ages, and we have Romaunt or Waldensian Tracts, entitled "Meczonia," "Goy de Paradis." Ambrose's tract De Superbia was also in great repute and circulation. In this there is the following passage: "Superbia, quod initium omnis peccati est, cujus primæ soboles septem nimirum principalia vitia ex hac virulenta radice proferuntur, scilicet inanis gloria, invidia, ira, tristitia, avaritia, ventris ingluvies, atque luxuria." Amb. Oper. Vol. 11. p. 493, pars II. Edit. Paris. 1690.

A Waldensian treatise, entitled "Lo pecca de la Superbia," reads thus: "Superbia es reyna de tuit li pecca de laqual di l'escriptura, superbia es comenczament de tot pecca. Sept principal pecca nayson en aquella, Ço es a saber: Vana Gloria, Envidia, Ira, Tristicia, Avaricia, Golicia, Luxuria." Dublin MS. C. 5. 22. fol. 118.

It is remarkable that we find, among the works attributed to St Gregory, passages which appear to be taken almost verbatim from St Ambrose, and which also correspond with the Romaunt translation just cited.

Ipsa namque vitiorum regina1 superbia. Radix quippe cuncti modi superbia est, de qua Scriptura1 attestante dicitur : Initium omnis peccati est superbia.' Primæ autem ejus soboles nimirum septem principalia vitia. De hac virulenta radice proferuntur scilicet inanis gloria, invidia, ira, tristitia, avaritia, ventris ingluvies, luxuria 3."

י

Another strong presumption that both the Romaunt translation of the New Testament, and the more ancient Waldensian treatises, which quote it almost verbatim, may be attributed to Waldo and his associates, rests on the fact that the Version exhibited in the Dublin, Grenoble,

1 66 Superbia es Reyna de tuit li pecca de laqual di l'escriptura." Wald. Treatise.

2 "Sept principal pecca nayson en aquella." Wald. Treatise. 3 Op. S. Greg. Lib. xxxi. Vol. 1. p. 1035. Par. 1705.

and Zurich MSS. offers internal evidence of having been translated by persons who had had access to four ancient manuscripts of the gospels, one of which was preserved at Lyons, and the other three in Lombardy.

Now Waldo himself was a Lyonnese, and his two assistants, Bernard of Ydros, and Stephen of Ansa, though resident in Lyons, were, as their names indicate, natives of Lombardy Ydros and Ansa being towns in the north of Italy.

There is also reason to believe that Waldo had a third associate from Lombardy. Pilichdorf tells us that there was one "Johannes, qui erat de Lugduno'," who joined himself to Waldo. Reiner makes mention of a "Johannes de Lugduno," who was afterwards a Ketter or Catharan Bishop of Bergamo in Lombardy2; and in the "Summa Fratris Reinerii," which is thought to be an abridgment of Reiner's treatise, he is called "Johannes de Lugio," (Lugo or Lugano in Lombardy?) Now the Romaunt Version, among other marks of great care and erudition, and of faithful adherence to the best Latin copies of the New Testament, as I have pointed out in my notes to the Gospel of St John,—contains proofs that its compilers, at the same time that they used the Vulgate of Jerome for their text, did not adhere to it servilely, but consulted the remains of the old "Versio Itala,” and adopted the readings of that version, whenever they saw reason to prefer them to those of Jerome. I have taken some pains to collate the portion of the Romaunt Version printed in this volume, with the ancient MSS. contained in the "Evangelarium Quadruplex," of Blanchini, and the "Bibliorum Antiquæ Versiones, seu Vetus Itala" of Sabatier, and I find that in the Gospel of St John the translators chose readings preserved in the manuscripts of Verona, Vercelli, and Brescia, eighteen times,

1 Bib. Patr. 4. 779. Paris, 1624.

2 Reiner, cont. Wald. Bib. Patr. Iv. c. 6. Edit. Paris.

I

in preference to those of Jerome's Vulgate : and that in this preference they were supported by other authorities. have also traced three notable variations from Jerome's text to the Græco-Latinus of the Codex Bezæ Cantabrigiensis, which was originally in the custody of the clergy of the Church of St Irenæus at Lyons'. These are indications, not to be lightly esteemed, of anxious discrimination on the part of the translators of the Romaunt Version, who, like the translators of our own Authorized Version, omitted no opportunity of comparing their own work with the various texts, and interpretations of the original, within their reach.

