950, repr.
71
1978), pp. 59-60.
313
Did Medieval Monks Actually Speak Latin?
it would fit very well the monastic virtue of
simplicitas
, which perhaps playedin favour of the vernacular rather than potentially suspect Latin flourishes.
72
So Latin was not necessarily used, even by people who knew it very well,even to discuss lofty theological questions. This may be what was hinted at ina distinction made by Bruno of Cologne in his commentary on Psalm 55,11(“
In Deo laudabo verbum; in Domino laudabo sermonem
”):
But
verbum
and
sermo
are not the same thing. It is accurate and usual to speak of
sermo
when somebody addresses somebody else, or a few people, not in a wellcomposed text, or an ornate and carefully constructed speech, but in the vernacular.
Verbum
, on the other hand, is when someone intends an elaborate composition inLatin or the vernacular to be for the benefit of the general public.
73
To the learned master turned Carthusian, it seems that the significant distinc-tion between two types of speech, one loftier than the other, was not the lan-guage that was used, but the care with which the speech was prepared, as wellas the audience for which it was intended.Another possible source of varying
equilibria
between vernacular andLatin was the status of those vernaculars. If they were well on their way to becoming literary languages, as in tenth-century Germany or England, it is possible that they were more tolerated in monastic communication. Even dis-tinguished Latinists such as Gerhoch of Reichensberg developed enough pridein their mother tongue to find it superior to others when it came to singing theLord’s praise. He pointedly writes about laypeople, but there is further evi-
74
dence, such as the mid-ninth-century letter of Otfrid of Weissenburg to Liut- bert, bishop of Mainz. The Alsatian Benedictine introduced there his harmony
Speaking too well could indeed be considered with suspicion. Unsurprisingly Peter
72
Damiancomesuphere;heoftencondemnedscholarlyrefinements,andmorespecificallyattackedinaletterthosemonkswho practisedanornateandelegantLatin, remindinghisnephewMarinusthat “fromaChristian mouth should resound Christ’s simplicity, and not Cicero’sLatinity”(
Die Briefe des Petrus Damiani
, ed. K. R
EINDEL
, 4 vols. (München, 1983-1993:
MGH
Die Briefe der deutschen Kaiserzeit
4.1-4), 3, Letter 132, p. 452).“
Sedverbum etsermodifferunt. Sermonemnamqueproprieetusualiterdicimusquando
73
aliquisnoncompositoscripto,velcompositaetornatalocutione,sedvulgariloquituradaliquem,aut ad plures. Verbum autem quando compositione Latinitatis seu vulgaris locutionis ad quem-libet agit
” (ed.
PL
152, col. 892).“
AtqueinoreChristomilitantiumlaicorumlausDeicrebrescit,quianonestintotoregno
74
Christiano, qui turpes cantilenas cantare in publico audeat, sed ut diximus, tota terra jubilat inChristi laudibus etiam per cantilenas linguae vulgaris, maxime in Teutonicis, quorum linguamagis apta est concinnis canticis
” (ed.
PL
193, col. 1436 (1148)).
314
JULIE BARRAU
of the Gospels in German rhyming couplets, and he clearly believed that hismother tongue, however uncouth, deserved to be made more grammatical andcivilized. In the tenth century the translating and glossing activity of Notker
75
Labeo at St. Gall contributed to the legitimatization of Old High German; it
76
has been argued that this was done not against Latin, but somehowalongwith,and supported by, a developing Latin literature.
77
In contrast, in Romance speaking countries the vernacular took much lon-ger to become a ‘proper’ language, conceptually distinct fromits Latin forebear and cousin. Nothing to be compared with Aelfric’s grammar, or Notker’s
78
translations, or even the fully-glossed Anglo-Saxon Psalters can be found inthose regions until the thirteenth century. This may be because such creations
79
were less needed, but it prevented Romance languages from becoming literaryand scholarly languages for centuries. From the late twelfth century onwards,French nevertheless acquired some
gravitas
; it is no surprise that this happenedfirst in post-Conquest England, where it was what the elite spoke. There, as wehave seen with Benedict of Peterborough and Odo of Battle, the monks, whenthey spoke Anglo-Norman, would still distinguishthemselves fromthe
vulgum pecus
. By the time Stephen of Lexington visited the Irish Cistercian houses
80
in 1228-9, French and Latin were put on the same footing in his recommenda-tions. He forbade the acceptance of postulants who could not confess their sinsin Latin or French; the chapter meetings also had to be held in Latin or
81
Otfrids Evangelienbuch
, ed. L. W
OLFF
, 7th edn. (Tübingen, 1973). F. Magoun’s trans-
75
lation ofthelettertoLiutbertcanbeconvenientlyfoundinP.D
UTTON
,
Carolingian Civilization, A Reader
(Peterborough, ON, 1993), pp. 419-422.OnOld High German literature and Notker’s legacy, see the synthesis by J. W
EST
, “Late
76
Old High German prose”, in:
German Literature of the Early Middle Ages
, ed. B. M
URDOCH
(Rochester, NY, 2004), pp. 227-245; also J.L. F
LOOD
, “Literary theory and practice in early-medieval Germany”, in:
The Cambridge History of Literary Criticism
, 2,
The Middle Ages
, ed.A. M
INNIS
and I. J
OHNSON
(Cambridge, 2005), pp. 324-332.G. M
EISSBURGER
,
Grundlagen zum Verständnis der
Deutschen
Mönchsdichtung im 11.
