U.CBERKEI 
 
 •^ ■•TT 
 
 iL^iLji 
 
 ivARi 
 
 UC-NRL P 
 
 711. 
 
 B 3 571 
 
 b'^? 
 
THE SISTER'S SON AND THE CONTE 
 DEL GRAAL 
 
 WILLIAM A. NITZE 
 
 Preprinted for private circulation from 
 Modern Philology. Vol. IX, No. 3, January 1912 
 
Modern Philology 
 
 Vol. IX January ^ igi2 No. 3 
 
 THE SISTER'S SON AND THE CONTE DEL GRAAL 
 
 In my article on the "Fisher King"^ I drew attention to the fact 
 that the ancient mysteries speak rituahstically of the tribal god as 
 iraTifjp^ and suggested that a similar concept underlay the Grail 
 story. I propose now to consider this question more fully, and inci- 
 dentally to show its bearing on the plot of Crestien's romance. 
 
 1 PMLA, XXIV (1909), 385-398 S. 
 
 2 Cf., especially, Famell, Cults of the Greek States, IV, 36 flf. In addition to my previ- 
 ous references, see H. Zimmer, Der babylonische Gott Tamuz, Teubner, 1909; Jane Ellen 
 Harrison, Prolegomena to the Study of Greek Religion, Cambridge, 1903, especially pp. 263, 
 where possible matriarchal aspects of the Demeter-Kore cult are brought out; R. Pischel, 
 Der Ursprung des christlichen Fischsymbols, BerUn, 1905; especially, I. Scheftelowitz, 
 "Das Fisch-Symbol im Judenthum u. Christenth.," in the Archiv f. Religionswiss., XIV 
 (1911), 1-53; 321-392; J. J. Bachofen, Das Mutterrecht, Basel, 1897; though B.'s theory 
 of matriarchy as a political system is not accepted by ethnologists, his work contains 
 an unusual wealth of classical material bearing on our subject. Thus he observes the 
 association of the male principle with the water, Poseidon, Dionysos, Osiris, Orpheus; 
 see pp. 39 flf. "So erscheinen [p. 43] zu Dodona Dione- Venus und Zeus-Acheloos, jene 
 die stoflaiche Erde, der Friichte Mutter (Apollod. apud Schol. Od. 3, 91, 11. 5, 370, 16, 
 233 sq; Ser. Aen. 3, 466; Cic. N.D. 3, 23; Hesiod Theog. 353, etc., etc.), dieser die 
 zeugende Wasserkraft, die erst in der Gebiu-t, also in der machtigen, hochgewipfelten 
 Eiche, zur Darstellung gelangt"; fiu"ther, pp. 239 fif. 
 
 On the graU problem itself additional matter is adduced by L. von Schroeder in his 
 Wurzeln der Sage vom heiligen Gral (Vienna Academy, Vol. CLXVI) — cf. also "Der 
 arische Naturkult" in the Bayreuther Blatter, July, 1911. S. equates the Fisher King 
 with the Norse Hymir: "Es ist [p. 66] Thorr, der Gewittergott, der den Bierkessel der 
 Gotter von Hymir erobert, wie Indra, der Gewittergott, es ist, der den Soma oder die 
 Somakufe dem Vrita, Vala, Vivasvant usw. abgewinnt." S. also mentions the caldron 
 of Odherir (cf. my remarks, op. cit., 405, on Mimir, to whom it belongs), of which he says: 
 "Man wird Hymirs Kessel imd den Odherirs niemals verwechseln und kann doch erken- 
 nen, dass beide auf dieselbe Grundvorstellimg zuriickgehen." I am not qualified to dis- 
 cuss the crux of Schroeder' s theory that the fundamental idea which underlies these 
 parallels is "altarische Sage [p. 92] — in der uralten Vorstellung von Sonne und Mond als 
 wunderbaren hinunlischen Gefiissen." One is naturally skeptical about such generaliza- 
 tions. Certainly, it is a bold leap from Crestien's silver trencher to the moon, and from 
 the grail to the sun (p. 91), and yet I do not wish to prejudge the case. On the whole, I 
 am more inclined to agree with the statement: "nicht nur mythische, sondem auch 
 291] 1 [MoDEEX Philology, January, 1912 
 
 254870 
 
'2**" '•' ■ William A. Nitze 
 
 I 
 
 Perceval, whose name is not revealed until after the grail-visit, is 
 called by Crestien in vs. 74* (Pot. II, vss. 1920 fif.) : li fi{l)z a la veve 
 dame (the term is repeated in Perlesvaus, p. 156). In accordance with 
 her general theory of mystical origin, Miss Weston saw^ in the term 
 a proof of ritualism. Modern mystics informed her that "Sons of 
 the Widow is a very wide-spread synonym for Initiates"; and its 
 occurrence in ancient Egypt and among the sect of the Manicheans 
 would give probability to this suggestion. But without following 
 Miss Weston into the hazy realm of modern mysticism, we must 
 admit that the term is striking and whether mystical or not is a 
 suitable appellation for one concerned, that is, in Crestien, with the 
 affairs of his mother's kin. This fact will appear clearly as we 
 proceed. 
 
 In the passage following vs. 340 the hero says: " J'ai nom biax 
 fi(l)z." This corresponds to Wolfram' (Parzival), § 140, 6: "bon 
 fiz, scher fiz, bea fiz, alsus hat mich genennet der mich da heime erken- 
 net" — and whatever other significance it has, illustrates well the 
 hero's ingenuousness, for thus any child might be called by its 
 parent.^ Directly after the first grail-visit, the hero's cousin ger- 
 mane (Wolfram's Sigune) again asks his name, and there follows this 
 passage (vss. 3534 ff.) : 
 
 E oil qui son nom ne savoit 
 Devine et dit que il avoit 
 
 kultliche Wurzoln, die in die urarische Zeit ziiruckreichen, werden wir daher Im Hinter- 
 grunde der mannigfachen keltischien Erzahhmgen von Zaubergefassen iind ihrer Gewin- 
 nung vermuten mtissen." Most important is the Rig- Veda (8, 66) version of Indra's 
 conquest of the celestial Soma, the guardians of which are the Gandharvon, in whom 
 Hillebrandt (V'ed. Myth., I, 427, note) had seen the "genius of fertility." Without 
 denying this feature, S. considers the Gandharven as the soul awaiting incarnation and 
 compared him to Lohengrin (on whom, see Pestalozzi, Ilabiliiationsschrift, 1908). In 
 any case, he it is who opposes Indra, who shoots him with bis arrow "und durchbohrt 
 den Gandharven Im bodenlosen Luftraum." The Soma-offering was an invocation for 
 rain {Reyemauber) and Indra's conquest liberated the streams and rivers (cf. PMLA, 
 XXIV [1909], 395 ff.)- But it is incorrect to say that the rain-making feature of the 
 grail ceremony Is "ein bisher ganz dunkcl gebUeboner Zug der Graldichtungen." 
 
 > I cite from the Baist text throughout. 
 
 ' iStr Perceval, II, 306 ff. "Perceval's title, perfectly natural given the donnies of 
 the legend, suggested that he was an Initiate, and he stepped into Gawain's shoes." 
 
 * Cf. also the Bel-Inconnu group of poems, which I intend to consider in a separate 
 article. 
 
 * On the use of biaus frere, see W. A. Stowell, Old French Titles of Respect, 147 ff., and 
 Tappolet, Verwandlscha/tsnamen. 
 
 292 
 
The Sister's Son and the Conte del Graal 3 
 
 Percevat^s li galois a non, 
 E ne set s'il dit voir ou non, 
 E il dit voir si ne le sot. 
 
 That is, he divines or guesses his real name. Whereupon the cousin 
 informs him that the Fisher King 
 
 "eiist regaaigniez 
 Les membres e terre tenist, 
 E[i]nsi granz biens an avenist," 
 
 if Perceval had asked about the lance and the grail. She explains 
 further that his failure to ask was due to the sin he committed in 
 deserting his mother: 
 
 "Per le pechie ce saches tu 
 De ta mere t' est avenu." (Vss. 3555 ff.) 
 
 Now, lest the fact escape us, it should be noted at once that 
 Perceval, simple-minded as he is in the poem, is yet fully conscious 
 of his duty toward his mother. The neglect of this filial obhgation, 
 owing to his enthusiasm for chivalry (Perceval's desmesure), is one 
 of the romantic situations of the Perceval legend which has not always 
 been brought out by the commentators. Having been invested with 
 
 La plus haute ordre avoec I'espee 
 Que Dews a faite e comandee 
 C'est I'ordre de chevalerie, 
 
 Perceval is first of all desirous : 
 
 Que a sa mere venir puisse 
 
 E que sain[ne] e vive la truisse. (Vss. 1676 ff.) 
 
 And later, when Blanchefleur has wooed and won him so that he 
 vanquishes Clamadeus and his seneschal, the excuse he gives for his 
 sudden departure is that he must seek his mother: 
 
 "Que ge ma mere veoir vois." (Vss. 2915 ff.) 
 
 But his absence is to be temporary; he will return with his mother: 
 
 il lor met an covenant 
 S'il trueve sa mere \avant 
 Que avoec lui I'an amanra 
 E d'iluec [anl avant tanra 
 La terre, ce sachiez de fi ; 
 E se ele est morte autresi.^ 
 
 > The last linejis significant as indicating the conclusion of the story. 
 
 293 
 
4 William A. Nitze 
 
 Again, he is imploring God for a glimpse of her (vs. 2944: Qu'il 
 doint veoir sa mere) when he meets the fisherman who directs him 
 to the Castle of the Grail. 
 
