Skip to main content

Full text of "The Latin Text of the Paris Psalter: A Collation and Some Conclusions"

See other formats


STOP 



Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world by JSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 
purposes. 

Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- 
journal-content . 



JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 
contact support@jstor.org. 



IV.— THE LATIN TEXT OP THE PARIS PSALTER: 
A COLLATION AND SOME CONCLUSIONS. 

The Latin text of the famous "Paris Psalter" (Ms. 8824, 
fonds latin, Bibliotheque Nationale, Paris) has not been re- 
printed since the editio princeps by Thorpe 1 in 1835, altho both 
the " West-Saxon Psalms," 2 the accompanying translation of 
the first " fifty," and the " Anglian Psalms," 8 the metrical 
version of pss. li-cl, have had modern editions. Meanwhile a 
number of students of these important Old English documents 
have used Thorpe's Latin text, with a pathetic reliance upon its 
fidelity and accuracy, for the determination of the exact relation 
of the versions juxtaposed in the Paris Psalter and their place 
in the succession of medieval Latin and Anglo-Saxon psalters; 
and far-reaching inferences have been drawn in recent years. 
In order to show the true character of Thorpe's text, a collation 
of the first fifty psalms with the original manuscript is hereby 
presented, together with an outline of alterations which seem to 
be indicated in certain over-hasty earlier conclusions. 

A suspicion might well have been awakened by the fact that 
Thorpe does not vouch for the accuracy of the Latin portion of 
his work, as he does for the Anglo-Saxon. Of the latter he says 
in his " Praefatio " (p. vii) : " Errores quos apertum est ex 
incuria librarii provenisse corrigere non dubitavi, omnes locos 
ad finem voluminis notans quibus emendandis operam navavi." 
This promise is fairly well fulfilled, considering the date of the 
edition, altho a glance at the textual notes of the latest editors 
will show that frequently, tho doubtless unintentionally, Thorpe 
has failed to reproduce the manuscript text of the West-Saxon 

1 Libri Psalmorum Versio Antiqua; cum Paraphrasi Anglo-Saxonica, 
partim soluta, oratione, partim metrice composita. Nunc primum . . . 
descripsit et edidit Benjamin Thorpe. Oxonii, MDCCCXXXV. 

'Liber Psalmorum: The West-Saxon Psalms, being the Prose Portion, 
or the "First Fifty," of the so-called Paris Psalter, J. W. Bright and 
R. L. Ramsay. Advanced Edition, 1907. (The Complete Edition, with 
Introduction and appended matter, will shortly be issued.) 

'Bibliothek der angelsachsischen Poesie, C. W. M. Grein and R. P. 
Wttlker. III. Band, 2. Halfte, 1898. 

147 



148 AMERICAN JOURNAL OF PHILOLOGY. 

and Anglian Psalms. But with regard to the adjoining Latin 
he merely remarks (p. vi) : " Versio Latina nullam mihi cog- 
nitam omnino refert, sed ad ' Versionem Antiquam Eomanam ' 
proxime aeeedit," adding in a footnote, " Ex hac versione lacu- 
nas textus Latini explevi." This last statement refers to the 
numerous lost pages for which Thorpe has supplied the missing 
Latin. His additions are inserted without indicating where 
they begin and end, and they are, as a matter of fact, far from 
being accurate transcripts of the Eoman version; apparently 
they are merely copied in from the Vulgate, with occasional 
Eoman readings. Nor did he hesitate, where the manuscript 
text lay before him, to make an extraordinary number of seem- 
ingly arbitrary changes. 

In the collation made by Tanger 4 in 1883, a considerable 
proportion, tho by no means all, of these changes were revealed. 
Tanger's useful article should have served as a warning to future 
investigators of the danger of implicit confidence in Thorpe's 
text ; yet strange to say practically all of them have disregarded 
his warning and his list of variant readings alike. 

The most serious offender is the latest to publish a study of 
the Anglo-Saxon psalters and their inter-relations. In his able 
and otherwise scholarly article, 5 Wildhagen places the Latin 
text of the Paris Psalter latest in the succession of the English 
copies of the " Psalterium Eomanum " furnisht with vernacular 
renderings, on the ground of its containing the largest number 
of Gallican readings. Wildhagen's theory postulates a peculiar 
national text of the Latin Psalter going back in all probability 
to a single copy brought by the earliest Eoman missionaries. 
This primitive Anglican text followed the Eoman version in the 
main, but contained a very considerable number of distinctive 
variants, — partly readings carried in from pre-Hieronymian 
versions, partly certain Gallican readings that were presumably 

'Gustav Tanger, "Collation dea Pariser Altenglischen Psalters mit 
Thorpe's Ausgabe," Anglia VI ( 1883) , Angeiger, 125-141. 

B Karl Wildhagen, Studien eum Psalterium Romanum in England und 
eu seinen Olossierungen (in gesdkiehtUcher Entwieklung) , Studien zur 
englischen Philologie, L, 417-472, 1913. See also Wildhagen's earlier 
publications: Der Psalter des Eadwvne von Canterbury, Stud. z. eng. 
Phil., XII, 1905; Archiv f. neuere Sprachen, CXVI, 159-163, 1906; 
Deutsche Literatur-Zeitung, 1909, 3106 f.; Der Cambridger Psalter: I. 
Text mit ErhVPrungen, Grein-Wttlker, Bibl. d. angels. Prosa, VII, 1910. 



THE LATIN TEXT OF TEE PARIS PSALTER. 149 

found in the original English copy, 6 and partly readings that 
are found only in the English manuscripts. Wildhagen at- 
tempts to determine this Anglican text in large part by a com- 
parison of the nine Anglo-Saxon Psalters of the type that sur- 
vive: namely, those known as the Salaberga, Blickling, Ves- 
pasian, Junius, Koyal, Eadwine's, Bosworth, Cambridge, and 
Paris Psalters, which he ranges approximately in this order of 
time, from the end of the seventh to the beginning of the 
eleventh century. 7 The succession is markt by a gradual re- 
placement of what Wildhagen considers the primitive Anglican 
readings by the ordinary Eoman text, and by the introduction 
of new readings, mainly Gallican. At the end of this process 
Wildhagen places the Latin text of the Paris Psalter, when the 
Eoman text was being superseded altogether by the Gallican 
under the influence of the Benedictine Eeform and the rising 
Norman influence ; and accordingly it belongs in order of devel- 
opment just between the two traditions, Eoman and Gallican, 
the latter being represented in England by the Spelman (or 
Stowe), Vitellius, Tiberius, Lambeth, Arundel, and Salisbury 

6 With regard to these Gallican variants found from the beginning in 
the English psalters of the Eoman type, Wildhagen seems to have 
changed his opinion. In his study of Eadwine's Psalter (p. 213) he 
maintained that they are only apparently Gallican, — that each instance 
really goes back to one of the Old Latin or pre-Hieronymian versions, 
or else to a patristic rendering that was carried into the original 
English copy, any coincidence with the Gallican being purely accidental. 
The same explanation is strenuously defended in his review of Boeder's 
edition of the Royal Psalter (Arohiv f. n. Spr. CXVI, 159 f.). But in 
the Studien sum Psalterium Romanum (p. 421), he admits the possi- 
bility of early direct influence from the Gallican version, — "vielleicht 
hier und da in Anlehnung an Hieronymus' Bearbeitung des Psalterium 
Gallicanum — die von Gallien aus eingefiihrt bis zum 6. Jahrhundert in 
England verbreitet gewesen war." Here lies the weakest point in his 
whole theory; for if there were numerous Gallican readings already 
present in the Anglican archetype, manifestly it becomes impossible to 
say whether a Gallican reading found only in two or three, or even in 
one, of the surviving psalters, is a part of the primitive stock that has 
been regularized out of the others, or an instance of the rising tide of 
later Gallican influence. 

' Eadwme's Psalter is, of course, in its present form a product of the 
early twelfth century; but Wildhagen holds that its Latin text was 
fixt about 950, approximately about the same time as the Koyal, and a 
little earlier than the Cambridge Psalter. 
4 



150 AMERICAN JOURNAL OF PHILOLOGY. 

Psalters. 8 Wildhagen's order, while of course not wholly, is 
largely derived from a detailed comparison of readings ; and this 
is especially the case with his conclusions about the Paris 
Psalter. As an important corollary, he holds that the prose 
West-Saxon Psalms and the Latin text were copied into the 
Paris Psalter from the same manuscript, both alike having been 
produced at Malmesbury, whereas the Anglian Psalms were 
taken from another manuscript brought to Malmesbury from 
Mercian territory, probably from Worcester. 9 

But when we discover, as the present collation will show, and 
as Wildhagen might have learned in large part from Tanger, 
that the extensive concessions to 6a supposedly found in P are 
non-existent, an important part of the foundation for his theory 
disappears. For example, Wildhagen illustrates the supposed 
character of the P text by citing (p. 466) the following Ga 
readings from Thorpe's edition: 

ix. 23, quaeret (inquiret Eo) ; xi. 7, argentum igne examina- 
tum probatum (Eo om. probatum) terrae; xii. 5, in misericordia 
tua speravi (in tua misericordia sperabo Eo) ; xvi. 12, eripe 
animam meam ab impio f rameam tuam ab inimicis manus tuae 
(. . . f rameam inimicorum de manu tua Eo) ; xvi. 13, a paucis 
de terra divide eos in vita eorum (a paucis a terra dispertire eos 
et supplanta eos in vita ipsorum Eo) ; xxiv. 18, non erubescam 
(domine non confundar Eo) ; lvii. 4, et venifici incantantis 
sapienter (et veneficia quae incantantur a sapiente Eo) ; xcv. 9, 
dicite in gentibus quia (Eo om. quia) dominus regnavit. 

But of the eight supposed Ga readings only one 10 is actually 
found in the Paris manuscript ! Thorpe alone is responsible for 
the other departures cited; and furthermore, except for one 

•The usual abbreviations are used in this article: A = Vespasian, 
B = Junius, C = Cambridge, D = Eoyal, E = Eadwine's, D 1 = Blick- 
ling, L = Bosworth, P = Paris; F = Spelman, G = Vitellius, H = Tibe- 
rius, I = Lambeth, J = Arundel, K = Salisbury; Ro = Roman Version, 
Ga = Gallican Version. 