77
und 12. Jahrhundert
(München, 1970).R. W
RIGHT
, “La période de transition du latin, de la lingua romana et du français”,
78
Médiévales
45 (2003), pp. 11-24.See the presentation of the Psalters glossed in Old English in M. G
RETSCH
,
The
79
Intellectual Foundations of the English Benedictine Reform
(Cambridge, 1999), pp. 17-27.On the linguistic situation of post-Conquest England, see I. S
HORT
, “On bilingualism in
80
Anglo-NormanEngland”,
RomancePhilology
33(1979-1980),pp.467-479,and
ID
.,“Languageand literature”, in:
A Companion to the Anglo-Norman World
, ed. C. H
ARPER
-B
ILL
and E.
VAN
H
OUTS
(Woodbridge, 2003), p. 199.“
Nullusadmittatur in monachum, cuiuscumque videlicet fuerit nationis, nisi quiculpam
81
suam sciverit confiteri gallice vel latine, ut, cum venerint visitatores et ordinis correctors, et
315
Did Medieval Monks Actually Speak Latin?
French. Even more remarkably, he insists that the Rule should be explained
82
in French; no Latin is mentionedat this point. This is not to say that it was no
83
longer considered a good thing for members of the regular orders to speak Latin, but it still seems that the decreasing numbers of
monachi laici
did not
84
necessarily increase the amount of Latin spoken in the cloisters. From the thir-teenth century Latinity and literacy are no longer synonymous, not just in thesecular world but probably also in monasteries.
85
Sensibly, provisions were made were made for the
laici
. For instance in theearly twelfth century a Cistercian statute made sure that no illiterate monk would receive the last rites in Latin. The presence of important groups of
86
illiterate monks would also give a new importance and meaning to the transla-tion of important monastic texts, in regions where the local vernacular was not
intelligent et intelligantur ab ipsis
”(
Registrum epistolarum Stephani de Lexinton
, ed. B. G
RIES
-
SER
,
Analecta Sacri Ordinis Cisterciensis
2 (1946), p. 93). Stephen informed the abbot of Clairvaux of his decision (
ibid.
, p. 47); Stephen added that the Irish should send their postulantsto Oxford or Paris to learn “letters and skills of speech” (“
litteras et loquele peritiam
”). TheIrishmen were probably to learn both Latin and French there, and it is tempting to translate
litteras
by Latin and
loquele
by French. Stephen also imposed his statute upon the abbot of Tracton,aWelsh-speakinghouse(
ibid.
,p.57).Heemphasizesthenecessityforamonktobeableto meditate on the Word ofGod “ifonlyalittle”(“
in lege Dei meditari saltim modicum quiddamdie aut nocte
”). See also B. O’D
WYER
, “The problem of education in the Cistercian Order”,
Journal of Religious History
3.3 (1965), pp. 238-245.“
Item regula de cetero non nisi gallice exponatur nec capitulum monachorum teneatur
82
nisi gallice vel latine
” (
Registrum epistolarum Stephani de Lexinton
, p. 93).
Ibid.
, p. 47: “
Unde Regula jam apud Mellifontemetaliam filiam vestram de Beatitudine
83
insimulquampluresdomosHibernienonnisigallicetamexponiturnecdeceteroexponitur
”.Thiswas also repeated in the letter to the abbot of Tracton, whom he reminded that the Rule must beunderstood by all (
ibid.
, p. 57).For instance the clerical members of the Carmelite order were required by the general
84
chapter in 1294 to speak in Latin. See F. A
NDREWS
,
The Other Friars: The Carmelite, Augus-tinian,SackandPiedFriarsintheMiddleAges
(Rochester,NY,2006),p.20.Thisreferencewasgiven by Frances Andrews to Elisabeth van Houts who most kindly passed it on to me.Brunetto Latini, the Italian author of
Li Livres dou tresor
, said in the 1260s of the
85
languagehe chose to write in: “[French] speech is the most pleasant and the most widely spokenamonglanguages”.Thisquotewasfound inS. L
USIGNAN
,
Parlervulgairement:Lesintellectuelsetla languefrançaise au
XIII
et
XIV
siècles
(Paris, 1986), p. 131. SeealsoL
ODGE
,
From Dialect
e e
to Standard
, pp. 100-101, 113-114.
Usus antiquiores ordinis cisterciensis
, ed.
PL
166, col. 1471: “
Si conversus est vel
86
monachus qui non intelligit litteras idem illi Romane exponat sacerdos: et conversus romaneconfiteatur se peccasse cogitatione, locutione et opere. Similiter monachus romane confiteatur, si suum nescieritConfiteor
”.Itisclearherethat
non intelligerelitteras
meansnot understandingLatin.
