Miscellanea Giovanni Mercati. Paleografia. Bibliografia. Varia [Vol. 6] 8821004821, 9788821004827

Citation preview

STUDI E TESTI ------------------ 126 -------------------

MISCELLANEA GIOVANNI MERCATI V

o lum e

VI.

PALEOGRAFIA - BIBLIOGRAFIA - VARIA

CITTÀ DEL VATICANO B IB L IO T E C A A P O S T O L IC A V A T IC A N A MCMXLVI

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

MISCELLANEA GIOVANNI MERCATI PUBBLICATA

SOTTO

GLI

SU A S A N T IT À IN

O CCASIO NE

AUSPICI

DI

PIO X I I

D E L L ’ O T T A N T E S IM O N A T A L IZ IO

D E LL’E.MO CABDINALE BIBLIOTECAKIO E AECHIVISTA D I S A N T A R O M A N A C H IE S A

V

o l u m e

VI.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

STUDI E TESTI 126

MISCELLANEA GIOVANNI MERCATI V

o lu m e

V I.

PALEOGRAFIA ■ BIBLIOGRAFIA - VARIA

CITTÀ DEL VATICANO B IB L IO T E C A A P O S T O L IC A V A T IC A N A M C M X LV I

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

INDICE

Bea , A ugustin, S.

I . (R e k to r des P â p stlich es B ib e lin s titu ts ,

Bona). D i e E n ts te h u n g des A lphabets-, e in e k ritis c h e U e b e rs i c h t ................................................................. ........................... P a g .

L owe, Elias A very

(P ro fe s s o r in th e In s titu te fo r a d va n ced

stu d y, P r in c e to n , N e w J e rsey ). i n D a t in m a n u s c rip ts :

th e ir

T h e oldest o m is s io n sign s

o r ig in

s ig n ific a n c e (w ith

and

8 p l a t e s ) ...........................................................................................

J ones, L eslie

W . (In s titu te fo r

N e w J e rs e y ). P r i c k i n g

T horndike , L yn n N ew

Y o rk

a d va n ced

System s i n

(P ro fe s s o r in

C it y ).

The

the

P r o b le m

Medea

36-79

stu d y, P rin c e to n ,

N e w Y o r k M a n u s c r ip t s

C o lu m b ia

o f the

80-92

U n iv e r s ity ,

co m p o s ite

m anu­

s c r ip t ..................................................................................................

N orsa,

1-35

93-104

(P r o f. d e ll’U n iv e rs ità d i F ir e n z e - Is titu to p a ­

p ir o lo g ic o ).

A n a lo g ie

e co in c id e n z e

tr a

s c rittu re

greche

e

la tin e n e i p a p i r i (con 8 t a v o l e ) .................................................... 105-121

Ce r lin i , Aldo fra n ce s e

(P r o f. d e ll’ U n iv e rs ità d i R o m a ).

di

ta c h ig ra fi

nel

secolo

ix

(con 2

Una

scu ola

illu s tra z io n i

e 2 t a v o l e ) ..............................................................................................122-146

W ehmer, Carl

(M ü n chen). D i e S ch re ib m e is te rb là tte r des spà ten

M itte la lte r s (m it 2 T a f e l n ) .............................

Beeson, Charles H enry

(P ro fe sso r

in

th e

147-161 U n iv e r s it y

of

C h ica go ). T h e P a lim p s e s ts o f B o b b i o ...................................... ·

D old, A lban , O. S. R este

z w e ie r

fiir

B. d ie

(B eu ron ,

H o h e n z o lle rn ).

G esch ich te

von

B o b b io

162-184

R a ts e lv o lle b ea ch tlich e r

D o k u m e n te ( e rh a lte n a u f d em V o rs a tz b la tt des G od ex V a t i ­ ca n u s L a t in u s 5 7 6 3 )

(m it 2A b b i l d u n g e n ) ............................

185-207

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

vi

INDICE

K ahle , P aul

(P ro fe s s o r in th e U n iv e r s ity o f L o n d o n ). T h e

Chester B e a tty M a n u s c r ip t o f the E a r k le a n G ospels

L eroquais, V ictor (t 1 mars 1946).

L e s E v a n g ile s

.

Pag.

de J o u a r r e

(avec 4 planches)..................................................................

(u n

m a n u s c rit

in c o n n u

A very , Myr tilla

de l ’école

208-233

fr a n c o -s a x o n n e )

234-257

(N e w Y o r k ). T h e re la tio n o f the C asanatensc

P o n t if ic a l (M s . C a sa n a t. 724 B 1 1 3 ) to ten th ce n tu ry changes i n the o r d in a tio n rite s at B orn e (w ith 8 ch a rts o f w h ic h 3 in section s) . . . . . . . . . . ......................

I nguAnez, Mauro,

O. S. B . (A r c h iv is ta d e ll’A b b a z ia d i M on te-

cassin o). F r a m m e n t i d i c o d ic i abruzzesi (con 5 illu s tra z io n i)

Battelli , Giulio

258-271

272-281

(A rc h iv is ta d e ll’A r c h iv io S egreto V a tic a n o ).

I l L e z io n a r io d i S . S o fia d i B e n e v e n to .................... ....

282-291

B alio , Carlo,

O. F . M . (P re s id e n te d e lla «C o m m is sio S cotis ta », B orn a). S e g n i e n o te c ritic h e nette op ere d i G io v a n n i B u n s S c o t o ................................................... .... ..........................

L azzarini, V ittorio

(P r o f.

d e ll’ U n iv e rs ità

di

P a d o v a ).

Il

m a estro ca rta ro P a c e da F a b r ia n o a T r e v is o ..........................

F ava , D omenico

292-323

324-335

(D ire tto re d ella B ib lio te c a U n iv e rs ita ria d i

B o lo g n a ). M o d e n a e s a n B e r n a r d in o da S ie n a .....................

D e Marinis , T ammaro D onati, L amberto fig u ra te d ella

(F ir e n z e ). N o t e b ib lio g r a fic h e .....................

(C on servatore

d e lle

in cisio n i

e

336-343 344-349

stam pe

B ib lio te c a V a tic a n a ). D i u n a n u o v a « B o ta

p a s ch a lis » ca lco g ra fica del x v

Bertola, Maria (Borna).

secolo (con 6 t a v o l e ) . . . .

In c u n a b u li

e s isten ti

a lla

B ib lio te c a

V a tic a n a d u ra n te i l see. x v .......................................................

Galb iati , Giovanni

(P r e fe tto

della

B ib lio te c a

(A ssisten te

d ella

B ib lio te c a

398-408

A m brosiana,,

M ila n o ). L a p r im a s ta m p a i n a r a b o ......................................

Mazzini, Giovanni

350-397

409-413

V a tic a n a ).

L ’« A s tr o la b iu m » d i L e o n a r d T h u rn e is s e r z u m T h u r n (con

2

t a v o le e

Berra , L uigi

1

d o p p ia ) ...........................................................»

(S c ritto re

d e lla

B ib l.

V a tic a n a ).

.

414-431

U n a ig n o ta

ra p p re s e n ta z io n e sa cra d i M o n d o v ì: « L o iu d ic io de la fin e del m o n d o » B

is h o p

, W

il l ia m

.................................................................................... W

a r n e r

432-441

(L ib r a r ia n em eritu s o f th e G en eral

L ib r a r y o f th e U n iv e r s ity o f M ich iga n ; A n n A r b o r ).

The

d is tr ib u tio n o f e a rly B r it is h p rin te d A m e r ic a n a i n A m e r ic a n lib r a r ie s ..............................................................................................

442-451

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

VII

INDICE

A llgeier, A rthur

(U n iv e rs ità ts p ro fe s s o r,

B ib lio th e k s g e s cliich tlich e K a r d in a ls

G a ra m p i

N a c h r ic h te n

m it

F ü r s ia b t

im

F r e ib u rg

i. B .).

B rie fw e ch s e l

M a r tin

G erbert

des von

S t. B la s ie n ........................................................................... P a g .

T isserant, E ugène,

452-478

C ard. (S e c ré ta ire de la S. C o n gréga tion

pou r l ’ É g lis e o rie n ta le ). L e ttre s de C o n s ta n tin v o n T is c h e n d o rf à C a rlo

V e r c e l l o n e ..............................................................

San N icolò, Mariano

(U n iv ersità tsp ro fes so r,

4 79 -498

M ünchen).

B e s te llu n g eines P f ô r t n e r s f i i r das A M t u -F e s t h a u s d er I s t a r von

U r u k .....................................

R ohlfs, Gerhard

(U n iv e rs ità ts p ro fe s s o r,

499-508 M ü n ch en ).

Feues

aus « G re c ia o tr a n tin a » ...............................................................

509-520

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

A U G U S T IN B E A , S. I.

DIE ENTSTEHÛNG DES ALPHABETS E IN E K R IT IS C H E ÜBERSICH T

D e r v e r e b r t e G e le b r t e , d e m

d ie s e S tu d i en g e w id m e t sin d , b a t

w â h r e n d s e in e r la n g e n w is s e n s c h a ftlic h e n T â t i g k e i t d e m G e b ie t d e r la te in is c b e n

und

g r ie c b is c h e n

m e r k s a m k e it g e w id m e t .

P a lâ o g r a p h ie

s e in e

beson d ere

A u f-

W i r e n t fe r n e n u ns d a h e r n ic h t a llz u w e it

v o n s e in e m A r b e it s g e b ie t , w e n n w ir in d ie s e m B e it r a g v e r s u c h e n , d ie Q u e lle n des

S c b r ifts y s te m s

z u fin d e n , d essen le t z t e A u s là u fe r d ie

la te in is c h e u n d g r ie c h is c h e S c h r ift sin d .

AJ s D o n G io v a n n i M e r c a ti

im J a b r e 1891 s e in e e r s te n w is s e n s c h a ftb e h e n F o r s c b u n g e n v e r ô ffe n t lic h te , w a r d ie P r a g e n a c h d e m T Jrsp ru n g u n s eres in t ie fe s D u n k eJ g e b ü llt.

A lp h a b e t s

n och

D a m a is la g e n n u r z w e i a lt e Z e u g e n d e r

p b o n iz is c h -p a lâ s tin is c b e n S e b r ift v o r , b e id e aus

d e r e r s te n H a l i t e

d es e r s te n J a b rta u s e n d s , d e r M e S a s te in (1851 b e z w . 1864 e n t d e c k t ) . aus d e m 9. J a h r h u n d e r t u n d d ie S ilo e - I n s c h r ift (188 0 a u fg e fû n d e n ) aus d e m E n d e d es 8. J a h r h u n d e r ts .

U eber

d ie

p h ô n iz is c h -k a n a -

n à is c h e S e b r ift des 2. J a b rta u s e n d s v . C b r. w u s s te m a n n ic b ts , J u b u s W e llb a u s e n g la u b t e n oeb im J a b r e 1914 s c h re ib e n z u d ü rfe n : “ V o r ib r e r A n s ie d lu n g in P a lâ s t in a b a t t e n d ie Is r a e lit e n k e in e S e b r ift ; sie ü b e rn a b m e n s ie e r s t v o n d e n K a n a a n it e n , b e i d e n e n s ie ü b r ig e n s a u c b in d e r M it t e d es z w e it e n J a b rta u s e n d s v . C b r. n o e b n ic h t in G e b r a u c b w a r ” . 1 In z w is c b e n b a b e n z a h lr e ic h e P u n d e aus d e n v e r s e b ie d e n s te n G e g e n d e n P a lâ s t in a s u n d S y r ie n s d ie s e A n s c b a u u n g w id e r le g t u n d g e z e ig t , dass d ie a lp b a b e tis c b e S e b r ift ta ts â c h J ic h s e b o n v o r d e r M it t e d es z w e it e n J a b rta u s e n d s b e k a n n t u n d in G e b r a u c b w a r , u n d dass. s ic b d ie g r ie c h is c b -ita lis e h e S e b r ift w o h l n o e b im z w e it e n J a b r ta u s e n d v o n ih r a b z w e ig t e . D a s s d ie g r i e e b i s e b e

A lp b ab etseh rift

aus d e r n o rd s e m i-

tis e b e n (p b ò n iz is e b e n ) K o n s o n a n t e n s c b r ift a b g e le it e t is t , is t e in e a lt e g r ie e b is e b e T r a d i t i o n , 2 d ie b e u te v o lla u f b e s t à t ig t w ir d d u re h d en 1 Israelilische und jüdische Geschichte, 7. Auil., 1914, S. 7S. 2 Vgl. Herod., V, 58 ( Φοινικήια [-γράμματα); Inscriptiones Graecae Antiquis­ simae, ed. R oehx, 497, B 37 s. = D ittexberger. Sylloge3, n. 38, lin. 40. Miscellanea G. Mercati. V I.

1

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

9

Miscellanea Giovanni Mercati. V I.

Hinweis auf die altesten Formen der griechischen Buchstaben, auf den Lautwert der Zeiehen, auf die gleiche Beihenfolge der Buch­ staben im nordsemitischen und im griechischen Alphabet, die. durch deren Verwendung als Zahlzeichen gesicbert ist, endlich auf die ISJamen der griechischen Buchstaben, die die gleichen sind wie die der nordsemitischen, allerdings in aramàischer Aussprache. W a n n die tibernahme des phonizischen Alphabets durch die Griechen erfolgt ist, ist noch nicht endgiiltig geklart; der unterete Ansatz, der heute vertreten wird, nennt das Jahr 720 v. Chr., 3 wàhrend andere Forscher bedeutend weiter zurückgehen. B. L. Ullmann 4 vertritt die Ansicht, dass die Übernahme des Alphabets schon im 11. oder 12. Jhdt. oder noch friiher (bis nm die Mitte des 2. Jahrtausends) erfolgt sei. Ein Yergleich der phonizischen Buchstabenformen mit den altesten griechischen Formen scheint fiir das Vorbild des grie­ chischen Alphabets die Zeit zwischen dem X II. Jahrhundert (Jehimilk-Insehrift) 5 nnd dem IX . Jahrhundert (MeSastein) zu fordern, also das X L oder X . Jahrhundert; « ob aber naher der oberen oder der unteren Grenze, ist bislang nicht zu unterscheiden ». 6 In die gleiche Zeit weist die Priifung der innergriechischen Entwick lung Jdes Alphabets in den altesten griechischen Schriftdenkmàlern. 7 Viel Yerwickelter ist dagegen die Frage nach dem Ursprung und der EntwicMung des phonizischen Alphabets selbst und des Alpha­ bets iiberhaupt, die hier behandelt werden soli. In den letzten z-wei Jahrzehnten haben sich die Funde alphabetisch geschriebener Texte so gemehrt, dass es heute moglich ist, die Problème wenigstens Marer zu -sehen, wenn auch eine endgiiltige Antwort — um dies von vornherein zu sagen — vorlaufig noch nicht gegeben werden kann. Funde, die fiir die Frage der Entstehung der Alphabetschrift von Bedeutung sind, sind im ganzen Bereich des Kiistenstrichs vom Boten Meer bis zum Golf von Issos gemacht worden. Sie zeigen, dass der Gedanke einer Buchstabenschrift in diesem Baum entstanden ist und an einer oder vielleicht fast gleichzeitig an mehreren Stellen dieses Baumes seine erste Verwirklichung fand.

3 R i i ï s C ar pe n te r , The Antiquity of the Greek Alphabet, in Am. Jburn, of Arch., 37 (1933), S-29. 3 How old is the Greek Alphabet!, ebd., 3S (1934), 359-381. 5 Ygl. indes S. 16 die Ausftilirungen iiber das A lter dieser Inschrift. 0 A. R e h m im Handbuch der Archeologie, I. Band, München, 1939, S. 194. 1 Ygl. R e h m , a. a. O., 195-197.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

A . B ea, Die Entstehung des Alphabets

I

3

Teii. - Die Funde

Die bis jetzt gemachten Funde verteilen sich geographisch auf drei Gegenden: die Sinaihalbinsel, Siidwest- und Mittelpalastina und Pbònizien-Syrien. Da die Veroffentlichungen der einzelnen Funde an den verschiedensten Stellen, meist in Zeitschriften oder Grabungsbericbten, erfolgt sind, so dass es aucb fiir den Facbmann scbwer ist, einen Gesamtiiberblick iiber das Fundmaterial zu gewinnen, sei hier zunàchst einmal eine Liste aller in Frage kommenden Schriftdenkmaler gegeben, innerhalb der einzelnen Gruppen moglichst in chronologiseher Ordnung (A). Dann wird in gleicher Keihenfolge zu den einzelnen Funden das Notwendige gesagt; bibliographische Nachweise sollen eine weitere Orientierung ermoglichen (B). A)

L

is t e d e r in

F

S c h r if t d e n k m a l e r 8

rage kom menden

I. Aus der Sinaihalbinsel A lter

Protosinaitische Inschriften

.

.

.

. | 19.-18. Jhrdt. J16?J

II. Aus Süd- und Zentralpalàstina

3. 4. 5.

6. 7. 8. 9. 10

11

. .

12 .

13. 14.

Fragment aus G e ze r........................... “ Plakette ” aus Sichem...................... Fragment aus S ic h e m ...................... Bronzedolch aus Lachis (“ dagger ” ) . Ostrakon aus B eth S e m e g .................. Schale ( “ bowl ” ) . . Ί Scherbe ( “ potsherd” ) aus Lachis. . Krug ( “ ewer ” ). . J Binginschrift aus M egiddo.................. Baucherschale ( “ censerlid” ) aus Lachis Topf von . E l- A g g u l.................. '. . Inschrift aus Tell e l- H e s i.................. B àlû 'a-S tele........................................

17.-16. Jhrdt. I 17. [16?] Jhrdt. [17?] 16.-15. Jhrdt. 15.-14. [13.-12?] Jhrdt. Ì ungefâhr 13. Jhrdt. j (1400-1200) 14.-13(?) Jhrdt. 13. Jhrdt. 13. Jhrdt. 14-1200 (12. Jhrdt.?) 14. -12. Jhrdt. (?)

' Die Liste schliesst sich an die yon J. W. F l ig h t in The Haverford Sympo­ sium on Archaeology and the Bible (1938), S. 123 f. gegebene an (mit den notwendigen Erweiterungen und Anderungen).

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

4

Miscellanea Giovanni Mercati. V I.

I II.

Aus PMnizien Alter

15. 16. 17. 18. 19. 20. 21 .

22 .

“ Halbhieroglypische ” Inschrift aus B y b lo s .................................... Alpbabetiscber Text aus Byblos. Bronzespatel aus Byblos . . . Ahîrâm-Sarkophag 1 aus Byblos . Jehimilk-Ins cbrilt j Speerspitzen-Inschrift aus Bueisse Abibaal-Inscbrift aus Byblos Elibaal -InscbrTft IV.

23. 24.

vor dem 22. Jhrdt. 1500 (1400?) 13. (11?) Jhrdt. 11. (13?) Jhrdt. 10. (12?! Jhrdt. um 1200 (10. Jhrdt.?)

IJ 10. Jhrdt.

Das Ras-Samra-Alphabet

Keilschriftalphabet von Bas Samra Ugaritische Sehrift in BethSemeS . B ) B em erkungen

i

z u d e n e in z e l n e n

16.-14. Jhrdt.