 Yet, as we saw, the fact that he deserted her sealed his lips in 
 the presence of the grail, and as a consequence great suffering will 
 befall mankind ("enui an avandra toi e autrui"; of. vss. 3553 ff.). 
 It appears then that the return to the mother (the 7notif of desertion) 
 is of more than incidental importance. Later romances of the 
 Perceval cycle, such as the Perlesvaus and even Syr Percyvelle, 
 bear out this conclusion. The "return" is an obligation the hero 
 is bound to fulfil, since failing to fulfil it he fails ipso facto — in Cres- 
 tien — in his most important adventure. The dramatic interest of 
 our poem lies between these two poles of action: the obligation to 
 the mother and the visit to the Grail Castle, and it seems improb- 
 able that a genuine Perceval tale ever existed without them. In 
 Syr Percyvelle, which to be sure some scholars have considered primi- 
 tive, the bond which united Perceval to the Fisher King seems to 
 me to have been transferred to King Arthur, in my opinion, probably 
 a later development.^ Moreover, the Good Friday episode, that is, 
 the interview with the hermit-uncle, is now seen to be the necessary 
 consequence of what has preceded. The mother having died, Per- 
 ceval is obliged to atone for her death if the curse for which he is 
 responsible is to be removed. Hence the hermit's solicitude, ex- 
 pressed in the words :2 
 
 "Or te vuel anjoindrc e doner 
 Penitance de ce peclii^." 
 
 > Cf. below, p. 26. In Crestien the connection with Arthur's court appears loosely 
 knit. At the same time, the connection may antedate Crestien; he mentions Perceval as 
 an Arthurian knight in his Erec, vs. 1525. The Arthurian portion of the Conte del Graal 
 repeats virtually the technique of the Ivain: (1) motif ot vengeance; (2) fight with a red 
 knight; (3) wooing of the hero — "otherworld" visit; (4) search of the hero by the 
 Arthurian court; (5) unhorsing of Keus; (6) return of the hero to court; (7) messenger 
 denounces the hero; (8) hero departs and wanders distraught. 
 
 ' Vss. 1394-1395. In Wolfram, Book IX, the motive for the question is pity, a feeling 
 still foreign to the youtliful Parzival's heart. Cf. § 473, 15: 
 
 daz er niht zcm wirto sprach 
 
 umbon kumber den cr an im sach. 
 According to Wolfram's ethical interpretation the hero thus commits a fresh sin, since he 
 was evidently free to ask. Cf. § 473, IS: 
 
 doch muoz er siindo engolten, 
 
 daz cr niht fragte des wirtes schaden. 
 This sin he atones for, together with that of his mother's death, at Trevrizcnt's cell. 
 But Wolfram no longer makes the mother's death responsible for the hero's failure. At 
 
 294 
 
The Sister's Son and the Conte del Graal 5 
 
 But why, we ask, should the success of the grail ceremony and the 
 welfare of its two kings depend on Perceval's conduct toward his 
 parent ? For the simple reason, I believe, that Perceval, like Gawain 
 in the main Arthurian tradition, is a sister's son. 
 In the English ballad^ Arthur says to his nephew : 
 
 " Come here, cuzen Gawaine so gay. 
 My sisters sonne be yea." 
 
 With less emphasis but implying as much, the hermit^ explains to the 
 remorseful Perceval: 
 
 "Cil cui Fan sert fu mes frere, 
 Ma suer e soe fu ta mere." 
 
 In other words, the Grail King is our hero's maternal uncle, and a 
 closer male relative Perceval could not have had. This concept is, 
 I think, the basis of Crestien's plot. Through youthful ardor 
 Perceval neglects his kin — that same kin, in the person of his maternal 
 uncle, stands in the way of his final success until the former misdeed 
 has been duly expiated. Not only was the success of the grail- 
 visit to benefit the kin, but its failure prolongs the suffering of its 
 chief representative, and inflicts positive harm on those connected 
 with it. Therefore the messenger proclaims to Arthur's court in 
 
 vss. 4640 ff . : 
 
 "Dames an perdront lor mariz, 
 Terres an seront esilli^es, 
 E puceles desconseilliees 
 Qui orfelines remandront, 
 E maint chevalier an morront, 
 E tuit avront le mal par toi."^ 
 
 the same time the kinship ties are essentially those mentioned by Crestien (see below). 
 Trevrizent calls Parzival, § 475, 19: "lieber swester suon," and adds a significant detail 
 not fomid in Crestien ; namely, that in killing the Red Knight (Ither, according to Wol- 
 fram) Perceval has slain his own kin ("din eigen verch erslagn"), for Ither is Parzival's 
 cousin (on the paternal side, § 498, 13), himself the sister's son of Uther Pendragon, 
 § 145, 12. His death Parzival is also to reqiiite. Thus Wolfram has complicated further 
 the bonds of relationship, whether on his own initiative or not is impossible to tell, making 
 the connection between the Arthurian court and the Grail dynasty more intimate than 
 in Crestien. On the other hand, his hero has advanced a step in that the motive of his 
 silence is no longer an external "taboo" but a youthful lack of human. Christian sym- 
 pathy. 
 
 1 King Arthur and King Cornwall; see Sargent and Kittredge, Eng. and Scot. Pop. 
 Ballads (Boston, 1904), p. 50. Compare the interesting article of Professor Gummere in 
 the Furnivall Miscellany, where additional instances may be found. 
 
 2 Vss. 6377 fif. 
 
 ' Cf. Parzival, §§ 316 ff., where there is no trace of this general effect in the mes- 
 senger's imprecation. See above, p. 4, note. 
 
 295 
 
6 William A. Nitze 
 
 Indeed, the messenger's reference to Fortune i^ 
 
 "Ha! Percevaus fortune est chauve 
 Derriers e devant chevelue, 
 E dahez ait qui te salue!" — 
 
 which the author of the Perlesvaus^ elaborated, is singularly appro- 
 priate to this dramatic moment. There is almost a touch of Greek 
 feeling in the maimer in which the poet here portrays the inexor- 
 ableness of fate. 
 
 In agreement with this fundamental notion of the sanctity of 
 kinship, we find that the poem carefully explains all the kinship ties, 
 and that with one exception these are all on the maternal side. 
 Like the Grail King, the hermit to whom Perceval does penance is a 
 maternal uncle, the Fisher King is a cousin germane, likewise the 
 damsel whom Perceval meets outside the Grail Castle (vs. 3562). 
 Blanchefieur, of course, belongs to a different gens, but her uncle 
 is Gornemanz,^ whose brother germane was slain by Anguinguerron 
 (vs. 2270), their common enemy. The exception is found in the 
 account given of the hero's father by Perceval's mother (vss. 398- 
 468; Pot., II, vss. 1607-1682). The father was feared An totes Us 
 isles de mer; the son may boast, says the mother: 
 "Que vos ne descheez de rien 
 
 De son lignage ne del mien, 
 
 Car je fui de chevaliers nee 
 
 Des mellors de ceste contree. 
 
 Es isles de mer n'ot lignage 
 
 Meillor del mien an mon aage." (Vss. 401-406.) 
 
 The passage I have italicized shows where the emphasis is placed; 
 that is, although the mother is speaking of her husband she stresses 
 her own lineage.^ Besides, it would have been unusual had Crestien 
 made no reference to Perceval's father (cf. his other romances, 
 especially Cliges). As I have stated elsewhere the manner of the 
 father's wounding {parmi les handles) is so similar to that of the 
 Fisher King that it possesses little originality.^ While, then, I see 
 
 > Vss. 4608 fl. Cf. MLN, VIII (1893), 230-38. for the history of the expression. 
 » Pot., I, 24 fr. 
 
 • In Peredur Gomemanz is confused with the hero's own uncle; see Loth, Mabinog., 
 II, 59; Hertz, Parz.*, 479. 
 
 • Cf. PerleavauB, p. 185. 
 
 » MLN, XXV (1910). 249. See, also Geoffrey of Monmouth, Historia, XI, 1-2, 
 where Arthur himself is wounded Uialiler and carried to Avalon to be healed. 1 tried 
 
 296 
 
The Sister's Son and the Conte del Graal 7 
 
 no reason for agreeing with NewelP that "the passage, intended to 
 emphasize the woes of the widow seems .... obviously to be the 
 work of a later hand," nevertheless Wolfram's statement that the 
 mother was already a widow at the time of her flight to the woods 
 appears to me closer to the original situation. To what apparent 
 absurdities the uncertainty as to Perceval's father could lead is seen 
 in the Perlesvaus,^ where the father is still living when the son leaves 
 home. The other references Crestien makes — in this connection — 
 to Uther Pendragon (pere le hon roi Artu), the King of (Es)cavalon, 
 and Ban de Gomeret,^ are perhaps inspired by the desire to connect 
 the tale more intimately with the Arthurian setting, though it is 
 well not to affirm this too strongly. In any case, since Crestien 
 has not given the father a name, and does not so much as mention 
 him again, we may assume that to all intents and purposes he was 
 nameless, and that his having once existed is a fact of no genuine 
 importance to the plot of the romance. Thus it follows that a 
 primary condition of the story is kinship, and that this kinship is 
 matrilinear or matriarchal. 
 