My collation of the Romaunt Version of St John's Gospel with the various readings of the Latin text, found in Sabatier's "Versiones Antiquæ, seu Vetus Itala," Blanchini's "Evangelarium Quadruplex," and the "GræcoLatinus Codex," or "Codex Bezæ Cantabrigiensis," gives the following results.

The Romaunt Version, using the Vulgate of Jerome as its text, agrees with the "Versio Antiqua ex Colb. MS." in preference to the Vulgate, six times.

Agrees with the Græco-Latin Beza MS., in preference to the Vulgate, three times.

Agrees with the Vercelli MS., in preference to the Vulgate, seven times.

Agrees with the Verona MS., in preference to the Vulgate, six times.

Agrees with the Brescia MS., in preference to the Vulgate, five times.

The examples will be found in the notes at the end of the Gospel of St John.

I may add, there is yet another ground on which we are justified in assuming that Lombard grammarians and scholars had something to do with the Romaunt Version, 1 See Kipling's Preface to Beza's Codex, p. xxi.

as it reads in the Dublin, Grenoble, and Zurich copies. The dialect is less purely Provençal than that of the Paris and Lyons copies, and partakes more of the Italian than of the Gallic Romaunt. For proofs of this I refer the reader to Mr Cornewall Lewis' able Essay on the Romaunt Languages, and to his distinctive characteristics of the Italian1, Provençal, Spanish, and old French dialects; and request that the Dublin and Paris MSS. may be examined, with the aid of the examples adduced by Mr Lewis.

The specimens which I have given of the six remaining copies of the Romaunt Version, taking the first chapter of St John's Gospel from the Dublin, Grenoble, Zurich, Lyons, and Paris Manuscripts, will assist the critical reader in forming his own opinion as to the antiquity and comparative value of those Manuscripts. The Paris MS. 8086 will be found to present many features different from those of the Dublin, Grenoble, and Zurich copies, which vary but little from each other. Its construction and dialect approach nearer to those of the oldest Troubadour Poems. It marks the sujet and the régime by the distinctive s2. As a translation it is more loose than the Dublin.

The Lyons MS. No. 60, and the Provençal MS. No. 6833, resemble the Paris 8086 more closely than those of Dublin, Grenoble, and Zurich, and exhibit the peculiarities of the Provençal dialect.

The Paris MS. 8086 is not improbably a transcript of the earliest copy produced by Waldo, and it may have preserved many passages from older partial translations afloat when Waldo commenced his work, as our Own Authorized Version retains many sentences which appear in Tyndale's, and in Wiclif's translations. The Dublin, Zurich,

1 See Essay on Romance Languages, pp. 68, 69, 118, 119, 120, 261. 2 E. G. Paris MS. John 1. Dieus v. 1. Oms v. 6. Carns v. 14. Johans v. 15.

and Grenoble MSS. display, in my opinion, marks of a revised and improved edition, which is more literal, and closer to the Latin text, and may have been put forth after Waldo's journey to Italy, and when he mixed, as Stephen of Borbon tells us he did, with the Lombard separatists from Rome, many of whom were allowed to be indefatigable students. Matthew Paris relates, in an anecdote sub anno 1243, that the seceders of the North of Italy sent their most promising youths to the University of Paris, that they might become accomplished theologians and disputants1.

It may reasonably be presumed that Waldo's later production, having been undertaken with the care described by Stephen de Borbon to render it a faithful translation, became the text-book for subsequent versions among the earliest reformers of France, Spain, and Italy.

The six Romaunt copies still extant agree in so many material points, especially in their being simple and literal translations, and not capricious paraphrases in some places, and glosses in others, like productions after the manner of Comestor and Guiart des Moulins, that it is manifest they are all of one family, and proceed from one prototype.

Some of them were transcribed, as I have observed in another place, while the copyists, and those for whom they were written, were still in communion with the dominant Church; others were copied, and received some slight alterations after the separation from Rome had begun. The Dublin, the Zurich, and Grenoble MSS., contain a version, which may be traced through the hands of the Waldenses of Piedmont2, and exhibit the received text of their Church. But my more matured judgment, and recent investigations, will not longer to the opinion I once

1 Matt. Paris. p. 413. Edit. 1644.

permit me to adhere any held, that the translation 2 See pages xxiii. xxv., supra.