F

unden

I - Die protosinaitischen Inschriften 9 Die erste Entdeckung der protosinaitischen Inschriften liegt vier Jahrzehnte zuriick: im Jahre 1904-05 entdeckte Flinders Petrie in den Minen von Serâbit el-Hâdem eine grosse Anzahl von In ­ schriften, 101meist in âgyptischer Schrift und Sprache; es fanden sich aber auch solche, die nicht âgyptisch abgefasst waren, sondern in bisher unbekannten Schriftzeichen, die sich mit einer gewissen Begelmâssigkeit wiederholten, also auf ein Alphabet schliessen liessen. P Spâtere Expeditionen (1927, 1929, 1930, 1935)12 vermehrten das Material noch um eine Anzahl weiterer Inschriften. Im ganzen zâhlt man jetzt deren 30 (Vr. 345-375; aber Î7r. 369 ist âgyptisch). Die ,J Eine Bibliographie iiber diese Inschriften gibt R. F. B u t in in Harv. Theol. Rev., 25 (1932), 130-132. 10 F l . P etrie , Researches in Sinai, 1906, S. 129-132. 11 11 Inschriften, die spater, in der Inschriftensammlung von A. H. G ardiner und E. P eet , The Inscriptions of Sinai, London, 1917, die Nummern 34Ó-SÓÓ erhielten. Flinders Petrie (a. a. O. S. 131) sagt iiber diese Schriftzeichen : “ I am disposed to see in this one o f the many alphabets which were in use in the Mediterranean lands long before the fixed alphabet selected by the Phoenicians ” . 12 Vgl. Harvard Theol. Rev., 21 (1928), 1-67; 25 (1932), 95-203; F. R. S. Starr and R. F. B u t in , Excavations and Protosinaitic Inscriptions at Serabit el Ehadem, London, 1936 ( = Kirsopp Lake and Silva Lake, Studies and Documents, V I).

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

A.

B ea,

δ

D i e E n t s t e h u n g d es A lp h a b e t s

Bedeutung der Aufschriften war von FI. Petrie nicht klar erkannt worden; erst A. H. Gardiner13 und K. Sethe 14 haben unabhàngig Yoneinander und fast gleichzeitig in den Zeicben die Bucbstaben einer alphabetischen Schrift festgestellt und in diesem « protosinaitisehen » Alphabet den Ausgangspunkt fiir die Entwicklung des phônizischen Alphabets erbliekt. Über die E n t s t e h u n g s z e i t dieser Inschriften herrscht bis heute grosse Meinungsverschiedenheit. PI. Petrie selbst schrieb die Objekte, auf denen sie angebracht sind, der X V III. Dynastie zu (etwa 1500 V. Chr.), wie es noch heute angesehene Eorseher, wie H. Bauer,15 H. Grimme16 u. a. tun. A. H. Gardiner dagegen vertritt den Ursprung der Inschriften aus der Zeit der X II. Dynastie (etwa 20001785 V. Chr.); ihm haben sich in der neueren Zeit zahlreiche andere Forscher angeschlossen,17 besonders aueh deshalb, weil feststeht, dass in der Zeit der X II. Dynastie, genauer unter Amenemhêt I I I (etwa 1849-1801), Semiten (Retenu und ‘A'amu, wohl aus Palâstina und Phonizien) in den Sinaibergwerken arbeiteten. 18 Über die E n t z i f f e r u n g und I n t e r p r e t a t i on der Inschriften ist bis heute keine Einigkeit erzielt. Die Hauptarbeit haben H. Grimme19 und R. Butin20 geleistet; wertvolle Beitrage sind auch von A. E. Cowley,21J. Leibovitch,22 M. Sprengling23 u. a.geliefert worden. W ie versch ied en d ie R e s u lta te sind, m ô g e ein B e isp iel zeigen . D ie w ich tig e, a llerd in g s sch w ierige, In s c h rift 349 w ird v o n B u t in 24 fo lgen d er13

T h e E g y p tia n O r ig in o f the S e m itic A lp h a b e t, in T o w n , o f E g y p t. A rc h ., 3

(1916), 1-1G. 11 D e r U rs p ru tig des A lp h a b ets , in N a c h r. d er K . Geseliseh. d. W issensch. zu G o ttin g e n ( G e s c h iiftl. M i t t . ), 1910, S7-161. 15 D e r U rs p ru n y des A lp h a b e ts (D e r A lte O rie n t, 36. Bd., H eft 1-2, 1937), S. 24

(ohne weitere Begriindung). . ,e A lts in a itis c h e F o rs c h u n g e n , 1937, 100-105 (in Übereinstimmung mit seiner, von den Faehgenossen fast einstimmig abgelehnten, Interpretation der Texte). 17 Vgl. H a r v . T h e o l. R e v ., 25 (1932), 133-135. 1! Vgl. J. H B ueasted, A n c ie n t R e cord s o f E g y p t, I, 1906, nn. 713-73S; R u t i n , in H a rv . T h e o l. R e v ., 25 (1932), 135 f. ,B A lth c b r a is c h e In s c h r if t e n v o m S in a i, 1923; vgl. auch: D ie Liisu n g des S in a iproblem s, die a ltth a m u d ische S c h r ift , 1926; D ie a lts in a itis c h e n B u ch s ta b e n in s ch riften, 1929; A lts in a itis c h e F o rs ch u n g e n , 1937. 20 Vgl. oben S. 4, Anm. 12. 21 T h e S in a itic In s c r ip tio n s , in J o u rn . o f E g y p t. A r c h ., 15 (1929), 200-21S. 22 D ie P e tr ie 's c h e n S in a i-S c h riftd c n l:m < i 1er, in Z e its . d. d eutsch. M o rg . Ges., S4 (1930), 1-14. 23 T h e A lp h a b et.

I t s R is e and D e v e lo p m e n t f r o m

th e S in a i In s c rip tio n s

(O r.

h is t: C o m m ., N° 12), 1931. 24 H a r v . T h e o l. S tu d ., 25 (1932), 167-170.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

6

M is c e lla n e a

Giovanni Mercati. V I .

massen übersetzt (wobei der Übersetzer selbst darauf hinweist, dass es sich nur run einen Versuch handle und die Uebersetzung manche zweifelhafte Elemente enthalte): “ This A-N-T (mine or sleeping shelter) is occu­ pied by the head stonesetter, M-SH. an ‘E-B-I (te) from L-B-ÏT (?) and b y ... A-H (b y ... the brother of) the prince of his tribe, and by B-N (his son)...” . H. Grimme25 übersetzt: “ Ich bin Hatsepsumos, |Verwalter des Erzgesteins und des heiligen Bezirks (von Sinai?), |Schreiber der fronarbeitenden Beute auf Sinai. |Sie hatten (— Man hatte — ) vermutet: Siehe, seine Seele ist verzweifelt, | Da hast du mich gegriffen heraus ans dem Nile (?) und |ich habe mich gestützt auf | jemand, der mir Peind (— Feindin? — ) war ” . Sprengling übersetzt die ersten drei Zeilen wie folgt: *‘ This offering laid down (St) |the foreman of monument-makers from the | goats (Se'irlm\ or -(from the barley), S6'6rlm\ or possibly er Diurnus, in SitzungsT>erichte der Kaiserlich. Alaci d. TFiss. in Wien,· C X V I I (1SS9), p a rt. v ìi, p. 26, N . 2. 21 Codex Fuldensts. Novum Testamentum latine interprete Hieronymo ex manuscripto Victoris Capuani etc., p. 472 in r e f. to p. 99, 22-24. M a rb u rg & L e ip z ig ,

1868.

22 nid.,

p. 4S9 in r e f. to p. 274, 24.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe, The oldest omission signs in Latin manuscripts

11

hd equal hoc deest, hs hoc sequitur.23 He fails to notice, howeveiy that hd and hs change places according to the position of the inser­ tion. The symbols found in Theodulf’s Bible (Paris Lat. 9380), S. Berger interprets thus: hp = haec pone, hd = haec desunt, and sr — sequitur. 242 5 For the editors of the Leon manuscript of the Lex Romana VIsigothorum, hd stands for hic desunt, hs for hie sunt.35 In the oldest manuscript of Pliny’s Natural History (St. Paul in .Carinthia MS X X V , 23), hd and hs are used. K. Dziatzko interprets hs as hic sup­ plendum, “ oder ahnliches ” 26 and rejects the impossible meaning hoc signo given by F. Mone in J. Sillig’s edition of P lin y .27 Mone goes from bad to worse: he reads ho where the manuscript has hd, and is guilty of forcing this nOn-existent form to mean h(oc sign)o. C. M. Franken ventures to interpret JDjR as deleantur, SR as scri­ bantur·, hd also as hic deleantur and hp as hic ponantur. 28 In his short paper entitled: “ On some symbols of Omission in Livian MSS ” 29 F. C. F. Walters gives interesting examples of mis­ understood hd creeping into the text as haud and hs as his; and concludes by saying (p. 162): “ The meaning of the letters can be supplied according to individual fancy. H is a common siglum for hie; d may represent some part of deesse or deficere, and s may re­ present supple ” . How little such signs were understood even by medieval scribes may be seen from the senseless way in which they come to form part of the text. In the Prolegomena (p. 20) to his edition of Auctor ad Herennium30 F. Marx calls attention to suck a corruption of the text. I reproduce it below, Ho. 154. The Visigothic MS of Isidore’s Etymologies (Madrid, Tolet. 15.8, s. viii-ix) has dh in the text and spr, sup, or sp in the margin, which R. Beer in his preface to the facsimile edition explains as deest or desunt hic 33 Corpus scriptorum ecclesiasticorum latinorum, X IX , p a r t I , p. x x v i, V ien n a, 1890. 28 H istoire de la Vulgate, p. 165, P a r is , 1893. See b e lo w on N o . 150. 25 Legis Romanae Wisigothorum fragmenta ex codice palimpsesto sanctae Legionensis ecclesiae, etc. p rooem iu m , p. x x v , M a d rid , 1S96. 28 Untersuchungen iïber ausgewdhlte Kapitel des antilcen Buchwesen, pp. 110 ft'., L e ip z ig , 1900. 27 C. P lin ii Secundi naturalis historiae lib ri X X X V I, v o i. V I , p. 15, i. 24, G oth a, 1855. 28 He Lucani versibus suspectis, in Mnemosyne, n. s. X V I I I (1890), p. 11. 28 Classical Review, X V I I (1903), pp. 161 ff. 30 In certi auctoris de ratione dicendi ad C. Herennium libri IV , L e ip z ig , 1894.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

42

and suppleatur or suppleantur.31 In the celebrated Codex Bobiensis o f the Gospels, known as k (Turin MS G. V II. 15, s. v), hd marks the omissions and hs the insertions. The half-uncial or cursive cha­ racter of these two symbols was misinterpreted by Wordsworth, who read ha for hd and hv for hs. 32 C. Cipolla corrects hv to hs, which he explains as hic sequitur, but retains the incorrect ha, which he ex­ pands to hic addatur.33 Professor P. C. Burkitt in a written commu­ nication (23rd April, 1913) suggested hie desunt for hd, hos sermones for hs and haee verba or vocabula for the presumed hv. The Codex Vercellensis of the Gospels, like the Bobbio Gospels, has hd and hs. Cardinal Gasquet, its latest editor, makes hs = hie supple, and mis­ reads hd for ha, which he explains as hic adde.34 Professor A. 0. Clark devotes a chapter to omission marks.35 His interpretations are: h = hie, hd — hie deest, hs — hie supple, hp = hie pone, hi = hie lege, hm = hie minus, sr = super, dm = dimissum. Por most of these interpretations he has manuscript authority but is misrepresenting the facts when he says (p. 34): “ The usual method is to employ a pair of signs, e. g. hd in the text and dh in the margin (or elsewhere) before the addition. ” As a matter of fact, manuscript usage shows that the customary insertion symbol answering hd is hs and not dh. L. Havet, in his important manual, barely touches the subject. He gives hie pone for hp and hie deest for hd.36 The same interpretation of hd and hp is given in A. Cappelli’s dictionary of abbreviations.37 But it is a curious comment on this work that the earliest instances known to Cappelli date from the 14th century, when a number of examples might be cited that come from pre-Caroline scribes or cor­ rectors. When Cappelli gives hie or hoc supple ov scribas for hs, he is probably quoting Wattenbach, as he gives no date for his example. More recently J. P. Gilson, describing a plate of the fifth century manuscript of Probus in the Urbinas Collection of the Vaticana3 5 4 2 1 31 Codices graeci et

latini photographiée depicti,

Tom .

X III,

p.

x x i,

Ley­

d e n , 1909.

32 Old Latin Biblical Texts, I I , fo lio s 41, 45, 46T, 57, 67, 73, 80, a n d 90 o f p r in t e d text, O x fo r d , 18S6, V a c s , in

C.L.A.

I V . 465.

33 I I codice evangelico k della biblioteca universitaria nazionale di Torino r i­

prodotto in facsim ile, pp. 24-26, T u r in , 1913. 34 Codex Vercellensis iamdudum ab Irico et BiancMno bis editus, etc., in Collectanea Biblica Latina ( I I , pp. 21 a n d 158), notes t o fo li. 526 a n d 351b, K o m e , 1914. 35 The Descent of Manuscripts, p p . 32-52, O x fo r d , 1918. “ Manuel de critique verbale, p. 368, P a r i s , 1911. ’ · Dizionario di abbreviature latine ed italiane2, pp. 15S, 164, 165, M ila n , 1912.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe, The oldest omission signs in Latin manuscripts

Ί3

(see below, No. 19), explained hs in tbe text as Me supple and hd after the insertion in the upper margin as Me deficit.38 The improbability of this interpretation is obvious; for why should hie deficit be put after the insertion? The correct meaning of hd and hs would have become manifest to the editor, had he realized that the symbols change place according to the position of the insertion. Lastly, in liis catalogue of the Codices Eeginenses, the late Dom Wilmart took note of the symbols d and h in one of his late manuscripts. Apparently he was quite convinced that d stood for deest and h for habendum. They are thus registered in iris index.39 I t is clear from the above that considerable divergence of opinion prevails regarding the significance of these symbols. Nowhere is there any indication that one set of symbols preceded another set in time, or that there was such a thing as local usage — with the one notable exception of Professor. W. M. Lindsay in his article on the Laon script. 40 Lindsay not only gives useful statistics but calls attention for the first time to the possibility of determining local usage. He rightly warns against hasty conclusions. My own interest in the subject was first aroused by the unusual methods employed by the Codex Bezae (see below, No. 13). Unusual, that is, in Latin manuscripts. The symbols there used are the anchora superior (re­ sembling an arrow pointing upward) in the text· where the omission occurred, and the anchora inferior (like an arrow pointing downward) placed after the supplied omission in the margin (see plate). Only four other Latin manuscripts are known to me which use this me­ thod. One of these, the Florentine manuscript of Justinian’s Digests, is probably a Byzantine product. The use of these arrow-symbols in a Graeco-Latin manuscript — symbols not ordinarily found in Latin manuscripts — naturally made me suspect that they are of Greek origin. The Greek evidence cited below is based on papyri and ma­ nuscripts that have come to my notice. They will, I think, suffice to indicate the source of the Latin practice. There was no attempt at a thorough investigation of the Greek practice, which must be left to other hands (cf. W. Schubart, Einführung in die Papyruskunde, p. 52 and Pas Buch bei den Griechen und Romern, p. 92, pi. 18). I now proceed to give the manuscript evidence. “ New Palaeographical Society, I I , pi. 18S, L o n d o n , 1930. ai Codices Reginenses Latini, pp. 799 f . , V a t ic a n C it y , 1937.

*° The Laon AZ-Type,

in

Revue des Bibliothèques,

X X IV

(1914), pp. 16-20.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

U

Miscellanea Giovanni M ercati. V I.

EVID E N CE OP TH E M ANUSCRIPTS Nearly all the Latin manuscripts have been examined in the original. The exceptions are marked with a dagger. For reasons of economy the abbreviation stroke over certain symbols is omitted, and no attempt is made to reproduce the exact form of the arrows or anchors, obeloi, and other critical signs. The dotted line stands for supplied omission* large or small. The position of the symbol shows whether it precedes or follows the insertion. Unless otherwise stated, it may be assumed that the in­ sertions are supplied by the scribe or by a contemporary corrector. A serial number is given to each manuscript in order to facilitate reference. In each group the manuscripts are arranged roughly in chronological order. Greek Usage In our oldest Greek papyri, apart from the obeloi, two kinds of omission and insertion signs are used. The anchora superior and inferior (I shall speak of them also as arrows which they get to re­ semble) form one kind, and άνω and κάτω another. These two sets are used separately or jointly. The signs άνω and κάτω indicate that the insertion belongs above or below. It is, therefore, the posi­ tion of the insertion that determines where άνω and κάτω are to stand both in the text and in the margin. The arrow-heads are used pre­ cisely like άνω and κάτω, so that when the insertion is in the upper margin it is marked by the up-pointing arrow, when in the lower margin by the down-pointing arrow, the point in the text being marked each time by the opposing arrow in the margin opposite. 1. L ondon . B e it . M us . P a p .

c v ii

Homer

s. i

b .c.

Omission is marked by f in the text and by f in the margin where the insertion is made. Note that both arrows point in the same direction, which is not the usual practice. Facs. in Palaeographical Society, II, pi. 64. The editors do not state the date of the corrector; presumably he is contemporary with the scribe.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe, The oldest omission signs in Latin manuscripts

2. Ox y b h . P a p . 1793

45

Callimachus

s. i ex.

Omission is marked by άνω in the text. The end of the in­ sertion which is in the upper margin is now lost. There presu­ mably κάτω stood. See B. P. Geeneell and A. S. H unt , The Oxyrhynchus Papyri, XV, 102. Oxford, 1922. 3. B e r l in . P a p . B eeol . P. 6845

Homer

s. i -i i (see plate 1.1) The omission, supplied in the lower margin, has άνω after the insertion. Pacs. in W. Sohubaet, Papyri Graecae Berolinenses, pi. 19c. Bonn, 1911.

4. L ondon . Chestee B e a t t y P apyru s

Xumer.-Heuter.

s. n

Omission is marked by opposing arrows: f in left margin opposite text, answered by j to left of insertion in lower margin. Pacs. in P. G. K enyon , The Chester Beatty Biblical Papyri, pi. V, p. 2, col. 7. London, 1935. 5. L ondon . B e it . M us . P a p . x cvn i

Hyperides

s. π or m

The place where the omission occurs is marked by άνω, the insertion in the upper margin by κάτω. See F. G. K enyon , Greek Papyri in the British Museum (1893) and E. M. T hompson, Introduction to Greek and Latin Palaeography, p. 63. Oxford, 1912. 6. Ox y e h . P a p . 223

Homer

s. in

Omission and insertion are marked by corresponding arrows and by άνω answering κάτω, thus: f in left margin opposite place of omission and κάτω to the right, I in lower margin before insertion and άνω after it.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

46

I f the insertion is in the upper margin, άνω and κάτω change places and the arrow points up in the upper margin and down opposite the text. See Oxyrh. Pap. II, pp. 100, 101,102, 112. Oxford, 1899.

7. V a t ic ,

ge .

1209

Codex Vatic. IB)

s. iv (see plate 1 . 2)

Correctors show great variety in marking omission and in­ sertion. Most of them use άνω and κάτω and the corresponding arrows in the manner illustrated in the previous item. Generally they also use ·/· after the insertion and at the point where the omission occurs. Complete facsimile edition in Codices e Vaticanis selecti, 4 voli. (1899-1904).

8. L ondon . B e it . M us .

Codex Sinaiticus

s. iv (see plate I. 3)

This famous Biblical MS, like the one preceding, abounds in omissions, άνω and κάτω and the corresponding arrows are used in precisely the same way as illustrated above. One corrector who busied himself, chiefly with the Old Testament uses arrows pointing in the same direction. Profuse use is also made of the obelos (·/·)· Complete details and illustrations will be found in J. Η. M. Milne and T. C. Skeat, Scribes and Correctors of the Codex Sinaiticus, 1930. Complete facsimile edition, K. L ake , Codex Sinaiticus Petropolitanus (New Testam. 1911, Old Testam. 1922). 9. M anchestee . J ohn B tla n d s P a p .

i

Deuter.

s. iv

Omission is marked by f in text and I at the insertion in the lower margin. See A. S. H unt , Catalogue of the Greek Papyri in the John Bylands Library, I. Manchester, 1911.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe, The oldest omission signs in Latin manuscripts

47

10. A mherst P a p y r i h , N o. x x iv Demosthenes s. iv (This is a vellum codex) κτ = κάτω stands after insertion in upper margin, .{· pre­ cedes the line where the omltsion occurs. See B . P .

Grenfell

an d A . S.

H unt ,

T h e A m h e rs t P a p y r i, II .

p. 24, pi. Y . L o n d o n , 1901.