 The full significance of this I shall bring out presently. It may 
 be noted first that the outline of the story conforms entirely to this 
 situation: A youth of uncertain fatherhood fares forth into the 
 world to win renown, and an inheritance. His ignorance of life 
 appears in various foolish or ill-advised acts he commits. But he 
 is valiant and strong, and obtains assistance (instruction) from those 
 who require his support. He frees a luckless maiden from oppres- 
 sors and then weds her. The maiden virtually offers herself to him. 
 He comes within reach of his goal, and is on the point of being recog- 
 nized and established by his nearest kin (the maternal uncle), but 
 fails at first because his mother has died through his neglect. 
 
 to make clear in my "Fisher King" that the GraU King, and not his son, seems to have 
 been originally the "important" person. In Crestien he is stiU the uncle, though it is his 
 son (i.e., his earthly representative) who is to be cured. Cf. the reviews of my article: 
 F. Lot, Bibl. de I'ecole des chartes, LXX (1909); A. Nutt, Folk-Lore, XXI (1910), 112 ff.; 
 E. Brugger, ZffS., XXXVI (2-4) (1910), 71-74; L. Jordan, Literaturbl., XXXII (1911), 
 335-37. 
 
 ^King Arthur and the Table Round (Boston 1897), II, 252. 
 
 2 Pot. I, 19. In Perlesvaus the hero is descended through his father from Glais 
 (patronymic of Glastonbury, see MP, I [1903], 248) and through him from Nicodemus. 
 
 ' Cf . Erec, VS. 1811: V usage Pendragon man pere .... doi je garder e maintenir; 
 vs. 1975: vint li rois Bans de Gomeret; on Escavalon, see the Gawain part of the Conte del 
 Graal. 
 
 297 
 
8 William A. Nitze 
 
 As is well known, Crestien did not complete his romance. In 
 fact, in MS 794 the text breaks off in the middle of a sentence. But 
 from what has been said it is sufficiently clear that having atoned for 
 his fault, Perceval was doubtless to return to the Grail, perhaps to 
 be initiated into its mystery,^ probably to succeed the Fisher King, and 
 of course to be reunited to Blanchefleur. Any other plan on Cres- 
 tien's part seems to me precluded. The continuation of the poem 
 found in MS Bern 113 — the so-called Rochat Perceval^ — practically 
 ends in this way :^ the Fisher King before dying says : 
 
 "Ore, biaus ni^s, si est bien drois, 
 ains que vos avant en sacois, 
 que vos corone d'or port^s, 
 sor vostre cief, et rois seres, 
 car ne vivrai mais que tier ior, 
 ensi plaist il a creator." 
 
 So much for the problem presented to us. The matriarchal idea 
 is evidently the Leitmotiv of Crestien's work in contradistinction to 
 Wolfram and the later romancers, whose hero is actuated by Chris- 
 tian ideals — in Wolfram by "pity," in the Quest-versions by 
 "purity," to which the author of the Perlesvaus adds the Augustinian 
 idea of grace ("Sire, fit li hermites, or n'oubliez pas a demander, se 
 Diex le vos veust consantir, ce que li autres chevaliers oubha*). 
 
 Let us now inquire into matriarchy as a system, the evidence of 
 its survival in the territory from which Crestien's work may have 
 come, the references to the idea in other story-material of the 
 twelfth century, and the significance of Crestien's Conte del Graal as 
 an illustration of the system. 
 
 II 
 
 Generally speaking, matriarchy is the system of tracing family 
 descent through the mother's line, and is not to be confused with 
 
 > See my "Fisher King," PMLA, XXIV (1909). 365 ff. 
 
 ' Cf. A. Rochat, Ueber einen unbekannten Percheral le Gallois. Zurich, 1855, p. 91; in 
 this version the Fisher King is P.'s uncle; see Hcinzel, Franzda. Gralromane, 58 fl. 
 
 » Compare also the ending of the Parzital, §§ 781 fl.: Parzival joins his wife and two 
 sons at the Grail Castle, asks the question, thus releasing Amfortas, and becomes ruler of 
 the Grail liingdom. Also Didot-Perceval, Hucher I, 484-485; Weston, Sir Perceval. II, 84. 
 
 * Pot., I, 83; cf. also 130: "Et Damediex li doint tel volenti et vos autresi; que vos 
 puissiez feire la volenti au Sauvfior"; also 5, 89 ff. "Pity" Gawain has in Perlesvaus, 
 89: "si an a grant piti6. et ne U souvient d'autro chose que de la doulor que cil rois 
 soufre" — yet this does not suffice. 
 
 298 
 
The Sister's Son and the Conte del Graal 9 
 
 gynocracy or the supremacy of women. It is defined with German 
 precision by Schrader^ as "jene, wie die Volkerkunde lehrt, noch 
 heute bei gewissen Volkern des Erdballs iibhche Familienordnung, 
 die zwar den Begriff des Ehemanns (auf langere oder kiirzere Dauer) 
 nicht aber den des Vaters kennt, weil eben die Kinder nicht dem 
 Vater, sondern der Mutter gehoren und nicht den Vater oder Vater- 
 bruder, sondern die Mutter, beziiglich den Mutterbruder oder 
 miitterhchen Grossoheim beerben." Essentially then, matriarchy is 
 a law of descent based on "the larger social fact, including the bio- 
 logical one, that the bond between mother and child is the closest in 
 Nature."- In recent writers on this subject it is made clear that 
 while the expression of the male power is obscured, the principle of 
 male authority is always in force, and the tribal matriarchal group 
 may be defined as "a fighting male organization living in a group of 
 females."^ Hence it is not inconsistent to find in the Perceval story 
 man physically and politically supreme and yet filiation taking place 
 on the side of woman. 
 
 While power is thus vested in the male, the leader of the group is 
 not the blood father but the mother's nearest male relative, usually 
 her oldest brother.* This necessarily resulted from the ephemeral 
 nature of primitive marriage and from the custom that the husband 
 during the period of cohabitation resided with his wife's kin.^ In 
 
 1 Indogermanen (1911), pp. 75 fl. 
 
 2 From W. I. Thomas, Sex and Society (Chicago, 1907), p. 66. The sociological 
 literature on the subject is large and is constantly growing. In the main, I have followed 
 the following authorities: E. Westermarck, The History of Human Marriage, 39-100 ff; 
 L. von Dargun, Mutterrecht u. Raubehe, Breslau, 1883; idem, Mutterrecht u. Vaterrecht, 
 Leipzig, 1892; J. J. Bachofen, Das Mutterrecht, Stuttgart, 1861; E. W. Hopkins, Journal 
 of American Oriental Society, XIII, 56fif. ; B. Delbriick, " Indogermanische Verwandt- 
 schaftsnamen," Saxon Academy of Sciences, XI, 586 flf.; O. Schrader, Prehistoric Antiq- 
 uities of the Aryan People (tr. Jevons), (London, 1890,) pp. 395 fif. ; G. Wilken, Das Matri- 
 arcat bei den alten Arabern, Leipzig, 1884; W. Robertson Smith, Kinship and Marriage 
 in Early Arabia, 1885; H. Ploss, Das Weib in der Natur- und Volkerkunde^, II, 379; E. B, 
 Tylor, "Matriarchal Family System," Nineteenth Century, 1896, p. 89; M. A. Potter. 
 Sohrab and Rustem (London, 1902), pp. 107 fl.; E. S. Hartland, Primitive Paternity (Lon- 
 don, 1909), I, chap, iv, "Motherright." I was vmable to procure Otto Hoffmann, "Die 
 Verwandtschaft mit der Sippe der Frau," in the Festschrift zur Jahrhundertfeier der Uni- 
 versitdt zu Breslau. 
 
 8 Cf. Thomas, op. cit., 69, note. Perhaps it is safer to say, instead of "a group of 
 females," a group of which the female is the imit of descent. I would guard against the 
 danger of overstating the case. See, for example, in reference to the Greeks, Fame 11, 
 Archivf. Religionswiss., VII (1904), 70 and H. J. Rose, Folk-Lore, XXII (1911), 277. 
 
 ^ Dargun, op. cit., 56-57; Hartland, Primitive Paternity, I, 99. 
 
 5 Hartland, op. eit., I, 59 ff. 
 
 299 
 
 N 
 
 X 
 
10 William A. Nitze 
 
 this way the sister gains great prominence in the social system^ 
 since it is her son who succeeds to the potestas of the family. "The 
 practice of the Wamoima [in Africa]," says Thomas,^ "where the son 
 of the sister is preferred in legacies, because 'a man's own son is 
 only the son of his wife/ is typical." Examples have been gathered 
 from almost every part of the world which establish the correctness 
 of this statement. Wherever matriarchy prevails or prevailed, and 
 its occurrence is exceedingly widespread, the tendency constantly 
 is to accord the nephew or niece more importance than the direct 
 offspring. A few striking cases may be cited; for others I refer to 
 the authorities cited in the footnotes. 
 
 On the island of Efate in the New Hebrides a kindred or family reckon- 
 ing descent from the same mother in female line is called nakainanga 
 
 Hence it was the duty of a man to instruct his sister's son, not his own 
 son, because he was not of the same nakainanga and the father would not 
 be responsible for him. The chief of a village has the right to appoint his 
 successor. He appoints not his own son, "but in preference to all others his 
 sister's son, who by the law of the nakainanga is considered nearer and dearer 
 to him than his own son, and to be his proper heir."^ 
 
 Among the Tahl-tan of British Columbia "kinship so far as mar- 
 riage or inheritance of property goes, is with the mother exclusively; 
 and the father is not considered a relative by blood."'* The Wyandot 
 Indians, according to J. W. Powell,^ recognize four groups: the 
 family, the gens, the phratry, and the tribe. 
 
 The gens is an organized body of consanguineal kindred in the female 
 line. "The woman carries the gens" is the formulated statement by which 
 a Wyandot expresses the idea that descent is in the female line. Each gens 
 has the name of some animal, the ancient of such animal being its tutelar 
 
 god Each gens is allied to other gentes by consanguineal kinship 
 
 through the male line, and by affinity through marriage Children, 
 
 irrespective of sex, belong to the gens of the mother." 
 