" Studien mm Psalterium Bomanum, pp. 469 f. 

10 At xi. 7 the manuscript does have the word probatum omitted in 
Ro. In the other seven passages it is faithful to Ro, with some varia- 
tions found also in the other Psalters: thus at xxiv. 18 P reads non 
confundar, omitting domine, as do also A and B; and at lvii. 4, 
according to Tanger, P reads et ueneflci que incantantur a sapiente, 
with A, B, C, D, and E. 



TEE LATIN TEXT OF TEE PARIS PSALTER. 151 

minor change of order, the correct reading had already been 
supplied by Tanger. On another page (p. 469, note 4), Wild- 
hagen affirms that frequently P gives a mixed reading, blending 
Bo and Ga. The passages he cites as examples are ix. 24, xi. 3, 
xxvi. 5, xxx. 4, xlvi. 2. In every case the apparent mixture is 
the work of Thorpe; and again Thorpe's changes had already 
been corrected by Tanger. 

Wildhagen does not profess to have given a complete list of 
the Ga readings he has found in P. He reserves a complete 
treatment (p. 469) for the second volume, which has not ap- 
peared, of his edition of the Cambridge Psalter. But in the first 
volume, " Text mit Erklarungen," publisht in 1910, he has very 
frequently cited the Latin text of P, with equally unfortunate 
results. His references, which are all taken from Thorpe, are 
wrong in the following instances, 11 as may be seen by compar- 
ing them with the corrections made below : 

ii. 13; vii. 15; viii. 8; ix. 7, 13, 24, 25, 31; xiii. 6; xvi. 1; 
xvii. 5, 13, 33, 45; xviii. 9; xxi. 21; xxv. 7, 8, 9; xxviii. 9; 
xxix. 13; xxxi. 4; xxxiii. 15; xxxiv. 7; xxxvi. 21, 25, 36; 
xxxvii. 7, 14, 16, 20; xxxviii. 7; xxxix. 5; xl. 2, 3, 7; xlv. 3; 
xlviii. 8, 15, 16; xlix. 4, 9, 21, 23; 1. 9. For the second and 
third fifties the following should be compared with Tanger: 
lvi. 10; lvii. 6; lxvii. 8, 22; lxviii. 36, 37; lxix. 4; lxxi. 6; 
lxxxv. 17; lxxxix. 9; ciii. 2, 11, 19; cxvii. 8; cxviii. 171; 
cxxxi. 11 ; cxlvii. 18. Most of these are cases where, instead of 
having a Ga reading, as Wildhagen had gathered from Thorpe, 
P really agrees with Eo or with one or more of the other English 
copies of Eo. We even find Wildhagen gravely citing as P the 
substitutes which Thorpe provided, as he explains in the passage 
from the Praefatio quoted above, for the pages cut out of the 
manuscript (e. g., at xx. 7, 9, 13 ; xxxviii. 2 ; 1. 16, 18 ; lxvii. 36 ; 
etc.), — an example of laborious futility that almost justifies 
some of the strictures made of late upon Teutonic scholarship. 

But Wildhagen was not the first to rely blindly upon Thorpe 
and ignore Tanger. Wichmann 12 in 1889, in his study of the 

u The verse numbering followed by Wildhagen is that of Sweet in Ms 
edition of the Vespasian Psalter, and frequently differs from that of 
Thorpe. 

"J. Wichmann, "Konig Aelfred's angelsachsische Uebertragung der 
Psalmen I-LI excl.," Anglia XI (1889), p. 42 f. 



152 AMERICAN JOURNAL OF PEILOLOGY. 

authorship and character of the West-Saxon Psalms, attempted 
to decide the important question whether the Anglo-Saxon 
prose version is translated from the accompanying Latin text 
or not, hy citing nineteen evident discrepancies. Among them 
are twelve in which Thorpe's text is incorrect : 13 vi. 8 ; vii. 9, 
10 ; ix. 12 ; xxvi. 5 (two cases) ; xxvii. 1 ; xxxvi. 36 (two cases) ; 
xxxviii. 7, 9 ; xl. 2 ; and in at least nine of these when the cor- 
rect text is secured the discrepancy disappears. Similarly Dr. 
Bruce, 14 in comparing the Anglian Psalms with the Latin text 
for the same purpose, has cited twelve discrepancies, three of 
which (lvii. 4; cxv. 2; cxvii. 4) a consultation of Tanger would 
have removed. The conclusions drawn by Wichmann and Bruce, 
namely that neither West-Saxon nor Anglian Psalms are based 
upon the accompanying Latin, are both certainly correct, and I 
shall attempt later in this article to support them; but their 
reliance upon Thorpe is reliance upon a broken reed. In view, 
therefore, of the numerous cases in which scholars have been 
misled by Thorpe's so-called edition, 15 it seems worth while to 
publish a collation of his Latin text with the original manu- 
script. 

The present collation was made by the writer in the summer 
of 1905 at the Bibliotheque Rationale. It included both a veri- 
fication of Tanger's readings, 16 and an independent comparison 
of Thorpe's text with the original Paris Psalter. I have at- 
tempted to list all the variations from Thorpe, even those merely 
orthographical, except his constant substitution of v for u and 
j for i. Tanger, altho usually accurate, needs correction in a 
number of cases, and he is far from complete. Not only does 
he disregard smaller differences of orthography such as the 
variations between ae, se, e, and e, ch and c, h and ch, and most 
cases of the omission or addition of h, all of which, as Wildhagen 

13 Thorpe's verse numbering. 

"J. D. Bruce, The Anglo-Saxon Version of the Book of Psalms 
commonly known as the Paris Psalter. 1894, pp. 123-126 (PMLA IX). 

15 Thorpe is also mentioned, strangely enough, as an important source 
for the reconstitution of the Roman text, by A. Eahlfs, Der Septuaginta- 
Psalter, Septuaginta Studien, Heft 2, Gottingen, 1907; but Eahlfe 
makes no detailed citations. 

"Tanger's readings were independently verified during the same 
summer by Professor James W. Bright, whose assistance is hereby 
gratefully acknowledged. 



THE LATIN TEXT OF TEE PARIS PSALTER. 153 

has demonstrated, are significant for the exact determination of 
the relation of texts, but he has overlookt a considerable number 
of larger changes introduced by Thorpe. In all, for the first 
fifty psalms Tanger notes about 225 distinct alterations of the 
text, and I have added about 75, making some 300 in all; the 
number of merely orthographical changes which I have noted is 
considerably larger. For the sake of completeness, all of Tan- 
ger's readings are here reproduced (followed in each case by a 
T), with any corrections that seem necessary. 

The collation was limited to psalms i-1, or that portion of the 
Latin text that corresponds to the West-Saxon Psalms. If, as 
I believe will become evident, the Latin has only a fortuitous 
connection with either of the adjoining Anglo-Saxon versions, 
and merely offers another copy of the same general type of text 
as those found in the Vespasian, Junius, Eoyal, Badwine's, and 
Cambridge Psalters, which have already been publisht, it will 
hardly be worth any future editor's while to reprint it ; and per- 
haps the exact picture of the original which I trust this collation 
will supply for the first third of its extent will give sufficient 
basis for the determination of its relations to the other copies of 
the Psalterium Eomanum as used by the Anglo-Saxon Church. 

Ps. i. 1. " Von BEATUS ist B noch deutlich, E nur noch 
schwach sichtbar; von A ist wenig mehr als ein stuck des quer- 
striches ubrig; darauf folgt ein brauner quadratischer fleck, der 
etwa den raum eines T bedeckt; im ubrigen ist keine spur von 
anderen buchstaben erhalten" (T). habiit (T). cathedra 
(T) ; rather cathedra, pestilentie. 2. ac for last et. 4. decidet 
(T). quecunque. fecerit (T). 5. proici &. ventos. terre. 6. 
resurgunt (T). 

Ps. ii. 1. fremuerun (T). 2. adstiterunt. terre. 3. Dis- 
rumpamus. proiciamus. 4. celis. inridebit. 5. conturbauit (T). 
6. preceptum. Domini for last ejus (T). 8. terre. 9. figuli. 
10. intellegite. omnes before qui (T). 12. Adprehendite. 13. 
in eum (T). 

Ps. iii. 1. anime mee. 5. milia. 6. michi. 

Ps. iv. 1. iustitie m§§. me for mihi (T). 2. michi. 3. gra- 
vis (T). diligitis (T). queritis. 4. dum for cum. clamarem 
(T). 5. que for quae. & before in cubilibus (T). conpungi- 
mini. 6. iustitie. 8. letitiam. 



154 AMERICAN JOURNAL OF PHILOLOGY. 

Ps. v. 1. intellege. mee for meae. 3. adstabo. 5. locuntur. 
7. misericordie tue. 12. letentur. eternum. 13. bone. tue. 

Ps. vi. 2. michi for mei(T). 3. Et for Sed. Domine (sec- 
ond time) omitted (T). 5. singulos (T). 6. pre. 8. retror- 
sum after auertantur(T). 

Ps. vii. 1. Dne without sign of abbreviation (T). 4. miehi. 
5. conprehendat. earn after conprehendat(T). 6. tuorum for 
meorum (T). 7. Et omitted (T). 8. This verse ineludes iu- 
dica me domine from verse 9(T). 9. super me after mea- 
rum(T). 10. Consummetur. dirige(T). 

13. paravit (second time) omitted(T). eflecit(T). 14. par- 
turit(T). 15. effodit(T). incidit(T). 16. capite(T). de- 
scendit(T). 

Ps. viii. 2. celos. lactantium(T). 3. destruam(T). 4. 
celos. tuos omitted(T). quas(T). 5. aud for ant. 7. uni- 
uersa(T). 8. eeli. 