11. A mherst P a p y r i N o.

iv

Ascensio Isaiae

s. v

f marks an omission supplied in upper margin, j marks the place in the text where omission occurred. Pa cs. in T h e A m h e r s t P a p y r i , I . p i. I V . L o n d o n , 1900.

L a t in U sage

Corresponding Arrows as Omission Symbols 12. V atic . P a i .,

lat.

1631

Vergil

cap. rustic.

s. iv-v

On fol. 192 there is an omission due to the fact that verses 500 and 501 of Book I X of the Aeneid begin with the same let­ ters. The omission is marked by an up-pointing arrow f in the left margin opposite the text and a down-pointing arrow j be­ fore the insertion in the lower margin. C om p lete facs. in C od ices e V a tic a n is s e le cti, V o l. X I V

(1928).

13. Cambridge U n iv . L ie r . NN. 11.41 Codex Bezae une.

s. v (see plate n ) An omission of two lines due to homoioteleuton in the Greek text as well as the Latin (fol. 59v-60) is marked by an up-point­ ing arrow in the text and a down-pointing arrow after insertion in lower margin. The corrector originally supplied a word too many in both the Greek and the Latin sides, which accounts for traces of the second arrow, especially clear on the Greek side. Only the shaft is visible on the Latin side. C o m p lete facs. in C o d e x B eza e C a n ta b rig ie n s is . C a m b rid ge, 1909; cf. also G . L . A . I I . 140. O x fo r d , 1935.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

48

14. V a t ic . U r b in . L at . 1154

Probus

une.

s. v ex. (see plate ιν. 3-5)

Omissions on fol. 137v, 146v, and 195 are marked by f in text answered by | after the insertion in the lower margin. But the arrows point in the reverse direction on fol. 117 where I in the text is answered by f after the insertion in the upper margin. On fol. 36v an insertion in the left margin is preceded by corresponding to —> in the text, all of which is Greek scribal practice. G.L.A.

P a ce, in

15. N

aples

L

at

I . 117. O x fo r d , 1934.

. 2 (V iN D O b . 1 6 )

P rob u s

q u a r te r u ne.

s. v

An omission on fol. 97 is supplied in the upper margin. The insertion is preceded by an up-pointing arrow f , and a down­ pointing arrow | is placed in the left margin opposite the text where the omission occurred. Same method as in preceding item.

16. F

lo renc e

L

aurenz.

s. n .

lu s t . D ig e s t a

u ne. a n d

s. v i

half-une. (see plate m ) There are numerous omissions. Some are marked by simple signes de renvoi; many by means of corresponding arrows with a point on either side of the arrow shaft. The arrows always point in the opposite direction, as in the previous items. C o m p le te fa c s im ile e d itio n in Digestorum seu Pandectarum Godeκ Florentinus, olim Pisanus. B orn e, 1902-1911. F a cs . o f fo l. 325 in G.L.A. I I I . 295. O x fo r d , 1938.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe, The oldest omission signs in Latin manuscripts

49

hd-hs: SYM BOLS W IT H LO C A TIV E SIG N IFIC AN CE The position of the symbols in the following manuscripts un­ equivocally demonstrates the locative significance of h d and hs. 17.

V erona x m

(1 1 )

H ila r iu s in P s .

une.

s. v (s e e p la t e i v . 1 -2 )

hd hs

in text with ·/· in margin opposite (fol. 103,121v, 193,194, etc.) ·/........ hs in lower margin in text with ·/· in left margin opposite (fol. *96v, 123, 186T, 217, 267v, etc.) 7....... hd in upper margin * See G.L.A. IV. 484.

Ou fol. 221, instead of the usual hs we have hssu, the more obvious abbreviation of h(ie) s(ur)su(m). To call attention to a lacuna at the bottom of col. 1, of fol. 279v, a contemporary cor­ rector wrote in the margin opposite the lines left blank: hie deest quod addendum est. 18.

B ologna U n iv . 701

L a c t a n t iu s

u ne.

s. v 2

hd in text (fol. 34v, 49, *49v, 50v, 60, 82, 232) .... hs in lower margin * The omitted text is Greek and the insertion on fol. 49v is in excellent Greek cursive. in text (fol. 153, 213, etc.) hs in lower margin hs in text (fol. 6V) hd in upper margin hs in text (fol. 230v, 231, 235, 260v) ..... hd in upper margin hd

On fol. 49v hd in text is answered by hd before insertion in left margin, the symbols being used as signes de renvoi. On fol. 244v col. 2 the scribe made two omissions. The one Miscellanea G. M ercati. V I.

4

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V L

50

near the top of the page he supplied in the upper margin and inadvertently wrote hs after the insertion which was later cor­ rected to hd. The second omission is supplied below with the symbol hs after the insertion. The correction of hs to h d is sig­ nificant. 19. Y a t . U r b in .

lat.

1154

Probus

une.

s. v ex.

(see plate iv. 3-5) hd in text (fol. 16, 146v with answering sign cut off) .... hs in lower margin hs in text (fol. *15V, 36) .... hd in upper margin. *Facs. in New Pal. Soc. II. pi. 188. This MS also has omissions and insertions marked by cor­ responding arrows, as described above. See No. 14. 20. F u l d a B o nif . 1

Evangelia

une.

ante a. 546 (see plate v) hd in text (fol. 210v, 235v, 236v, 246v, 308v, 310v, 329v) .... hs in lower margin hd in text (fol. 103v) hs .... in lower margin hs in text (fol. 448) hd .... in upper margin

On fol. 449v an Anglo-Saxon scribe writing uncial marked omission in the typical Anglo-Saxon manner, namely by d in the text and h before insertion in upper margin. He further betrayed his nationality by the Insular form of autem. On fol. 448v a corrector writing Anglo-Saxon minuscule marked omis­ sion by d in the text and h before insertion in lower margin.2 1 21. L

ondon

B. M. H

arley

1775

E v a n g e lia

u n e.

s. v i

(see piate v i) hd in text .... hs in lower margin hs in text .... hd in upper margin See fol. 50v, 61v, 140v, 168v, 326v, 438v.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe, The oldest omission signs in Latin manuscripts 22. V

at

. L

at

. 5758

A u g u s tin u s

u n e.

51

s.

v i- v n

M to the right of text (p. 255) .... hs in lower margin hs in text (p. 22) .... hd in upper m a r g in On p. 12, apparently in a contemporary uncial, dh in text is answered by hds after insertion in lower margin. This has a Visigothic flavor. See below, Wo. 142. 23. L y o n s 426 (352)

August.

une. and half-unc.

s.

v i -v ii

hd in text .... hs in lower margin hd in text hs .... in lower margin hs in text .... hd in upper margin See fol.

9 V, 2 2 v, 26 , 3 0 v, 3 8 v, 4 4 v, 4 6 v, 47, 4 9 , 6 0 v, 67, 7 2 v,

74, 75, 8 4 v, 85, 88, 8 8 v, 1 0 0 v, 1 3 2 v, 1 3 9 , 143, 147, 1 4 7 v, 1 6 7 v, 176, 1 7 6 v, 177, 1 7 7 v, 178.

24. L

yons

hd

607 (52 3 b is )

A u g u s tin u s

h a lf-u n c .

s. v u

in t e x t

hs ... in lower margin (fol. 1, *34v) hs in text hd ... .. in upper margin (fol. *69v) * See faes. in E . A . L o w e , Codices Lugdunenses Antiquis­ simi, p. 43, pi. X X V II. Lyons, 1924. The promise there made to treat the subject more fully is now being fulfilled. 25. V atig . L a t . 5763

Isidorus

cursive min.

s. vm

hd in text hs in lower margin hs in text (fol. 33) hd .... in upper margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V L

52

hd-hs: LO C ATIV E SIG N IFIC AN C E POSSIBLE B U T N O T C E R T A IN I next proceed to enumerate cases in which hd in the text is an­ swered by hs after (or before) the insertion in the lower margin. It so happens that most omissions are supplied in the lower margin, so that the scribe seems to have used the symbols correctly although he may have lost all sense of the original locative significance. The following cases do not, therefore, contradict the locative interpre­ tation of hd and hs, but neither do they confirm it. Gradually hs shifted its place. In the oldest MSS it appears always after the inser­ tion; in later MSS it is often before the insertion. 26. V er c elli B i b l .

cap . s .

n.

Evangelia

une.

s. rv (ca. a. 371)

[hd] in text (pag. 619, 526, 599, 609) ....; hs in lower margin 27. T u r in G.

15

v ii

Codex Bobiensis

une.

s. iv-v

h d i n t e x t ( f o l . 41, 4 6 v, 80, 8 7 ) .....

hs in lo w e r m a r g in

C o m p lete fa csim ile: I l v e r s ita r ia N a z io n a le

G od ìce E v a n g e lic o

d i T o r in o ,

T u rin

k d e lla

1913;

also

B ib lio te c a G .L .A .

U n i­

I V . 465,

(O x fo r d , 1946). 28.

Ch u r a n d S t . G a l l 1 39 4 e tc .

E v a n g e lia

u n e.

s. v

hd in text (fol. 1) ....... hs. in lower margin 29.

V

a t ic

. L

at

. 5750

Schol. Bob.

in C ie e r.

u n e.

s. v

h d in t e x t (p . 2 6 ) ..... hs in lo w e r m a r g in

30. St . P a u l

in

Ca r inth ia x x v 2.36

Pliny

une.

s. v ex.

hd in text (p. 112) .... hs in lower margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe, The oldest omission signs in Latin manuscripts

31. V er o na

x iv

(12)

Hilarius

une.

53

s. y ex.

M in text (fol. 12, 120, 258, *274v) he ..... in lower margin * Facs. in C h âtelain, U n c ia lis

32. V erona

lv

(53)

Scriptura,

p i. I X , P a ris 1901.

Didascalia Apostol.

half-unc.

s. v ex.

hd in text (fol. 70 v, 83 v, 97) hs .... in lower margin 33. P aris L a t . 8907

Hilarius

une.

s. ν-νι

HD in text (fol. 187v) HS ... .. in lower margin hd in text (fol. 16, 54, *59, 61) hs .... in lower or lateral margin * F a cs . in

Ch

â t e l a in

,

Uncialis Scriptura,

34. V a t ic . B asilicanus D. 182

Hilarius

p i. X I .

half-unc.

a. 509-10

hd in text (fol. 149) .... hs in lower margin hd in text (fol. *294v) hs ... .. in lower margin * C o m p lete facs. in Collegium codicibus rescriptis evulgandis: Cor­ pus Extravagantium codicum. C o n legit e t e d id it J. L . P e r u g i, R o m e , 1922; also in N e w Pal. S o c . I I , pl. 187, w h e re hd is e x p a n d e d as “ h ic d ee st ” , a n d hp (m is rea d in g fo r hs) as “ h ic p o n e ” .

35. V e r o n a x x x v m (36)

Sulpicius Severus

half-unc.

a. 517

hd in text (fol. 45) hs ..... in right margin 36. P aris L a t . 2235

Hieronymus

une.

s. vi in.

hd in text (fol. 10v, 31, 92v) .... hs in lower margin T h e in s e r t io n is o c c a s io n a lly p r e c e d e d b y E

(=

r e tr o ? ).

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

54

37. P a r is L a t . 8084

Prudentius

cap. rustie.

antea. 527

HD (in rustic cap.) in text, at end of verse (fol. 12v) hie ..... H S (in rustic cap.) in lower margin An important witness for the meaning of h before an insertion. 38. W hd

o le e n b ü tt e l

36. 23 Aug.2° (2403)

une.

s. viin .

in text (fol. 99v) hs .... in lower margin

39. V e r o na x x x v n (35) hd

Agrimen.

Clemens, Recognitiones

s. vi

half-une.

in text (fol. 37v, 103, *120v, 124, 151v, 208, 311v, 312v, 313) ..... hs in lower margin

* A propos of this MS, Châtelain, Uncialis Scriptura, p. 133, incorrectly observes that the normal position of hs is before the insertion and not after. 40. L yo n s 478 (408) hd

Augustinus

une.

in text (fol. 45v, 47v, 48, 51, 200v) .... hs in lower margin by une. and half-unc. hand, s. v m

41. Cam b r id g e C orpus Christi C old. 286 hd

s. v i

Evang.

une.

s. v i2

in text (fol. 58v) hs .... in lower margin

On fol. 151, 159v the insertions are preceded by R with a slanting stroke through the letter, standing probably for retro. (Of. above, No. 36). An Anglo-Saxon corrector, s. vm , marks omissions by h in text and d before lower insertion (fol. 121,154).4 2 42. M

il a n

H . 78 su p . a n d T

u r in

G .V . 15

A m b ros.

h a lf-u n c .

hd in te x t (fol. 6 4 v, 8 3 v) ..... hs in low er m argin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

s. v i

E. A. L owe, The oldest omission signs in Latin manuscripts

43. St . Ga l l 722

Hilarius

half-unc.

55

s. v i

hd in text (fol. 171) .... hs in lower margin An omission on fol. 161 lias hs after lower insertion; the hd in the text seems deleted. 44. P aris L a t . 12214

Augustinus

half-unc.

s. v i2

hd in text (fol. 44v) hs .... in lower margin This is the exception; normally hd stands in text, and hd before and after insertion in lower margin (156v, 195, 200, 207, 245, 255v); or before and after upper insertion (70v); or merely before upper (233) or after upper (185v). On fol. 145v hd in text is matched by hd in left margin with another hd after insertion in lower margin. Here hd has practically become a mere signe de renvoi. 45. V er o na

χ χ π

(20)

Hieronymus

half-unc.

post a. 555

hd in text (fol. 93) .... hs in lower margin ·/. in text (fol. 50) d ..... in lower margin What d stands for here is not clear. 46. V a t . L a t . 1322

Synod. Chalcedon.

half-unc.

s. v i

hd in text (fol. 34)* hs .... in lower margin * Pacs. in E. Carusi-W. M. L indsay , Monumenti Paleografici Ve­ ronesi, I , pi. 16. B orn e, 1929. 47. M onte Cassino 150

Ambrosiaster

half-unc.

ante a. 570

hd in text .... hs in lower margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

56

Miscellanea Giovanni M ercati. Y I.

48. B am ber g B.

iv

21

Hieron.-Gennadius

half-mu.

S. V I

hd in text (fol. 125v) ..... hs in lower margin 49. R ome V ttt.E m an . Sessor. 55 (2099) August, half-une.

s. Vi ex.

hd in text (fol. 41v, 55v, 59) .... hs in lower margin 50. P akis L a t . 9533

Augustinus

half-une.

s

V I-V II

hd in text (fol. 15v) hs .... hs in lower margin B y a different hand from that which supplied most of the omissions. See below under «Symptoms », No. 143. 51. V a t . L a t . 5750

Concilia

s. vu

half-unc.

hd in text (p. 43, 130) .... hs in lower margin 52. V a t . P a l . L a t . 210

August.

une. and half-unc.

s . vi-vri

hd in text (fol. 262v, 263) .... hs in lower margin hd in text (fol. 192) hp .... hs in lower margin hd in text (fol. 173v, 223v) hp .... . in lower margin 53. V erona

l ix

(57)

Vigilius Thapsensis

half-unc.

i.

V I-V II

hd in text (fol. 9V) hs in lower margin5 4 54. Cologne 212

Canones

half-unc. and une.

s. vu in.

hd in te x t (fol. 10) ..... hs in low er m argin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe, The oldest omission signs in Latin manuscripts

55. L e o n Ca th ed r al 15

Brev. Alarici, etc.

une.

37

a. 546-669

M in text (fol. *91 et passim) .... hs in lower margin * Facs. in C. U. Clark , Collecta,ne,a Hispanica, pi. 3, Paris, 1920. 56. B oulogne 37 (32)

Ambrosius

une.

s. νπ

hd in text (foil. 69v, 71, 144, 181v) .... hs in lower margin 57. E scorial Camarxn d e las K e liq u ia s

A u gu st,

une.

s. v n

half-une.

s. vn

hd in text (fol. 136v) .... hs in lower margin 58. L yons 468 (397) and P aris N. A. 602

Hieron.

hd in text (fol. 47v of Lyons MS and fol. 57v of Paris MS) hs .... in lower margin 59. L yo n s 604 (521) and P aris N.A. 1594 August, half-unc.

s. vn

hd in text (fol. 14v, 63) .... hs in lower margin hd in text (fol. 49) hd ... .. in lower margin A later hand marks omission on fol. 41, 41v by Me minus habet. 60. M onte Cassino 271

Augustinus

une.

s. vn

une.

s. vn

hd in text (now faded) (p. 155) .... hs in lower margin 61. O xford B odl . A uct . D. 2.14

Evangelia

hd in text (fol. 96) .... hs in lower margin

B u t on fol. 8 dh in text; h ..... sd in lower m argin. The sym-

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

58

Miscellanea Giovanni M ercati. V I.

bol sd is a remarkable departure from the normal. The h before the insertion is by a later hand. Other omission marks, d an­ swering d, are by an Anglo-Saxon hand, s. xi. 62. P aris L a t . 1625

Origenes

une.

s. vu

An omission supplied in the exemplar was copied blindly into the text along with hd (before) and hs (after; the insertion (fol. 48). 63. V

at

.

O ttob . L

at

. 319

A u g u s tin u s

u n e.

s. v ii

hd (or hs *) in text (fol. X V I I v, X V III) .... hs in lower margin * Curious misuse of hs, probably a mere slip. 64 . V

ercelli

B

ib

. Ca p .

c l v iii

C le m e n s

u n e.

s. v n

hd in text (fol. 64) .... hs in lower margin 65. X ew Y

ork

M organ L ib r . M. 334

Augustinus

une.

a. 669

hd in text (fol. *38v, 53v) hs .... in lower margin * On same page omission is also marked by signes de renvoi 66. M il a n I. 61 sup.

Evangelia

* Irish majuscule

s. v n 8

hd in text (5V, 33v, 36, 38v, 50v, 58v) .... hs in lower margin The insertion is sometimes preceded by h (60v), sometimes also followed by it (36, 58v). * Written however in Italy, doubtless at Bobbio. 67. L

yons

602 (619)

Hieronymus

une. and half-une.

s. v n e x .

hd in text (fol. 21) ..... hs in lower margin in min. s. vrn

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe, The oldest omission signs in Latin manuscripts

59

M in text (fol. 15) hs .... in lower margin in min. s. v m ■68. P a m s L a t . 17226

Evangelia

une.

s.

v n ex.

hd in text (fol. 44v) [h]s .... in lower margin 69. P a m s L a t . 6400 G Isidorus (foil. 112v-146) hd in text (fol. 115 v, 139) hs .... in lower margin

une. and half-une.

s. vn-vni

70. P a m s L a t . 11641

une. and half-une.

s. vn-vm

August.

hd in text (fol. 13v) hs .... in lower margin 71. I vmoa 1 (I)

Gregorius

Luxeuil min.

s.

v n -v m

hd in text .... hs in lower margin (fol. 65) òr hs .... in lower margin (fol. 42v) See also fol. 80v. 72. L ondon B.M. A d d . 11878

Gregorius

Luxeuil min.

s. vm in.

hd in text (fol. 36v, 55v, 67v, 68v) hs .... in lower margin 73. St . P a u l i n Ca b in t h ia x x v . 2.36

Hieron.

Luxeuil min.

s. vm

hd in text (pag. 13, 20) hs .... in lower margin 74. P a m s N.A. L a t . 1063

Nov. Testam.

Cursive min.

s. v r a

hd in text (fol. 117v) hs ... .. in lower margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

60

Miscellanea Giovanni M ercati. V I.

