Part 3

(4) In the Scottish collection of legends, said to be Barbour’s. Cf. Barbour’s _Legendensammlung_, ed. C. Horstmann, Heilbronn, 82, ii. p. 12.

(5) In the old Engl. translation of the _Legenda aurea_, see Horstm., _l.c._, p. cxxxv. Caxton’s edition of the legend, No. 196.

(6) The complete text of the legend printed in Horstmann’s above-mentioned collection, _Altengl. Legendensamml._, p. 211 ff.

(7) _St. Eustas_, by I. Partridge, see Gibbs’ above-mentioned edition, and Horstm., _l.c._ p. 472 ff.

With this legend are connected, more or less, the following poems, which it is necessary to speak of in turn:

(1) The Pseudo-Chrestien epic poem, _Guillaume d’Engleterre_.[6]

[Footnote 6: _Guill. d’Engleterre_, ed. Fr. Michel, _Chron. Anglo-Norm._, III. 39-172. On the authorship of this poem see C. Hofmann, _Sitzungsberichte der Münch. Akad._, 1870, II. p. 51, and P. Meyer, _Romania_, VIII. p. 815 f.]

(2) The two Middle High German poems, _Die gute Frau_,[7] and (3) _Der Graf von Savoyen_.[8]

[Footnote 7: _Die gute Frau_, ed. E. Sommer in Haupt’s _Zeitschrift für deutsches Alterthum_, II. 389.]

[Footnote 8: _Der Graf v. Savoyen_, ed. F. H. v. d. Hagen, _Minnesinger_, IV. 640, and Eschenburg, _Denkmäler altdeutscher Dichtkunst_, Bremen, 1799.]

(4) The romances of _Isumbras_; (5) of _Octavian_; (6) last, _Syr Eglamour of Artois_, and (7) _Sir Torrent of Portugal_.

The first five have been treated by Holland in his book, _Chrestien de Troies_, Tübingen, 1854.

According to Holland’s opinion, all of these are derived from the legend of Eustache. He has not exactly inquired into each of them, but restricts himself to a detailed account of their contents. A critical inquiry into these poems, except the romance of _Octavian_, has been recently published by J. Steinbach: _Der einfluss des Crestien de Troies auf die altenglische literatur_. Leipzig, 1886, p. 41 ff. As to the French and the two German poems, it may be sufficient to refer to this exhaustive essay, since it is only by the same legendary origin that they are connected with _Sir Torrent_; otherwise they are quite different.

But of the English romances of _Sir Isumbras_ and of _Octavian_ it is necessary to treat more minutely. _Isumbras_ was edited first by Utterson in his _Select Pieces of Early Popular Poetry_, London, 1817; secondly by Halliwell in _The Thornton Romances_, from the Lincoln MS. A. i. 17. AÂ critical edition of this poem has long been promised by Prof. Zupitza.

In this romance the legend of Eustache can be most clearly recognized. Its contents are, indeed, somewhat transformed according to the taste of the later Middle Ages: the Roman captain is changed into a Christian knight, who performs wonders in fighting against the infidels; he finds his wife as queen of a heathen country; they end their lives as mighty princes, and so on. The legendary style has been supplanted by the romantic diction,[9] but the leading features remain the same. In his above-mentioned essay, pp. 46-48, Steinbach concludes, from a detailed comparison of the contents, that the author of _Isumbras_ did not derive his story from the epic poem, _Guillaume d’Engleterre_, but from an original which bore a still greater resemblance to the legend of Eustache, and, at the same time, contained many of those additions which are to be found in all versions of the legend. Whether this original was composed in Latin, French, or Anglo-Norman, Steinbach does not pretend to determine.

[Footnote 9: On this text see Sarrazin, _Octav._, p. xlv; he speaks of “die entstellte, spielmannsmässig zersungene Form, in der die Thornton Ms uns die legende überliefert. . . . . . dasselbe Pathos, dieselbe Sentimentalität und Frömmelei, aber auch dieselbe anschauliche und lebhafte Erzählungsweise (sc. as in Oct.).”]