Ps. ix. 2. L&tabor. 3. perient(T). 4. aequitatem. 5. eter- 
num. seculnm saeeuli. 6. T reads def ecerun ; but ms. has -nt. 
framea(T). 7. eternum. 8. equitate. 12. Adnuntiate. ora- 
tionem for clamorem. 13. " St. ut hat der Cod. Put ; dock ist 
wol das goldene U nachtrdglich aus versehen dahin geraten, so 
dass es scheint, als ob hinter mortis ein neuer vers anfangt, im 
widerspruch zum ae., welches zu Thorpe stimmt." (T) ; no new 
verse is indicated here, filiae. 14. infixe. conprehensus. 15. 
conprehensus. 16. qui(T). 18. preualeat. 21. conprehendun- 
tur. 22. laudator (T). anime sue. iniqua(T). benedice- 
tur(T). 23. Irritabit. dominus(T). ire sue. inquir& for non 
quaeret(T). 24. pulluuntur. uie. eius for illius(T). 28. sedet 
omitted(T). 29. sicut for quasi. 30. adtrahit. "St. in laqueo 
hat ms. Jin; der fall ist hier derselbe wie oben v. 18 "(T) ; no 
new verse. humiliauit(T). inclinauit(T). 31. usque before 
in finem(T). 34. dolorum(T). in manibus tuis(T). 36. 
eternum. seculum seculi. 38. adponat. 

Ps. x. 1. anime mee. 2. tetenderunt(T). 3. que for quae. 
4. celo. 5. palpebre. 6. hodit for odit(T). 7. Pluit(T). et 
omitted before sulphur. 8. justitiam(T). equitatem. 

Ps. xi. 1. diminute. 2. Vana loeuti sunt unusquisque 

(mala omitted) (T). loeuti sunt mala(T). 3. maliloquam(T). 

4. est before dominus(T). 5. exsurgam. 7. terre. 8. eternum. 

Ps. xii. 1. VSQUEQuo for Quousque(T). quousque for 



THE LATIN TEXT OF THE PARIS PSALTER. 155 

usquequo(T). avertis(T). 2. animam meam(T). 4. " 4 um- 
fasst im Codex auch noch 5 bis adversus eum, ebenso ae. bis 
(>onne he; v. 5 beginnt also im Codex mit Qui und £>a; die 
beiden initialen fehlen jedoch "(T) ; Tanger seems wrong here: 
a new verse (no. 5) begins with Ne quando as in Thorpe, only 
the capitals have been omitted; v. 6 begins with Qui, and vs. 7 
with Exultabit. in tua misericordia. sperabo(T). 6. michi. 

Ps. xiii. 1. abhominabiles. 3. celo. 6. uelociter(T). 9. deum 
for Dominum(T). 10. deus for first Dominus(T). juxta for 
justa(T). confudisti. deus for Dominus (second time). 11. 
sue. 12. Letetur. 

Ps. xiv. 1. habitauit(T). 4. mala(T). proximum suum(T). 
5. nichilum. 7. hec. commovebitur(T). 

Ps. xv. 1. indies (=indiges?) for eges(T). 3. Multipli- 
eati(T). enim after sunt(T). 4. ero omitted(T). illorum 
for second eorum(T). 5. hereditatis mee. meis for mei(T). 
michi. hereditatem. 6. michi. preclaris. hereditas. preclara. 
michi. 7. michi. 8. ad for a(T). michi. nee for ne. 10. 
infernum(T). 11. michi. vite. letitia. 

Ps. xvi. 1. deprecationi mee(T). 2. equitatem. 6. michi. 
7. "7 umfasst im lat. auch noch v. 8 bis tue"(T). dextere. 
tu§. 8. adflixerunt. 10. Proicientes. 11. predam. 12. Ex- 
surge. preueni(T). framea inimicorum de manu tua(T). 
13. a terra dispartire eos et supplanta eos in uita ipsorum; 
" damit schliesst der vers; das folgende ist im Codex zu v. H 
gezogen"(T). impletum(T). 14. porcina(T). que for quae. 

Ps. xvii. 3. mef. et before adjutor omitted(T). 4. iniqui- 
tates(T). 5. preuenerunt. 8. exardescet(T). 9. celos. cali- 
go(T). 10. cherubin(T). 12. fulgora(T), carbones(T). 13. 
celo. Altimus(T). 16. ire tue. 17. adsumpsit. 18. hode- 
runt. 19. Preuenerunt. mee. 20. michi. michi. 22. reppuli. 

or or 

23. c, um (cum corrected to corum; T reads c, am). 24. 
michi. 26. facias. 27. inluminas. inlumina. 28. temptatione. 
29. inpolluta(T). uie(T). 30. preter. preter. 31. precinxit. 
uirtutem(T). inmaculatam. 33. prelium. posuit. ereum. 34. 
michi. 36. " Von cadent bis meos . . . im Codex zu v. 37 
gezogen"(T). 37. preeinxisti. 38. michi. hodientes. 40. illos 
for first eos (T). 41. capud(T). 42. eognouit(T). michi. ab 
for in(T). michi. 43. mentiti(T). michi. 44. mee for meae. 
45. vindictam(T). 



156 AMERICAN JOURNAL OF PHILOLOGY. 

Ps. xviii. 2. eructuat(T). 3. loquele. 4. terrae. 6. gi- 
gans(T). cek. se. 7. prestans. iustitie. recte. letificantes. pre- 
ceptum. inluminans. " Von justitiae bis oculos . . . ist im 
Codex zu v. 8 gezogen"(T) ; Tanger is wrong: a new verse 
begins with Timor, as in Thorpe, but the gold of the T is faint. 
8. " permanens : von ns ist nur der erste grundstrich des n noch 
schwach sichtbar, dahinter eine Meine teclce"(T) ; ms. has per- 
manet. seculum seculi. 10. eustodi&(T). ea for illa(T). in 
custo ilia (sic) (T). 11. domine after me(T). 12. inmacu- 
latus. 

Ps. xix. 2. santo(T). 3. omnes(T). 5. Letabimur. 6.1m- 
pleat for faciat(T). faciat for fecit (T). celo. dextere. 7. Hii 
for hi (twice)(T). 8. resurreximus(T). 

Ps. xx. 1. letabitur. exultauit. 2. anime. voluntatem. 3. 
preuenisti. 4. in seculum et omitted (T). 5. " Mit est gloria 
. . . schliesst fol. 20 "(T). 

Ps. xxi. 1. "Die worte Verba delictorum meorum beginnen 
im Codex v. 2"(T). 2. michi. 5. obprobrium. 8. mee. me§. 
13. adhesit. 14. concilium. 17. aspice. 18. Deus omitted{T). 
20. fratribus. ecclesie. 21. magnificate for glorificate(T). 23. 
aecclesia. 24. "&4 umfasst im Codex auch noch 25 bis se- 
culi"(T). 25. universe, terrae. 26. patrie. 27. terre. 29. 
Adnuntiabitur. adnuntiabunt. celi. 

u 

Ps. xxii. 1. nichil. michi. pascue. 3. semitam(T). i,stitie. 
4. umbre. es(T). 7. Inpinguasti. preclarum. 8. uite mee. 

Ps. xxiii. 1. ea for eo. 2. maria. eam(T). illam(T). 3. 
ascendit (T) . 5. domino for Deo (T) . deo for Domino (T) . 6. 
Hec. querentium. 7. T reads porte for portas wrongly; ms. 

has portas. porte for postes. eternales. glorie. 8. Quis, iste 
for Quis est iste. glorie. prelio. 9. porte eternales. glorie. 
10. glorie. glorie. 

xxiv. 2. inrideant. 3. michi. 5. Eeminiscere miserationum 
tuarum, domine, et misericordie tue, que a seculo sunt(T). 
6. iuuentutis mee. ignorantie mee. " Hint er memor esto mei 
hat Codex noch deus, womit dieser vers schliesst; der rest des- 
selben bei Thorpe bildet im Codex einen selbstandigen vers, dem 
im Codex Thorpe's ae. v. 7 gegenubersteht. Thorpe's lat. v. 7 
ist im Codex mit Thorpe's ae. v. 8 bis eac rihtwis zusammen- 
qestellt . . . Thorpe's lat. v. 8 steht zusammen mit dem rest 
seines ae. v. 8 von Ealle bis lufiaS; . . . Thorpe hat die hier 



THE LATIN TEXT OF TEE PARIS PSALTER. 157 

vorliegende unordnung trotz seiner willkurlichen dbweichung 
vom ms. nicht beseitigt; sie ist jedoch leicht zu heben, wenn 
man die ersten drei zeilen seines ae. v. 8: ' For jnnre godnesse, 
drihten, pu eart swete, and wynsum, and eac rihtwis ' dem vers 
6 anschliesst, wie Thorpe das im lot. mit ' propter bonitatem 
tuam, domine : dulcis et rectus dominus ' getan hat. In vers 7 
ist seine anordnung richtig, und von v. 8 db herrscht auch im 
Codex wieder ordnung."(T). The confusion may be more clearly 
described as follows: in the Latin v. 7 should begin with Propter 
bonitatem tuam, v. 8 with Propter hoc legem, and v. 9 with 
Uniuerse me Domini; in the Old English Thorpe's verse divi- 
sion correctly reproduces the ms., except that v. 9 should begin 
with Ealle Godes wegas; and the correspondence in sense is as 
follows: Latin v. 7 = Old English v. 8, Latin v. 8 = Old Eng- 

o 

lish v. 7, Latin v. 9 •= Old English v. 9. 13. ms. has laque, 
pedes (o inserted in corrector's black ink). 14. "Ik- und 15 
scheinen im Codex nur einen vers zu bilden, doch fehlen die 
initialen von Tribulationes und von And"(T) ; ms. has ribula- 
tionis, but clearly a new verse is intended. 15. dilatate. 16. 
omnia peccata for universa delicta(T). 17. hodio. hoderunt. 
18. confundar for erubescam(T). 19. adheserunt michi. 

Ps. xxv. 1. infirmabo. 2. tempta. 4. concilio. 6. circuibo 
for circumdabo (cf. Ps. xxvi. 7) (T). 7. laudis tue. et 
omitted(T). 8. domus tue. tabernaculis(T). glorie tue. 9. 
Deus omitted. "Mit iniquitates sunt . . . schliesst im Codex 
fol.86"(T). 

Ps. xxvi. 1. inluminatio. 2. defensor for protector (T). uite 
mee. 3. innocentes for nocentes(T). 4. proelium. 5. uite 
mee. "Die liicke im ae. v. 5 ist im Codex nicht bezeichnet, 
vielmehr steht And geseon etc. in gleicher hohe mit Unam etc. 
Der schluss von ut videam bis ejus fehlt im Codex "(T). 7. 
capud(T). 10. Et omitted before ne auertas(T). et present 
before ne declines. 12. adsumpsit. 13. michi. semita recta (T). 
14. persequentium for tribulantium(T). mentita(T). 15. ms. 

ouid 

has Cred, ere. " 15 und 16 lat. und ae. bilden im Codex nur 
einen vers"(T). 