75. M

odena

A r c h . C a p . 0 . 1 . 12 C o lle c t io c a n o n .

u n e.

s. v m

hd in text (fol. 21v) hs in lower margin 76. P a m s L a t . 6400 G (foli. 146v-193)

Isidorns, De Off.

une.

s. v m in.

hd in text (fol. 150) hs in lower margin 77. V

erona

x l ii

(4 0 )

G r e g o r iu s

h a lf -u ne.

s. v m

hd in text (fol. 8) .... hs in lower margin

OMISSION IN V ISIG O TH IC M AN USCRIPTS Nearly all extant Visigothic manuscripts have been examined, some to be sure cursorily, and my notes on their omission marks may not be complete. The usage recorded below, however, may safely be taken as typical. A glance at the sixty odd items shows that a fairly uniform method existed among Visigothic scribes and that the distinctive feature was the use of dh in correspondance with SR or its variants, SUR, SUP, SP, SPR (occasionally written out as supra or super), and SRS, SRSS (standing for ‘ sursum ’ or ‘ sursus ’ ). Obviously SR and its variants or synonyms have locative signifi­ cance. Item 137 proves it beyond all doubt, for in this manuscript SR changes places according to the position of the insertion, pre­ cisely as does hs in the oldest Italian manuscripts. (See pp. 49-51). It is worth noting moreover that Spanish scribes follow ancient Italian tradition not only in keeping alive the locative sense of the insertion mark, but in being the only ones to preserve its post-posi­ tive character. I f the sense of SR is identical with that of hs, one is tempted to conclude that dh is the Spanish equivalent of hd and stands for ‘ deorsum hic ’. This may have been the original mean­ ing; but if it was, it did not last. Por we find dh in the lower in­ sertion where ‘ deorsum ’ is out of place. The same may be said of Ih and h i. Originally the barred i-longa must have meant ‘ infra ’. In item 81 the word is actually written out, and furthermore i-longa

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe , The oldest omission signs in Latin manuscripts

61

is placed after an insertion in the upper margin, precisely where hd appears in the oldest Italian manuscripts. Other examples, however, prove that Ih is also used as a mere signe de renvoi, quite regardless of any locative sense. The meaning of RS (cf. Kos. 92, 93, 130, 135 twice) is puzzling at first; but the use of RS as a variant of RS (Ko. 93) suggests ‘ retrorsum ’, another w a y of saying ‘ sursum ’ . 78.

E scobia I ì R . n . 18

Is id o r u s

une.

s. v n

ex.

dh in t e x t (f o l. 2 0 v)

.... SRP (surely error for SPR) in lower margin

79.

A

tjtun

27

C o m m e n t, in G e n e s .

s. v m (s e e p la t e v u . 1 )

db

80 .

M

in text (fol. 6 3 v) .... SUR in lower margin (SUR below supplement)

a d b id

Bib . X

ac

.

T ol . 15. 8 (H h . 3 )

Is id o r u s

s. v m - i x

d h in t e x t (p a s s im ) ...... s p r

(or

sp, sup, su p r) in lo w e r m a r g in

There is great variety in this MS. The above is typical. C o m p lete facs. in G od ices G r a e c i et L a t i n i p h o to g ra p h ié e d e p ic ti. Y o l. X I I I , p. x x i, L e y d e n 1909.

81.

M onte Cassino 1 9

Augustinus

s. vm -ix

This MS. is a star witness for the correctness of the inter­ pretations h — ‘ hie I = ‘ infra ’, and SRSS = ‘ sursus ’, an­ other form of ‘ sursum ’. in text (p. 72) .... SR in lower margin dh in text and Infra dh in margin opposite to the left of text (p. .174). ·/· h'e ···■· SuR in lower margin dh in text (replaced by a later signe de renvoi) h ic .... srss in lower margin dh

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

62

An insertion in the upper margin of p. 72 has -f- over the first word followed by a barred i-longa, now erased, which doubt­ less stood for ‘ infra 82. A

lb i

29

Isidorus, Augustinus

Carol, and Yisig.

s.

v m - ix

dh in text (fol. 34v) .... sur in lower margin 83. E scorial T. Π. 25

Isidorus

a. 795-843

dh in text (foil. 25, 75v) .... SR in lower margin 84. P

a r is

L at. 4667

L e g e s V is ig .

a. 828

Ih in text (foil. 18v, 4 l v, 93v, 109v, 131v) Ih ... .. in lower or lateral margin, the symbols apparently used as mere signes de renvoi. 85. E scorial e. I. 33

Concilia et Decret.

s. ix in.

ih in text (fol. 44v) Ih ... .. in l o w e r m a r g in See preceding item. 86. P aris L a t . 609

Computi

a. 778-815

dh in text (fol. 56) dh .... sr in lower margin. The symbols dh apparently reduced to mere signes de renvoi. 87. P aris L a t . 5387 dh in text (fol. 72) dh ... .. in lower margin.

Vitae PP.

s. ix

See preceding items.

88. P aris 3ST.A. L a t . 2170. Part I.

Cassiani Collât.

dh in text (quire to fol. I I v) s .... in lower margin dh in text — .... in lower margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

s. ix

E. A . L owe, The oldest omission signs in Latin manuscripts

89. P aris L at . 10877

Regula S. Fructuosi

03 s. ix

dh in text· (fol. i vj .... SR in lower margin 90. P

a r is

L

at

dh in text .... SR dh in text .... SP

. 12254.

Gregor.

M.

s. i x

(fol. 10) in lower margin (fol. 211) in lower margin

Both insertions are in ordinary minuscule, 91.

P

a r is

L iâ t .

2994 A (foil. 73v-194)

Isidores

s. i x

dh in text (foil. 94, 124v) ..... SR in lower margin ./. in text (fol. 75v) .... supra in margin The word ‘ supra ’ written out below the insertion on fol. 75v supplies us with the meaning of SR in other parts. 92. P aris N.A. L a t . 2167

Gregor. M.

s. ix

dh in text (fol. 3V, 4V, 9, 52v, 100, 120v, 125v) .... SR in lower margin dh in text (fol. 63, 102) RS .... SuR in lower margin dh in text (fol. 109v) Rs .... SuR in lateral margin ·/. dh in text (fol. 82v, 93v) without any corresponding sign in margin. RS throws light on RS: the expansion ‘ retrorsum ’ makes sense and accounts for the suprascript o over R. 93. P aris L a t . 10876

Reg. S. Isidori

s. ix

Text and omission sign erased (fol. 2) .... SR in lower margin dh in t e x t

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

64

.... RS in lower margin RS may be a slip for SR or it may stand for ‘ retrorsum 94.

P

a e is

L

at

8093 (foil. 1-38)

.

Sedulius, Dracontius

s. ix

Rufinus-Eusebius

s. ix

dh in text (fol. 4V) .... SR in lower margin 95. L e o n . Ca t h ed r a l 15

This is an important witness for the meaning of sr, spr, sp after the insertion in lower margin. I t is exceptional in using the P before h. Ih in text (fol. 82v) Ph .... super (written out) in lower margin Facsimile in C. IT. Cl a r k , Collectanea Hispanica, pi. 2. The P preceding the h may be a monogram of PL. But the seeming monogram is more likely P with an ornamental foot, in which case it could signify 1pone ’, a synonym for j‘ retro ’ (cf. bio. 93). The end of the insertion has super written out, which most Yisigothic scribes abbreviate by means of sur, sr, spr, etc. 96. M

a d r id

B i b . ÎT a c . T

o l.

dh in text dh ..... in lower margin. 97. M

a d r id

A

cad

. H

is t

Bible

32

s. i x

Apparently mere signes de renvoi.

. 44 (2 )

T a io

s. i x

dh in text (fol. 53 etc.) sr ..... in lower margin t 98. L

eyden

dh

Y

oss.

F

ol.

Ill

(c x i)

A u s o n iu s

in t e x t

.... SR in lower margin Omission also marked by “ hie minus habet ” .

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

s. i x

E. A. L owe , The oldest omission signs in Latin manuscripts

99.

L

yons

443 (372) only in part Yisig.

65

Origenes

in text (fol. 27) ..... SUR in lower margin dh in text (fol. 69, 82) answered by dh in lower margin. de renvoi.

s. is

dh

Apparently mere signes

Omission also marked by “ hie minus habet ” . See No. 143. 100. E

s c o r ia i,

dh

P . I. 8

Is id o r u s

s. i x

in text (fol. 258v) “ h ie le o d e s in d u s m in u s" h a b e t ” in m a r g in

Another omission on fol. 230 is marked in margin by “ hie minus habet I I I capitula ” . 101. E scoriai, & I. 14

Isidorus

dh in text (fol. 103) .... SP in lo w e r m a r g in 102. E

s c o r ia i,

dh

103.

S. I . 17

(w it h

PR

above

s. ix

SR)

S e n te n tia e P a t r u m

s. r x

in text (fol. 12) .... SR in lower margin

E scoriai, S. I . 16

A u g u s tin u s

s. i x

ex.

dh in t e x t (fo il. 4 1 v, 9 7 v, 163, 207, 208, 2 0 8 v, 222, 2 23 )

.... SR in lower margin 104. L ondon B.M. A d d . 30852

Orationale Gothicum

·/. in text .... SRS in lower margin SRS c a n h a r d ly s t a n d f o r sus ’ . 105. L

eon

Ih

a n y t h in g b u t ‘ s u rs u m

s. ix ex.

’

or

1s u r ­

s. i x

ex.

O f. a b o v e N o . 81. C a t h e d r a l 22 (2 )

E u g e n iu s T o le t .

in text (foil. 50, 52v, 55v) .... SR in lower margin

Miscellanea G. M ercati. V I.

5

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

66

106. L ondon B.M. A d d . 30854

Gregorius

s. ix ex.

Ih or ·/■ in text Ih or 7· in lower margin, the symbols becoming mere si­ gnes de renvoi.

107. P aeis "N.A. L a t . 2168

Gregor. M.

s. ix ex.

Ih in t e x t (fol. 25v) h i ...... in lower margin

108. E scobial P. I. 7

Isidores

a. 848-912

dh in text (foil. 10 9V, 126, 262v)

.... SR in lower margin Regula 8. Benedicti

109. E scobial & I II . 13

s. ιχ-χ

An omission on fol. 30v is marked by “ hie minus habet ” . 110. E

e. I . 12

s c o b ia l

C o n c ilia

s. i x - x

C h ry s o s to m u s

s. i x - x

dh in t e x t dh .... SR in lower margin 111. M

a d b id

A

cad

.H

is t

. 27

dh in text (foil. 19v, 31v) dh ... .. sr in lower margin

112. M a d b id B i b . N ac . T ol . 10.25 (now 10.007, olim Hh 68)

Valerius, Vitae PP.

a. 902

dh in text (foil. 5, 6, 41v)

.... SR in lo wer margin 113. B abcelona R ip o l l . 49

Taio

a. 911

dh in text (fol. 84v) dh ... .. in lower margin, the symbols becoming mere signes de renvoi.

114. M anchesteb J ohn R yla n d s 83 (93) Greg. M.

a. 914 (S e e p la t e v h . 2)

dh in text (foil. 29v, 38 et passim) dh .... SR or SuR in lower margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe , The oldest omission signs in Latin manuscripts

67

dh in text (fol. 274v) dh .... SuR in lower margin 115.

M ad r id B i b . N ac . T ol . 15.12

a. 915

Is id o r u s

(now 10.067, olim Hh. 5) dh in text (fol. 22) .... SR in lower margin dh Θ in text (fol. 14v) Θ .... SR hoc in lower margin dh in text (fol. 15v, 65) dh ... .. in lower margin 116.

M ad r id B i b . K A c . T (n o w

ol.

14.22

D r a c o n tiu s

s. x in .

1 0 .0 2 9 )

dh in text dh .... SPA in lower margin (with R below the P) in text (or other arbitrary signs) .... SPA in lower margin (with R below the P) The monogram at the end of the insertion obviously stands for ‘ supra ’ . 117. M

a d r id

dh

B ib . N

ac

. T

ol.

Hh

58

B e a tu s

s. x in .

in t e x t

.... SR in lower margin 118. M a d r i d A

dh

cad.

H

is t .

20

B ib le

s. x in .

in t e x t (fo l. 2 2 3 )

.... SR in lower margin dh in text (fol. 194) dh .... SR in lower margin 119. L

eon

C a t h e d r a l 6 (2 )

B ib le

a. 920

dh in text (fol. 47v) .... SR in lower margin Ih in t e x t Ih .... SR in lower margin Ih used here as simple signe de renvoi.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

68

120. F ew Y ork M organ L ib r a r y M. 644

Beatus

a. 922 (see plate vm . 1)

dh in text dh .... SR in lower margin Exactly the same symbols are used to mark as many as 20 omissions occurring on folios 36v, 41v, 45v, 47, 48v, 49, 73, 90v, 102v, 106v, 110, 116v, 127v, 128, 130, 131, 165, 172, 192. 121. M ad r id A cad . H ist . 1007 H

Patristica

a. 932

dh in text (foil. 37v, 42v, 44, 54) .... SR ./. in text (fol. 13v) ./....... Suit in lower margin 1 122. U rged B ib l . C ap . dh in text .... SR 123.

Beatus

a. 938

in lower margin

M a d r id B ib . H ac . T

od.

1 1 .3

G r e g o r iu s

a. 945

(now Vitr. 14.2) dh in text dh .... SR (or sr) in lower margin I f a second omission occurs on the same page, an arbitrary sign is used in the text and SR follows the insertion. 124. M a d r id A cad . H ist . 25

Isidorus

a. 946

dh in text dh SR in lower margin 125. M ad r id B ib . H ac . T ol . 15.16 (now 10.041)

Concilia

a. 948

dh in text (fol. 20v, 21) dh (intercolumnar insertion) fl2 6 . M anchester J ohn R y b an d s 89 (99) Cassiod.

a. 949 (see plate vm . 2)

hi in text (fol. 140) ·/....... SR in lower margin (SR below supplement)

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A. L owe , The oldest omission signs in Latin manuscripts

127. T oledo Ca p . 2.2

Bible

69

s. x

db in text .... SR in lower margin 128. P aris N.A. L a t . 2179

Yitae SS.

s. x

in text (fol. 181) ~ SR in lower margin 129. M ad rid A cad . H ist . 33

Beatus

s. x

dh in text (fol. 176v) .... SR in lower margin 130. E scoriar a. I. 13

Valerius in Psalm.

s. x

dh in text .... RS (in monogram, standing for ‘ retrorsum ’ or else a slip for SR) in lower margin 131. E scoriar T. II. 24

Isidores

s. x

dh in text dh in lower margin dh used as signe de renvoi 132. P aris N.A. L a t . 2170. Part I I

Cassian

s. x

lb in text; the corresponding sign to mark the insertion now obliterated. 133. P aris N.A. L a t . 2180

Yitae SS.

ante a. 992

dh in text (fol, 135v) .... SR in lower margin 134. E scoriar & II. 5

Beatus

s. x ex.

dh in text (foil. 37, 39, 70v) .... SR in lower margin

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni Mercati. V I.

70

135. L

ondon

B.M. A dd . 30853

Sermones & Homiliae

s. x-xi

RS in text RS in lower margin The symbols seem mere signes de renvoi. ‘ Retrorsum ’ makes sense before the insertion but not in the text. 136. P aris ST.A. LAT. 2178

Vitae SS.

s. x-xi

dh in text (fol. 40, 60, 199) .... SR in lower margin dh in text (fol. 101v, 274v) .... SRS in lower margin Here SR and SRS perform the same function. The latter stands for ‘ sursum’ or ‘ sursus’ , the former for ‘ super’ or ‘ supra ’ . Both indicate location. 137. E scoriai. & I. 3

Isidores

a. 1047

SR in text (col. 2, line 7, fol. 216v) dh ... .. in upper margin dh in text .... SR in lower margin (foil. 98v, 101v, 123v, 126v, 207v, 236 v) dh in text (fol: 193v) dh .... din in lower margin df in text (foil. 61v, 141v,156v, 171v, 200, 238v) Note how the position of SR depends on the position of the insertion. The locative sense of the symbol is clear. The exact expansion of din on fol. 193v and of df must remain conjectural. An interpretation like ‘ deest inser ’ and ‘ deficit ’ would con­ stitute a distinct departure from the locative sense normally given by Visigothic symbols, dh answering dh on fol. 193v seems a mere signe de renvoi. 138. P aris N.A. L a t . 2171

Liber Comicus

a. 1067

X

dh in text (p. 148) X dh .... SR in lower margin dh in text (p. 280) .... SR in lower margin in red

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

B. A. L owe, The oldest omission signs in Latin manuscripts

139. P a r is IT.A. L a t . 2169

Isidorus

71

a. 1072

dh in text (fol. 30) .... in lateral margin 140. L o n d o n B.M. A d d . 30851

Liturgica

s. x i

dh in text .... Ih in lower margin Ih in text answered by Ih in margin OMISSION M A R K S AS V IS IG O TH IC “ SYMPTOMS » There is no clue too insignificant for the palaeographer. By pay­ ing attention to such trifling details as the omission marks used in a manuscript he may be enabled to draw specific conclusions regard­ ing the home of the exemplar or archetype, or gain some new light regarding his manuscript’s later vicissitudes. W e have seen above (Nos. 20 and 41) that two of the famous sixth century vulgate ma­ nuscripts — Fulda Bonif. 1 and Cambridge C. C. C. 286 — had omis­ sions marked in the normal Italian manner, but the omission marks employed by the correctors of a century and a half later were quite different and betrayed Insular practice. I t so happens that the nationality of the new correctors is amply proven by the script and abbreviations. But had these been absent, the new omission marks (d answered by h and h answered by d) would have been quite enough to betray the hand of the Insular reader. The following examples are submitted because they all either give a hint of the influence at work in the centre which produced the manuscript, or furnish some clue as to its probable ancestor. They are confined to manuscripts betraying Yisigothic “ symptoms 141. P a r is L a t . 9533

Augustinus in Ps.

half-une.

s. v i- v n

The home of this MB is unknown. The abbreviation of bus (b') and the manner of marking omitted m {— ) suggest Spain. The symbol dh used by the original scribe confirms the im­ pression. I t is, to my knowledge, the oldest purely 'Visigothic omission sign. E.g.:

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. VI.

72

dh in text (fol. 142 v, 204, 166) .... ds dh in text (fol. 211v, 212v) .... des The symbols ds and des probably stand for 4desuper 142. V a t . L a t . 5758

Augustinus in Ps.

une.

s. vi-vn

This MS comes from Bobbio and is certainly of Italian origin. An omission on p. 12, supplied apparently by a con­ temporary hand, has dh in the text and hds after insertion in lower margin. These marks are in the same style as in the pre­ ceding item. Since h = hie, ds is the main insertion mark and should mean 4desuper ’ . A t least that makes sense here. Note that both the Vatican and Paris MSS contain the same author and even the same work. This may have some bearing on the transmission of this text. Châtelain mentions hds in a MS of Lucretius. See above p. 4. 143. L y o n s 443 (372)

Origenes

half-unc.

s. vn

The minuscule corrector of the half uncial manuscript of Origen, Lyons 443 (372), s. vn — part of it is Paris N. A. Lat. 1591 — was evidently acquainted with Visigothic practice, for an omission supplied by him on fol. 27 ends with SUR, and other omissions on folio 69 and 69v are marked by dh in the text answered by dh before the marginal insertion. These are familiar Visigothic symbols. See No. 99. Note that part of the manuscript (foil. 7 - llv, 77 1. 9 to bottom) is actually in Visigothic script contemporary with the ordinary minuscule. Cf. E. A. L o w e , Codices Lugdunenses Antiquissimi, p. 37, Lyon 1924. 144. L

ondon

B. M. B u r n e y 340

Origenes

une.

s. vu

ds in text (fol. 5) dh ... .. sur in lower margin dr in text (fol. 21v) s .... in lower margin The manuscript comes from Corbie and was doubtless writ­ ten in a French centre, yet the manner of marking the omission

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

1 - Pap. Berci. P. 0845. See No. 3.