To _Isumbras_ I join a few remarks on the romance of _Octavian_, which was edited by Halliwell for the Percy Society, _The Romance of the Emperor Octavian_, London, 1844; and by Sarrazin, _Zwei mittelengl. Versionen der Octaviansage_, in Koelbing’s _Altengl. Bibliothek_, Band III. As for its contents, cf. Sarrazin, as above, p. xviii ff. Concerning the origin of the story, he agrees in general with Holland, only he shows a still nearer connection between _Isumbras_ and _Octavian_, taking the former for a mere imitation of the latter. This opinion, however, cannot be proved. As I cannot enter into detail, I only observe that the contents of _Octavian_ are a great deal more complicated and copious than those of _Isumbras_, which is simple in its plot and style, and shows the nearest resemblance to the old Eustache legend, whilst _Octavian_ is a refined and adorned version of the legendary tale with considerable change in the plan. Isumbras, of course, bears a strict resemblance to Eustache, but not to the Emperor Octavian, who has but little of the character of a suffering saint, as he does not become an outlaw himself, nor is to lose his earthly goods. Even those of his adventures which are conformable to the original--the separation from his family, the rape of the children, the final reunion--are exhibited in a different manner.

The principal contents of the romance of _Octavian_ bear internal evidence of its later origin, as it treats chiefly of the adventures and exploits of Florent, Octavian’s son; especially in the second half of the story, exploits of Florent so prevail that the romance might justly bear his name on the title instead of his father’s. I therefore believe that Sarrazin’s opinion, that _Isumbras_ is nothing but a bad imitation of _Octavian_, is wrong; and I am rather inclined to think the two poems were composed independently from each other, after French originals, as is evidently the case with _Octavian_, and probably with _Isumbras_. See Halliwell, _Thornt. Rom._, p. xviii. Sarrazin, moreover, supposes, p. xlv, both poems to be due to the same author, in consequence of the conformity of the dialect and style, and of some literal coincidences. But the fact that both of these romances are written in the same dialect is not sufficient to prove the identity of the authors, nor is the style, which is nearly stereotyped in all of these romances. As to the literal coincidences, only three of the nine passages quoted by Sarrazin seem to me to be of any importance. See _Octavian_, notes on ll. 382, 397, 481. But even these only show that the writer of _Octavian_ knew _Isumbras_, or _vice versâ_.

As to the relation between _Octavian_ and our poem, these two romances have no other affinity than the same legendary origin, and the motive of the woman innocently persecuted, which may very well have been introduced independently by two different authors. In all other particulars they are quite different.

The heroes bear little resemblance to their legendary models; in _Octavian_ the Emperor of Rome; in _Torrent_ the young, hardy knight who encounters marvellous struggles to win the hand of his spouse. Also in the treatment of the other motive, each romance has taken its own course. In _Octavian_, Florence is calumniated by her mother-in-law; in _Torrent_, Desonelle is persecuted by her father. The causes are consequently quite different: there the jealousy of the mother-in-law against the mighty Empress; here Calamond’s hatred against Torrent. These differences, now only alluded to, cause a great number of others, and produce a general difference of the two poems, which renders the opinion of a nearer connection between them altogether illusory.

Of all the poems mentioned above, the last, _Syr Eglamour of Artois_, is most nearly related to _Sir Torrent_, a fact found out by Halliwell,[10] who, however, thought that there was no necessity for him to prove a similarity which would be at once detected by the reader; still, he takes it for certain that the romance of _Torrent_ is younger than and partly founded on _Sir Eglamour_. As he gives no proof for this opinion, it will be worth while to enter once more into this question, in order to see whether he is right or not.

[Footnote 10: _The Thornton Romances_, p. xxii f. ‘The romance of _Torrent_ is partly founded upon the story related in _Sir Eglamour_. The names are changed, but the resemblance is too striking to have been the result of chance. The treachery of the sovereign, the prowess of the knight, the indiscretions and misfortunes of the lady, and the happy conclusion of her misfortunes, these form the leading incidents of each romance . . . . there is, perhaps, a secret history attached to the source of these romances that remains to be unravelled.’]

Upon it, the MSS. do not help us. The earliest MS. that can have contained _Sir Eglamour_ is the parchment one of the Duke of Sutherland,[11] written about the end of the 14th century. The other four MSS. of it[12] are still later. The only MS. of _Sir Torrent_ belongs to the 15th century, so that neither of these romances can be traced very far back.

[Footnote 11: Cf. Koelbing’s _Englische Studien_, vii. p. 191 ff.]