Ps. xxvii. 1. nequando taceas a me omitted(T). 2. mee. 5. 
nequitia(T). secundum opera manuum eorum tribue illis 
omitted(T). 7. Destrue(T). edificabis. mee. 9. sue. 10. 
hereditati tue. usque before in seculum. 



158 AMERICAN JOURNAL OF PHILOLOGY. 

Ps. xxviii. 1. adferte (twice). 2. adferte (twice). 6. soli- 

tudinem for desertum. 7. preparantis. reualauit(T). 8. 

habit&(T). eternum. 9. uirtutum(T). Dominus omitted, et 

before benedicet. 
Ps. xxix. 3. memorie. 4. in omitted before indignatione(T). 

5. letitia. 6. in meahabundantia(T). eternum. 7. prestitisti. 

8. Que. 9. adnuntiabit. 10. michi for J mei(T). 11. michi. 

precinxisti. letitia. cantem(T). 12. eternum. 
Ps. xxx. 1. eternum. 2. " Inclinadme rait trennungsstrichen 

vor und hinter ad"(T); I found no marks visible before or 

after ad as per T. 3. michi. 4. fortitudo mea et omitted(T). 

et refugium meum after firmamentum meum(T). michi. 5. 

michi. 7. letabor. 8. me omitted. manus(T). 9. michi. 13. 

obprobrium. 19. Inlumina. 20. que. in omitted before con- 

temptu(T). 21. tue. 24. circumstantie. 26. mee. 27. ha- 

bundantur(T). 

Ps. xxxi. 1. remisse. 4. erumna. confringitur for configi- 

tur(T). spina (T). 7. orauit. adproximabunt. 8. michi, pre- 
sura(T). que. 10. equis(T). 13. Letamini. 
Ps. xxxii. 2. cordarum. ei for illi. 4. rectum. 5. celi. 6. 

thesauros(T). 8. Quoniam for Quia(T). 10. eternum. 11. 
hereditatem. 12. celo. preparato. 13. singillatim(T). in be- 
fore omnia. 14. gigans. sue. 15. habundantia(T). sue. 16. 
"Hinter eum ist im Codex ein punht, und dahinter heisst es: 
sperantes autem in misericordia eius " (for et in eis, qui sperant 
super misericordia ejus) (T). 17. "Hinter noster est im 
Codex ein punht; dahinter: & in ipso letabitur u. s. w."(T). 

Ps. xxxiii. 2. letentur. 3. in inuicem. 4. Inquisiui(T). 5. 
inluminamini. 7. Inmittit(T). 9. nichil. 12. uidere(T). 
13. Coibe. 14. Deuerte. sequere for persequere. 18. his for 
iis(T). 19. Multe. 21. hoderunt. 22. in eum(T). 

Ps. xxxiv. 1. inpugnantes. 2. Adprehende. exsurge. michi. 
3. persecuntur. anime mee. 4. querunt. 5. michi. 6. adfli- 
gens(T). 7. uie. tenebre. 8. michi. in before interitum(T). 
supervacue omitted(T). 9. T reads ign6rat, but ms. has ig- 
norant. laqueo(T). incidant(T). in idipsum(T). 10. exul- 
tauit(T). 11. inopum(T). egenum(T). 12. Exsurgentes. 
que. michi. et after bonis (T). sterelitatem. anime mee. 13. 
dum for cum(T). michi. et before humiliabam. 14. ita for 
sic (first time), tamquam for quasi, ita for sic (second 



THE LATIN TEXT OF THE PARIS PSALTER. 159 

time)(T). 15. letati. ignorauerunt(T). 17. eorum after male 
faetis(T). 19. michi. hoderunt. annuebant(T). 20. michi. 
22. Exsurge. 23. insultent in me for supergaudeant mihi(T). 
anime nostre. 24. pudore omitted after erubescant(T). pudore 
for confusione(T). 

Ps. xxxv. 3. intellegere. 4. Adstitit. ui§. bone, hodiuit. 
5. celo. 8. tue (twice). 9. aput. uite. 10. Pretende. 11. 
michi. superbie. 12. omnes omitted. 

Ps. xxxvi. 1. emulari. emulatus. 2. tamquam for sicnt(T). 
fenum. sicut for quemadmodum(T). holera(T). 6. tamquam 
for quasi. 8. emuleris. 9. hereditatem. 10. et non for nec(T). 
queris(T). nee for et non (T). 12. fremit(T). inridebit. 13. 
et omitted before tetenderunt(T). deiciant. inopem(T). truei- 
dant(T). 14. conteretur(T). 17. inmaeulatorum. hereditas. 
eternum. 20. soluet for eommodat(T). commodat for tri- 
buit(T). 22. nimis. 24. Juvenior(T). et for etenim(T). 
25. commodat(T). 27. aeternum. 29. J of Justi omitted^). 
hereditatem. seculum. 32. querit. dampnabit. 34. libani(T). 

35. Et omitted oefore transiui(T). et omitted oefore quesiui. 

36. ueritatem for innoeentiam(T). equitatem for veritatem(T). 
reliquie. hominum(T). 37. reliquie. 39. eripiet for eruet(T). 

Ps. xxxvii. 2. sagitte tue. infixe. michi. 3. ire tue. et oefore 
non(T). 4. mee. honus. grauate. 5. Conputruerunt. cica- 
trices (T). mee. insipientie mee. 6. turbatus for curvatus(T). 
7. conpleta. inlusionibus. 8. " Vv. 8 and 9 bilden im Codex 
nur einen vers"(T). 9. et for Domine(T). 11. adpropia- 
uerunt for adpropinquaverunt(T). 12. querebant. michi. 13. 
uel ut for tamquam (T). sicut for velut(T). aperuit(T). 14. 
ut for velut(T). 15. me omitted. 16. ne aliquando for Ne- 
quando(T). insultent for supergaudeant (T). 17. ad for in(T). 
19. Initial I missing(T). uiuent(T). hoderunt. 20. michi. 
michi. 21. me§. 

Ps. xxxviii. 6. " Nach dem ae. argument ist das letzte funftel 
von fol. 45 v unbeschrieben. Auf fol. Jf6 r fdngt das Lat. an mit 
tas omnis homo, das Ae. wie bei Thorpe, nur dass die McJcen- 
bezeichnung im Codex fehlt. Soweit v. 6 iiberhaupt vorhanden, 
ist er im Codex mit v. 7 zusammengeschrieben " (T) . 7. in 
omitted(T) ; where in should be there has been an erasure. 
imaginem(T). dei after imaginem. conturbabitur(T). 8. " Vv. 
8 und 9 sind ebenfalls im Codex zusammengeschrieben "(T). 



160 AMERICAN JOURNAL OF PHILOLOGY. 

Thesaurizat for Congregat(T). ignorat(T). congregat for con- 
gregabit(T). 9. que. nichil. ante for apud(T). 10. obpro- 
briiim. 11. T reads tua for tuas wrongly; ms. has tuas. tua 
for tuae. 13. Uerumptamen. " Vv. IS und 14 im Codex zusam- 
mengezogen"(T). 14. lacrimas. 15. aput(T). 16. miehi. 
priusquam. 
Ps. xxxix. 1. miserie. feeis. 2. inmisit. 4. uanitate(T). 

6. adnuntiaui. michi. 9. aecclesia. "Die worte Domine tu 
cognovisti bilden im Codex den anfang von v. 10 "(TU). 10. 
The ms. has a raised dot after Justitiam tuam, and no mark of 
punctuation after corde meo. abscondi for eelavi(T). 11. celaui 
for abscondi(T). 12. semper omitted. 13. conprehenderunt. 
meae. 15. Conplaceat. eripias(T). 16. querunt. 17. miehi. 
18. michi. 19. letentur. querunt. 

Ps. xl. 1. intellegit. liberauit(T). 2. faciet(T). 3. uniuersi 
strati {Thorpe's Note). 5. michi. periet(T). 6. si omitted (T). 
The ms. has no mark after viderent, out a raised dot after vana. 

7. " 7 umfasst im Codex von v. 8 auch noch die worte: in unum 
susurrabunt; ae. wie bei Thorpe "(T). 8. susurrabunt(T). 
michi. 9. adicife. meg. edebant(T). subplantationem. 10. illis 
for eis(T). 11. quoniam for quia(T). quia for quoniam(T). 
12. eternum. 

Ps. xli. 2. fontem omitted (T). 3. michi lacryme me§. michi 
cotidie. 4. Haec. 8. cataratarum(T). 10. uite me§. 11. 
reppulisti. tristis for tristatus(T). adfligit. 12. omnia before 
ossa(T). michi. 

Ps. xlii. 2. reppulisti. adfligit. 4. letificat. 

Ps. xliii. 1. adnuntiauerunt. 3. Initial M missing (T). adflix- 
isti. et omitted before expulisti. 5. inluminatio. complacuit tibi 
(apparently altered from complacuisti). 9. adfligentibus. hode- 
runt. 11. reppulisti. 12. pre. nos hoderunt for oderunt nos. 
diripuebant(T). 15. in before obprobrium. 19. Haec. 20. adflic- 
tionis. 23. estimati. 24. Exsurge (twice). 25. obliuiscens for 
oblivisceris. 26. adhesit. 27. Exsurge. 

Ps. xliv. 2. scribe. 3. pre. V. k in the ms. begins with propterea 
benedixit(T). aeternum. 7. Sagitte tue acute, in omitted before 
corda(T). 8. Sedis(T). et omitted. 9. hodisti. letitie pre. 10. 
gradibus for domibus. filie. 11. ad dextris. de aurata(T). 13. 
filie. 14. filie. 15. fimbreis(T). " Von Adducentur und von 
Eala kyning ab im Codex zuv. 16 gezogen "(T). 16. Afferentur 



TEE LATIN TEXT OF THE PARIS PSALTER. 161 

omitted(T). letitia. 19. asternum (digraph used here for first 
time). 