τ

a

ο y e γ Α ργτογτ^ ^ Τ Η Μ ^ γ τ ρ γ ΊΓΟΊΠ6 0 , t \ c i \ è y r w r j ò f η m *»h Ko y c e n n o x y t A y n h« m e r'i λ γ τ ά > kt À i n f p τ ~ ό γ ^ ^ * . N I H AM TCU Μ ί C Α ' Γ ρ Τ · / c i e a e c θ a i À y - r o μ n o >; T e d i a Vj A p e c e V ê i h o i m ' r o y c i i i x u j r a c i a g i rwû. o » r a c ΐ λ έ γ ό ρ * r ò A O f «a μ M A O iC K À ir iG j’C A ic 'n y ni  A l O f l C M O Μ K A I C T A Γ Μ M A M O K A G I A é y C C T M rM j o y / v ê » r i A ^ A A A A ^ A I 'ΓΟ^ - T e 6 R A C I A G y C € I l l f I i 15' H Γ Α r c j M TC \M A A A/I M \ M I I V _ " G Al G R A A O Al t I C T O Al A AK K O M T ( Û N A G O W TU) Al KAl C I Π G A fO K A C IA ty CTUJ A A W iH X Ô o d c o y c o c y A AT-J-ey^GI c GM A G A Ê X " ' A y T b c e i i t a g »T À i c t - i ^ H M G r * G A A I A I e»OAI KTÀIG

,r.

J, ^ ty^c«en*YÂ^>|ir«>iJiz.^u.·^ N· y i r t A À r - ^ i i y r - 'H y

2 - Vat. Gr. 1209, i>. 1222, col. I. See No. 7.

π » ί γ ^ χ ι t t > t T | A r t ·» Γ

Μ O H t l J ,C XK fO fiY*e K fcf*?v jaa) e n t i f o ' ,v n i o r ic jn r p M h cV ^ 1 Mi l f f H C F N Κ λΙ C H i ^ α Χ ίΌ Ν Η

i

.Ip ';

K K ifN Q N ^ C T H N « M

Ì

w

* ip v X c ^ e ^

V K H if a iK n u E o c s o ^ ^ J > \ p f& ir c * ?

k m

* * > 1^

6 | ia t< P y w if^ o G f^ > A p i è l N Al A C T » f ? * r A tO IVAN £ H OJ &AJ ’ R A f* J A&Mf>| 9 J H X IYM 3 Yt^lY^ ;

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

t' »§ >e-l S.'%- I ιI *"**· V* *11 iL 1 fc V I-t, f pun Ji «*.* g am Λ Vv V^ Im iken κι VqI Vu5»t τ4 c^u 6 ep.ucn c-o ^

C ,1

p u m p ì r e n ò e o s is ty m v is p u e p ix T e M e h t

ir

T u p b A e c x c i i cum a a i xb u. ep tit φ χτ e ^ n « , ' t N b upMjoa.V«?s i ^ u u m i i s x cju i l i χ ικ ιπ π ι a n

e

Ff

r e c n b u c m A v A i c o e p x e x T s i / ep A A u rY cjxu A

VI1

u e t é c i s*se b i c e i u pi u b i c u u n c u m μ οχλ(

il

fr e o n »O K ie g e ò b tiu p L b x e c A C T io p e c e M m

P‘ CT

■I? .,:

*

nrt t

y ç u m i p i P A M u m i v s l O M C s e x ii e l

m a

t axofi «.

l’x u l u s Î i B p i ' M O K i o b e e i x n o A t N c b i c r t i c T x p p . x t f ^ T e p u A e c N i m i u T i Ι ι τ χ τ i S A e i r i c n v i : ι ο ί K» u o e

\

I m b ic iu a u ie p 5 x tu p tp i λ m u s L15p.o tu Ct-KS tan oTepxtoAcle Λ i U*

i t A U U S M t l AP’e O t 'O M .î j î M S i i t l A i n t s c )U lb u «V > λλ'λ

t t i e p p e g p u w ^ m M v n e u o i p e c e p i j aux

c i

^ v a e u i e e e p t i v i t A c i \ s i i p s t u i ι α ι ο Ν ά ^ 1^ * H U c l i KIOÎM Î A H v'v'pei p e r e «V u O 1 i O A O U ) U ο ύ

\MS«?pu le b p c û u a u t c Uw

Γ*

pxcx 11cv?t>y4tt **v

et

S e n t i i e b p i > t x M O N > ' C I t i o i i s l o e u s c j u i peci pi a çu eux u m m o M u clipeo eiai Meet i Acn »;o P p Λ u> cvvel il a ·>e o cj u e w o m isi eex i xcnsepti Ubp*

n n A t

ce

t « Φ Cm .i x t u p e s s e x j î s c p u u V m x I m ï u u i e p u s e o p t i s p ACT a p t i o >c o t ! 1 p e s i u n e x p u s p e e e p u ; c u i r » p e e n s v x ’ M c *k i i n i t e l l e v i i u p. n i a c t * s t p w i

« *X CM

recnpusvitôepiu upLel^ciipecisye ; y tx x u

> ^ ]i«? a 7 i> > n o M iF u i

■Ί«.It

« -j « t

(

/

py r ^ t r * f■“" v

» ex

p.

s

«»LLeo[-'fpfKuciw tfjvôlCXV» ■^^r* crei ' um e h ^ m « r r .n i «

ÆJ 1-2 - Flor. Laurent. Cod. Justin., ff. 177, 308.

gee No. 16.

T

av. III

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

r r A F O N I A Fo A O H C hC

A C I I I . I ^ U Ì ì M U A v*

\ ò l 'x< u ' C t y M ì > t n \ Wviiii.i XIII (111. f. 1

,

C o l s . I. 11.

V ofiifjT~-': ·**

ç J

t i r f -

.& y a i

t la iH m ifJ i,

tjU A iiì ifn ^ ^ irtu o c ip -x iS w M ^ ^ iM iv ffn ^ ' ix/UTfJÙfyì >

^ι~· >nn

>

f i é n *T T u T f ‘t r x h ο π ο Ι έ ΰ

fi^ 4 i

( γ * f J >*f fL*y *u ( x ^ & Y

~ —

M i> iiil| v iA q iitÌu im iN ^ M in iA le n v 5 ig t; u

ia

|u » * o ( 11

r ii| u | i- ‘ ‘ , I 1t

- r

r U H ìu j^ i iA t ^ - 1 ir n îm 11 ο |Ι| γϊ>ι Ι # μι

Μ I s i 1p O R p .X c I O K ! O S i V>A.

L x ’ >

,N üN ilc r ie N •‘' R i c i r i P ^ i U ' o l 110^ Ι ολ

h. J

O rescissesi q u u io o it"

H O C C Λ ϋ ΐ Ι β Ο Ν Γ ’ λ ' - i l '4t* L ^ i o p . e s c u o r i *! ? S ’ T S e £ *

β, ΈΑ"

tv i s KJUIO |>X Ï ï U lì H i 1 i l

K i J

11 W il' iil.nl I

1-2 - Fulda lion if. 1, ff. 44S, 233. See No. 20. T

av

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

.

V

c*~rc V c e j t l «

ί

U Ç

o

SUlÊ p(ΊΟΚ

»

'ICR RA C M J O I K C S \p K 'N i I ) Λ(Ι >S νΙ,Ο OlONIS

ì

•.Λ .

. .. M P I

1 4 -

C a m b r. C orp. C h rlstl C o ll. 197, p. 203.

T

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

av

. V III

E. A . L owe, The oldest omission signs in Latin manuscripts

73

on fol. 5 smacks of Spain. A contemporary half-uncial hand wrote ds in the text and d h .... sur in the lower margin. And an omission in fol. 21v is marked by dr in the text and s before insertion in the lower margin. There can be no doubt of the locative significance of most of the symbols. 145. Chartres 40 (2)

Gregorius

une.

s. vm

The manuscript comes from Elenry where evidently A'isigothie practice was known, for on fol. 37 an omission is marked by dh in the text and by SR after the insertion in the lower margin — pure Visigothic practice. On fol. 28 v a contemporary minuscule hand has dh in text and h p .... ss in the lower margin. But on foil. 1 and 52v we find dh in text and h p ....rs. On fol. 30v an omission is supplied apparently by first hand thus: dh in text and hp ... RS in lower margin. But an omission inserted on the same page in the upper margin has hp before and ri after it. The same hp — ri enclose an insertion in the upper margin of fol. 43. From the above it seems clear that SR (or RS) and ri have locative significance, s must stand for “ supra” and i for “ infra R probably stands for “ require ” . 146. C a m b r id g e C o r p u s Ch r is t i C o l l . 304

Juvencus

une.

s. v m

An omission on fol. 20 is marked by hd in text answered by dh before insertion in lower margin written in early mi­ nuscule, s. IX in. The MS has other Yisigothic symptoms. * It later fell into English hands and the omission marks made by the Anglo Saxon corrector are d in the text and h in lower margin — pure Anglo Saxon practice. * Cf. G.L.A. II. 127. 147. M il a n A mbros. B. 159sup.

Gregor. M.

une.

s. v m med.

The MS. was written at Bobbio. On fol. 170v an almost contemporary corrector uses the Visigothic omission signs dh in text and SR following the insertion in the lower margin, manifestly suggesting Spanish influence in North Italy.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni M ercati. V I.

74

148. M o n t p e l l ie r . B i b l . d e

la

V il l e 3

Evang.

b type

s. v m 2

This liturgical Gospel book from the abbey of St. Guillemle-Desert, or Gellone near Montpellier, could not have been written there, for the abbey was founded in 804 and the MS was written fully a generation before in Northeast France. Its minuscule resembles that of Autun 20 and Vatic. Regin. Lat. 316. On fol. 67v an omission is marked by dh in the text and answered by dh before lower insertion and SR after it. Similarly on fol. 135 dr in text is answered by SR after inser­ tion in lower margin. The insertions are in late ninth century Caroline minuscule, yet the symbols must be due to Visigothic influence. 149. P a r is L a t . 11710

Canones

Carol, min.

a. 804-05

The script is not Visigothic, but the MS has Visigothic symp­ toms (aum for ‘ autem ’ ). I t also marks omission by dh in text and SR in lower margin. f 150. P a r is L a t . 9380

Bible

Carol, min. s. ix in.

This is the famous Theodulph Bible. Everyone knows how greatly it depends on Visigothic models. According to Berger (Histoire de la Vulgate, p. 165) two variants are placed between hp and sr and the margin has hd. The description is rather vague, but it is clear that omission marks are used. The use of SR suggests Spain at once. For Berger hp signifies “ haec pone ” , sr “ sequitur ” , and hd “ haec deest ” . I find all three inacceptable. 151. P a r is L a t . 1796

Fulgentius, etc.

Carol, and Visig.

s. ix

The Visigothic minuscule is seen on foil. 77-232. Omission even in the Caroline portion is marked by dh in the text and sr after the insertion in the lower margin. The centre which produced the MS was manifestly following Visigothic practice. 152. V e r c e l l i co n

Isidorus, Etymol.

pre-Carol, min.

s. i x in.

On fol. 84, 84v, the omission symbols dh and HR occur. An omission on fol. 86 is supplied in the lower margin with HR

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe , The oldest omission signs in Latin manuscripts

75

before and after the insertion. This has a distinctly Visigothic flavor. HR probably signifies “ hie retro ” . 153. P aris L a t . 11709

Coll. Canon. H isp..

Car. min.

s. ix

On fol. 24v an omission is marked by dh in text and SR after lower insertion. On fol. 26v it is marked by X in text and X .... RS in lower margin (possibly a slip tor SR or else signifying “ require supra ” or even “ retrorsum ” ). One sus­ pects Yisigothic influence. This is confirmed by certain Visi­ gothic abbreviations (nsi for nostri, etc.) and by a few addi­ tions in a pure Visigothic hand (fol. 198v, 199, 201v, 209). Pacs. in my article “ Nugae Palaeographicae ” in Persecution and Liberty, Essays in Honor of George Lincoln Burr, pp. 65f. and plate. New York, 1931. 154. P a r is L a t . 7714

Auctor ad Herennium

Carol, min.

s. ix

Misunderstood omission signs occasionally found their way into the text. Thus on fol. 52v of this MS we read: Item quem senatus damnarit, quern .hd. omnium exi­ stimatio damnarit, eum vos sententiis vestris absolutis, hp populus damnarit quem SR. It is clear that an ancestor of this MS had omitted three words due to homoioteleuton. The omission was marked in the text by hd and the insertion in the lower margin stood between hp and SR. The final symbol clearly suggests a Visi­ gothic exemplar. 155. P a r is L a t . 1796 Fulgentius (foil. 97-232 v) dh in text (foil. 118, 190v) .... SR in lower margin

Carol, and Visig. min.

s. i x

The MS is not Visigothic throughout. It is in the non-Visi­ gothic portion that the omission occurs on fol. 190v. The in­ sertion likewise is in non-Visigothic script. The corrector, how­ ever, followed Visigothic traditions in this as in the use of nsm (instead of nrm) for * nostrum ’.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

76

Miscellanea Giovanni M ercati. V I.

OMISSION SYMBOLS IN IN S U L A E M ANUSCEIPTS In their manner of marking omission Insular scribes vary con­ siderably. The method most characteristic of them is: d in the text answered by h in the margin, be it upper, lower, or lateral.

See pi. vni.

I t is the Anglo-Saxon rather than the Irish scribes wüo show a predilection for this method. I t doubtless dates back to the seventh century. For we find this pair of symbols in the restoration of the Book of Kings in the famous Plautus palimpsest in the portion where the restoring hand is Irish majuscule of the seventh century.41 By the end of the seventh century, if not earlier, this method was also known in England, for we find it in the Lindisfarne Gospels written about A. D. 700.42 But whereas English scribes in the homeland as well as on the Continent make constant use of d and h in the man­ ner described above, Irish scribes on the whole use signes de renvoi, such as obeloi, crosses, asterisks, and similar combinations of lines and dots. W. M. Lindsay, who was the first to call attention to Insular usage, gave a small number of examples in illustration. 43 These can now be multiplied many times. But further evidence seems unne­ cessary here, since anyone can verify the facts by consulting the volumes of G.L.A., especially the first three, where many Insular manuscripts are described. Two interesting facts emerge from a consideration of Insular practice: 1) the Insular symbols show no trace whatever of locative significance; 44 2) found in non-Insular

41 Eacs. in C.L.A. II I . 344 b. 45 Ibid., II , 187. 43 See his article on Laon A Z script mentioned above, p. 8. 44 I t is significant that in one o f the finest and most calligraphic products o f England o f about A. D. 700, when the specificaUy Insular symbols were already known, an entirely novel set o f signs are employed. I refer to the Codex Amiatinus, where we find hR in the text answered by hp or hd in the lower margin; hp in the text and hR in the margin. I t would be interesting to know the source and the meaning o f these symbols. That the scribe also knew Italian practice is evidenced by his use o f hd in the text and hp in the lower margin (foil. 436T, 466T). On fol. 44 he has h in text and barred d (now erased) in the margin which is only a variation o f the more usual Anglo-Saxon symbols.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe, The oldest omission signs in Latin manuscripts

77

manuscripts, they may reasonably be taken to be Insular symp­ toms, especially in the pre-Caroline period (see p. 71). In manuscripts showing other Insular symptoms the presence of d and h as omission marks is definite confirmation of Insular influence of one sort or another. M E A N IN G OP TH E SYMBOLS hd and hs are the only symbols in the oldest Latin manuscripts of Italian origin whose meaning is beyond dispute. The only symbols in Spanish manuscripts are SR (and its variants sur, sup, spr) and, for a time, Ih. In all of these combinations h stood for the adverb ‘ hie ’ (see Nos. 37 and 81). 45 Its object was to pull the reader up, to issue a warning. The accompanying d stood for ‘ deorsum ’ (κάτω of the Greek papyri) and s stood for ‘ sursum ’ (άνω). The h takes the place of the obelos or arrow in the Greek. I t may be illustrated thus: hd = ■/· or κάτω hs = or I άνω

t

To my knowledge ‘ deorsum ’ is never found written out. Nor for that matter ‘ sursum ’. But of ‘ sursum ’ there are at least ab­ breviated forms which are unmistakable (see No. 17). Also, in the Visigothic, where the ancient locative sense of the symbols is at first preserved, we find instances o f.abbreviated ‘ sursum ’ or ‘ sur­ sus ’ (see Nos. 81, 104, 136). Visigothic scribes, moreover, may be said to confirm the * deorsum ’ interpretation of d, for they use Ih (‘ infra hie ’) where the Italian has hd, and occasionally ‘ infra ’ oc­ curs written out (see No. 81). Visigothic Ih must originally have meant ‘ iusum hie ’ 46 or better, ‘ infra hie ’ to contrast with the opposite ‘ supra ’ or ‘ super ’ which is also occasionally written out (see Nos. 91, 95, and the monogram in 116). One cannot be certain what precise meaning was attached to hd and hs, dh, Ih, d, and h, once the locative sense was lost. A reason­ 45 II is an ancient abbreviation o f “ hie See L. Traube, Neues Ardivo, X X V II (1901), p. 271. I t is not infrequently written out: “ hie deest ” , “ hie minus e s t” , “ hic d im itte” , etc., where there is trouble in the text. “ In an 11th century manuscript o f Vergil letters are used like neums to guide the Cantor: s= “ sursum ” and 1= “ iusum ” = “ deorsum See Studi Medievali, V (1932), p. 73.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

78

Miscellanea Giovanni M ercati. V I.

able guess for h d is * hie deest ’, since the words occasionally appear written out and as far back as the fifth century (see No. 17). I refrain from guessing what hs meant in later manuscripts. Nor can one be certain what R signifies before an insertion. I have come across two sixth century instances, both probably Itab'an (Nos. 36 and 41). Normally R stands for ‘ require ’, but here, if we consider that the locative significance of these symbols was still alive, the expansion ‘ retro ’ or ‘ retrorsum ’ seems reasonable, especially as it is sup-

__ _

·

©

ported by Visigothic usage. "We find R S and R S in several Visigothic manuscripts (Nos. 92, 93, 130, 1351, where ‘ retrorsum’ is logical, since it accompanies insertion marks meaning ‘ supra ’ or ‘ sursum ’. The phrase ‘ sursum versum retroque ’ occurs in Cicero (Partitiones Orat. 24). I t may be crediting our medieval scribes with too much learning to expect them to know Cicero, but tradition has a way of fingering on. H R is used interchangeably with hp by the scribe of the Codex Amiatinus (see p. 41, n. 44), whether in the locative sense or not, it is hard to say. But if H R = H P , then P may stand for the adverb ‘ pone ’, the synonym of ‘ retro ’. B y the eighth century, however, to some scribes at least, hp stood for ‘ hic ponas ’ (see above, p. 40, n. 20). The sense of the symbols des, ds, and dr is far from clear. I t should be borne in mind, however, that they are exceptional and that all my instances are from manuscripts under Visigothic influence (Nos. 141, 142, 144). This circumstance makes their locative significance quite probable. ‘ desuper ’ makes sense in No. 141 but not in 144 where ‘ deorsum ’ is the logical expansion both for ds and dr. We must leave the problem unsolved till further evidence turns up.

