The Paris Psalter is the only true translation of the Psalms into Old English, the other O.E. psalters being interlinear glossings. (note) Each page of the ms. has two columns, the Latin on the left and the English on the right. The psalms are introduced by Latin rubrics, and the first fifty (except 1, 21, and 26) also have a series of "Arguments" written in English. (note) The verse translation of Psalms 51-150 was edited in the Anglo-Saxon Poetic Records. (note) The title to Thorpe's edition of the first fifty calls the English a "paraphrase" of the psalms, but that term is not really appropriate to the prose. It is true that the English of a verse is frequently much longer than the Latin, but the reason is often simply that the translator needs more words for a clear English equivalent. Putting þam þe min ehtað for persequentibus, as he does at 7:2, is not paraphrase. Nor is a translator unfaithful to the original when purging it of stylistic traits which will seem ungainly in the target language. Thus, as I will argue in the first section of this introduction, the paratactic quality of the Hebrew, which the Latin retains at least in spirit, does not survive the transition to good English. The crucial point is that the translator does not ignore individual words in his original, as a "paraphrast" would do. He does abandon the literal meaning of the Latin from time to time, but in these cases his purpose is consistently to guide the reader to one of the two kinds of meaning which he sees in the psalms. First, he will often phrase a sentence so as to make it amenable to the specific allegorical understanding ascribed to it by the exegetical tradition. (note) Secondly, each psalm is translated so as to emphasize the major themes announced by the Argument which introduces it, as I will demonstrate in the third section of this introduction.
For the most part, the translator has successfully domesticated the syntax of his original. The Latin present participle regularly becomes a relative clause, (note) and can even give rise to a new principal clause in the translation. (note) Similarly, past participles are often resolved into either noun clauses with se þe (note) or adjective clauses with þe. (note) At 38:12, the infinitive in tabescere fecisti is resolved to gedest þæt he aswint. Nouns are especially susceptible to such resolutions; noun groups with a genitive agent often become full clauses in the English, (note) and other genitive constructions are regularly separated into either noun-plus-modifier (note) or two nouns in the same case. (note) At 8:9, pisces maris and semitas maris become the compounds sæfiscas and sæwegas. Furthermore, verbal and other nouns are resolved to noun clauses with se þe, (note) þæt, (note) hwæt, (note) and hu, (note) or to principal clauses. (note) Even an adjective may be resolved; gentibus iracundis (17:48) gives þam þeodum ðe wið me yrsiað. Finally, numquid (49:13) and ecce (50:7,8) give wene ge þæt . . . , þu wast þæt . . . , and ic wat þæt . . . , respectively. (note)
More importantly, the translator consistently brings to the surface the relationships among sentences and sentence-members which are suppressed by the paratactic style of the Latin. The opening verses of Ps. 49 demonstrate some of the techniques by which he makes his translation less paratactic:
1]Deus deorum Dominus locutus est, et vocavit terram,
1]Dryhtna Drihten wæs sprecende þæt he wolde cuman to eorðan swa he eft dyde, and cliopode eorðlice men to geleafan,
A solis ortu usque ad occasum. 2]Ex Sion species decoris eius.
Fram sunnan upgange oð hire setlgang. 2]Of Sion aras se wlite his andwlitan.
3]Deus manifeste veniet, Deus noster, et non silebit.
3]And eft cymð se ylca God swiðe openlice, þæt ys ure God, and he þonne naht ne swugað.
When the translator returns to God's name in the third verse, the purpose of se ylca is not, we may be sure, to insult the reader's intelligence. Rather, he wishes to make explicit the understood links that connect the three verses. (note) Similarly, þæt ys externalizes the relationship between Deus veniet and Deus noster. (note) Another significant addition to the third verse is þonne, which seems here to suggest both futurity and consequence (as it does at 13:1, 36:4,10,27, and 7:5,8). And of course, the And that bridges verses 2 and 3 is used for that purpose innumerable times throughout the prose. (note)
4]Et vocavit celum sursum, et terram ut discerneret populum suum.