 « See J. J. Bachofon, Antiquarische Brief e, I, 144-209, for the intimate ties between 
 brother and sister. 
 
 • Op. cit., 62; from J. Lippert. Kulturgeachichte, II, 57. 
 
 » See Hartland. I. 291. from Rev. D. Macdonald in Rep. Auslr. Assoc, IV, 722-23. 
 
 * Hartland, op. cit., I, 280, from Dawson, Am. Rep. Geol. Survey Canada, 1887, 
 pp. 7 ff. 
 
 ' Reports of the Bureau of Ethnology, I, 59 (T. 
 
 300 
 
The Sister's Son and the Conte del Graal 11 
 
 In northeast India "the son does not succeed his father, but the 
 raja's neglected offspring may become a common peasant or laborer; 
 the sister's son succeeds to rank, and is heir to the property."^ 
 
 In Loango [in Africa] the uncle is addressed as Tate (father). He exer- 
 cises paternal authority over his nephew, whom he can even sell. The 
 father has no power; and if the husband and wife separate the children 
 foUow the mother as belonging to her brother. They inherit from their 
 mother; the father's property, on the other hand, goes at his death to his 
 brother (by the same mother) or to his sister's sons.^ 
 
 Possible traces of the system are found in the Bible: Sarah is 
 related to Abraham only on the paternal side (Gen. 20:12); Tamar 
 could have become the wife of Amnon, her half-brother (II Sam. 
 13:13); Laban tells Jacob: "These daughters are my daughters, 
 and these children are my children" (Gen. 31:43), and the well- 
 known marriage injunction is: "Therefore shall a man leave his 
 father and his mother, and shall cleave unto his wife" (Gen. 2:24). 
 The laws of Solon in Athens also permitted the marriage of brother 
 and sister who were not of the same mother.^ Herodotus^ says: 
 "Ask a Lycian who he is, and he will answer by giving his own name, 
 that of his mother, and so on in the female line." According to W. 
 Robertson Smith^ the ancient Arab sentiment held the sanctity of 
 women to be inviolate, the greatest of insults being an insult to 
 them — an idea which he traces to female kinship. 
 
 Accordingly, the bond between brother and sister is scarcely 
 second to that uniting mother and son.^ Antigone in Sophocles 
 endures for Polynices toil and suffering that she would not have 
 undergone for husband or children.'^ The wife of Intaphernes, 
 Herodotus^ tells us, when allowed by Darius to claim the life of a 
 single man of her kin, chose her brother, sajdng that husband and 
 children could all be replaced. In this connection the student of 
 
 'Tylor, op. cit., p. 87. 
 
 2 Hartland, op. cit., I, 281. From A. Bastian, Deutsche Exped. an der Loango-Kustt 
 (Jena. 1874-75), I, 166. 
 
 3 Wilken, op. cit., 41. 
 
 * Rawlinson, I, 173; cf. Thomas, op. cit., 64. 
 
 » Op. cit., 100 ff. 
 
 ' See especially J. J. Bachofen, Antiquarische Brief e, I, 144-209, and Potter, 
 op. cit., 196. 
 
 ' Thomas, op. ct7., 65. » Rawlinson, III, 119. 
 
 301 
 
12 William A. Nitze 
 
 Arthurian romance is at once reminded of the significant role taken 
 by Perceval's sister in the Quest-versions^ and the striking passage 
 in the Conte del Graal (vss. 8697 ff.) where Sir Gawain plays an 
 important part in the affairs of his female kindred.^ So, too, Pro- 
 fessor Gummere finds in his ballad studies^ that brother and sister 
 afford older instances of confidence and affection than husband and 
 wife or lover and sweetheart. Ballad literature records the same 
 preference of the sister's son over a man's own child which we noted 
 above apropos of African savages — a condition which Tacitus^ posits 
 of the ancient Germans in the words: "quidam sanctiorem arti- 
 oremque hunc nexum sanguinis arbitrantur et in accipiendis obsidi- 
 bus magis exigunt, tamquam et animum firmius et domus latins 
 teneant." Gummere cites an excellent example,^ probably the best 
 on record, of the concentration of this kinship bond, in the Danish 
 ballad Nilus og HilleliUe; here the marriage of Sir Nilus places in 
 mortal conflict brother and sister, two sister's sons, and the maternal 
 uncle. 
 
 While, therefore, the matriarchal system put the maternal uncle 
 in an exalted place, as the chief of the clan — the juridical iruT-qp of 
 the family — granting him control over its children, the physical 
 father, if recognized at all, was quite a subordinate personage. 
 This practice obtained, as has frequently been shown, as long as 
 women controlled the marital arrangement, as long as they exercised 
 the primary right of selection and possessed, as they did in ancient 
 Arabia,^ the privilege of dismissing their husbands and choosing 
 others. Says Hartland:^ 
 
 Where the matrilineal clan is in full force, or where the family has been 
 formed within the larger organisation of the clan but has not yet succeeded 
 in supplanting it for effective social government, the husband remains 
 subordinate to the wife's male kinsmen, her uncles or her brothers. 
 
 > See Jessie L. Weston. Sir Perceval, II, chap, v; Brugger. ZffS., XXX (1904) (6-8), 
 126 ff. 
 
 » Quant ele do fl Ic savra 
 Qu' cle est sa suer et il ses frere 
 S 'an avra grant joie sa mere 
 Autre que ele n'i atant. (Vss. 9034 ff.) 
 Cf. also Weston, op. cii., I. 209. 
 
 • The Popular Ballad, 182 ff. * Germ. 20. 9. Op. cil., 183. 
 
 • J. Robertson Smith, op. cil., 65 ff. ' Primitive Paternity, II. 94. 
 
 302 
 
The Sister's Son and the Conte del Graal 13 
 
 And even when the family as such has become firmly established, 
 the sons sometimes still pass at an early age to their uncle's care or 
 must be purchased from the wife if they are to continue in the 
 father's control.^ The rise of the paternal supremacy is a moot 
 question which camiot be discussed here. Nor does it affect our 
 problem directly, though it may enter into the larger question of the 
 primary motifs of Arthurian romance. But for the present that is 
 neither here nor there, and it is enough to note that primitive man 
 is averse to sudden, radical change, and even long after the triumph 
 of the male, the social sanctions of the past obtain to a considerable 
 extent. So that we may find "a formal elevation of woman to author- 
 ity in groups where the actual control is in the hands of men."^ 
 Moreover, we should not forget that the word ''clan" is essentially 
 a sociological term. Only in a general way is it synonymous with 
 blood-relationship.^ While the blood-bond is considered to unite 
 the various members of a clan, it can be acquired through ceremony, 
 or by sucking the blood of a member of the clan, as Cuchulinn sucks 
 that of Dervorgil, thus becoming her blood-brother.^ In this way, 
 "many savage peoples are organized as totemic clans, each clan 
 bearing the name of an animal or plant supposed to be akin to the 
 human members of the clan,"^ The clan once constituted, each 
 member shared in the privileges and deprivations to which any 
 member thereof was liable. He was entitled to protection, but he 
 was compelled to guard the clan against attack and to take part in 
 its feuds and avenge injuries to it. But his greatest restriction was 
 "the prohibition to marry or have sexual relations with any woman 
 
 1 Hartland, op. cit., II, 99. 2 Thomas, op. cit., 73. 
 
 3 See the clear distinction made between the "physiological" and the "social" by 
 Van Gennep in his Rites de passage, Paris, 1909. "C'est [p. 4] le fait meme de vlvre 
 qui necessite les passages successifs d'mie societe speciale a une autre et d'une situation 
 sociale a une autre: en sorte que la vie individueUe consiste en une succession d'etapes 
 dont les fins et commencements forment des ensembles de meme ordre : naissance, puberte 
 sociale, mariage, patemite, progression de classe, speciahsation d'occupation, mort. 
 Et a chacun de ces ensembles se rapportent des ceremonies dont I'objet est identique: 
 faire passer I'individu d'une situation determinee a ime autre situation tout aussi dfiter- 
 minee. " And in speaking of initiations proper (p. 97): "11 convient done de distinguer 
 de la puberU physique la puberte sociale, de meme qu'on distingue entre une parents 
 physique (consanguinite) et une parente sociale, entre une maturite physique et une maturitS 
 sociale (majorite), etc." I follow in the presentation above the discussion found in 
 Hartland, op. cit., I, 257 fl. 
 
 * Eleanor Hull, Cuchullin Saga, 82. 
 
 s Hartland, op. cit., I, 257. 
 
 303 
 
14 William A. Nitze 
 
 of the kin." Consequently we find marriage taking place with 
 women of a different kin. For obvious reasons, the theories on the 
 origin of exogamy camiot be discussed here.^ Our purpose, however, 
 is served in recording its existence, practically the world over, in 
 connection with the primitive clan organization. Endogamous 
 clans there were, but the characteristic tribal life was exogamous. 
 Upon this all ethnologists agree. ^ 
 
 III 
 
 In considering now the evidence presented by Arthurian litera- 
 ture, we naturally look first of all to the Celts, especially as we have 
 good reasons for considering the grail ceremony to be in the main 
 of Celtic origin.^ Relations between the French (Normans and 
 Angevins) and the Celts of Brittany and Wales were intimate from 
 the time of the Conqueror until the end of Henry the Second's reign, 
 while there was inter-communication between Ireland and the Con- 
 tinent as well as between Ireland and Britain long before and during 
 this period.* We have ample reason therefore to take note of what 
 the ancient Irish law^ has to say with reference to the duties of the 
 
 • Cf. G. E. Howard, History of Matrimonial Institutions, I, 117, on the "problem 
 of exogamy." Frazcr, J. G., in his Totemism and Exogamy (New York, 1910), argues for 
 thp separation of the problem of exogamy from that of totemism. See with respect to 
 his views the discussion in Folk-Lore, XXII (1911), 48-81, by Westermarck, A. Lang, and 
 Van Gennep. 
 