CONCLUSIONS The tangible results of the above investigation may be summa­ rized as follows: 1) Greek manuscripts were the models for the earliest Latin usage. Early Graeco-Latin manuscripts show the Greek method. 2) The symbols hd and hs originally had locative significance, and the original position of the insertion mark, be it hs or hd like άνω and κάτω, was after the insertion in the margin and not before.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

E. A . L owe, The oldest omission signs in Latin manuscripts

79

3) The symbols hd and hs are the usual symbols in onr oldest Latin manuscripts. 4) W ith time, the old symbols are modified or altered, and en­ tirely new ones arise. The original locative meaning is lost in all countries but Spain. 5) In Spain the locative sense of the original symbols survives, but the symbols themselves appear in new guise. 6) Insular and Spanish scribes develop distinctive omission signs. Visigothic or Insular influence may therefore be suspected wherever these distinctive omission signs occur in non-Visigothic or non-Insular manuscripts.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

LESLIE WEBBER JONES

PR ICKING SYSTEMS IN N E W YORK M ANUSCRIPTS

Ancient scribes knew the value of ruled lines as guides for their unsteady hands. A t first, prompted by aesthetic considerations, scribes ruled their lines à la pointe sèche ; in the twelfth century and later they used lead and finally ink. Whatever the nature of the ruling, it is obvious that two points are required for the proper placing o f each line. Now the most desirable sort of point is that which can be made on more than one page at a time and. is at once visible and invisible—in other words, a simple pricking made by some sharp instrument.1 Such prickings Avere universally adopted in mediaeval manuscripts to direct the ruling of the horizontal text-lines and of the vertical lines that bound the text on each side. The monks Avho made these little punctures unwittingly gave valuable clues to palaeographers. In the first place, the positions of the punctures on a particular page vary someAvhat from century to century, from country to country, from scriptorium to scrip­ torium, and occasionally from scribe to scribe. A survey of these changes, therefore, enables one to draw a number of conclusions concerning the date and provenience of manuscripts— conclusions based on objective criteria. Thus, it is obvious that the occurrence of the ‘ outer-marginal ’ system2 in a very early manuscript indi­

1 On the nature o f the Instrument—a punotorium (stylus or awl) ; a circinus (compass or dividers) ; a thin, narrow knife (fo r making slits rather than round punctures) ; a complete frame; a ' comb ’ ; or a wheel—I hope to have more to say later. 2 For a definition o f this term and o f the others which follow, as well as a plate showing the various positions o f pricking on a page and a complete account

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. W. Jones, Pricking systems in New Y ork manuscripts

81

cates a date later than 450 A . D. Again, prickings in both margins or on the outer bounding lines in both inner and outer margins of books with non-Insular script are proof of Insular influence. Prick­ ings in the inner margin alone, ceteris paribus, may in themselves indicate this same influence, since the prickings in the outer margin are often trimmed off by the binder. The employment of the ‘ insidetext ’ system, moreover, is the mark of a date before 800 A . D. The appearance of intercolumnar prickings as late as the ninth and tenth century is a sign of Yisigothic origin. The placing of punc­ tures for the text-lines and for the vertical bounding lines as well at or near the extreme edge in the various margins indicates a date at least as late as the tenth century. There are, finally, a number of such interesting phenomena as the simultaneous use at the end of the eighth century in St. Gall of Insular prickings in all the Winithar-manuscripts and of the 'outer-marginal’ system in all manuscripts written exclusively by other scribes. In the second place, there is considerable variation in the number of folios or bifolia, which are pricked in a single operation and in the arrangement of these folios or bifolia before, during, and after the pricking. Seven distinct systems of arrangement have thus far been determined— systems which have proved of use in dating manuscripts of Tours,8 Beauvais (?), a Franco-Saxoii center (or *3

o f the survey on which the present conclusions are based, see L. W. Jones, Where Are the Prickingsl Transactions of the American Philological Association (1944).— Since the making o f the survey I have noted only the following additional phenomena which are worth recording. In Morgan S19 (900-930 A. D.) the prickings for the vertical bounding-lines vary from positions considerably below the upper horizontal text-line and above the lower horizontal text-line to positions on these two lines, beyond them, or at the extreme edge o f the leaf. A t this time most prickings are at or near the edge o f the le a f; exceptions to this rule occur in Morgan 755 (East Franconia, saec. x), which has the pricking on the upper and lower text-lines, and Morgan 748 (Greek, saec. x), whose prickings vary from positions on the text-lines to positions well out in the margins. In two mss., Morgan 1 (late saec. ix ) and New York Public Library 1 (saec. x), the prickings for the text-lines are placed on an extra outer bounding-line (along 0-0) ; the po­ sition is normal, though the extra line is unusual. 3 On Tours mss. (which employ only six systems) see E. K. R and, Traces de piqûres dans quelques manuscrits du haut moyen âge, Académie des Inscriptions et Belles-Lettres, Comptes Rendus, 1939, pp. 420-429; E. K. R and, Prickings in a Manuscript of Orléans, Transactions o f the American Philological Association, 70 (1939), pp. 329-337; and L. W. J ones, P in Pricks at the Morgan Library, ibid., pp. 318-326. On all the other centers cited in this sentence see L. W. Jones, Pin Pricks, etc. Miscellanea G. M ercati. V I.

6

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

82

M is c e lla n e a

Giovanni M ercati.

V I.

Aachen), Luxeuil, Reims (or Liège?), St. Bertin’s at St. Omer, Corvey in East Franconia, St. Gall, Constantinople (for Greek manuscripts), Ireland (?), and northern Spain, and which may eventually prove of use in placing manuscripts as well. To the data on Morgan manuscripts which was published at the beginning of W orld W ar I I 4 I now propose to add similar data on 23 New York manuscripts, many of which were until recent months unavailable for study for the simple reason that they had been taken to bomb-proof shelters in the country. The new group includes 13 early manuscripts from the Morgan Library, one from the New York Academy of Medicine, two from the New York Public Library, and seven from the Plimpton Collection at the Columbia University Library. I have also inspected 15 fragments in the Plimpton Col­ lection and have set down the information which they provide. The palaeographer first seeks to discover, by noting individual variations in the prickings, whether one or more Mfolia have been pricked at a time. In theory this problem is difficult; in practice, easy. Prickings are large or small, round or oblong (often mere slits); when arranged in what is intended to be a vertical column, some of them regularly deviate and form a definite pattern.5 The palaeographer then attempts to find out how the folios or Hfolia of each gathering were arranged before, during, and after the pricking. In his search he has as his guides the seven systems of arrangement already mentioned. Since the nature of the systems is somewhat difficult to comprehend, I shall not only describe them as carefully as possible but I shall also attach diagrams. Among the manuscripts in which the first three systems are found are the earliest products of Tours. For system 1 all the lifolia, of a quire are set in position one above another and folded : 6 prickings (P ) are then made in a single set of operations straight through all the leaves of a quire— the uppermost (fol. 1) first and

1 L. W. J ones, P in Pricks, etc. 6 There is a not uncommon situation in Morgan 1 : the eleventh pricking from the bottom in the outside vertical line o f prickings is double (i. e., there are two prickings close together) on both fol. 211r and 2181' (the first and last folios o f quire 30, which is pricked in accordance with system 2, described below) and on no other folios o f the quire. An inspection reveals that 2l l r was without doubt pricked directly over 218r. e In the accompanying diagram the folded hifolia are represented by 1-8, 2-7, 3-6, and 4-5.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. W . Jones, Pricking systems in New Y o rk manuscripts

the under leaves (2, 3, 4, 5, 6, 7, 8) next. illustrates the procedure :

83

The diagram below

_ _ _ _ _ _ 1 P . ______ ____ 2 .

.

______

3

.

. . ______

4

. . . ______ 5 . . ______ 6 . _ _ _ _ _ 7 _______1_____ 8

y

I f the quire be turned upside down, the prickings w ill be made first on folio 8T and then on V , 6T, 5T, 4T, 3T, 2T, and 1T. A t Tours this method is apparently used down into the seventh century and perhaps through it and is revived (or evolved independently again) about 825 À. D. System 2 introduces a brilliant invention by some eighth-century scribe : after one of the four bifolia of a gathering (1-8, e. g.) is folded, one set of prickings is made through its two leaves (1. and 8) ; the bifolmm is then unfolded, laid out over the remaining three Infolia, and all four are ruled together. The diag­ ram below illustrates both pricking and ruling (R ) : 1

P

8V

8

f

6V 5V

1R 1 Y Y

ir 2Γ 3* 4Γ

V-

System 3 is a refinement of system 2 : when the four Ijifolia are outspread, they are all pricked a second time through the half of the upper bifolmm which, after the bifolia have been folded, becomes folio 1 recto. A diagram follows : 1

Ρ

8

Y

Ρ ? SV 7V fiv Y

■ I BI Y . Y

5V _ -

.

1Γ Ρ ?,Γ 3Γ 4Γ Y

When this system is used, the prickings of only one half of the quire (folios l r, 2r, 3r, and 4r) are established and those on folio 1

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

84

Miscellanea Giovanni M ercati. V I.

recto reinforced, though sometimes those of the other half (folios 8\ T , 6T, and 5T) may be established by reprickings through what later becomes folio 8 verso. Apparently the repricking is done to fix the prickings definitely throughout the quire in case the ruling lines do not show well and have to be redrawn. System 3 is exceed­ ingly rare in manuscripts of Tours written in the eighth century. About the beginning of the ninth century two more systems begin to appear at Tours and elsewhere. In the first, system 4, any two Mfolia (è. g., 1-8 and 3-6) are selected ; each is folded! separately and its two single leaves pricked one above the other from the top ( l r on 8T, and 3r on 6T) ; each is then unfolded and spread over an­ other M f olirmi, and the two are ruled together. The procedure is as follows : •ο

_____ l

p

_____ 8

Y

____

3

„ P

_____ 6

Y

IB I 8v ____ I . 1 ____i r 7V_____ . ______ 2T IB I 6 v _ _ X . 1____3r 5V____ - . _____4r

This system, it should be noted, occurs during the period in which leaves are normally ruled two at a time and are arranged in accordance with what is known as Old Style.7 System 5 is some­ what different : two Mfolia (e. g., 1-8 and 3-6) are placed ope above the other, folded together, and pricked from the top while folded ( l r on 3r on 6r on 8r) ; each of the two Mfolia is then unfolded and spread over another Mfolium and ruled against it. 1 P 3 6 8 7 1r

1E 1 Y Y 8V 7V ---- . —

_lr 2r

1E 1

6V 5V

Y

Y

3r 4τ

7 On methods o f ruling in general see E. K. R and, A Survey of the Manu­ scripts o f Tours, Studies in the Script of Tours, I (Cambridge, Mass. : The Me­ diaeval Academy o f America, 1929), pp. 11-18; and L. W. J ones, The Script of Cologne from Hildehald to Hermann (Cambridge, Mass. : The Mediaeval Academy o f America, 1932), pp. 8-10.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. W . Jones, Pricking systems in New Y ork manuscripts

85

This time-saving system is uncommon. In the early ninth cen­ tury at Tours the usual methods are systems 3 and 4. A b ou t 825 A . D ., when system 1 reappears a t Tours, as noted in the second paragraph above, s till another system— 6— is used in several m anuscripts : the prick in gs a re m ade as in system 1, but several quires are pricked togeth er, one below another. N o special d iagram is necessary to illu stra te this procedure. Perhaps the p ric k ­ ings Avere m ade Avith a double-fram ed instrum ent, sharp points being set in one o f the tAvo fram es, but th is is m ere conjecture.

In system 7, not encountered at all at Tours, the prickings are applied from the top (8T- l r) to four 1>if olia Avhich have been out­ spread one above another. P

Y

8V____ _ . _____ l r qv_____ _______ 2r 6V_____ . _____ 3r 5V__ __ . _____ 4r

P

Y

The use of system 7 obviously consumes more time than the use of systems 1, 2, 3, and 6, but it is especially suited to 4 (2) Old Style ruling. Before I describe my findings I must speak a word of caution. Three éditions de luxe/ one édition de grand lux'e,® and six other books10 have their leaves ruled separately. N oav it is obvious that manuscripts Avith leaves so ruled have no opportunity for using five of the seven methods described. Systems 2, 3, and 7 may normally be used only when four leaves are to be ruled at a time ;8 9

8 Éditions de luxe and éditions de grand luxe normally have their leaves ruled separately. The leaves are so ruled in Morgan SIS (Gospels, s. x), Morgan 647 (a Lectionary o f the Gospels, in Greek; s. xii), and Morgan 692 (also a Leetionary o f the Gospels, in Greek; oa. 1200 A.D.), all o f which have fine script and illumination. Morgan 748 (Gospels, in Greek; s. x), however, which is also an édition de luxe in script and illumination, is ruled 4 (2) O. S. (outside or inside) at times; at other times each bifolium is folded and its two folios are ruled together from the top (on the hair side) in the manner employed in Morgan 655. 9 New York Public Library 1 (Evangelistary, s. x), which has excellent min­ iatures and a number o f pages written in gold letters on a purple ground. M or­ gan 1 (s. x), however, which has illuminations on purple, is ruled 4 (occasionally-*!) O. S. (outside). 18 Morgan 897 (Greek), 764, 765, and 714 (Greek); and Plim pton 264-950(2) and 220-1100 ( 68).

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

86

Miscellanea Giovanni M ercati. V I.

system 4 and 5 only when two leaves are to be ruled at a time. System 1 and its variant, 6, however, are the sole systems which may be used for any type of ruling—one, two, three, or four l·if olia at a time, or even such irregular rulings as the Luxeuil type (two or three or four single leaves at a time). One must bear these facts in mind in considering thè data which I am about to present. Even though my information comes from books diverse in date and provenience, it does not necessarily indicate the normal situa­ tion in every scriptorium. I present below a detailed account of my studies.11 In this report, as in my previous reports, I have noted no instances whatever of system 3; since this system, as I have indicated, is rare even at Tours, no argument is to be adduced from its absence. I have, however, discovered a new method of pricking, which I shall call system 8 : any two S folta of a gathering.are placed together, folded,*4

11 Though the following fifteen fragments (now a part o f the Plimpton Col­ lection at thé Columbia University Library) consist in each case o f too few leaves -to reveal their 'system s’ o f prickings, they do in some instances reveal other interesting facts which have to do with prickings: s. v iii: Ms. 54. 1 fol., Commen­ tum, in lot) (written in Anglo-Saxon minuscule, probably in England; the trim­ ming o f the margins has removed the prickings; the ruling is on the hair side). Ms. 129, 1 fol., I sidore, Synonyma I 65-08 (written in Anglo-Saxon minuscule, pos­ sibly in or near Würzburg).—s. ix : Ms. 27, 1 fol., 3 cols., Joshua ΧΧ Ι-Χ Χ Π Ι (Visigothic minuscule; the ruling is on the hair side, now blank; the only prickings —a vertical line o f them—appear between columns 2 and 3 o f the script; the appearance o f this ancient usage as late as the ninth century is a sign o f the conservatism o f Visigothic practice). Ms. 53, 2 fols., Gregory the Great, Homiliae in Ezechielem (written in Beneventan script). Ms. -}9A, J erome, Tractatus de Psalmis (X C III). Ms. 55, 2 fols., a Lectionary plus parts o f two o f Bede’s ser­ mons (Germany). Ms. 57, 1 fol., F austus R eiensis , Sermons (possibly written in France). Ms. 32 (s. viii eas.-ix in.7), 2 fols., a Lectionary with sermons. Ms. 127, 4 fols., I sidore, Etymologiae (written in Irish minuscule; though the outside mar­ gin o f each leaf has been cut off and though prickings now show near the inside margin only (just inside the text), it is obvious that the leaves were originally ruled one at a time between guiding punctures in both inside and outside margins ; placing o f the punctures on both sides o f the text o f a single folio is an Insular practice).—s. x : Ms. 56, an unidentified commentary on I . Cor. x and Bede’ s Expositio in Lucam (possibly written in France). Ms. 4SA, 1 fol., Pseudo-Augus­ tine’s Sermo (Appendix) I (Germany).—s. x ii : Ms. 121, 3 fols., Statius ’ Thebaid (probably written in France; the prickings along the outer vertical line which bounds the text conform to ninth-century practice). An unnumbered ms., 2 fols., Passio Sanctorum L u ci et Clementani (the outside margin contains prickings in the form o f horizontal slits).—s. x iii : Ms. 122, 4 fols., Statius ’ Thebaid (probably written in France).

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. W . Jones, Pricking systems in New Y ork manuscripts

87

and pricked from the top ; the process is then repeated in the re­ maining two bifolia of the gathering.12 A diagram follows : 3 p 4

7

5 6

00

1 p 2

r

y

The earliest books on my present list belong to the ninth cen­ tury.13 The first, a remarkable miscellany (containing works of Alcuin, Gennadius, Isidore, Leo the Great, and other writers) 14 was written in France ca. 820 A. D. and had its leaves ruled 4 (3) Old Style (outside). The second, a very interesting ancient cook­ book (Apicius, De re culinaria) ,15 was written probably in Germany at the middle of the century ; its leaves were ruled one bifolium at a time on the hair side before folding. Both of these books employ pricking system 1 ( l r-8r). A third manuscript, a Gospel Book w rit­ ten at Landevennec in the diocese of Quimper (Brittany) in the second half of the century,16 uses throughout the rare system 7, which has hitherto been noticed in only a single quire of a single manuscript— a Hesychius,, also written in France, possibly at Beauvais, in the latter half of the century.17 A fourth manuscript,

12 This system occurs in only one gathering (14) o f one ms., Morgan 655. 12 In an earlier ms., Morgan 17 (Augustine and Jerome, Sermones et epistu­ lae X X V I I ), which was written in France in Merovingian minuscules in s. vii or viii, the rulings are faint, the system o f ruling is uncertain, and whatever prickings were originally present have been cut off.—My article, P in Pricks at the Morgan Library, Transactions of the American Philological Association 70 (1939), pp. 318-326, reported pricking system 1 in Morgan 462 (ca. 500 A.D.), Morgan S3.) (669 A.D .), and Morgan 776 (s. v iii ex.) ; and system 1 or 5 in Morgan 56.) (ca. 800 A.D .). I t also reported system 1 for four ninth-century mss.: Morgan 640 (câ. 882-900), Morgan 728, Morgan 835 (ca. 825-850), and Morgan 23; and other systems for Morgan 191 (system 5; 4 when the parchment is unusually thick) and Morgan 768 (see note 17 below). 12 Plim pton 230-820 (58). 15 New York Academy of Medicine 1. ’ * New York Public Library 115. Its ruling is 4 O. S. (outside). 1T L. W. Jones, P in Pricks, etc., p. 324 : Morgan 768. Some leaves are ruled separately and others are ruled 2 (4) O. S. No less than four systems o f pricking are used, all without reference to any particular type o f ruling: system 1 most commonly, system 2 in eleven quires, and systems 5 and 7 in one quire each.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

88

M is c e lla n e a

Giovanni M ercati.

V I.

the magnificent Morgan Gospels,'18 written possibly in Switzerland toward the end of the century, is satisfied with no less than three systems of pricking : it chooses system 2 when the ruling is 4 Old Style (outside)19 and system 4 or 5 when the ruling is 2 Old Style (outside). A fifth codex,20 which uses 2 Old Style (outside) ruling throughout— St. Gregory’ s Liber regulae 'pastoralis curae (written in northern France ; possibly at Beauvais under the influence of Tours)—also employs both systems 421 and 5.22 The pricker in each o f the last two books apparently was not much affected by the fact that system 5 took less time than system 4. I f for each book we assume that there was only one pricker and not one for the system-5 quires and a second for the system-4 quires, the pricker may have been an old scribe who knew 5 but preferred the more familiar 4 ; or, again, he may have felt that 4 was worth the extra work inasmuch as the lighter prickings which it produced Avere aesthet­ ically superior to those produced by system 5. Four Greek manuscripts belong to the tenth century.23 One24 (St. Basil or St. John Ohrisostom, Homilies on Genesis), AiTitten probably in southern Italy, has a rather unusual method of ruling : after each bifolium has been folded, a sharp point is used to line the hair side of the top folio so that both folios may be ruled in one operation. This manuscript employs pricking system 1 ( l r-8r) in all its gatherings except one (quire 14) ; in that one it uses system 8, possibly because the parchment is thick. A Gospel Book,25*likeAvise Avritten in Southern Italy, has the same unusual method of ruling in at least six of its quires (12-17) ; in the others in which a system can be made out20 it has 4 (2) Old Style (outside or inside). A

18 Morgan 1. 18 Except in a single quire (25), which has 2 O. S. (outside) ruling. 20 Morgan SS6, s. ix-x. 21 Quires 10, 11 (probably), 14, 15, and 18-22. 22 Quires 1-7, 9 (probably), 12, 16 (probably), and 17. In quires 8 and 13 no clear prickings are visible. 23 My article, P in Pricks., etc., pp. 324-325, reported pricking systems 1 in three tenth-century Latin mss.—Morgan 91, Morgan 644, and Morgan 755—and in one tenth-century Greek ms.r Morgan 652. M Morgan 655, 880-910 A. D. The ruling here described is used in all the quires except 19 and 25 (both of which have their leaves ruled separately), 27 (4 O. S., outside), and 28 (in which the three outside leaves are ruled 3 O. S. but the inside leaves, 212r and 213\ ruled separately). ” Morgan 748. 2‘ In quires 2, 5, 8, 18-20, and 22 the system o f ruling is uncertain.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. W. Jones, Pricking systems in New Y ork manuscripts

89

Lectionary of the Gospels (Evangelion) 272 8of ca. 950 A. D. has its leaves ruled one at a time on the hair side. Similar in ruling is a miscellany,25 written in the tenth or eleventh century in southern Italy. As we should expect, all three manuscripts employ pricking system 1 (Γ-8Γ). Four Latin manuscripts Avere also written at various times dur­ ing the tenth century.29 The earliest30 is a Gospel Book written ca. 900-930 A. D. in northern France or Belgium (possibly at Saint-Amand). The second31 is an Evangelistary written in Ger­ many, possibly at Reichenau, in the middle of the century. The gold letters on a purple ground, the full-page miniatures, and the illuminations make it one of the most precious possessions of the Netv York Public Library. The last tAvo manuscripts 32 in the present group are parts of a single Avork— St. Gregory’ s Moralia in lo i) ; they Avere Avritten in a single center in southern Germany. A ll four manuscripts have their leaves ruled separately on the hair side. The first tAvo employ pricking system l , 33 Avhile the two companion manuscripts employ system 6 throughout. I t is interest­ ing to discover this second system, Avhich appears in some Tours codices of ca. 825 A. I), and later, but Avhich I failed to find in my previous study of the manuscripts noAV in N cav York City. Two, three, or four quires are pricked at a time.34 Future investigation Avili show Avhether system 6 appears in manuscripts Avritten after the tenth century.