4]And he cleopað to þæm heofone; hæt hine þæt he hine fealde swa swa boc, and he bebyt þære eorðan þæt heo todæle hyre folc,
5]Congregate illic sanctos eius, qui ordinaverunt testamentum eius super sacrificia.
5]And gegadrie on þa swyðran hand his halgan, þa þe heoldon his bebodu ofer ælcere offrunga.
Here, the translator knits together two quite separate sentences, making v. 5 a continuation of the clause introduced by þæt in v. 4. We find the same kind of intensified hypotaxis at 7:4-5, where a sequence of si-clauses in the Latin becomes a set of subjunctives organized under a single gif, and at 25:6-7, which arranges its statements into þæt-clauses dependent on ic wilnode symle. Interrogative sentences are especially susceptible to this kind of extension. Nonne cognoscent, 13:4, inspires a complex series of interrovatives with hwi ne ongitað in the subsequent English verses. (note) The most common way of reducing parataxis is simply to subordinate one of a pair of principal clauses, by using se, (note) þy læs, (note) þeah . . . þeah, (note) þa hwile þe, (note) to þam þæt, (note) þæt (="so that," with a subjunctive), (note) and forþam and forþy. (note)
Despite his skill with syntax, the translator sometimes remains content with Latin word order, especially when it puts an important word into an emphatic position in the sentence. Thus, he follows his original by retaining meus est . . . and similar constructions, (note) by removing a direct object to final position, (note) and by suppressing the verbs in some paratactic structures. (note) Rather less felicitous is the translator's fidelity to a Latin dative where one might expect a prepositional phrase (abscondisti timentibus te / hæfst gehyd and gehealden þam þe þe ondrædað). (note) Similarly, the translator sometimes suppresses the definite article. Genitive constructions are especially unlikely to acquire the article when they pass into English (1:1 in consilio impiorum / on geþeaht unrihtwisra), (note) and in several places the translator chooses to omit a definite article in the nominative. (note) On the other hand, se (without þe) sometimes renders qui, forming a quite latinate construction. Et de his omnibus liberavit eos (33:20), for example, gives and of eallum þæm hi alysð. (note)
Finally, it must be said that the translator is fairly casual with the tense and mood of verbs, (note) and nouns are very frequently pluralized when they pass into the English. (note) He also tends to turn direct discourse into indirect. (note)
Continue reading with the next section, on the text's relationship to the Roman and Gallican psalters, or skip ahead to the third section, on the relationship of the translation to the arguments, or return to the main menu.
Dick Stracke, Augusta College, rstracke@admin.ac.edu
Note: Ms. Bibliothèque Nationale fonds latin 8824. The hymns, litany, and prayers which occupy the last 11 folios are not translated. The ms. is described and photographically reproduced in The Paris Psalter, ed. Bertram Colgrave (Copenhagen: Rosenkild and Bagger, 1958). The entire ms. was edited by Benjamin Thorpe, under the title Libri Psalmorum Versio Antiqua Latina (Oxford, 1835). The prose translations of Psalms 1-50 were edited by J. W. Bright and R. L. Ramsay, Liber Psalmorum: The West-Saxon Psalms (Boston and London: D. C. Heath, 1907).
Note: The Arguments are paraphrases of the argumenta which introduce each section of Manegold of Lautenbach's In Psalmorum Librum Exegesis (in Bede's Dubia et Spuria, vol. 93, 477-1098). See J. Douglas Bruce, "Immediate and Ultimate Source of the Rubrics and Introductions to the Psalms in the Paris Psalter," MLN 8 (1893), 72-82. A second copy of the Arguments, in MS. Cotton Vitellius E. XVIII, has been damaged by fire; Bright and Ramsay use it to supply omissions in the Paris Prose.
Note: Ed. George Krapp and Elliott V. K. Dobbie. New York: Columbia University Press, vol. 5, 1932.