 ' Frazer's argument (see above) is that exogamous rules sprang from an aversion to 
 marriages of near kin. This view Lang enlarges {op. cit., 84) by saying that the "aver- 
 sions to such unions, through the association of ideas, led to the proliibitions of marriage 
 between members of the same clan on account of the notion of intimacy connected with a 
 common descent and a common name." Thomas, op. cit.. 57, says, "aside from its origin, 
 exogamy is an energetic expression of the male nature. Natural selection favors the 
 process by sparing the groups which by breeding out have heightened their physical 
 vigor." 
 
 » See my previous articles in PMLA, XXIV (1909), and Elliott- Studies (1911). 
 
 « The evidence on the commimication with Ireland has recently been put together 
 by Dr. T. P. Cross in his study of Marie's Yonec in the Revue celtique, XXXI (1910). 
 424 ff., and it is unnecessary to repeat here. On Norman and Breton, and Norman- 
 Welsh relationships see the historians of the period: notably Freeman, Norman Con- 
 Quest'; Davis, England under the Normans and Angevins, London, 1905; J. E. Lloyd 
 History of Wales from the Earliest Times to the Edwardian Conquest, Vol. 11; D'Arbois de 
 Jubainville, Histoire des dues et comtes de Champagne, Vol. II; Zimmer, Gdtt. gel. Ameige 
 No. 20, 1890; J. Loth, Revue celtique, XIII (1892), 475-503; John C. Fox, Eng. Hist. 
 Rev., XXV (1910), 303-306; Zimmer, "Ueber directe Handelsverbindungen Westgalliens 
 mitlrland Im Alterthum und frUhen Mittelalter," Prussian Academy, 1910, II. 
 
 ' Cf. Ancient Laws of Ireland (Senchas mor), Dublin, 1865-1901; H. D'Arbois de 
 Jubainville, Risumi d'un cours de droit irlandais, Paris, 1888-92; idem. Etudes sur le 
 droit celtique, Paris, 1895 (.Cours de litt. celt., VII, VIII); idem. La famille celtique, Paris, 
 1905; Joyce, Social History of Ireland, 1903; O'Curry, Manners and Customs of th* 
 Ancient Irish, 3 vols. 
 
 301 
 
The Sister's Son and the Conte del Graal 15 
 
 kindred. The Irish law provides that when a sister's son has been 
 slain, the maternal uncle shall avenge him.^ Glasfine is the technical 
 name given him; that is, in D'Arbois' words,^ "famille grise et 
 bleue, parce que, dit-on, le pere est un etranger qui est arriv^ en 
 Irlande sur la mer grise et bleue; il n'a par consequent pas de famille 
 en Irlande; il ne peut done donner une famille a son fils, et celui-ci 
 est considere comme faisant partie de la famille de sa mere." In 
 harmony with this idea we find the Irish practice of tinnscra or male 
 dowTy (le douaire), essentially a form of the bride-price. A passage in 
 the Book of Leinster accounts for it on a legendary basis.^ It is by 
 this means that Conchobar becomes king of Ulster, for when Fergus, 
 son of Rogh, sues for the hand of Conchobar's mother, she stipulates 
 as a marriage portion that Conchobar should hold the kingdom for a 
 year so that his children may be known as the children of a king.* 
 
 The matrilinear side of Celtic tribal life is also seen in various 
 provisions of old Welsh law. For example, if a youth or maiden 
 imder twelve, because of the father's death, is placed under guardian- 
 ship, the guardian is of the maternal kin, so that he may not be 
 tempted to deprive his charge of his property or shorten his life.^ 
 Again, one version of the Vendotian Code^ provided that in case of 
 murder, one-third of the blood-money (galanas) had to be paid by the 
 murderer and his father and mother, if living; but two-thirds fell 
 on the kindred, which "was defined as 'from maternity to maternity 
 unto the seventh descent.'" The Welsh law, like the Irish, made 
 provision that when a girl married a non-tribesman,^ the responsi- 
 bility for her sons rested with the maternal kindred, who had to 
 provide an inheritance and pay the fine in case they committed a 
 
 1 Ancient Laws. IV, 244, 11. 20-22. 
 
 » Op. cit., VII, 187 ff. See also M. A. Potter, op. cii., 125 flf. 
 
 3 Cf. D'Arbois, op. cit., 233; O'Curry, Lectures on the MS Materials of Ancient Irish 
 History, 501. 
 
 * Ct. E. Hull, Cuchullin Saga; Lady Gregory, Cuchulain of Muirthemne; "Windisch 
 ed. of Book of Leinster," Saxon Academy, XXXVI (1884); D'Arbois, Epopee celtique 
 (Cours de lit., I). 
 
 6 Walter, Das alte Wales, § 199; also Welsh Medieval Law (Laws of Howel the Good), 
 ed. Wade-Evans, Oxford, 1909. 
 
 6 See F. Seebohm, Tribal System in Wales (London, 1895), p. 79 (for a different version 
 see p. 80). Consult also the evidence cited by Hartland, op. cit., I, 274; and A. Owen, 
 Ancient Laws and Institutes of Wales, 109 ff. 
 
 ' According to H. Lewis, Ancient Laws of Wales (1889), pp. 56-57, 197, marriage had 
 to be outside of the trev or kindred who hved in one inclosure. 
 
 305 
 
16 William A. Nitze 
 
 crime.^ "Das Wichtigste," says Walter,^ "aber war dass solche 
 Sohne ihren miitterlichen Grossvater neben den Briidern ihrer Mut- 
 ter beerbten, selbst wenn diese die Tochter des Grundherrn war, so 
 dass dann deren Sohn der Grundherr seines eigenen Vaters wurde." 
 Thus while the Celtic clan, at the time we can observe it in the 
 extant law tracts, was maintained on the agnatic principle, there is 
 ample evidence that it may have passed through the matriarchal 
 stage,' traces of which survive until relatively late. So that we may 
 agree with Lang^ that the presumption is that Celtic tribal society 
 developed much as the local tribes did elsewhere, on the basis of 
 kinship as first reckoned in the female line. 
 
 As for the Picts there is little doubt that royal succession never 
 went from father to son in early times. "Failing brothers," says 
 Lang, "the succession went to the son of the sister." Not that 
 women were politically dominant. "Im Gegentheil," says Zimmer,^ 
 "nirgends herrscht, soviel wir sehen eine Frau: die Mutter, also die 
 Geburt, bestimmt aber die Stammeszugehorigkeit. Auf einen 
 Piktenherrscher und seine Briider folgt nicht etwa der Sohn des 
 altesten, sondern der Sohn der Schwester, auf diesen und seine 
 eventuellen Briider von Mutterseite folgt wieder ein Schwestersohn 
 und so fort." 
 
 » Wade-Evans, op. cii., 211: "If a Cymraes (i.e., a Cymric woman] be given to an 
 alltud, her children shall have a share of land except the principal homestead; that they 
 are not to receive imtil the third generation; and therefrom originate cattle without 
 surety, because, if ho commits a crime, the mother's kindred pay the whole ot his galanas." 
 Cf. according to Walter, the Vendolian Code (Peniarth MS, 29T I, 97. 
 
 2 Op. cit., 165. 
 
 » See J. E. Lloyd, History of Wales, I, 284 tt. 
 
 * History of Scotland, I, 78 ff. Contrast D'Arbois, Cours de litt. celt., VII, 242 ff.; 
 also S. Ueinach, Revue celtique, XXVIII (1907), 233. " Le droit irlandais," says D'Arbois 
 (p. 246), "conserve a la puissance paternelle la durfie consacr6e par la coutumo primitive 
 celto-romaine." Yet, the son of a sister, the aonMAC, "avait droit a I'usufruit de I'hfiri- 
 tage maternelle." And: "lo privilege modesto accord6 au flls de la soeur par lo droit 
 Irlandais nous 61oigne du droit romain avec Icquel s'accorde le droit gaulois quand, en 
 rfegle g6n6ral, il fait durer la puissance patcrnelJe aussi longtemps que la vie du p6re." 
 
 ' Zeilsch. der Savigny-Sliftuno fUr Rechtsgeschichte, XV (1894), 218-19. The ethno- 
 logical theories, however, which Zimmer sets up on the basis of the above fact (see pp. 
 234 ff.) must be taken cum grano aalis. See Revue celtique, XVI (1895), 188-20: "La 
 loi irlandaisc du IX* sificle," says D'Arbois. "admet lo droit successoral des neveux par 
 les femmes." The Prussian Academy has recently published (Feb., 1911), a further 
 article by Zimmer, " Der kulturhistorischo Hintergrund in der altirischen Heldensage," 
 which elaborates that scholar's views; on this, in turn, consult Revue celtique, XXXII 
 (1911), 232 ff. 
 