27 Plim pton 264-950 (2). 28 Morgan 807 ( the story of the Btephanites and Ichnelates, the Physiologus, the life o f Aesop, Aesop’s Fables, Babrius, and Hierocles). AA7hile the ruling is generally on the hair side, it is occasionally on the flesh side. 28 Though I have also examined Morgan 539 (s. x med,), a fragment o f two folios containing part o f the Koran in Kufic (Arabic) script, I have found no trace whatever o f ruling lines or o f prickings. While prickings could have been trimmed off, the absence o f ruling lines is remarkable in view o f the unusual linear regularity o f the script. 88 Morgan 819. 31 New York Public Library 1. 32 Morgan 764 and 765, s. x. 38 l r-Sr in New York Public Library 1; V'-Sr or 8T-1V in Morgan 319. 84 Morgan 764- quires 1-4, 5-7, and 9-12 are pricked together; quires 8, 13, and 14 have in each case probably been pricked under other quires made for other manuscripts that contained Gregory’s long work (764 contains only Books I-V II and 765 only Books XXIII-XXX). Morgan 765: quires 1-3, 4-7, 8-11, 12-13, 15-16, and 18-19 are pricked together; in addition, 14, which seems at first pricked alone, is actually pricked over quire 17.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

90

Miscellànea Giovanni Mercati. V I.

Only two books belong definitely to the eleventh century.33 One,36 an Italian copy of Victorinus’ Commentum in Rhetoricam Ciceronis, employs 2 Old Style (outside) ruling for the most part.37 The other,38 a Greek copy of the Epistle to the Hebrews and the Apocalypse, has its leaves ruled separately on the hair side. Both codices use pricking system 1 ( l r-8r) ·39 The same system of pricking appears in all the manuscripts of thé twelfth century40— two Greek Lectionaries qf the Gospels41 (one with the lines of each page arranged in the shape of a cross),42 an English Gospel according to Matthew (with a glossary),43 a fragmentary copy of Isidore’ s Mysticorum expositiones sacramen­ torum ,44 a French codex of Ambrose’ s Expositio in Evangelium S. Lucaef ·5 and a computistical treatise containing works of Bede, Dionysius Exiguus, and Isidore.46 This concludes my list. I t w ill of course be necessary to examine the ‘ systems’ of pricking in more products of as many scriptoria as possible and to make detailed studies of these systems in manuscripts of such centers as Luxeuil, Corbie, Laon, and Reims before attempting to reach definite conclusions. Palaeographers are urged to continue 38 My article, Pin Pricks, etc., p. 325, reported pricking system 1 in two mss. assigned to the tenth or eleventh century : Morgan 758 and Morgan 627. 30 Plim pton 10S. " 2 0. S. (outside-inside) in quire 7 ; the leaves are ruled separately in the last quire (9); the ruling is doubtful in quire 1. 38 Morgan 7H , written late in s. xi. 39 Plim pton 103 has 8y-lT in its last quire. 40 My article, Pin Pricks, etc., p. 325, reported pricking system 1 in two twelfthcentury mss. : Morgan 61$ and Morgan 6S9. 41 Morgan 61ft and Morgan 692 (the latter ca. 1200 A. D.), both possibly written at Constantinople. Both have their leaves ruled separately on the hair side. 42 Morgan 692. 43 Plimpton 225-1150 (29). There is only one pricking (in the outside margin) for every three lines o f text. A similar practice appears in the fifth-century Codex Alexandrinus o f the Bible and in other early mss. 44 Plim pton 220-1100 (68) (from Genesis through I I I . Kings only). In addition to this work, known also as Quaestiones in Vetus Testamentum, the ms. contains an anonymous commentary on certain parts o f Numbers, Fulgentius’ homily on the prophecy o f Micah, and (in incomplete form) the life o f the 40 martyrs of Sebaste. The flesh side o f each leaf is ruled in brown ink. 45 Plim pton 220-1120 (50). The ms. also contains Augustine’s Sermo de Nativitate. Bach page (not folio) is ruled separately with lead. 48 Plim pton 251 (ca. 1130 A. D.) : Bede, De ratione temporum; Dionysius E xi­ guus, De titulis pascalibus (with tables as far as 1129) ; Isidore, De responsione mundi et astrorum ordinatione. Bach page (not folio) is ruled separately with lead.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. W . Jones, Pricking systems in New Y o rk manuscripts

91

the investigation. Meanwhile the tentative results thus far set forth may serve as guides in dating and possibly in placing manu­ scripts. Among the important new facts established by the present ar­ ticle are the following : there is a new method of pricking— system 8 ; system 2, which at Tours normally belongs to the eighth century, appears in the second half of the ninth century in France (Morgan 768) and possibly in Switzerland (Morgan 1) ; system 7 (New York Public Library 115) exists beyond the shadow of a doubt in the second half of the ninth century ; a single manuscript often uses several systems of pricking, sometimes because the types of ruling vary {Morgan 1 : systems 2, 4, and 5), sometimes because the parchment changes from thin to thick {Morgan 191 : system 5 for thin parchment and 4 for thick), and sometimes for no apparent reason at all {Morgan 768 : systems 1, 2, 5, and 7) ; a less efficient system is not necessarily discarded in favor of one more efficient {Morgan 856: 2 Old Style [outside] ; systems 4 and 5 are both used) ; and, finally, system 6 {Morgan 76Jf and 765) continues to be used at least as late as the tenth century. Incidentally, one should not forget two simple but useful points : first, the absence of prick­ ings in a single leaf 47 placed among leaves pricked systematically is frequently an indication that the leaf is not original ; and second, an inspection of the nature of the prickings may possibly permit the uniting of membra· disiecta of a quire or of an entire manuscript.48 Apparently less important at the moment but none the less in­ teresting are several sidelights thrown by the present study on prick­ ing practices. I t is obvious, for example, that gatherings are sometimes formed and pricked in advance of actual needs and that exigencies sometimes bring about the mixing of previously arranged gatherings.49 Special requirements are met in logical ways : thus, in a manuscript pricked in accordance with system l , 50 the addition-

47 New York Public Library 1, pp. 217-218 ( = the first leaf o f quire 16). 41 Cf. my Pin Pricks, etc., p. S26 and note 33. 43 Thus, an exigency arises in Morgan 319 when gathering 9 requires 12 leaves. Since this senion’s eight outside leaves (59-62 and 67-70) originally formed a qua­ ternion, they were pricked together in accordance with system 1 (59r-70r). The four additional inner leaves (63-66), however, which presumably formed part o f a second original quaternion, were naturally given their own set o f prickings (63r on 64r on 65r on 66r). 50 New York Academy o f Medicine 1, quire 1.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

92

Miscellanea Giovanni M ercati. V I.

al punctures (near the center of the folios) demanded by the two columns of text on folios 2r and 2Tof a quire whose other leaves have only one column are punched through all the leaves ; this is a timesaving operation. Again, if some or all of the original prickings are misplaced, corrections are made, frequently by the use of prick­ ings of a contrasting type.51 Other equally profitable observations can easily be made.

C1 Thus, in the last quire (14) o f Morgan 764 the first set o f prickings (round holes) was slightly misplaced and a second set (nearly vertical slits) was placed in the proper position and used for the ruling.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L Y N N T H O R N D IK E

THE PROBLEM OF THE COMPOSITE MANUSCRIPT In the making and copying of medieval Latin manuscripts two opposite processes may be distinguished. Standard works of con­ siderable length, which were used as textbooks in the universities or in the convent schools of the friar orders, needed to be multiplied in many copies, but at the same time the original text should be preserved unaltered. Accordingly an exemplar or standard text was maintained in an unbound, loose-leaf state consisting of sections of eight pages each. A student or copyist could borrow one of these peciae for a small fee and get the next one when he returned the former. Thus approximately one hundred persons could be makingcopies of one text of 800 pages at the same time.1 The other process of manuscript makiing was that of the indi­ vidual who desired to copy or to have copied for him in a single volume a number of relatively brief treatises by different authors of his own selection. Since this was his own affair, he might alter the word order or even the wording of his authors to make the mean­ ing clearer or the style more acceptable to himself, or just because he was a bit careless and indifferent as to such matters. He might purposely omit some of the text which did not seem to him worth the trouble of copying, or condense it a little, or expand it a little, or embody a previous marginal note in the text, or add a new note of his own, or make such other alterations as he chose. He might. 1 See Jean D estrez, La Pecia dans les manuscrits universitaires du X I I I e et du X IV e siècle, Paris, 1935. For earlier poetical florilegio, see Dom A. W ilmart , Le florilège de Saint-Qatien, Bevile bénédictine, X I/V III (1936), 1-40, 147-181, 235-258. For a type o f MS compiled “ for the use o f English lawyers during the period o f the first three Edwards ” , see George E. W oodbine, F ou r Thirteenth Century Law Tracts, 1910, pp. 1-3.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

94

Miscellanea Giovanni M ercati. V I.

know the authors and titles of the works which he was copying so well that he would not bother to record them. He might insert anonymously a composition of his own. So, while the exemplar was a single text which aimed to be standard divided into handy pieces, the ordinary learned Latin manuscript is often a composite of different ,treatises in all sorts of permutations and combinations, not merely of selection and arrangement of the component treatises, but in the characteristics and peculiarities of the text of each. This situation raises several questions. What do such composite manuscripts reveal apart from their particular constituents ? How may they best be classified and catalogued? How was the choice of the component treatises exer­ cised, limited and determined, especially in the case of those bearing upon a common topic or field? What scope was possible in the selection ? The problem of the cataloguing and analysis of complex scientific manuscripts may be illustrated by the example of Sloane 3457, an important alchemical collection in the British Museum.2 The old long-hand catalogue of the Sloane manuscripts3 divided this manu­ script into 62 component parts, whereas Mrs. Singer’ s more recent catalogue4 distinguished 91 items, some of which, however, com­ bined items which had been listed separately before. Even so, both cataloguers left large stretches of the text virtually uncatalogued and unidentified. These portions consist in the main of numerous recipes, operations, extracts and passages which would require several hundred separate entries for their complete and satisfactory description. I t took Mr. W . J. Wilson 42 pages thus to catalogue the single Lehigh University alchemical manuscript of Arnald of Brussels5 and almost two hundred pages to describe it more fu lly.6 Therefore this would bë a colossal task, if undertaken for a large

2 For a fuller discussion o f it see L ynn T horndike , A Study in the Analysis o f Complex Scientific Manuscripts, Isis, XXIX (1938), 377-92. 3 A continuation in 9 vols, o f Proofs of a projected, catalogue of Sloane MSS. 1-1091, circa 1837 : sbelfmark, Circle S4a. * D orothea W aley S inger , assisted by Annie Anderson, Catalogue o f Latin and Vernacular Alchemical Manuscripts in Great B ritain and Ireland, dating from before the X V I Century, Union Académique Internationale, Brussels, 3 vols., 1928, 1930, 1931. 5 Catalogue of Latin and Vernacular Alchemical Manuscripts in the United States and Canada, Osiris, V I (1939), 83G pp., at pp. 473-514. 3 An Alchemical Manuscript by Amaldus de Bruæella, Osiris, I I (1936), 220-405.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. Thorndike, The problem o f the composite manuscript

95

number of manuscripts. But it is doubtful whether such material could be adequately catalogued.in any other way. For in the past history of chemistry, metallurgy and technology, or of medi­ cine, surgery and pharmacy, the particular process or operation, secret or recipe or cure, may have as great importance as a longer theoretical treatise or an unoriginal compilation from previous authors. Cataloguers of manuscripts have long since been accusto­ med to itemize state papers in detail, or, in the case of a series of sermons or collection of poems to list each homily or sonnet sepa­ rately with its incipit. I t is high time that they recorded in at least equal detail the learned and scientific writing of the past which possesses some real content and is not mere words. I f a composite manuscript is all written in the same hand, or if it is clear from colophon or note that all the component parts were copied for the use of a single individual, the codex obviously reflects the personal interest and knowledge of its maker or orderer, and also those of a particular period. W e have, for instance, the handbooks of practising physicians and alchemists, such as that, of Leonard of Mauperg described by Professor Corbett,7 containing not merely medical or alchemical tracts but, as already noted, in­ numerable recipes taken from a multiplicity of sources and perhaps including some experiences or experiments or discoveries of the writer. Such manuscripts combine within one cover a working library with a laboratory notebook or a clinical record of cases. Or they may reflect the professional interests of a lawyer or logician or theologian or astronomer and astrologer. Others are common­ place books displaying a miscellaneous literary interest. In all these instances what is left out may be as significant and revealing as what is included. The personal liberties which the maker of the new manuscript has taken with the old texts also have their import. There is a field open to conjecture not only why certain treatises have been included but why this or that particular extract from a past work was made. I f the manuscript was not put together for professional purposes and does not deal with any one special field of knowledge exclusively, its combination of subjects provides fur­ ther food for thought as to the type of mind back of this conglo’ J ames C orbett, L ’alchimiste Léonard de Mauperg {X IV e siècle) : sa collection de recettes et ses voyages, Bibliothèque de l ’École des Chartes, X C V II (1936), 132-41.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

Miscellanea Giovanni Mercati. V L

96

meration of ideas. Why vvere these treatises on different topics thus brought together ? This last question may also be asked in the case of manuscripts written in different hands or of which the component parts were bound together at a later date. W hy were they bound together ? I t may be of some value in determining past intellectual relation­ ships and associations of ideas to note in what sort of manuscripts a given treatise is found and what are the other treatises which most often accompany it; in other words, to observe its manuscript setting and the company which it keeps. Take the anti-clerical and anti-ec­ clesiastical Defensor pads, coposed in 1324 by Marsiglio of Padua, with its advocacy of the lay state and leanings towards popular sovereignty, whose author had to flee from Paris to Louis of Ba­ varia. Unlike his Defensor minor, it is found in a number of manu­ scripts. W e are not surprised to find it in one with W yclif on civil lordship,8 or in another with a dialogue by Ockham and erroneous articles of W y c lif,9 or even in a third with Gerson on the ecclesia­ stical power and Hervaeus Natalis on ecclesiastical jurisdiction.10 But it is a bit surprising to And it immediately following the work of St. Bernard on contempt of the world in two fourteenth century codices,11 both of which open with these two works, although their subsequent contents are quite different. Quite different, I say, and yet their respective owners were apparently in accord in liking to familiarize themselves with varied or almost diametrically opposed viewpoints. The one continues with the treatise of Richard FitzRalph against sturdy beggars, an anonymous tract on the papal power, the moral compendium of Raoul Presles on the republic, an epistle of Lucifer dated 1351, and a bibliography of the writings of Aquinas. The other includes two translations of the Koran, works of Nicholas of Lyra against the Jews and the Moslems, letters of the emperor of China to the pope and of Christians living in Cambakéth, and disclosures by converted Waldensians. There may further be significance in the order in which related works occur, although, so far as my knowledge goes, there is likely to be more variety than uniformity in this respect. This may be

* Paris, 5 Paris, 10 Paris, 11 Paris,

Bibliothèque Nationale, Latin MS 15869, 14th century. Bibl. Nat. Latin 14619, 15th century. B N 14620, 15th century. BN 14503, 15690.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. Thorndike, The problem o f the composite manuscript

97

illustrated by a number of manuscripts, for the most part of the thirteenth and fourteenth centuries, containing works of'Galen in Latin translation. Each such work is indicated by a capital or small letter of the alphabet as follows : A B C D E F G H I J K L M N O P Q

De utilitate pulsus De voce et anhelitu Anathomia Megategni De interioribus Secreta De diiferentiis febrium De utilitate respirationis De causis respirationis De icteri cura De dispensantibus De bono habitu De dignotione in somniis De duodecim portis De marasmo De rigore et tremore Liber pharmacorum

a b c d e f g h i j k 1 m n o p q

R 5 T TJ W X Y Z 6

De febribus Optima constructio De secretis secretorum De sententiis De temporibus De temporibus paroxismorum Megapulsus Peri craseos Podagra

r s t u V

w X y z

De elementis De virtutibus naturalibus De complexionibus De malitia complexionis De criticis diebus De crisi De signis interiorum Liber secretorum De dispermate De accidenti et morbo De causa morbi De ingenio sanitatis De iuvamento membrorum De sectis De spermate De iuvamento anhelitus De differentia pulsuum (Megapulsus?) De tactu pulsus De compendiositate pulsus Diagnosticon De simplici medicina De causis pulsus De regimine sanitatis De cibis (or. alimentis) De motibus liquidis lib. ii ad Glauconem

They occur in the order indicated in the following manuscripts : E jD Z Boulogne-sur-Mer 197, 13th century beyxGoNTnvrAUdpwP Bourges 299, 14th century cuawbEdjkvGpmeyBqrsxlf Breslau University IV . F. 25 ltuvqwxyz& Cesena dextr. X X III, 1, 13th century abcdefghij Cesena dextr. XXV, 1, 13th century lcinnopdqrsj Cesena, dextr. XXV, 2, 13th century acudmlrAyBCjDEefoF Chartres 284, 13th century jfE dlzG ebH IFoJK LM SW X Chartres 293, 14th century Miscellanea Gl·. M ercati. V I.