[Footnote 12: Cf. _The Thornton Romances_, p. xxv ff., and p. xxxvi.]

_Sir Eglamour_ was printed several times in the beginning of the 16th century, and edited anew by Halliwell from the Cambridge MS. in his well-known collection. To judge from the numerous readings of the Lincoln, Cotton, and Cambridge MSS. which he has quoted, the Lincoln MS. shows best the original dialect, and offers in several passages a reading preferable as to rhyme and meaning.[13] Even slight differences in the contents occur now and then.[14]

[Footnote 13: See the following passages which Halliwell has quoted in the notes: _Eglam._ 54, 96, 107, 111, 122, 128, 139, 153, 177, 195, 213, 247, 337, 347, 399, 445, 572, 605, 614, 737, 740, 765, 858, 883, 945, 985, 1081, 1143, 1206, 1216.]

[Footnote 14: See _Eglamour_, notes on ll. 1064, 1082, 1267.]

The metre and probably the dialect are the same in both romances; they are composed in the tail-rhymed twelve-line stanzas, and written in a North Midland dialect. In both of them the style is alike swelled with the habitual phrases; only the long prayers and pious reflections so frequent in _Torrent_ are not to be met with in _Eglamour_. On the other hand, the poet is wont to predict the fates of his heroes (ll. 204, 951); he often demands attention (ll. 15, 39, 343, 634, 904); he never omits, in describing the festivals, to mention the performances of the minstrels, and to praise the liberality of the lords. These characteristics render it probable that the author of _Eglamour_ was a minstrel, not a clerk or monk, as I suppose the author of _Sir Torrent_ to be.

I now pass on to compare the contents of the two poems. The principal features of the plot are the same in both. A young knight who seeks the hand of a princess engages to win her by valiant exploits. The princess’s father opposes his wooing, jealous as he is of the hero’s renown. The knight vanquishes all the giants and other monsters against which he is told to fight, and at length gains his spouse. A few weeks after their marriage, he sets out again on adventurous expeditions. While he stays abroad, his wife is delivered of twins. Her father sends her to sea in a leaky boat; she lands on a foreign shore, where her children are carried off by wild beasts; but they are saved in a marvellous manner, and brought up at royal courts, whilst she herself lives for a long time at a foreign court. As the hero, when he comes home again, doesn’t find her, he goes into the Holy Land to fight with the infidels. After various adventures he finds his wife and children after a tournament at a foreign court. They return home gladly, and celebrate their nuptials by great festivals. The cruel father is duly punished.

On entering into details, however, we find considerable discrepancies between the two romances. First, the names are altogether different. (Eglamour = Torrent. Crystyabelle = Desonelle. Prynsamour = Calamond. Organata = Gendres. Degrabelle = Antony fice Greffoun.) The stage of the plot is in _Eglamour_ Artois, Rome, and Egypt; in _Torrent_ Portugal, Norway, and Calabre. Only the Holy Land is mentioned in both. There the children are carried off by wild beasts, saved by princes and brought up; there the hero fights against the infidels.

The differences of the plot itself are the following:

1. Eglamour confesses his love to Crystyabelle before his deeds; a squire is the go-between in his suit; Eglamour finds love in return. In _Torrent_ Desonelle does not know that she is adored by the hero till after his first exploit. See ll. 109, 448.

2. Accordingly, Eglamour, setting out on adventures, receives two greyhounds and a sword of St. Paul from Crystyabelle as presents, whereas Torrent gets an ambler from his lady love, but not till after his first deed.

3. Prynsamour charges Eglamour with three deeds by which he is to gain Crystyabelle. Torrent is obliged to undertake not less than five combats.

4. In _Torrent_ the combats of the hero are enlarged and adorned by additions not to be found in _Eglamour_. The latter does not release the daughters and sons of kings, nor does he find precious swords in the castles of the giants, nor is he deceived by a king’s counterfeit letter, which causes Torrent a dangerous struggle and the rivalry of a foreign prince. Only in _Eglamour_ (ll. 40-48) some knights are mentioned who came to win Crystyabelle by jousting, but were all vanquished by Eglamour.

The greatest differences are found in the second halves of the stories.

5. Crystyabelle has one child by Eglamour; Desonelle has two by Torrent.

6. Crystyabelle is driven away into Egypt, where she is graciously received by the king. Desonelle finds refuge in the court of the king of Nazareth.