 306 
 
The Sister's Son and the Conte del Graal 17 
 
 Turning now to Irish literature, we find that Conchobar is com- 
 monly known by his matronymic alone, as Conchobar mac Nessa.^ 
 This fact Nutt has discussed in an interesting manner in his treat- 
 ment of the supernatural birth.^ "He [Conchobar] was the son of a 
 god who incarnated himself in the same way as did Lug and Etain; 
 this is probably the oldest form, and it may be owing to the fact that 
 the father's name was unknown that Conchobar is usually described 
 by the matronymic alone," Perhaps the case is more correctly 
 stated by saying, that here, as elsewhere among primitive peoples 
 (with few exceptions), kinship being reckoned through the mother, 
 the question of paternity, i.e., actual paternity, was little regarded, 
 except that a great hero must naturally have sprung from a father of 
 illustrious rank. Hence the motherhood is fixed while the fatherhood 
 varies; and supernatural birth occurs most commonly where the 
 principle of matriarchy prevails.^ 
 
 This fact is quite evident in the birth-story of Conall Cernach — 
 the primitive character of which is obvious. In the Coir Anmann* 
 we read : 
 
 When his mother's brother, Get Magach, heard that his sister would bear 
 a child that should slay more than half the men of Connaught, he continued 
 
 protecting his sister until she should bring forth her boy Druids 
 
 came to baptize the child into heathenism, and they sang the heathen 
 baptism over the httle child, and they said: " Never shall be born a boy more 
 impious than this one toward the men of Connaught [his mother's kin] ; not 
 a night shall he be without a Connaughtman's head on his belt." Then Get 
 drew the Uttle child towards him, and put it under his heel and bruised its 
 neck, but did not break its spinal marrow. Whereupon its mother exclaims 
 to Get: "Wolfish [conda] is the treachery [fell] thou workest, brother!" 
 
 1 D'Arbois, Cours de litt. celt., V, 4 fif. (transl. by Dottin); Thurneysen, Sagen aus 
 dem alten Irland, 63; Kimo Meyer, Revue celtique, VI (1885), 173. So, too. Lug is 
 called Lug mac Ethlenn (the name of his mother), cf. Revue celtique XVI (1895), 298. 
 
 2 Nutt, Voyage of Bran, II, 72 flf. 
 
 » Hartland, op. cit., II, 283 ff. H. rejects the view that motherright is foimded on the 
 uncertainty of paternity, and concludes that "whereas motherright was foimded on the 
 recognition of a common blood, fatherright was traceable to social and economic causes 
 of a different character, that no assertion of a common blood was implied in fatherright, 
 but that it was an artificial organization formed upon the analogy of the organization of 
 motherright which it supplanted." I am, of course, unable to judge the extent to which 
 Hartland's views may be accepted, but that the primitive blood-tie, and all that the term 
 implies, was through the mother, can hardly be doubted. 
 
 * See Whitley Stokes, Iriache Texte, III. 2 (Leipzig, 1897), summary by Nutt, op. cit., 
 II. 74. 
 
 307 
 
18 William A. Nitze 
 
 "True," saj^s Cet, "let Conall [Con-feall] be his name henceforward." And 
 he gave her son back to her. Whence he is called wry-necked Conall. 
 
 Again, Bres,' the son of Eri, is on his maternal side a kinsman 
 of the Tuatha D6; for even if his mother is both the wife and the 
 sister of Elatha, king of the Fomorians, she and Elatha are the chil- 
 dren of Delbaeth, a king of the Tuatha De Danaan;^ accordingly the 
 text^ speaks of her as " a woman of the Tuath De." And when Nuada 
 falls ill, we find Bres succeeding him. "A contention as to the sov- 
 ranty of the men of Ireland arose between the Tuath D6 and their 
 women; because Nuada, after his hand had been stricken off, was 
 disqualified to be king. They said it would be fitter for them (to 
 bestow) the kingdom on Bres, son of Elatha, on their o^vn adopted 
 son; and that giving the kingdom to him would bind tho alliance of 
 the Fomorians to them." But Bres maltreats them sorely and they 
 depose him. Thereupon, with the help of his mother, he seeks the 
 aid of the Fomorians. And thus the Battle of Moytura is brought 
 about, in which the Tuatha De are victorious. 
 
 But perhaps the most notable example of matrilinear kinship 
 in Irish is the bond which unites Cuchulinn to Conchobar. While 
 what seems to be the earliest version of the Cuchulinn (birth) story^ 
 represents the great Ulster hero as a rebirth of Lug, the constant 
 feature of his parentage is the name of his mother Dechtire, the sister 
 of Conchobar — his father is variously stated to have been Lug, or 
 an unknown or unnamed god of Faery, or Conchobar himself, or 
 
 > See "The Second Battle of Moytura" tr. by W. Stokes, Rerue celtique. XII (1891), 
 57 fl. ; also the comparison D'Arbois makes between Bres and Kronos, op. cil., II 
 chap. ix. 
 
 ' See D'Arbois. Cours, II, 182 (and passim): "Cette parents n'a rien qui doive nous 
 surprendre. Bress, Fomore, est le gendre de Dagde, I'un des chefs des Tuatha De Danaan. 
 Nous avons deja vu que Lug, un autre des chefs des Tflatha De Danaan, est par sa m6re, 
 petit-flis de Balar, un des chefs de Fomor6. De m6me Brian, luchar et lucharba, trois 
 personnages que des textes appellent les trois dieux du gfinie au de Dana, irt d&i Dana, 
 tri die Danand, c'est 5, dire les trois chefs principaux des TQatha De Danaan, sont flls du 
 FomOrg Bress, et c'est seulement par leur mere Brigit, fllle de Dagdfi, qu'ils appartien- 
 nont aux TQatha Dfi Danaan." Thus Zeus, son of Rhea, mother of the Olympians, 
 combats Kronos, chief of the Titans. Cf. further analogues in Revue celtique, XXVIII 
 (1907), 24 IT. Manifestly the principle of succession here is maternal, and TQatha D6 
 and Fomori are thought of as exogamous tribes intermarrying. 
 
 " Op. cit., Gl. 
 
 * Windisch. Irische Texle, I, 140. 143; transl. by Duvan. in D'Arbois, Coura, V, 22 ff.; 
 in German by Thumeysen, Sagen aua dem alien Irland, 58 fl. ; summary by Nutt, op. cit., 
 II, chap, xiv; Modern English version in Lady Gregory, Cuchulain. 
 
 308 
 
The Sister's Son and the Conte del Graal 19 
 
 finally the Ulster chief Sualtam.^ When Cuchulinn is about to 
 be born, the men of Ulster are gathered about Dechtire. They fall 
 asleep and when they awake behold a little child having features of 
 Conchobar. Finnchoem, Conchobar's other sister, at once loves the 
 child and to her he is intrusted. When they have returned to 
 Emain, Morann speaks this judgment: ''It is for Conchobar to help 
 the child to a good name, for he is next of kin to him."^ 
 
 According to the Tain bo Cualnge^ Cuchulinn was brought up 
 in the house of his mother at Mag Muirthemne. The people there 
 tell him of Conchobar's court at Emain Macha and of the games 
 that go on there among the noble youths. He longs to go to the court 
 but his mother urges him to wait until one of Conchobar's warriors 
 can accompany him: "pour te proteger contre les jeunes gargons ou 
 te venger s'il y a lieu." He persists in his request, and as she tells 
 hun where the court is — "le mont Fuad est entre Emain et toi" — 
 he sets forth to j&nd it, taking with him "his hurling stick,^ his silver 
 ball and his little dart and spear," with which he shortens his journey. 
 
 Arriving on the plain of Emain he finds "three fifties" (cent 
 cinquante) of noble youths playing games. He goes among them 
 and with both feet hurls his ball beyond the goal at which the youths 
 are aiming. This enrages them and led by Conchobar's son, Folio- 
 man, they try to kill him. But he defends himself skilfully — "il 
 
 1 See Nutt's discussion, loc. cit. The son of a brottier and sister is common in 
 mythology. Thus in the Norse Wolsimg story Sinf joth is the son of Sigmimd and Signy 
 (cf. G. Holz, Der Sagenkreis der Nibelunge, 15 — a trait which Wagner employs in the 
 Walkure); in Egypt, Horus is the son of Osiris and Isis {Book of the Dead, tr. Budge, 
 passim), in Greece Zeus springs from Kronos and Rhea (Hesiod Theog. 475), Virgil {Aen. 
 I, 46) says of Juno: Jovisque et soror et coniux; etc. Tliis reafllrms the idea of the 
 Rebirth, and insures the perpetuation of the divine race in a pure form. But even so 
 the son may still take up arms against his father. A survival of this trait is the Huth 
 Merlin (I, 147) account of how Arthur becomes the father of Mordred: "Mais quoi 
 
 que elle fust sa suer, [si] n'en savoit elle riens Moult fist li rois Artus grant joie de 
 
 la dame et moult le festia et li et ses enfans .... en chelui terme il gut a li et engendra 
 en li Mordrec, par cui tant grant mal furent fait en la terre de Logres et en tout U monde." 
 Cf. also Malory. I, 64-65. On the parallehsm between Mo[r]dred and Mider see Rhys, 
 Arthurian Legend, 38 flf. 
 
 2 Lady Gregory, op. cit., 6; Thirmeysen, op. cit., 62: "Conchobar stehe es zu, ihn 
 den Erziehem zu ubergeben, weil er Dechtires ndchster Verwandter ist." In general, it 
 was the Irish custom not to bring up children beneath the paternal roof. Cf. Henderson, 
 Survivals of belief among the Celts (Glasgow, 1911), pp. 36ff. 
 
 ' Ed. by Windisch from the Book of Leinster; for the synopsis I have made see 
 D'Arbois' translation in the Rexue celtique, XXVIII (1907), 241 ff. 
 
 * "Son baton courbe de bronze, sa boule d'argent, son javelot, son baton brtll§ au 
 gros bout." 
 