Ps. xlv. 1. que. 2. conturbabuntur(T). 3. turbate. aque. 4. 
letificat. " 4 umfasst auch noch non eommovebitur von v. 5"(T). 

5. Conturbata. 7. que. 8. terre. eonburet. 

Ps. xlvi. 2. Bex magxms super omnem terram omitted(T). 
4. he hereditatem [sic](T). 5. tube. 7. terra. 9. ms reads 
conguenerunt, with the g partially erased between n and u; did 
the scribe start to write the Gall, congregati sunt ? dii. terre. 

Ps. xlvii. 2. exultationis uniuerse terre. syon. latere. 3. dinos- 
eitur. 4. terre. 5. adprehendit. 7. eternum. 8. medio omit- 
ted(T). 9. terre. 10. Letetur. syon. filiae Jude. 11. syon. 
12. Deus omitted (second time), eternum. 

Ps. xlviii. 1. hee. 2. terrigene. 4. aurem after similitudinem. 

6. multitudine omitted, in habundantiarum. 7. redemit [twice] 
(T). anrme sue. et for nec(T). laborauit(T). eternum. 8. 
morientes(T). 9. sepulchra. eternum. 12. Hec. 13. posita(T). 
depaseit(T). 14. matutina(T). ueterescent(T). a omitted before 
gloria. 15. Verumptamen. liberauit(T). 17. haee. descendit(T). 
19. eternum. 

Ps. xlix. 2. exion for ex Sion(T). 5. uocauit(T). celum. 7. 
celi. 8. quoniam before first Deus(T). 10. tuo for tua. hyreos. 
11. mee. fere. 12. celi. 13. terre. 14. hyreorum. 16. tue after 
tribulationis(T). 18. hodisti. T reads poste for postea, wrongly; 
ms. has post te. 20. T reads habundauit for abundauit; but 
ms. has habundabit. nequitiam(T). 21. tue. 22. Haec. iniqui- 
tatem for inique(T). tibi for tui(T). 23. illam after statuam. 
Intellegite haee. 24. T reads honoroficauit, but ms. has honorifi- 
cauit. in before quo(T). 

Ps. 1. 7. sapienti§ tue. michi. 8. hysopo. 9. letitiam. " Mit 
humiliata und blissian schliesst fol. 63 des Codex" '(T). "fol. 6Jf 
fangt an mit adjutorem sibi (Th. Ps. li, v. 6, mitte), welches in 
gleicher hohe steht mit den ae. anfangsworten fore aenigre"(T). 

For the rest of the Psalter I have collated only Tanger's notes 
on the loss of pages from the manuscript. 

Ps. lxvii. 28. " Codex fol. 79 schliesst mit gentes que . . . 
Zwischen diesem und dem folgenden blatte sind noch geringe 
spuren eines herausgerissenen blattes zu bemerken." '(T) . 

Ps. lxxix. 18. " Mit uirtutum . . . schliesst im Codex fol. 97. 
Der rest eines blattes vor fol. 98 ist noch sichtbar " (T) . 



162 AMERICAN JOURNAL OF PHILOLOGY. 

Ps. lxxx. 8. "gif J?u etc. steht im Codex Thorpe's lat. v. 9 
gegenuber, mit welchem fol. 98 r anfangt"(T). 

Ps. xcvi. 1. " Mit multe und gar ssecge schliesst im Codex fol. 
113; dahinter ist keine spur mehr von einem fehlenden blatte 
vorhanden " (T) . 

Ps. xcvii. 8. "Mit manibus und stundum beginnt im Codex 
fol. lip "(T). 

Ps. cviii. 30. " Dieser psalm schliesst db mit fol. 132, welches 
sellst eingeschnitten ist; fol. 133 fdngt an mit ruinas conquas- 
sauit [Th., ps. cix, v. 7] "(T). 

Ps. cl. 3. "Mit tube*lau und ae. beman schliesst im Codex 
fol. 175 v . Dieses, sowie das vorige blatt dicht am riichen weit 
eingeschnitten. Hinter fol. 175 ist ein olatt herausgeschnit- 
ten"(T). 

The following table will make clearer the facts about these 
losses. They occur at nine places in the manuscript, and involve 
the loss of the following sections of the text : 

1. After fol. 20 : xx. part of 5-13, the W-S. Introduction, and 
xxi. part of 1 (in all, about 9 verses and an Intro.). 

2. After fol. 26: xxv. part of 9, 10, 11, and Intro, to xxvi 
(2 + verses and an Intro.). 

3. After fol. 45 : xxxviii. 1-part of 6 (5 + verses). 

4. After fol. 63: 1. 10-20, li. 1-part of 6 (16 + verses). 

5. After fol. 79: lxvii. part of 28-31 (3 + verses). 

6. After fol. 97: lxxix, part of 18; lxxx. 1-8 (8+ verses). 

7. After "fol. 113 : xcvi. 2-12, xcvii. 1-part of 8 (18 + verses). 

8. After fol. 132: cix. 1-part of 7 (6 + verses). 

9. After fol. 175 : cl. part of 3-5 (2 + verses). 

Besides the losses in the text there is no doubt that some of 
the missing pages contained additional matter, chiefly illustra- 
tions, which furnisht the motive for their excision. The posi- 
tion probably held by these illustrations reveals the scheme of 
division that was adopted in the Paris Psalter, a matter the 
importance of which Wildhagen (p. 424 f.) was the first to 
emphasize. By comparing the amount of text on the lost folios 
we can estimate which ones had space left for illustrations. A 
count of several sections of the psalter shows that the average 
folio page held, recto and verso together, about 12-13 verses. 
In the first case listed above, the missing matter would have 



TEE LATIN TEXT OF THE PARIS PSALTER. 163 

filled easily both sides of the page, and it is unlikely that any 
illustration was present, the more so since none of the other 
Anglo-Saxon psalters divide at this point. This page was prob- 
ably lost by accident. In cases 2, 3, 5, 6, and 8, on the other 
hand, there was enough text to fill only one side of the page, 
the other side being doubtless occupied by the dividing illustra- 
tion. Cases 4 and 7 must each have involved the loss of two 
folio pages, as already remarkt by Bouterwek (see Wichmann, 
p. 41) ; hence naturally, as Tanger notes, no trace of the excision 
is left. The missing text would have filled three sides, leaving a 
fourth for colophon, title, etc., between the West-Saxon and 
Anglian Psalms after ps. 1, and an illustration after ps. xcvi. 
In the last case, the end of ps. cl and perhaps a colophon prob- 
ably occupied the recto of the missing page, and the verso may 
have contained one of the canticles. In the manuscript the next 
folio begins with the " Canticum Ezechie," which is elsewhere 
always preceded by the " Canticum Esaie Prophete," or, as it is 
otherwise called, the " Confitebur tibi." The " Confitebur tibi," 
which as Wildhagen notes (p. 469, note 6) is unaccountably 
missing from the Paris Psalter collection of liturgical pieces, 
has but 6 verses; and it is therefore not unlikely that it was 
originally present on the verso of the lost folio. 

The total loss was thus probably eleven folio pages. The 
points of division markt were before pss. xxvi, xxxviii, lxviii, 
lxxx, xcvii, and cix. Nowhere else, unless possibly before ps. li, 
is there any indication of division either in the Latin or the 
Anglo-Saxon texts. 

The conclusions to be drawn from an examination of the 
genuine Latin of P as above restored may be briefly indicated. 

1. The Latin text of which the Paris Psalter furnishes a late 
copy (first half of the eleventh century) belongs not late, but 
fairly early in the succession of Anglo-Saxon psalters based on 
the Eoman version. As a specimen of this version in its special 
Anglican form, it is most nearly allied to the Eoyal and Bos- 
worth Psalters of the early tenth century, but seems to be earlier 
than either, and has some features that connect it with the 
Vespasian Psalter of the early eighth century. 

Thorpe's alterations of the original are in large majority 
(about 250 of the 300) merely substitutions of the Vulgate, 



164 AMERICAN JOURNAL OF PHILOLOGY. 

which of course usually means the Gallican reading. With these 
spurious Ga readings removed, the text assumes a very different 
appearance from that which it presented to Wildhagen. A 
limited number of Ga readings, however, remain to be con- 
sidered. They fall into three groups : 

(a) Ga readings found also in all or several of the other 
psalters of the Roman type (A, B, C, D, and E only are avail- 
able in publisht form). On Wildhagen's theory these must be 
counted as part of the original stock of Ga readings found In 
the primitive Anglican text. The following cases for the first 
fifty psalms are collected from Wildhagen's notes to the Cam- 
bridge Psalter, after eliminating the mistakes into which he was 
led by Thorpe (Thorpe's verse numbers when different are added 
in each case in parenthesis) : vii. 13(12), 16(15) ; ix. 36(35) ; 
x. 8;xiv.5(6) ; xvii. 3 (2) ;xxi.l8(15) ;xxvi.3(4) ;xxvii.9(10) ; 
xxxi. 4; xxxiv. 13, 15; xxxvi. 21(20), 23(22); xliv. 5; 
xlvi. 9(8);xlix. 3. 17 

(b) Ga readings found only in the Latin of C and P. 
Wildhagen (p. 466, note 2) gives seventeen cases, which he con- 
siders proof of special later Ga influence on these two psalters. 
(" Teilweise finden sie sich zwar auch bei einem der Kirchen- 
vater. Da aber samtliche iibrigen englischen Texte ... an 
diesen Stellen geschlossen der Vorlage treubleiben, und anderer- 
seits nur der Pariser Psalter, der der neuen Fassung im latei- 
nischen Teil die weitesten Konzessionen macht, mit dem Cam- 
bridger Texte zusammengeht, so kann hier einzig und allein 
Beeinflussung durch das Psalterium Gallicanum vorliegen.") 
As a matter of fact, only nine of these readings actually occur 
in P, three in the first fifty psalms: xiii. 7(11) ; xvii. 7(5) ; 
xliv. 6(7), and six in the rest of the psalter: lxx. 22(20); 
lxxxix. 13(15), 17(19) ; cii. 3(2) ; cxi. 7(6) ; cxxxi. 11. There 
is no apparent reason why these cases, like the similar coinci- 

»Cf. the partial list given by Wildhagen, p. 421, note 1. To these 
clear departures from Ko may be added the following, which, tho 
departing from the standard Ko text as given in Migne, Pat. hat. xxix, 
are found in the Psalterium Romanum of Jac. Faber Stapulensis, 
Quimcuplex Psalterium, Paris, 1513 (denoted by Wildhagen Ro 1 ) : 
iii. 6(4); ix. 26(24), 33(32); x. 4(3); xvii. 23(22), 33(31), 34(32), 
40(37); xviii. 10(8); xx. 5(4) ; xxi. 3(2) ; xxvii. 4(5) ; xxxvii. 12(11) ; 
xxxix. 15(16); xliv. 9(10), 10(11), 11(12). 