7. Degrabelle, the son of Crystyabelle, is saved and brought up by the king of Israel; the sons of Desonelle by the kings of Greece and Jerusalem.

8. The father of Crystyabelle is not punished like Calamond in _Torrent_, immediately after the hero’s return, but he dies at the end of the poem, throwing himself down from the battlements.

9. Degrabelle is sent, when fifteen years old, into Egypt by his adoptive father to sue for a spouse. In a joust he gains the hand of his mother and marries her. On the very wedding-day the mother recognizes her son by his escutcheon, and the marriage is instantly dissolved. Quite differently does the story run in _Sir Torrent_. Leobertus, fifteen years old, marches by order of the king of Jerusalem against his father, and takes him prisoner, but at length solicits his release.

10. The tournament, which in both poems compasses the reunion of the separated family, is brought on in a different manner. In _Eglamour_ Degrabelle himself proposes the hand of his mother as the prize in the next tournament, to which his father comes. In _Torrent_ Desonelle, hearing of the victories of the strange knight, supposes him to be her spouse from his arms, and at her request a tournament is arranged. (Her hand seems to have been likewise the prize, as may be gleaned from l. 2440.)

11. At the very end of the poems two slight differences are to be noted: in _Eglamour_, Degrabelle marries Organata, daughter of the king of Sidon, whereas the sons of Torrent return into Greece and Jerusalem. Eglamour is crowned prince of Artois; Torrent is elected Emperor of Rome.

From this comparison we may conclude that _Torrent_ is not directly founded upon _Eglamour_, or _vice versâ_; the differences are too great to justify the supposition that either is drawn from the other. Especially is the opinion of Halliwell, which I mentioned above, to be rejected: _Sir Torrent_ cannot be founded on _Sir Eglamour_, simply because it agrees more closely with the old legendary tale than _Syr Eglamour_ does, and has preserved some essential features not to be found in _Eglamour_, in which these are supplanted by others. Desonelle, for instance, has two children according to the old legend, Crystyabelle one; Torrent must fight and suffer in heathen lands like Eustache, whereas Eglamour appears as a mere knight-errant. Further, neither in the Eustache legend nor in _Torrent_ do we find the history of the son who marries his mother, which motive the poet may have taken from the legend of Pope Gregory, or perhaps from the tale of _Syr Degaré_.

But how can the resemblance of the leading features and the discrepancies in particulars be explained? IÂ think the most probable conjecture is, that an old poem, now lost, existed, with which the authors of _Sir Eglamour_ and of _Sir Torrent_ were acquainted; but not having a MS. of it, or knowing it by heart, both of them made up their minds to rewrite the story in a well-known metre, changing, omitting, adding whatever they liked, even filling up the gaps in their memories by invention. Both of them recollected the first half of the story better than the second.

That this poem was an English one seems to be shown by a good many verbal coincidences in both poems; these I accordingly suppose to have belonged to the lost original. They are, indeed, too frequent to be counted simply amongst the large stock of conventional phrases which are to be met with in every poem of this kind. Here they are:--

_Eglamour._

The boke of Rome th{u}s can telle. 408, 561, 886

_Torrent._

As the boke of Rome tellys. 187, 924, 1450, 1924

Ther ys a jeaunt here besyde, 478 That sorowe doyth ferre and wyde. On us and odur moo.

There ys a gyante here besyde, In ale thys covntre fare and wyde, No man on lyve levythe hee. 960

And alle prayed for that knyght. 573

For hym aƚƚ they pray. 108

Alle that in the cyté ware. 598

Aƚƚ that in~ the sytte were. 1047

Alle that cuntrey was fulle fayne, 640 That he homeward was comyn ageyne.

Gentilmen were blith and ffayn~, 1098 That he in helth was comyn~ agayn~.

Aftur sopur, as y yow telle, He wendyd to chaumber with Crystyabelle. 670, 671

After mete, as I you teƚƚ, To speke with mayden Desoneƚƚ To her chamber he went. 1358-60

That lady was not for to hyde, 673-75 She sett hym on hur beddys syde, And welcomyd home that knyght.

The damyseƚƚ so moche of pride, Set hym on~ her bed-syde, And said ‘welcom~ verament.’ 1363

So gracyously he come hur tylle, 679 Of poyntes of armys he schewyd hur hys fylle, 680 That there they dwellyd alle nyȝt.