 309 
 
20 William A. Nitze 
 
 fit des contorsions . . . . il ferma un de ses yeux qui ne fut pas 
 plus large que le trou d'une aiguille, il ouvrit I'autre qui devint plus 
 grand qu'une coupe d'hydromel. II ^carta telleraent les machoires 
 
 que sa bouche atteignit les oreilles Du sommet de sa tdte 
 
 jaillit la lumiere qui atteste le h^ros."* Then he takes the offensive 
 and overthrows fifty of the youths, five of whom fall between Fergus 
 and Conehobar where they are playing chess. When the young 
 hero comes within reach, Conehobar seizes him by the arm. 
 
 When Conehobar taunts him for his "rudeness," the boy replies: 
 "I came as a stranger, and I did not get a stranger's welcome." 
 "Who are you?" says Conehobar. "I am the little Setanta, son 
 of Sualtam and of Dechtire, your sister," says the boy. Conehobar 
 then informs him that there is a "taboo" {magique defense) against 
 playing with the royal youths without their permission. So Con- 
 ehobar obtains the permission for him, and he being the stronger, 
 the royal youths are henceforth placed under his protection.^ 
 
 Another instance of the same motif of a hero seeking the protec- 
 tion of an uncle's court is the Welsh tale of Kulhwch and Olwen. 
 This story is probably independent of Crestien's influence, even if 
 we grant Loth's contention^ that some of its characters are "frott^s 
 de civilisation frangaise." 
 
 Here the young Kulhwch, who is the son of Arthur's mother's 
 sister, Goleuddydd, is left to the care of his nurse, whereas his mother, 
 who dies as the result of his birth, instructs his father not to marry 
 until he sees^ a "briar with two blossoms on her grave." At the 
 same time she tells her "preceptor" not to allow anything to grow 
 on it. 
 
 » From the Book of the Dun Cow, 59, col. 1. 42-43, according to D'Arbois. 
 
 ' Lady Gregory (p. 8) has somewhat rearranged the dialogue to read: " ' You did not 
 know then,' said Conchubar, ' that no one can play among the boy troop of Emain unless 
 ho gets their leave and their protection." ' I did not know that, or I would have asked it 
 of them,' he said. 'What is your name and yoiu- family 7' said Conchubar. 'My name 
 is Setanta, .son of Sualtim and Dechtire," he said. When Conchubar knew that he was 
 his sister's son, he gave him a great welcome, and he bade the boy troop to let him go 
 safe among them."" 
 
 ' See Les Mabinogion, I, 16, 18.5; contrast I. B. John, The Mabinogion, Nutt"s Pop. 
 Series (1901), p. 3, who says that the tale has "no affinity with any of the Arthurian 
 romances handed down to us in French or German." 
 
 * Her words are, lx)th, 188: "Ce serait cependant mal h toi de rulner ton flls; aussi 
 Je te demande de ne pas te remarier, que tu n'ales vu une ronce & deux tfites sur ma 
 tombe." 
 
 310 
 
The Sister's Son and the Conte del Graal 21 
 
 In the seventh year thereafter the "preceptor" having failed in 
 his duty, the king remarries. One day the new wife learns from a 
 "sorceress" that the king has a son. The boy is sent for, and the 
 queen tells him about Olwen, the daughter of Yspaddaden Penkawr, 
 whom Kulhwch accordingly craves. Then his father advises him: 
 "Arthur is thy cousin. Go, therefore unto Arthur, to cut thy hair,^ 
 and ask this of him as a boon." 
 
 So Kulhwch fares forth; in his hands are two spears of silver, a 
 gold-hilted sword is on his thigh, and his horn is of ivory. Before 
 his steed go two white-breasted greyhounds: the one on the left 
 bounds across to the right side, and the one on the right across to 
 the left side. The porter at Arthur's palace tries to dissuade him 
 from entering, for "the knife is in the meat, and drink is in the horn, 
 and there is revelry in Arthur's hall. "- But Kulhwch threatens to 
 bring disgrace upon Arthur and to set up three shouts which will 
 deprive women of their offspring and make them barren, unless he 
 is allowed to enter. So the porter obtains permission of Arthur. 
 
 Then Kulhwch, contrary to custom,^ does not dismount, but 
 seated on his charger rides into the hall and greets the king. He 
 comes not to eat, he says, but to ask a boon; and if it is not granted, 
 he vnW carry Arthur's dispraise to the four quarters of the world. 
 Except for certain reservations^ Arthur is ready to grant this request. 
 Kulhwch then asks Arthur to "bless his hair." Whereupon Arthur 
 inquires who he is : "For my heart warms unto thee, and I know thou 
 art come of my blood." When Kulhwch reveals his parentage, 
 
 1 This is one of the customs which Van Gennep classes among the rites de separation, 
 prevalent in primitive societies. See Rites de passages, 103 flf. Cf. also Loth, I, 190, 
 note; Lady Guest, Les Mahinogiorv, 260; Hartland, op. cit., I, 197 note; Keating, History 
 of Ireland, II, 173-175. 
 
 2 See Conte del Graal, vss. 2785 ff. for a passage where Arthur will not eat, "tantqu'a 
 ma cort novele viegne." See references to this custom mentioned by Hertz, Parz.*, 
 512; also Hist. litt.. XXX, 349. Professor Kittredge, "Arthur and Gorlagon"( Harvard 
 Studies, VIII), 210, note, calls attention to the Irish parallel in the shorter Fled Bri- 
 crend (Windisch, Ir. Texte, II, i, 174, 188): "It is not fitting to consume this feast of mine 
 without a brave deed of the Ulstermen in return for it." 
 
 3 Loth, op. cit., I, 199; Lady Guest, 211; cf. Conte del Graal, vs. 882: E li vaslez autre 
 a cheval An la sale. Consult Kittredge, op. cit., for this commonplace of Arthurian 
 romance. 
 
 < One of these, it is interesting to note, is Guenevere (Gwenhwy^'ar). Compare the 
 abduction motive in the Lancelot story; cf. also the Welsh Pwyll, Loth, I, 45 fif., and 
 elsewhere. Zimmer {Prussian Academy, 1911, p. 177) equates Guenevere with the Irish 
 
 Findabair. 
 
 311 
 
22 > William A. Nitze 
 
 Arthur says: "That is true; thou art my cousin." "Whatsoever 
 boon thou mayest ask, thou shalt receive, be it what it may be that 
 thy tongue shall name."^ 
 
 The occurrence of the sister's son in the French romances was 
 mentioned alike by Professor Gummere' and Mr. M. A. Potter.' 
 Further, and no less striking, material, however, is at hand. For 
 example, Tristan is Mark's sister's son — a fact which may throw light 
 on the moral issue arising when Tristan yields to the charms of Isolt. 
 As Bddier has observed, the Welsh law dealt lightly with the crime 
 of adultery.'' Yet Tristan is not, like Modred, a vile seducer — he 
 does not abduct the queen. "Jamais Iseut ne songe a quitter le 
 roi Marc, ni Tristan a la ravir."^ "II ne renie pas institution 
 sociale, il la respecte au contraire, il en souffre et seule, cette 
 souff ranee confere a ses actes la beauts." But, Bedier continues:^ 
 "II repugne a tout ce que nous savons des contes de Bretagne et de 
 leur transmission de supposer que les Celtes aient possede jamais 
 un grand roman d 'amour sur Tristan." The fatalistic love-theme, 
 however, is by no means foreign to the Celts: witness, for example, 
 the Fate of the Sons of Usnech.'' And it is not only possible but 
 probable that Tristan's conduct is influenced by the fact that he 
 remains conscious of his intimate kinship to the king. As Bedier 
 himself saj^s: "II est le neveu et le fils adoptif du roi Marc : il ne 
 conteste pas la loi de la reconnaissance, il la viole, et souffre de la 
 violer."^ The B^roul version is too fragmentary to throw much 
 light on the question. Yet in Tristan's repentance the true situation 
 becomes apparent : 
 
 "Dex! tant m'amast mes oncles chiers, 
 Se tant ne fus[s]e a lui mesfet!" (Vss. 2170-2171.) 
 
 » Loth. op. cit., cites (p. 191) a striking example of tlie hair-cutting ceremony from the 
 story of Vortigem (cf. Nennius). "Guortigern," he says, "ayant eu un flls de sa fllle, la 
 poussa a allor porter I'enfant Sl Germain, I'Cvfique, en disant qui! etait son pSre. Germain 
 dit h, I'enfant: 'Pater tibi ero, nee te permittam ntKi 7nihi novacula cum forcipe et pecline 
 delur, el ad patrem tuum carnalem tibi dare liceal.' L'enfant va droit a Guortigern, et lul 
 dit: 'Pater meus ea tu, caput meum tonde, et comam capitis mei pecte."' See above, p. 21. 
 
 ' Op. cit. ' Sohrab and Rustem, 193 ft. 
 
 * Roman de Tristan, II, 163: "Le trait le plus singulier de la vie celtique, c'cst la 
 fragility du lien conjugal." 
 
 ' Op. cit., 165. • P. 167. 
 
 ' Windlsch, Ir. Texte (first series), 67-82; D'Arbois, Cour de litt. celt., V, 236-86; 
 Stokes, Ir. Texte, II, 2, 109-78; Dottin, Revue celt., XVI (1895), 426. 
 
 » P. 166. 
 
 312 
 
The Sister's Son and the Conte del Graal 23 
 
 "A Deu, qui est sire du mont, 
 Cri je merci, que il me don(s)t 
 Itel corage que je lais 
 A mon oncle sa feme en pais." (Vss. 2185-2188.) 
 