TEE LATIN TEXT OF THE PARIS PSALTER. 165 

dences that occur between P and the earlier psalters, 18 should 
not be treated exactly as the cases listed under (a). 

(c) Ga readings found only in P. A comparison of the 
genuine text of P with Lindelof's convenient parallel edition 
of ten psalms from eleven psalters 19 shows P adhering closely 
to the text of the five Eoman psalters collated (A-E). Of the 
very numerous variants from these noted by Lindelof in the 
text of his six Gallican psalters (F-K), P agrees with F-K 
against A-E in only five instances: vii. 17(16); ix. 18(16), 
19(17); lxxxix. 11(13); cxxxvi. 7(9). 20 So small a total of 
peculiar Ga readings may be paralleled in the other psalters 
without difficulty. Most of them no doubt are part of the 
original readings that have happened to be eliminated from all 
other surviving copies. Some of them might possibly be elimi- 
nated from P by a further scrutiny of the manuscript, especially 
in the second and third " fifties." The rest are probably due 
to the latest copyist of the manuscript ("Wulfwine or his succes- 
sor? cf. Bruce, p. 10 f.). That Wulfwine, altho he copied a 
Eoman text, was familiar with the Gallican, which he probably 
knew by heart, is indicated by several slips that he has himself 
tried to correct : e. g. xvii. 24(23), cum (Ga) corrected to corum, 
for Eo coram ; xxxii. 4, Quoniam rectum est sermo domini (Eo, 
rectus ; scribe evidently thinking of Ga Quia rectum est uerbum 
domini; Thorpe changes rectum to rectus); xlvi. 10(9), Con- 
guenerunt, with g partially erased (Eo, conuenerunt; Ga, con- 

M Cf. the Ga readings found in PB: xxxix. 5(4); in PE: xxxiii. 
21(20) ; in PD: xxxvi. 14(13) ; in PAB: x. 8; in PDE: xlviii. 12(9) ; 
in PABDE: xxxiv. 13 and xxxv. 12(11) ; in PAC: Ivi. 5(4) ; in PCE: 
xxxvii. 4(3) and lxxxix. 2. 

"Uno LindelSf, Studien zu altenglischen Psalterglossen, Bonner 
Beitrage XIII, 1904. 

20 In the remaining psalms of the first fifty I have noted the following 
seventeen cases where P agrees with the Vulgate (hence presumably 
with Ga) against EoABCDE: v. 4(2), et om. before exaudies; vi. 7(5), 
singulos for -as; xi. 7, probatum before terrae; xiv. 5(6), innocentes 
for -em; xvi. 2, tui for mei after oculi; xvii. 16(15), terrarum for 
terrae; xvii. 21(20), puritatem for innocentiam; xxi. 12(9), quoniam 
for et before non est; xxi. 26(23), laus mea for laus mihi; xxi. 28(25), 
uniuersae for -si; xxxv. 13(12), omnes om.; xxxvi. 35(34), sicut for 
super; xlii. 5(6), adhuc before confitebor; xlvi. 5(4), in om. before 
hereditatem; xlvi. 10(9), populorum for populi; xlviii. 9(7), et for nee 
before pretium; xlix. 22(23), deum for dominum. 

5 



166 AMERICAN JOURNAL OF PHILOLOGY. 

gregati sunt) ; xliii. 4(5), complacuittibi, altered from compla- 
cuisti (apparently the scribe first wrote the Ga complacuisti, 
then altered the last three letters to form the Ro complacuit 
tibi). Other slips into the Gallican of the same sort may well 
have escaped his notice. It is surprizing that there are not 
more of them. 

Indeed, the Paris Latin text preserves the features of the 
primitive Anglican type with remarkable fidelity, rivaling in 
this respect even the Vespasian Psalter. Wildhagen has listed 
(on p. 421) 94 test passages, departures from the regular Eoman 
version which he believes were found in the original common 
source of the eight English psalters. All but nine of these 
(xxxix. 5(4) ; xliii. 23(24) ; xliv. 9(10) ; lxviii. 36(37) ; lxxri. 
13(10) ; lxxviii. 4; ciii. 32(30) ; cvi. 10(9) ; cxlii. 10(11)— are 
retained in P, aside from three others that happen to fall on the 
missing pages. Ten of them have disappeared from the Latin 
of the Vespasian Psalter. Twenty-eight of these peculiar read- 
ings have been eliminated in Thorpe's edition. 

Another group of peculiar readings are of special significance 
because they establish a connection between the Latin text of P 
and the psalters of the tenth century. Wildhagen (p. 452) cites 
a group of ten readings, neither Ro nor Ga, found in the Royal, 
Bosworth, Eadwine's, and Cambridge Psalters, but not in the 
earlier Vespasian or Junius. With some reason he finds in the 
introduction of these the effect of the rising tide of Benedictine 
influence which began to be felt in the first half of the tenth 
century. Seven of these are found also in P: xxv. 7; xxxi. 4; 
xxxiii. 15(14); lxii. 7(6), 11(8); lxvii. 19; lxviii. 16(15). 
Wildhagen sees influence from the Benedictine liturgy also in 
the numerous cases of the insertion of " Domine" in DLEC ; of 
this P has no less than nine examples : iv. 2 ; viii. 6 ; xxvii. 2 ; 
xxx. 5 ; ci. 14; cxviii. 4, 49, 142, 165. At the same time it must 
be noted that P does not go so far in this direction as the other 
psalters of the group: cf. the readings at xvii. 24, xxi. 9, and 
lvii. 2, found in DLEC, but not in P,— a fact that suggests that 
P dates from a little earlier period in the movement. 

Two other features of the Paris Psalter Latin text bear out 
these indications of its early character. They are the appended 
collection of liturgical pieces, and the system of psalter division. 



TEE LATIN TEXT OF TEE PARIS PSALTER. 167 

The liturgical matter regularly appended to the psalters 21 
affords, as Wildhagen has shown, a valuable criterion of date. 
Oldest and most general are the seven canticles from the Old 
Testament sung at matins, one for each day in the week. These 
alone are found in the seventh century Salaberga Psalter, the 
oldest surviving psalter of English origin; and they begin the 
collection in each of the others. The Vespasian adds the Bene- 
dictus and Magnificat, which came later into use for daily 
matins and vespers. The Eoyal and Bosworth add the Nunc 
Dimittis, used daily at compline, which was of course the last 
of the daily services to come into general use. The Eoyal has 
also the " Quicumque vult " and " Gloria in excelsis," the Bos- 
worth the " Quicumque vult " and " Te Deum," these three 
hymns, according to Wildhagen, having been introduced into 
England first in the tenth century under Benedictine influence. 
Eadwine's and the Cambridge Psalter are the most comprehen- 
sive : they both have all the thirteen pieces mentioned and also 
the Apostle's Creed, the spread of which in England was likewise 
connected with Benedictinism. The Paris Psalter has ten pieces 
(or eleven, if we assume that the hymn " Confitebur tibi " was 
originally present on one of the lost leaves, as suggested above). 
Like the Vespasian, it has the original seven Old Testament 
pieces, with the Benedictus and Magnificat; like the Bosworth, 
it adds the "Te Deum," and like Eoyal and Bosworth, the 
" Quicumque vult " ; but it has neither the compline hymn 
" Nunc Dimittis " nor the Apostles' Creed. It would thus seem 
to occupy a position after the Vespasian and before the Eoyal 
and Bosworth Psalters, — the same position which we have seen 
to be suggested by the textual readings. 

A similar result is obtained by comparing the systems of 
psalter division indicated by the position of the illuminations 
in the manuscripts. The meaning of these facts was first revealed 
by Goldschmidt, 22 and first applied to the English psalters by 
Wildhagen (p. 423 f.). Briefly outlined, the results are as 
follows : Four systems of division are found in the psalters that 

21 Found in all nine of the Roman psalters except the Junius and 
Blickling, from which manuscripts the latter portions are missing. 

'•Adorph Goldschmidt, Her Albani-Psalter in Eildesheim und seine 
Beziehung zur symbolischen Kirehenseulptw des XI. Jahrhtmderts, 
Berlin, 1895. 