Note: Bruce surveyed the translation's correspondences with exegeses that Migne had ascribed to pseudo-Bede, Remigius, and Cassiodorus, in "The Anglo-Saxon Version of the Book of Psalms Commonly Known as the Paris Psalter," PMLA 9 (1894), 43-164. Wichman found in the text passages similar to quotations from the Psalms in Alfred's translation of the Cura Pastoralis, and he suggested that Alfred also translated the Paris Prose ("König Aelfred's Angelsächsische Übertragung der Psalmen I-LI Excl.," Anglia 11 [1889], 39-96). But Bruce argues persuasively that there was no single exegetical source. Rather, he argues, the translator's comments seem to be the sort of thing a churchman would carry in his head. John D. Tinkler studied hard words in the prose in the light of various exegetical works in A critical commentary on the vocabulary and syntax of the O.E. version in the Paris Psalter (Dissertation: Stanford University, 1964), and his dissertation makes it very likely that the translator had such sources on his desk, even though none of them has dominated the translation. In the present edition, I append to each psalm a set of notes that will refer the reader to four writers representative of the exegetical tradition: Manegold of Lautenbach (in Bede's Dubia et Spuria, Migne 93), pseudo-Jerome (Migne 26), the twelfth-century work ascribed to Remigius by Migne (131), and the Glossa Ordinaria (Migne 113).
Note: The most common resolution uses a noun clause with se þe. Thus: persequentibus / þam þe min ehtað (7:2), timentes Dominum / þone þe Godes ege hæfð (14:4). See also the translations of insurgentes, 17:40 and 43:6; hodientes, 17:41; insurgentibus, 17:49; timentibus, 21:26 and 30:20; observantes, 30:7; persequentibus, 30:16; circumdantibus, 31:7; timentes and sperantes, 32:18; timentium, 33:8; inquirentes, 33:11; facientes, 33:17 and 36:1; rapientibus 34:10; benedicentes and maledicentes, 36:22. This pattern is sometimes varied by a clause with ælc (or ænig) þæra þe. Thus: requirentibus testamentum eius / ælcum þæra þe his æ secað, 24:10; timentibus eum / ælces þæra þe hine ondræt, 24:14; ut videat si est intellegens aut requirens Deum / hwæðer he geseo ænigne þæra þe hine sece, 13:2. Compare sperantibus in te / him . . . þæra þe to þe hopað, 30:20. More commonly, the definite article will be replaced (as by þu at 7:10: scrutans corda / þu Drihten þe smeast heortan, and by the genitive nouns at 30:14; vituperationem multorum circumhabitantium / manegra manna edwit þe me ymbutan budon) or expanded to a noun phrase (such as se ylca God þe nan unriht nelt for non volens Deus iniquitatem, 5:5; or for þara ansyne þe min ehtað for a facie inimici, 43:17. See also the translations of generatio querentium Dominum, 23:6; inique facientes, 24:4; animas persequentium, 26:12; a descendentibus, 29:4; timentibus, 22:10; homo non audiens, 37:15; and a voce exprobrantis, 43:17.)
Present participles can also become adverbial clauses, as in 2:6: constitutus sum rex . . . predicans preceptum Domini / ic eam þeah cincg geset . . . to þam þæt ic lære his willan. The effect is to rationalize and clarify the sometimes confusingly elliptical style of the psalms. The translator gains the same effect by resolving present participles elsewhere to adverbial clauses with þeah (3:7), forþam (9:4, 18:8,9), þonne (="when" 21:14), sona swa (47:6), and swa swa (18:6, 47:8).
Note: For example, inluminans oculos gives hit onliht þa eagan, 18:9. The translator follows this pattern for permanens, 18:10; sperans, 25:1; intercidentis, 28:7; concutientis, 28:8; preparantis, 28:9; congregans and ponens, 32:7; eripiens, 34:10; exsurgentes, 34:11; dilatans, 47:3. The translations of habitans (16:12), procedens (18:6), and faciente (36:7) use essentially this same pattern. Magnificans and faciens at 17:51 become imperatives, and adfligens and persequens (34:5-6) are accomodated to an ongoing series of subjunctives.
Note: Adversus christum at 2:2, for example, becomes wið þam þe he to hlaforde geceas. Also see the renderings of occultis meis, 18:13; notis meis, 30:12; and mortuus a corde, 30:13.