Such gestenyng he a-right, That there he dwellid aƚƚ nyȝt With that lady gent. 1364-66

A golde rynge y schalle geve the, 715 Kepe yt wele my lady free, Yf Cryste sende the a chylde! 717

Thes gold rynges I shaƚƚ yeve the, Kepe them weƚƚ, my lady ffre, Yf god a child vs send! 1396-1398

Doghtur, into the see schalt thou, 803 Yn a schypp alone, And that bastard that to the ys dere!

There fore thou shalt in to the see And that bastard with-in the! 1793

Sche prayed hur gentylwomen so free, Grete wele my lord, whon ye hym see! 826, 827

She said ‘knyghtis and ladyes gent, Grete weƚƚ my lord sir Torrent, Yeff ye hym~ euer sene! 1837-39

Hur yonge sone away he bare. 842

A way he bare her yong son~. 1871

Thys chylde ys comyn of gentylle blode, Where that ever that he was tane. 863

This chylde is come of gentiƚƚ teme, Where euer this beest hym~ ffond. 1923

Kepe we thys lady whyte as flowre, And speke we of syr Egyllamowre. 950

Leve we now that lady gent, And speke we of sir Torrent. 2080-81

The knyght swownyd in that tyde. 975

Swith on sownyng there he feƚƚ. 2093

Be the XV yerys were comyn and gone, The chylde that the grype hath tane, Waxe bothe bold and stronge. 1018-20

And be the VII yere were gone, The child that the liberd had tane, Found hym his fiƚƚ off ffyght. 2233-35

Yn yustyng ne in turnament 1021 Ther myght no man withsytt hys dynte, But to the erthe them thronge. 1023

With heve tymbyr and ovyrryde 40 Ther myght no man~ hys dent abyde, But to the erthe he them strake. 42

Be thre wekys were comyn to ȝende, Yn the londe of Egypt can they lende. 1057

But ore thre wekes were comÌ yn~ to end, To Portynggaƚƚ gan he wend. 373

Gentilmen that herde of thys crye, Thedur come they redylye. 1195-96

Gret lordys that herith this crye, Theder come richely. 2431-32

Syr Egyllamour knelyd on his kne, ‘A Lorde God ȝylde hyt the! 1288-89

Torent knelid vppon~ his knee 2575 And said ‘God yeld you, lordys ffree!

In swounynge than felle that lady free, ‘Welcome, syr Eglamour, to me!

She said ‘welcom~, my lord sir Torent! And so be ye, my lady gent! In sownyng than feƚƚ she. 2505

_Eglamour_, Linc. MS. Note on 1267:

Grete lordis thane told scho sone.

Gret lordys told she sone. 2539

Perhaps some more light will be thrown on this question when we get the much-wanted critical edition of _Sir Eglamour_; but I fear that the ‘secret history attached to the source of these romances’ will even then remain to be unravelled. What I have proposed has no title to a better name than a conjecture.

§ 5. THE ARRANGEMENT OF THE EDITION.

As to the only MS. in which this romance has come down to us, I have mentioned before that it is exceedingly corrupt; many conjectures, more or less sure, were necessary in order to restore metre, rhyme, and meaning; the greater part of them seemed worthy to be entered in the text, the rest being offered in the notes. No attempt has been made to introduce a uniform character of dialect, considering the quite unsettled state of orthography in early times. The only exceptions are where the sounds are fully determined by the rhyme. In general the orthography of the MS. has been reproduced as accurately as possible. The contractions used by the scribe are expanded and printed in italics. At the beginning of a new period, or a proper name within the line, capitals have been introduced. From l. 1200, where the numeration of my text no longer coincides with that of Halliwell’s edition, the line-numbers of the latter are added in brackets.

The fragments which I have added as an appendix to the text have been consulted in all cases of difficulty, and proved of no little service in correcting the blunders of the manuscript; they contain indeed a somewhat better text than the MS., though they are by no means free from clerical errors. AÂ detailed comparison gives the following result:

In fifty-one lines the text of the fragments is evidently correcter than the MS.:

_Fragments._ _Manuscript._

The kyng of Nazareth sent hym me, Torent, I wot-saue hym on the. 466 The kyng of Portynggaƚƚ seyd, ‘So mot I the! Torrent, I wet-saffe of the.