168 AMERICAN JOURNAL OF PHILOLOGY. 

we have been considering. The simplest, and perhaps the oldest, 
is the two-fold division before ps. cix, between the portions of 
the psalter used in the nocturnal and diurnal services of the 
breviary. Equally old in England, and always combined with 
the first, is the Eoman system, which markt with special promi- 
nence the seven psalms used at vigils in the Eoman Office for 
each day of the week: viz., pss. i, xxvi, xxxviii, lii, lxviii, lxxx, 
and xcvii; sometimes pss. xvii and cxviii were also distinctly 
markt for similar liturgical purposes. Third, the Irish division, 
so called because it seems to have originated in Ireland and 
spread thence wherever the influence of Irish missionary effort 
extended, gave prominence to the beginnings of pss. i, li, and ci, 
thus dividing the psalter into three equal " fifties." Finally, the 
Benedictine system, resting upon the highly developt Benedic- 
tine liturgy, markt no less than sixteen points in the psalter: 
i, xx, xxvi, xlv, lix, lxviii, lxxiii, lxxxv, xcvii, cxviii, exix, exxxiv, 
cxxxviii, exli, cxliv. 10, cxlviii. In the south of England we 
find the Roman system originally prevailing, along with the 
division at ps. cix; and this is exemplified in the Vespasian 
Psalter. In the north, naturally, the Irish method was as old 
or older; and so in the two early northern psalters, Salaberga 
and Blickling, the Eoman and Irish systems appear united. At 
a later period the Irish system came into general use thruout 
England, introduced mainly, Wildhagen thinks, from the Conti- 
nent, where it had become entrenched by the influence of the 
great monasteries of Irish origin. The Junius Psalter has both 
Eoman and Irish marking, as well as special prominence for ps. 
cxix; the latter feature being perhaps a first sign of Benedictine 
influence. The other four psalters of our group agree in drop- 
ping the Roman system altogether. The Eoyal and Cambridge 
Psalters have only the Irish division at pss. i, li, and ci, together 
with cix ; Eadwine's divides only at ps. cix ; the Bosworth, most 
elaborate of all, unites the Irish with the complete Benedictine 
system. Here again the Paris Psalter stands conspicuously close 
to the Vespasian. As we have seen above from our study of the 
missing pages, the places markt were at pss. i, xxvi, xxxviii, lxviii, 
lxxx, xcvii, and cix (ps. lii is passed over, apparently by acci- 
dent) ; i. e., the primitive Eoman points of prominence are 
markt, and those only. There is not a trace of the Benedictine 
system, nor of the Irish. The absence in the Paris Psalter of 



THE LATIN TEXT OF TEE PARIS PSALTER. 169 

the Irish tripartite division, 23 so widespread in the later psalters 
of the Eoman group (and in nearly all of the Gallican) is espe- 
cially remarkable ; for we have in the accompanying West-Saxon 
Psalms the most conspicuous exemplification extant of this 
curious Irish custom. There can be little doubt that the limita- 
tion of the prose version to the first " fifty " rests originally in 
some way upon such a partition of the psalter. 24 Its absence 
in the Latin text is a strong indication both of the derivation 
of the Latin from a fairly early source, and of its entire lack 
of connection with the West-Saxon Psalms. 

2. The Latin text of the Paris Psalter is unconnected with 
the West-Saxon Psalms, which must have been translated from 
an altogether different original. Whereas the Latin text, as we 
have seen, supplies an early and primitive type of the Roman 
version as it was carried to England, the West-Saxon Psalms 
are clearly based on a very late type of this text, with many 
Gallican readings found in none of the other Roman psalters. 

A large proportion of Thorpe's changes were evidently made 
to obtain greater agreement between the adjoining Latin and 
Anglo-Saxon, especially when, as is very frequently the case, the 
West-Saxon Psalms follow a Gallican reading not to be found 
in the Latin. As a result of Thorpe's efforts the two texts in 
his edition do show a general agreement, tho even there far from 
a complete one. Wildhagen was led by this factitious corre- 
spondence to believe (p. 469) that they came from the same 
source, and that the differences were due to scribal changes in 
the Latin. That the truth is precisely opposite will appear from 
the following list of cases where Thorpe has turned an authentic 
disagreement into an induced agreement of the two texts. 
(P = the genuine Latin text; Th = Thorpe's Latin; W-S = 
version of the West-Saxon Psalms.) There are four groups: 

a. Where P = Ro, and Th = Ga (or Vulgate) = W-S. 

b. Where P differs from Ro but = ABCDE (i. e. has one of 

a The absence of any marking before ps. ci is decisive of this. There 
was some sort of separation between pss. 1 and li, but this was mani- 
festly due to the necessity here of marking the end of the West-Saxon 
and the beginning of the Anglian Psalms. 

"More fully discussed by the writer in the Zeitschrift filr celtische 
Philologie, 1912, p. 486 f. 



170 AMERICAN JOURNAL OF PHILOLOGY. 

the peculiar Anglican readings) ; and Th = Ga (some- 
times also = Eo) = W-S. 

e. Where P differs from both Eo and ABCDB (i.e. has a 
peculiar or manifestly mistaken reading ; and Th = 
GaEoABCDE = W-S. 

d. Where P = EoGaABCDE, and Th = W-S (i. e., Thorpe 
has deliberately, tho not always successfully, introduced 
a peculiar reading of his own to get closer agreement 
with the Anglo-Saxon). 

A few of the more striking examples under each head will be 
given in full (verse numbering Thorpe's) : 

a. 

vii. 6, in finibus inimicorum tuorum : on minra f eonda mearce 

(ThGa, meorum). 
vii. 9, secundum innocentiam manuum mearum super me : sefter 

minre unscaeSfulnesse (ThGa om. super me; Ga, secundum 

innocentiam meam). 
xii. 1, quousque auertis: hu lange wilt )?u ahwyrfan (ThGa, 

usquequo avertes). 
xxii. 3, super semitam : ofer ]>a wegas (ThGa, semitas) . 
xxviii. 6, solitudinem . . . desertum: ]>a westan eorSan . . . 

\>a westen stowe (ThGa, desertum . . . desertum). 
xxxii. 16, sperantes autem in misericordia eius: and ofer >a >e 

hopiaS to his mildheortnesse (ThGa, et in eis qui sperant 

super misericordia ejus), 
xxxiv. 15, ignorauerunt: ic nyste (ThGa, ignoravi). 
xxxvii. 9, Et: Drihten (ThGa, Domine). 
xxxviii. 9, substantia mea tanquam nichil ante te est: mid >e 

is eall min seht (Th, apud for ante; Ga, substantia mea apud 

te est). 
xl. 6, Et: }>eah (ThGa, Et si, taken by translator as Etsi). 
xli. 2, ad Deum uiuum : to Gode for )>am he is se libbenda wylle 

(ThGa, ad Deum fontem vivum). 
xliv. 10, a gradibus eburneis : of >inum elpanbaenenum husum 

(ThGa, domibus). 
xlviii. 6, in habundantiarum suarum (sic) : j^fflre mycelnesse 

hiora speda (Th, multitudine abundantiarum suarum ; Eo, in 

abundantia diuitiarum suarum; Ga, in multitudine divi- 



TEE LATIN TEXT OF TEE PARIS PSALTER. 1?1 

tiamm suarum. P is evidently intended for the Eo reading, 
whereas W-S translates Ga). 
xlix. 22, existimasti iniquitatem : J?u rseswedest swiSe nnryhte 
(ThGa, inique). Other examples may be found at ii. 6, 10; 
vi. 3 ; xv. 8 ; xxvi. 13 ; xxvii. 7 ; xxxiv. 19, 23 ; xxxvi. 24 J xl. 8. 



xiii. 10, confudisti (ABCDEP) : gedrefe ge (ThGaEo, confu- 
distis). 

xxix. 11, ut cantem tibi gloria mea (CDEP) : Ipsst min wuldor 
and min gylp J>e herige (ThGaBoAB, tit cantet). 

xxxvii. 6, turbatus (ABCDEP) : gebiged (ThGaEo, curvatus). 

xlviii. 7, Prater non redemit, redemit homo (ABCDEP) : J»ast 
nan broSor ojrres sawle nele alysan . . . gylde for J>y him 
sylf, and alyse his sawle (ThGaRo, redimit redimet; appar- 
ently the W-S translates the reading redimet redimet of Eo 1 ; 
see Wildhagen's note to this passage in his ed. of the Cam- 
bridge Psalter). 
Other examples may be found at ii. 13; vii. 16; xvii. 8; 

xxxi. 7; xxxiv. 10; xxxvi. 24; xxxvii. 16; xl. 1; xlix. 5. 

c. 

viii. 3, ut destruam: for Sam J?u towyrpest (ThGaEo ABODE, 

ut destruas). 
xxvi. 3, innocentes (sic) : mine fynd (ThGaBoABCDE, 

nocentes). 
xxxiv. 8, exprobrauerunt animam meam : idle hi wseron J>a hi me 

tseldon (ThGa, supervacue exprobaverunt animam meam; 

EoABCDE, uane for superuacue). 
xlvi. 2, Dominus summus terribilis super omnes deos: swyf>e 

heah God and swy]?e andrysnlic and swi)?e micel Cynincg ofer 

ealle oSre godas (Th adds, Eex magnus super omnem terram; 

Ga, Dominus excelsus terribilis rex magnus super omnem 

terram; EoABCDE, Deus summus terribilis et rex magnus 

super omnes deos). 

Other examples: ix. 23, 30; xiv. 1; xxx. 8; xxxv. 12; xxxix. 
12 ; xxxix. 12 ; xliii. 3, 25 ; xlvii. 8. 



172 AMERICAN JOURNAL OF PHILOLOGY. 



xxiii. 5, a Domino ... a Deo (so EoGaABCDE) : fram Gode 
... set Drihtne (Th, a Deo ... a Domino). 

xxiii. 7, 9, eleuamini, porte eternales . . . eleuamini, porte 
eteroales (so EoGaABCDE) : onhlida)> J>a ecan geata . . . 
onhlidaS eow, ge ecan geatu (Th, elevamini postes aaternales 
. . . elevamini, porta setemales. Thorpe's change here was a 
happy inspiration, for it would explain the curious difference 
in the W-S rendering of the two verses ; but I have found no 
source for his reading). 

xxiv. 6-8, cf. Thorpe's deliberate change of verse division. 

xxxviii. 8, Thesaurizat et ignorat cui congregat ea (so ABODE; 
Eo, congreget; Ga, congregabit) : hy gaderiaS feoh and nyton 
hwam hy hyt gadriaS (Th, Congregat et ignorat cui congre- 
gabit ea) . 

xlviii. 7, et laborauit in eternum (so EoAB; GaCDE, et labor a- 
bit) : hu he on ecnesse swincan msege (Th, nee laborabit in 
seternum; the passage is difficult, but Thorpe's change is 
unnecessary ; W-S probably followed Ga) . See also xxxvii. 13 
and xlviii. 15. 

It would, however, be wrong to leave the impression that 
Thorpe's changes are always made for the sake of getting a 
closer agreement with the Anglo-Saxon version. Frequently he 
alters merely to correct or smooth the Latin, or for no apparent 
reason, except to bring the text of P nearer to the Vulgate. 
Indeed, where the West-Saxon Psalms are translated from a 
Eoman or a peculiar Anglican reading, Thorpe's alterations 
often destroy an agreement that was originally present. Some 
of the clearest and most interesting of these cases may be cited : 

vi. 8, auertantur retrorsum: and gan hy on ear sling (ThGa, 
avertantur). 

viii. 7, oues et boues, uniuersa insuper et pecora campi (so also 
ABCDE) : sceap and hrytSera and ealle eorSan nytenu (ThGa- 
Eo, oves et boves universas, insuper et pecora campi). 

xvi. 12, framea inimicorum de manu tua (so B; EoACDE, 
frameam) : of J?Eere wrsece minra f eon da alys me mid Junre 
handa (ThGa, frameam tuam ab inimicis manus tuae). 