Note: De absconditis tuis / mid ‘re witnunga e u lange gehyd h‘fdest, 16:14; condensa / e ‘r w‘s ofer eaht, 28:9.
Note: For example, abiectio plebis becomes ic eom ut aworpen fram him at 21:7, and he translates eloquia oris mei at 18:15 by ðonne sprece ic. At 20:3, þæs þe he mid his weolorum wilnade renders voluntate labiorum eius. Phrases with a possessive pronoun can undergo the same change: propositionem meam / hwæs ic wylle ascian, 48:5; plenitudo eius / eall þæt heo mid gefyld is, 23:1.
Note: The modifier is most often an adjective clause. (Thus: nequitiam studiorum ipsorum / þam unrihte þe an swincað, 27:4; Deo vite mee / þam Gode þe me libbendne þanon gelædde, 41:9. The translator uses the same method to render Deus iustitie mee, 4:2; die adflictionis mee, 17:19; delicta . . . ignorantie mee, 24:7; conturbatione hominum, 30:21; lectum doloris eius, 40:4.)
Elsewhere, it may be an adjective (tabernaculi glorie tue / þines wuldorfæstan temples, 25:8; die malorum / þam yflan dagum, 26:5; vocem fletus mei / mine wependan stefne, 6:9), or a prepositional phrase (as in 36:4, petitionem cordis tui / þæt þu bidst on þinum mode; comparable methods are used for sacrificium iustitie at 4:6 and plebis at 21:7.) Also note the adverbial clause rendering in desideriis anime sue at 9:24.
Note: Deus salutaris meus (24:5 and 26:9) gives þu eart Drihten, min hælend. The translator's general avoidance of genitive constructions is well demonstrated by the separation of in nubibus æris (17:12) to on þam wolcnum and on þære lyfte, and of omnes patrie gentium to ealle þeoda and ælc cynn (21:28). This preference for groups with and extends even to the rendering of appositives and adjectives; thus, min God and min gefriðiend translates in Deum protectorem at 30:3, and insanias falsas (39:5) gives to leasungum ne to dysige; especially interesting in this regard is me sealdest . . . wlite and mægen for prestitisti decori meo virtutem (29:8), and the rearrangement of Que utilitas etc. at 29:10.
Note: Voluntate labiorum eius / þæs þe he mid his weolorum wilnade, 20:3; terrigine / ealle þa þe þæron acende synt, 48:3; compare exultatio mea / þu eart min frefrend, 31:7.
Note: Rectos decet conlaudatio / rihte hit gerist þæt hine ealle rihtwillende emnlice herian, 32:1; petitionem / þæt þu bidst, 36:4.
Note: Omne desiderium meum / eall hwæs ic wilnie, 37:10; propositionem meam / hwæs ic wylle ascian, 48:5.
Note: Vide humilitatem meam / geseoh hu earmne me habbað gedon, 9:14.
Note: At 18:7, egressio eius and occursus eius become Heo stihð. . .and þanon astihð. And at 29:6, fletus and letitia give (þeah) we wepon. . .we hlihhað. The translator works in the same fashion with scandalum, 48:14; and caligo, 17:10. Also note the formation from interitum, 34:7, of the adverbial clause, to þam þæt hi woldan me angefon.
Note: The preference for resolution often affects the rendering of prepositional phrases. Thus, usque in progeniem patrum suorum gives þær his foregengan beoð, 48:20. Compare the resolutions of in finem, 9:7; ab auditu auris, 17:45; ab uberibus matris mee, 21:10; in magnificentia, 28:4.
Note: He also uses se ylca for this purpose at 10:6.
Note: þæt ys also bridges members of sentences at 4:7 and 43:3. Compare how se Drihten se þæs . . . bridges the first two members of 10:5.
Note: Besides and, the translator uses an initial ne, functioning as "nor," to render non at 5:6 and 13:3. The word ac also appears without a lemma at 6:2-3, 11:5-6, 25:4-5,5-6, 27:1-2, 30:9,18 33:22, 34:16, 35:13, 38:9, 41:5, 48:14.