7

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

98

Miscellanea Giovanni M ercati. V L

DEujwGfeAyBbcdaToNmqr Erfurt, Amplon. F. 249, 13-14th century xBwGbf Florence, Laurentian Library, 73, xi, early 14th century IgGfewjcbAd Klosterneuberg 126, 14th century acbuh sqAEj C 12 Leipzig 1118, 13th century acdumbGorAByDljEefh(or F ?) Leipzig 1184, a mid-fourteenth cen­ tury list of “ Libri Galieni quos communiter habemus” acdbBrjfenyïnuEBwl Montpellier 18, 13th century aQnjwERcdfeSxbmv Munich, Nat. Bibl. Latin 5 abCoNFryBmEDjfeGvwxPcdul Oxford, Balliol College 231, 14th century acdujDfegb Oxford, Merton College 218,. 14th century acbmjEDeflu Paris, Bibl. Nat. Latin 14389, 14th century acudm lrAyBCjDEefFoN Paris, University 125, 13th century OlcdwbjGef Vatican, Palat, lat. 1095 DlpB Vatican, Palàt. lat. 1097 mYsqy Vatican,Palat, lat. 1099 cluj...EmfbG Vienna 2273 leJkGjdQBcRfe Vienna 2944 Somewhat similar results are seen in the case of the contents of some sixteen manuscripts of the Lullian alchemical corpus, and in manuscripts containing works of Nicolas Oresme and Henry of Hesse.13* On the other hand, the frequent association together of the same group of treatises may hold true not only of the works of a single author like Aristotle, or of textbooks on the same subject like Philaretus on the pulse, Theophilus on the urine, the Apho­ risms and Prognostics of Hippocrates, and other tracts making up the Ars medicinae,14 or like the Insolubilia, Impossibilia, Conse­ quentiae and so on of late medieval logic, but also of more advan­ ced works such as accompany the optical tracts of Alkindi, Tideus and pseudo-Euclid.15 Such strict unity in subject-matter is by no means always obser­

12 In the spaces left blank occur tracts not covered by the above alphabets, namely, Experimenta, De typo, De dinamidiis, De catarticis, in that order. 13 See my History of Magic and Experimental Science, IV , 619-23, and I I I , 743-47 respectively. 11 C harles H. H a s k in s , Studies in Medieval Science, 1924, p. 369, note 63. 13 See the contents o f the MSS as catalogued by B j Srnbo in Abhandlnngen zur Gescldchte der mathemaiischen Wissenscliaften, XXVI, 3 (1911), 123-47.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. Thorndike, The problem o f the eomposite manuscript

99

ved. Or at least what seem to us anomalies and inconsistencies creep in, but they should perhaps warn and inform us of a diffe­ rent mental outlook then. A fourteenth century manuscript at E rfu r t16 is in the main a collection of mathematical, astronomical, physical and optical treatises of a high order. I t has a Latin trans­ lation of Archimedes on burning mirrors, two tracts on weights by Jordanus, the work on the magnet of Petrus Peregrinus, the Perspectiva communis of John Peckham, the works on mirrors ascribed to Euclid and to Ptolemy, Thomas Bradwardine on pro­ portions, an Algorismus de integris, the Canons of John of Saxony on the Alfonsine Tables, an anonymous Theorica planetarum, a table by Petrus de Dacia of the positions of the moon, and a little known treatise by a James of Naples for the students of the Augustinian convent there. So far the contents would seem to reflect the purely scientific interests of some forerunner of modern speciali­ zation in the exact sciences. But we have not yet seen the whole picture. Between the work of Bradwardine on proportion and three books of Euclid’ s Elements with the commentary of Campanus of Novara come a table of interpretation of dreams and a tract on judging horses. Between Peregrinus on the magnet and the Theo­ rica planetarum occur tables of conjunctions of the planets sug­ gestive of astrology. A t the Bodleian in Oxford a membrane manuscript which is mostly of the early fourteenth century 17* has scattered among its 27 items, besides Latin poems—three of which are of a religious turn—and theological tracts, an arithmetic, several works on com­ putus, others on the calendar and astronomy, Peregrinus on the magnet, Secreta secretorum of pseudo Aristotle, the alchemical Speculum secretorum variously ascribed to Boger Bacon or Albertus Magnus, the astrological medicine of pseudo Hippocrates, the long passage on weather prediction from the 18th book of P lin y’ s 'Natural History,1S the Physiognomy of Polemon, the prophecies of Merlin, Marbod on gems, Peckam’ s Perspectiva, and Messahala, the

“ Amplon. Q. 387. 17 Digby 28. The first eight leaves, containing the Massa compoti o f Alexander o f V illa Dei, are early 13th century. '* It is found separately and· anonymously in most MSS as a Liber de presaglis tempestatum without Pliny’s opening paragraph and was so printed by error :in Appendix 18 to voi. I l l of A History o f Magic and. Experimental Science. Digby 176, fol. 60T has the opening paragraph.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

10 0

Miscellanea Giovanni M ercati. V L

Arabic astrologer, on the astrolabe. These Avorks of experimental and occult science constitute the fifth, tenth, eleventh, thirteenth, nineteenth and last six pieces in the manuscript. The two exclu­ sively theological tracts are the tAvelfth and twentieth items. SandAidehed in between these and one another come the other 14 Avorks indicative of a poetical or mathematical bent. Thus a broad mental outlook is suggested and the combination of varied modes of thought, but withal considerable curiosity as to the occult. From what diverse sources a composite manuscript might have been gathered may be illustrated from an astronomical and astro­ logical collection covering 119 leaves and including 32 items. I t dates from the fourteenth century, Avhen it belonged to W illiam Rede, bishop of Chichester. A note on the verso of the first leaf records that he was given part of it by master Nicholas of SandAvich ; purchased part of it from the executors of Thomas Bradwardiine, archbishop of Canterbury, Avho died in 1349; bought another section of it from the executors of Richard Camsale; AArrote part of it Avith his own hand and had a portion of it copied off for him from other manuscripts.19 It Avas a good thing that W illiam Rede formed this composite manuscript, for it alone has preserved and transmitted to posterity Avhat seems our earliest knoAvn detailed and systematic record of the Aveather over a considerable period, namely, seven wears.20 H oav much time and pains the putting together of a single com­ posite manuscript might take is further shoAvn by another codex of the last decade of the fourteenth century,21 much of Avhich seems to have been written out Avith his own hand by Donatus de Monte of Chiusi. On May 3. 1391 at the nineteenth hour he began to copfy a question by the famous doctor of medicine, Marsilius de Sancta Sophia, as to the multiplication of species, a topic treated by Roger Bacon in the previous century. Donatus did not finish copying it until the 27th of the month. On August 27 he completed a question by Albertinus of Piacenza on the contact of solid bodies, and on

'· Oxford, Bodleian, Digby 176. 20 M erle’s MS Considerationes Temperiei pro 7 annis... The earliest known journal o f the weather... lS37-lSJf4, reproduced and translated under the supervi­ sion o f G. .T. Symons, London, 1891, folio. For an earlier but briefer record in 1269-1270 see A History of Magic and Experimental Seienee, I I I , 141, note 3. 11 Oxford, Bodleian, Canon. Mise. 177.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. Thorndike, The problem o f the composite manuscript

10 1

September 3 at Padua the treatise on the first and last instant which John of Holland had composed at the university of Prague in 1369. During the next year, 1392, Donatus finished copying the work of Gregory of Bimini, general of the Augustinian order, who had died in 1358, on intension and remission of forms, and on December 29 he brought to a close the discussion of the same topic by Blasius of Parma, who was still living and not to die until 1416. A t the time of copying this treatise Donatus was already a doctor in arts and was studying medicine. On March 4, 1393, which was the first Sunday of Lent, he completed another treatise, on the theme of augmentation and diminution. The next recorded addition to the manuscript by him was on December 29, 1935, at Chiusi, whither he had returned to spend the Christmas holidays with his father and where he enriched his collection with the work on proportions of Thomas Bradwardine. On February 12, 1396, during very heavy rains at Padua, where he had now become a doctor of medicine as well as of arts, he finished copying a treatise on maximum and minimum, “ subtly composed” , he says, “ By Lineriis or Suiseth or Bosetus at some English university” . From October 15 to 19 lie wrote down a question on the elements by Marsilius de Sancta Sophia which bore upon the topic of intension and remission. This copying was done in the house of Jacobus de Vellegio, while Donatus was waiting to go to Venice to take a house there, and during heavy rains and a lunar eclipse. On February 4, 1399 he copied another treatise on the first and last instant, this time by W alter Burley, the English schoolman. The last record in this manuscript of works copied by Donatus was on January 29, 1401, when he completed the questions of John of Casali on action, of which the first dealt with the velocity of the motion of alteration. He adds that he began to copy them fourteen years ago at Padua “ in the time of my youth ” . He completed them at the turn of the century in Chiusi. The remaining treatises in the manuscript, interspersed between those already mentioned, are not specifically signed and dated by Donatus and are usually not in his handwriting. But they are on similar topics and sometimes are the compositions of professors at Padua, so that they may well have been collected by him or copied for him. A t first glance it might seem dubious whether the present manuscript is his original, since the treatises with his colo­ phons do not occur in strictly chronological order of their copying.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

102

Miscellanea Giovanni M ercati. V I.

As at present constituted, the codex begins with works which he copied in 1396 and 1399. I t appears, however, from the table of contents that it originally opened with two treatises by Jean Buridan and ended at leaf 213 with the table of contents. The works copied in 1396 and 1399 have somehow been substituted for those by Buridan, whereas they would more properly be added after leaf 213, as are the questions by J ohn de Casali which Donatus finished in 1401. I f space permitted, I might go on to describe other manuscripts that include a number of the works which Donatus thus put toge­ ther. But not precisely the same selection would be found in any two of them. A ll display a similar specialized interest, but it is expressed differently in each individual case.22 This grouping of several specialized treatments of the same field or related topics continued to be manifest in early printed col­ lections, which indeed perhaps sometimes simply reproduced the treatises in a single composite manuscript such as we have been describing. For example, the edition of Venice, 1505, which opens with Bassanus Politus De modaUbus, also contains a treatise on proportion introductory to the Calculationes of Suiseth, further the works on proportion of Bradwardine and Oresme, those on the latitude of forms of Oresme and of Blasius of Parma, an anony­ mous Tractatus sex inconvenientium, the Quaestio of John of Ca­ sali on the velocity of the movement of alteration, and that of Bla­ sius of Parma as to lifting two solid bodies which are in contact. In a period when Latin was the universal language of learning and culture, and, aside from its employment in the church service, was used almost exclusively by intellectuals, and when manuscripts were guarded more closely by their possessors than printed books are today even by professional librarians, we find many a striking continuity of tradition in subject of author, and many a concate­ nation of distinguished ownership of a given codex. John of Gmunden, the Vienna mathematician in the early fifteenth century, was still interested to compose a work on the instrument known as Albion, of which Richard of W allingford had written early in the previous century. Gmunden's treatise passed into the hands of

22 Venice, St. Marks V I, 62 (Valentinelli, XI, 20); V I, 155 (Valentinelli, XI, IS); V I, 96 (Valentinelli, XI, IT); V I, 149 (Valentinelli, XI, 23).

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

L. Thorndike, The problem o f the composite manuscript

103

Johann Virdung de Hassfurt, another mathematician,23 astro­ nomer and astrologer of the early sixteenth century. A final question which suggests itself is; W hy does the same work recur a number of times in the extensive libraries of men like Amplonius Ratinck de Berka (1360-1435) and Hartmann Schedel of Nürnberg (1440-1514), whose manuscripts are, or Avere, preserved at Erfurt and Munich respectively ?24 These men ivere not mere book collectors but learned physicians Avho acquired manuscripts with the aim of using them for professional and scholarly purposes. They AArere not princes or prelates AA’hose books Avere selected and ordered for them ; they purchased these themselves, often abroad, and sometimes made their copies Avith their OAvn hands. One reason Avhy the same treatise recurs so many times is that it is found in different combinations AA'ith other tracts, and that the manuscript AA'as presumably acquired for the sake of Avhat Avas neAV in it, despite the fact that much of its contents AAras already in the acquirer’ s library. HoAA’ever, the combinations, too, are in considerable mea­ sure duplications. And not only may there be several separate copies of the same single Avork, but in a single manuscript the same treatise may occur tAvice, sometimes even Awritten in the same hand. It is hard to believe that a scholar or professional man Avho had laboriously copied off a text for his oavii use Avould forget that he had done so and begin over again. But a professional copyist Avho executed a multiplicity of such orders Avithout interest in the content might do so, either inadvertently or fraudulently. Some­ times the second version breaks off after a page or tAAro has been Avritten, indicating that the duplication had been discovered. W ith regard to the larger question it may be surmised that the ncunirer either Avas interested in procuring different versions of a text AA’hich he might compare, or that he A\ras after all more or

23 Vatican, Palat, lat. 1369, 1444 A. D., fols. l r-53v. What may be an older copy than this which passed into Virdung’s possession is at Munich, cod. lat. 14583, 4to 15th century, fols. 340-(410), De compositione et usu instrumenti Albyon a. 1430 editus. Concerning Virdung’s MSS see further Isis, XXV (1936), 363-71; XXXIV (1943). 291-93. 24 Concerning Ratinck see W il h e l m Sc r u m , Beschreibendes Verzeichniss der Amplonianischen Handschriften-Sammung zu E rfu rt, Berlin, 18S7, especially pp. vlviii, « Gesehiehte des Amplonius und seiner Sammlung », and pp. 785-867, « Das von Amplonius eigenhiindig um 1412 angelegte Verzeichniss seiner Bibliothek ». Concerning Schedel see R ichard Stauber, Die Schedelsche Bibliothek, 1908.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

10 4

Miscellanea Giovanni M ercati. V I.

less a promiscuous collector. As time went on, he may have deve­ loped a passion for gathering as many copies as he could of certain favorite works. Or it may be that, after his library had attained a certain size, the law of diminishing returns began to operate so far as the obtaining of new works was concerned, and that he had either to cease adding to his collection of manuscripts or be sa­ tisfied for the most part with further copies of texts already in his possession. I f this supposition be true, we would have some measure of the extent and limits of the literary output in various fields at various times in various areas, and, against the impression that much has been lost or has not yet been discovered, which we get by inference from the survival in so many cases of only one manuscript of a given work, could set the failure to include such material by the most assiduous manuscript collectors of the medieval period itself. This would also have an important bearing on the fundamental problem wheter the extant manuscript remains are too scanty a percentage of what once existed to draw satisfactory conclusions from them, or whether they are by and large fairly representative of the thought and science, learning and letters, of that period, and this quantitatively as well as qualitatively.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

M E D E A N O R SA

ANALOGIE E COINCIDENZE TRA SCRITTURE GRECHE E LATINE NEI P A P IR I È ovvio e naturale che un reciproco influsso e interferenze visi­ bili tra le forme delle scritture latine e delle greche si possano r i­ scontrare in Egitto, dove per più secoli l ’ uso delle due lingue e delle due scritture fu contemporaneo, quantunque di fronte al greco, che era la parlata di tutti, la lingua ufficiale e popolare così dei greci come degli egiziani ellenizzati e degli asiatici, il latino rimanesse limitato alla cerchia militare e alle cancellerie di ordine superiore,1 a un numero dunque relativamente esiguo di individui. È noto che le legioni romane stanziate in Egitto furono dapprima tre, ridotte poi a due, se non proprio ad una sola, che la massima parte dei documenti latini provengono appunto dalla cerchia m ili­ tare e che ben rari sono i documenti delle alte sfere amministra­ tive di Alessandria dov’era usato il latino sebbene gli editti impe­ riali e le ordinanze dei prefetti fossero diffuse nel paese in lingua greca. I l greco era la lingua ufficiale delle cancellerie delle metro­ poli dei vari nomoi e tutti gli impiegati statali e comunali, dallo stratego in giù, ignoravano il latino.2 Essendo quindi tanto inade­ guato il campo delle due lingue e per conseguenza il numero delle scritture latine rispetto alle greche,3 il confronto non è facile nè 1 La cancelleria del prefetto, deirdppiiKacvjjc, dell’ioioc \6yoc, del inridicus e auelle degli epistrateghi. 2 A. Ste in , Untersuchungen sur Geschichte und Venraltung A egyptens unter Romischen Herrscliaft, pp. 149 sgg. , 166 sgg. ; U. W il c k e n , Ueber den Nutzen der lateinischen Papyri, in A tti del IV Congresso internazionale di Papirologia (F i­ renze, 1935), Milano, 1936, p. 102 sgg. 2 Si contano a decine di migliaia i papiri greci ; quelli latini invece sono ap­ pena 200-250 : si cfr. M ax I h m , Zentralblatt. f iir Bibliothekswesen, XVI, 1899 ; H. B. V an H oesen, Roman cursive W riting, Princeton, 1915; a questi si devono aggiungere

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

106

Miscellanea Giovanni M escati. V I.

può essere completo e definitivo ; può però tornar utile per vari riguardi^ sia per le scritture documentarie (corsiva e cancellere­ sca), sia per le librarie. S c r i t t u r e d o c u m e n t a r i e . — Consideriamo dapprima le scritture documentarie. Tra i latini stanziatisi in,Egitto non pochi dovettero imparare il greco e usarlo in pratica se già lo conoscevano. E come nella pronuncia di chi apprende una lingua straniera si notano tona­ lità e atteggiamenti speciali dovuti alle abitudini fonetiche prece­ denti, così è ovvio che avvenga anche per le forme della scrittura straniera che conserveranno talvolta un ductus dovuto alle movènze grafiche precedenti abitudinarie. Sono sfumature lievi, trapassi non sempre evidenti e che talvolta l ’ occhio sorvola, ma che non sfuggono a un esame più attento. Infatti valenti studiosi non mancarono di no­ tare, sia pur raramente, in qualche documento greco alcune forme grafiche che rivelano il ductus di chi è avvezzo a scrittura latina. Già nel 1901 Gregorio Zeretelii*4 pubblicava il facsimile del papiro di Ber­ lino P. 7007, notando che una delle due mani di scritto in quel docu­ mento rivela un ductus latineggiante, e il Wessely nelle sue Schrifttafeln sur dlteren lateinischen Palaeographie già nel 1898 aveva accennato tra i problemi che i nuovi ritrovamenti di Egitto avreb­ bero dovuto risolvere anche «d ie w.eehselseitige Beeinflussung der lateinischen und griechischen Schrift » (Intr. p. I). E a forme di scrittura greca con particolare atteggiamento dovuto a ductus la­ tino accenna più volte lo Schubart nella sua Griechische Palaeo­ graphie, ma usa con esitazione (p. 132) la parola « National typus ». I l Wilcken unisce alla sua pubblicazione dei Papiri di Brem a5 il facsimile della lettera di Paberius Mundus ad Apollonio, stratego dell’ Eptacomia (P . Brem. 5) mettendo in rilievo (p. 35) l ’interesse paleografico di quel papiro (che è degli anni 117-119 d. C.) sia per la formula di saluto aggiunta autografa da Faberius Mundus, in greco perchè diretta a uno stratego, ma scritta con spiccato ductus latino, sia perchè la lettera, di mano di uno scriba ufficiale, offre uno specimen notevole di scrittura calligrafica, uscita dalla can-

i papiri latini editi dopo il 1915 per cui si cfr. P. Collari, in Revue de Philologie, 1941. 1 G. Z ereteli, Ueber die Tsationaltypen in der Schrift der griech. Papyri, in Archiv f iir Papyrusforschung, I, pp. 336-338. 5 Die Bremer Papyri, Ahhandlungen der prensu. A had. der Wiss., 1936, Phil. Hist. Cl., n° 2.

www.torrossa.com - For non-commercial use by authorised users only. License restrictions apply.

M. N orsa, Analogie e coincidenze tra scritture greche e latine nei papiri

107

celleria di un funzionario romano. Nota inoltre il Wilcken (p. 33 sg.) che anche la lettera frammentaria di Ulpius Celer (P. Brem. 10) ha una scrittura « die jenen griechisch-lateinischen Mischtypus zeigt », simile a quello che egli ha già rilevato nella lettera n. 5 e per cui si può confrontare anche la lettera n. 6. Però il Wilcken, che faceva queste osservazioni nel 1935, si limita a rile­ vare la particolarità di queste scritture senza trarne alcuna con­ seguenza,6 mentre lo Zereteli, nel 1901, dalle sue giuste osserva­ zioni era portato, troppo intempestivamente, ad ammettere dei tipi di scritture nazionali nella corsiva greca. E va notato che il Wilcken non usa il termine « Nationaltvpus » ma dice « Misch­ typus », « Mischprodukt ». Infatti, sebbene anche oggi, pur con più ampia documentazione, può essere arrischiato fare afferma­ zioni troppo recise, si può tuttavia rilevare che allo stato odierno delle nostre cognizioni le peculiarità grafiche di questi documenti, peculiarità che solo sporadicamente ci sono attestate durante i primi tre secoli d. C., non ci autorizzano ad ammettere un tipo di scrittura latineggiante o romana nella corsiva greca, perchè si tratta soltanto di uno speciale atteggiamento che potè talvolta as-

* Conferma soltanto l’assegnazione al see. n d. C. fatta dallo Zereteli per il papiro di Berlino P. 7007. I l P. Brem. 5, riprodotto nella nostra tav. I, mostra chiaro il