TEE LATIN TEXT OF THE PARIS PSALTER. 173 

xvii. 12, Prae fulgora (for praefulgora, nom. plur. of adj. prae- 
fulgorus; so D; C, Prefulgorae or -ra; E, Prefulgorae; 
ABRoGa, Prae fulgore) in conspectu eius nubes transierunt: 
and pa urnan swa swa ligetu beforan his ansyne (Th, Prae 
fulgore; cf. Wildhagen's note to passage in Cambridge 

Psalter), 
xvii. 33, posuit: he gedyde (ThGa, posuisti). 
xxvii. 1, ne sileas a me: ne swuga (ThGa, ne sileas a me 

nequando taceas a me), 
xxvii. 5, Th adds, secundum opera manuum eorum tribue illis, 

with Ga ; om. in P and W-S. 
xxx. 4, firmamentum meum et refugium meum: min trymnes 

and min gebeorh (Th, fortitudo mea et firmamentum meum; 

Ga, fortitudo mea et refugium meum). 
xxxvii. 15, Tu exaudies, Domine (so ABCDE) : Gehyr Sis, Drih- 

ten (ThGaBo, Tu exaudies me, Domine). 
xxxviii. 7, Quamquam imaginem Dei ambulet homo (sic; in 

erased after Quamquam; ABCDE, in imagine): And swa 

J»eah selc man hasfS Godes anlicnesse on him (Th, Quamquam 

in imagine ambulet homo; Ro, Quamquam in imagine Dei 

ambulat homo; Ga, Verumtamen in imagine pertransit 

homo), 
xlix. 18, post te: under baec fram J>e (Th, postea; Ga, retror- 

sum; did the W-S translator read retrorsum post te?). 

Other examples : vii. 5, 7, 10, 14 ; viii. 4 ; ix. 12 ; xi. 2 ; xii. 5 ; 
xiii. 10 ; xvii. 3, 23 ; xix, 6 ; xx. 4 ; xxv. 9 ; xxvi. 14 ; xxviii. 7, 9 
xxxii. 17; xxxiv. 24; xxxvi. 36; xli. 12; xliv. 16; xlvii. 12 
xlviii. 4; xlix. 8, 16. Note also the changes in verse division 
which Thorpe has made at xvi. 7-8; xxi. 1-2; xxxix. 9-10; 
xl. 7-8; xlv. 4-5. 

The 56 discrepancies given above by no means exhaust the 
list. They include merely those affected by Thorpe's changes of 
the Latin text. A complete list, which would include from among 
the nineteen cases cited by Wichmann ten (viz., vii. 9, xvi. 15, 
xxvii. 2, xxx. 24, xxxi. 6, xxxix. 6, xli. 9, xxxviii. 9, xl. 2, xlv. 5) 
that are not vitiated by the recovery of the correct text, would 
be long and would necessitate another paper. The instances here 
collected are enough to show the utter divergence of the two 
texts, and to indicate the strong Gallican tincture (about 30 of 



174 AMERICAN JOURNAL OF PHILOLOGY. 

the W-S readings cited are distinctively Gallican where P is not 
Gallican) of the West-Saxon Psalms. Thus we get the curious 
result that, altho Wildhagen's conclusions with regard to the 
date and character of the Paris Psalter Latin are inadmissible, 
conclusions very similar seem required for the accompanying 
West-Saxon Psalms. 

3. The Paris Psalter Latin is also unconnected with the 
Anglian Psalms. These, however, are translated from a very 
similar type of text, which had comparatively few Gallican 
readings, and was distinctly earlier in character than the original 
of the West-Saxon Psalms. 

As we have seen above, Bruce has already demonstrated that 
the Anglian Psalms could not have been based on the accom- 
panying Latin, altho in three of the cases cited by him Tanger's 
restored text happens to agree with the Anglo-Saxon version. 
Much more often, however, Tanger's restorations reveal addi- 
tional divergencies which Thorpe's perversions had concealed. 
Accordingly the following discrepancies may be added to Bruce's 
list (Ang = Anglian Psalms) : 

liv. 13, Th adds the Ga, in domo Dei (om. by P) : on Godes 

huse (EoABCDE, in domo Domini), 
lv. 9, Th adds tota die (om. by PEoGaABCDE) : ealne daeg. 
lix. 4, electi tui (so EoABCDE) : leofe ]>ine (ThGa, dilecti tui). 
Ixi. 7, In Deo salutari meo (so ABC) : On Gode standeS min 

gearu hselu (ThGaBoDE, In Deo salutare meum). 
lxxii. 19, Tenuisti manum dexteram meam in uoluntate tua: 

J?u mine swy]?ran hand sylfa gename, and me mid >inon 

willan well gelaeddest (ThGaEoABCDE, Tenuisti manum 

dexteram meam, et in voluntate tua deduxisti me), 
lxxvii. 9, Et: J>st (ThGaEoABCDE, Ut). 
xcv. 9, in nationibus (so EoABCDE) : on cynnum and on 

cneorissum (ThGa, in gentibus). 
ciii. 18, in tempore (so ABCDE) : on ]?a maaran tid (ThGaEo, 

in tempora). 
cxviii. 147, in uerbum tuum (so CDE) : on Sinum wordum 

(ThGa, in verba tua; EoAB, in uerbo tuo). 
cxxxi. 13, super sedem meam (so EoABCDE) : on Jnnum setle 

(ThGa, super sedem tuam; cf. gloss to E, ofer setl Jrin). 



TEE LATIN TEXT OF THE PARIS PSALTER. 175 

exxxvi. 1, dum recordaremur tui, Sion (so RoABCDE) : }>onne 
we Sion gemunan swiSe georne (ThGa, dum recordaremur 
Sion). 

cxlvii. 7, flauit (so ACDE) : blaweS (ThGaRo, flabit). 

Brace's valid examples of divergence are nine in number, in 
all of which the Anglian Psalms follow the regular Eoman 
reading, while the Latin either has a reading peculiar to itself 
(ci. 25, cvi. 38), one shared by one or more of the group ABODE 
(lxv. 3, lxxvi. 2, lxxvii. 62, cviii. 28, cxiv. 4), or a Gallican 
reading (cxxxviii. 17, cxl. 3). Besides these Bruce mentions 
(p. 126) five other discrepancies in which the Anglian Psalms 
follow an Old Latin reading (lxxvi. 11, xc. 2, xci. 10, cxviii. 165, 
cxix. 4), and two (ciii. 14, cxviii. 151) in which they follow the 
Gallican; in all seven of these the Latin text has the regular 
Roman reading. Finally, four other cases remain to be cited : 

lxii. 4, leuabo manus meas (so ThGaRoABCDE) : ic . . . mine 

handa }nvea (probably merely the translator's mistake for 

lauabo; cf. Grein's note), 
ciii. 30, Qui respicit terram (so ThGaRo) : He on Sas eorSan 

ealle locaS (ABODE, Qui respicit in terram). 
cxxxi. 11, earn (so ThGaC) : hine (EoABDE, eum). 
cxxxi. 12, super sedem meam (so ThRoABCDE) : ofer }>rn 

heahsetl (Ga. super sedem tuam). 

Thus in all there are 32 cases of divergence between the two 
texts, — a number ample to show their independence, but small 
compared to the total that may be gathered from the first fifty 
psalms. In only nine of these cases do the Anglian Psalms 
follow Gallican readings not found in the Latin, a number 
which again is negligible compared with the constant depen- 
dence upon the Gallican version on the part of the translator of 
the West-Saxon Psalms. The facts suggest, tho alone they 
would of course in no way demonstrate, a date for the original 
of Ang later than P, but distinctly earlier than the original 
of W-S. 25 

23 The order indicated is borne out by a consideration of the very 
different exegetical sources used in the two Anglo-Saxon versions, an 
aspect which will be treated fully in the forthcoming edition of the 
West-Saxon Psalms. 



176 AMERICAN JOURNAL OF PHILOLOGY. 

The compiler of the Paris Psalter has thus united in his very 
composite manuscript three texts which certainly had no pre- 
vious connection whatever. One of his sources, clearly the 
oldest, was a copy of the Eoman version of the entire psalter 
having numerous primitive Anglican textual features, a very 
early system of psalter division, and a fairly early collection of 
liturgical addenda. This Latin original can hardly have been 
later than the beginning of the tenth century, and may have 
gone back to the ninth. His second source was an Anglo-Saxon 
metrical translation of the entire psalter, made fairly late, as 
the character of the meter shows, but from an early and distinct- 
tively Anglican type of the Eoman version. Thirdly, he had 
before him a recent translation of the first " fifty " in the late 
West-Saxon prose, based upon a Latin original which adhered 
to the Eoman text in the main, but which liberally admitted 
Galliean readings to a considerably greater extent than any 
other English copy or version of the Eoman psalter preserved 
to us. There is no evidence that this prose translation ever 
extended beyond its present limit. Apparently it was regarded 
by the compiler as his greatest treasure, for in its favor he 
discarded the first third of his metrical translation, parts of 
which were destined to survive in a copy of the Benedictine 
OfScium. A fourth volume in his scriptorium furnished him 
with his last ingredient,— a set of brief rubrics in Latin for the 
entire psalter. This " Collectio Argumentorum," made use of 
in several of the later Anglo-Saxon psalters, reveals a depen- 
dence upon much the same sources as the West-Saxon Psalms, 28 
and may indeed have been a sort of preliminary study of their 
translator's; but if so, these two are the only elements of the 
whole compilation with any inherent connection. To the student 
of the West-Saxon and Anglian Psalms, accordingly, their acci- 
dental companion the Latin text of the Paris Psalter has no 
futher interest than attaches to any copy of the Eoman version 
bequeathed to us from the Anglo-Saxon Church. 

BOBERT L. EAMSAY. 

University of Missouri. 



""See Bruce, pp. 17-24; and "Theodore of Mopsuestia in England and 
Ireland," Zeitschrift fur celtisehe PMlologie, 1912, pp. 488-497.