Note: Compare the repetitions of hwy at 3:2. A similar, if less spectacular, effect is the repetition of interrogative adverbs and adjectives within sentences--again, with the effect of keeping relationships between members explicit and external. See the repetitions of hwænne at 34:17, of (for)hwy at 9:22, 43:24, and 49:16, and of hwæt at 8:5.
Note: 7:13, 13:4-5, 47:7. He also uses þæs to sum up the content of a preceding sentence in secgon him þæs lof (28:9) and ne gesceamað me næfre þæs (30:2).
Note: 17:37,39, 19:6, 35:13, 42:4, 50:9(x2),10.
Note: 9:11, 29:8, and 39:4 (forþy for et). Of course, it is often hard to distinguish between forþam the adverb and forþam the conjunction; see 9:26(x3), 15:5, 17:44, 20:3, 23:7 (forþan þe), 26:6, 32:15, 35:2, 36:18, 37:8, 45:6, 48:15.
Note: Mee sunt and meus est, 49:10,12, give min synt and min is. Domini est gives Drihtnes synd at 21:29 and Drihtnes ys at 23:1. Similarly, the Latin's initial placement of the direct object is echoed in rihtwisnesse God lufað (30:24) and Unriht he byð smeagende (35:5). And compare 4:7, Signatum est / And is þeah geswutelod.
Note: See 30:8, salvam fecisti de necessitatibus animam meam / þu gedydest hale æt nydþearfe mine sawle; and 30:19, loquuntur adversus iustum iniquitatem / sprecað wið þone rihtwisan unriht.
Note: 30:20. So also gesette God æ scyldiendum (24:8), delictum meum cognitum tibi feci / ic þa gedyde mine scylda þe swyþe cuþe (31:5), and adferte Domino filii Dei / bringað eow sylfe Gode (28:1).
Note: Usually, it is the genitive plural that is rendered in this way, as in 1:1, in via peccatorum non stetit / ne on þam wege ne stent synfulra. (Compare the translations of refugium pauperum, 9:10; preces pauperum, 21:25; congregationem malignorum, 25:5; terra viventium, 26:13; vituperationem multorum, 30:14; filie regum, 44:14; and Deus deorum, 49:1.) In two places, the construction uses a genitive singular (flammam ignis, 28:7; protectionem salutis tue, 17:36). Ofermodra manna anweald (28:5) answers to nothing in the Latin, but might perhaps be a rendering of some commentary.
Note: See 18:2, 21:14, 23:1, 32:6,11. Note also Dem nu Drihten, þearfe þæs earman (9:39) and gesette . . . æ scyldiendum (24:8).
Note: Compare the translations of quorum os, 13:3; in quorum manibus, 25:10; and in quo ostendam, 49:23. Also note heriað Drihten þone ðe eardað for psallite Domino qui habitat, 9:12.
Note: For changes in tense, see section II below, p. v. Changes in mood are especially interesting. Tu es . . . becomes Hu ne eart þu . . . at 43:5. Quis est homo qui vult vitam gives Se þe libban wylle . . . gehyre hwæt ic secge at 33:13. The imperative often replaces a jussive subjunctive (7:10, 9:33, 26:14, 30:25), or even an indicative (syle for dabis at 50:10). It is also interesting that the translator so often constructs a present progressive to render a deponent (meditabitur, 1:2; meditatus est, 35:5; admirati sunt, 47:6; locutus est, 49:1).
Note: See 5:3, 6:8, 7:10, 9:38, 17:42, 19:4, 21:4,29, 27:3, 29:7, 30:5,10,19,23, 31:6,9, 32:20,21, 33:19, 34:13, 36:15,21, 38:7,12, 39:3,7, 41:9, 42:2, 46:6, 48:6, 49:7-9. Changes from plural to singular are less frequent: 11:2, 23:2, 27:3,8, 31:5, 35:8, 36:17, 37:5, 39:13, 45:10, 49:16, 50:7. In other cases, singular-plural alternation may reflect variant readings. See Tables I-III.