 Tristan s'apuie sur son arc : 
 
 Sovent regrete le roi Marc, 
 
 Son oncle, qui a fait tel tort, 
 
 Sa feme mise a tel descort. (Vss. 2195-2198.) 
 
 And in Thomas the trait appears clearly in the birth-story 
 which the Anglo-Norman poet has prefixed^ to his version. Here 
 Blanchefleur^ says to the loyal Foitenant: "Je vous confie I'enfant 
 qui va naitre de moi. Si vous avez aime mon seigneur Rivalin, en 
 souvenir de I'amour qu'il vous portait, prenez- le comme votre propre 
 enfant en votre protection. Gardez cet anneau; mon pere I'avait 
 donne au roi Marke; le roi me I'avait donne; s'il le revoit un jour, il 
 reconnaitra I'enfant ne de sa soeur."^ 
 
 Again, while Layamon insists that what makes Conan, the 
 murderer of Constantine in Geoffrey of Monmouth, such an 
 "accursed" villain'* is his being Constantine's sister's son, and that 
 for a like reason Mordred's betrayal of Arthur's confidence has no 
 redeeming quality, we find Wace^ already denouncing Mordred's 
 crime in these words: 
 
 "Oi^s quel honte e quel vilte: 
 Ses nies, fils sa soror estoit." 
 
 Of course Gawain is the exemplar of the type in its highest sense. 
 Thus it is with clear intent that Crestien makes Cliges the sister's 
 son of Arthur's loyal nephew. And in Wauchier's continuation of the 
 Conte del Graal,^ Gawain's own off-spring while ignorant of his father's 
 name has the striking appellation of le neveu son oncle.'' It is, as I 
 said above, the prerogative of the nephew to guard the honor, and 
 therefore to avenge the shame (honte) of the uncle, whereas the 
 
 » Cf. W. Golther, Tristan u. Isolde, 145. 
 Bedier, op. oil., I, 23. 
 
 Cf. also pp. 60-61, where Marke recognizes Tristan and knights him: "II appelle 
 lui Tristan par de douces paroles et I'embrasse tendrement comme son fils d'adoption 
 et son neveu. On the moral question in the Tristan, see J. Loth, Revue celt., XXX (1909), 
 270-82. 
 
 * Gummere, op. cit. 139 « Pot., vss. 20,671 ff. 
 
 « Brut, vss. 13,422-13.423. ' Mentioned by Potter, op. cit.. 49. 
 
 313 
 
24 William A. Nitze 
 
 latter affords the former his protection and assistance. This 
 mutual obligation is characteristic of Gawain, Perceval/ Clig^s, 
 Roland, etc. According to Crestien, Yvain's expedition to the 
 fountain is motived by a similar consideration •? 
 
 "Par mon chief," dist mes sire Yvains, 
 "Vos estes mes cosins ger mains, 
 
 Si nos devons mout antramer; 
 
 Mes de ce vos puis fol clamer. 
 
 Quant vos le m'avez tant cel6. 
 
 Se je vos ai fol apel^, 
 
 Je vos pri qu'il ne vos an poist; 
 
 Car se je puis et il me loist, 
 
 J'irai vostre honte vangier." 
 
 Thus, too, AioP is told by his ot\ti father of his claims on the 
 Emperor Louis. 
 
 "Car vous estes li nies I'empereour, 
 le sai bien a fiance, fiex sa serour." 
 
 Probably nowhere, however, do we find the strength of this tie 
 more clearly brought out than in Partonopeus de Bids — here the 
 hero is nephew to Clovis : 
 
 Un sien neveu avoit li rois, 
 
 Cuens fu d'Angieus et cuens de Blois; 
 
 Fils ert Lucrece sa seror, 
 
 Li rois I'amoit de tel amor 
 
 Qv£ nis son fil de sa moillier 
 
 N'avoit il de nient plus chier* 
 
 Further examples can certainly be found, especially if we go 
 outside our field into that of the epic and ballad. The Roland is 
 filled with the spirit of kinship on the maternal side,^ and to mention 
 
 ' The trait is well brought out in the Goon Desert (Partinel) episode of Manessler, 
 see Pot. V, vss. 34.935 ff. 
 » Yvain, vss. 581-89. 
 ' Aiol, ed. Foerster, vss. 190-91. 
 * Ed. Crapelet, I, 19. 
 
 » Richard le Vicill e sun nevuld Henri. (Vs. 171.) 
 Tedbald de Reins e Milun sun cusin. (Vs. 173.) 
 "Ensurquotut si ai jo vostre socr 
 Si'n ai un fllz, ja plus bels n'en estoet: 
 
 Guardez le bion." (Vss. 294-298.) 
 "Tenez, bels sire," dlst Rollanz il sun uncle, 
 "De trestuz reis vus present les curunes." (Vss. 387-389.) 
 
 314 
 
The Sister's Son and the Conte del Graal 25 
 
 one further striking instance,' Raoul de Cambrai is the sister's son of 
 Louis. But the above examples are sufficient to show what a hold 
 the matrilinear descent had on the minds of men in the twelfth 
 century, at least as a matter of tradition. We have seen, too, that 
 the tribal organization it represents must once have been known to 
 the Goedelic (and Brythonic) Celts, for survivals of the system appear 
 in their laws and in the main body of their heroic legend. As has 
 been previously shown,^ it is with this legend that the Conte del 
 Graal has evident points of contact. 
 
 IV 
 
 Reverting to the poem proper, it is noteworthy that, with the 
 possible exception of the Gawain section, the entire work seems 
 to bear the imprint of the primitive, tribal life. The advances 
 that Blanchefleur makes to Perceval, which Nutt sought to explain 
 in the light of Minnedienst,^ are referable rather to a cruder state 
 of society in which the wooing quite naturally fell to the part of 
 woman. As Nutt himself says: " In the great tragic tale of ancient 
 Ireland, .... Deirdre takes fate into her own hands, and wooes 
 
 "L'altre meitiet avrat Rollanz ses nies." (Vs. 473.) 
 E I'Algalifes sis uncles e sis fedeilz. (Vs. 505.) 
 As porz d'Espaigne ad laissiet sun nevuld. 
 Pitiet Ten prent, ne poet muer n'en plurt. (Vss. 824-825.) 
 
 Marsile also has a nephew : 
 
 "Bels sire reis, jo vus ai seroit tant 
 Si'n ai oiit a peines e ahans." (Vss. 863-864.) 
 
 £0 est Gualtiers Id cunquist Maelgut, 
 i nies DroUn, a I'vieill e a I'canut. (Vss. 2047-2048.) 
 "Rollanz mis nies hoi cest jur nus defalt." (Vs. 2107.) 
 "Se j'ai parenz, nen i ad nul si prud." (Vs. 2905.) 
 The Oxford text does not mention the name of Roland's mother. But it probably 
 was Gisle or Gille (cf. G. Paris, Histoire poetique, 407). The pseudo-Turpin calls her 
 Berte, and the same name occurs in Philippe Mousket (vss. 2706 fl.): 
 
 S'ot Charles ime autre sereur 
 Bertain: cele prist a seigneur 
 Milan d' Anglers, s'en ot Rollant. 
 
 For an "Enfance" story, in which the boy Roland seizes the goblet of Charles, cf. 
 the Venice MS XIII and the Reali di Francia (G. Paris, op. cit.. 409 ff.). In certain 
 versions, notably the Karlamagnus-Saga (I, 36), Roland is the son of Charles and the 
 latter's sister Gille. Cf. above p. 18, for other examples of this motif. On the episode 
 itself, see G. Paris, op. cit., chap. viii. In Spanish literature, Bernardo del Carpio was 
 considered a sister's son of Charlemagne, later of Alfonso II; see Baist, Grober's Grund., 
 II, 2, 392. 
 
 1 Garin le Loherain, tr. Paulin Paris, p. 333. 
 
 2 Cf. Zimmer, Gdtt. gel. Anzeigen (1890), XII, 519 flf.; Keltische Studien, Berlin, 1884, 
 II, 200; also articles cited above. 
 
 ' Studies on the Legend of the H. Grail, 228 fl. 
 
 315 
 
26 William A. Nitze 
 
 Noisi with outspoken passionate frankness." Doubtless the situation 
 is romantic, and that may account for its late persistence, but 
 the Fands, Viviens, and Orgueilleuses — to some extent — have 
 counterparts among primitive peoples today.' The manner in 
 which Lancelot becomes the father of Galaad in the Prose Lancelot 
 cycle is a curious survival of this time-old custom.^ Moreover, the 
 matriarchal idea is typical of the grail romances. "Das ist das 
 Wesentliche," says Brugger,^ "das den Joseph (und den Didot-Perceval) 
 isoliert : dass der Gralheld hier vaterlichseits, in alien andern Versionen 
 miitterlichseits mit dem Gralhiiter verwandt ist." The Syr Percy velle, 
 which lacks the grail adventure, retains this arrangement. Here the 
 hero's mother (Acheflour) is the sister of King Arthur, and not of 
 the Grail King. A priori this appears to be a substitution, since 
 Arthur's traditional nephews are of course Gawain* and Ivlordred, and 
 the names in Syr Percy velle are French and not Welsh. The very 
 instructions the hero receives from time to time, such as:^ (1) to 
 respect the advice of elders; (2) to succor women in distress; (3) to 
 greet cordially those he meets; (4) to refrain from useless speech; 
 though probably determined as to form by the mediaeval school 
 learning and by their connection with chivalry (to which the poem 
 alludes^), are characteristic of the tribal life as observed elsewhere. 
 Thus Howatt affirms that among the southeastern Australians, 
 before the conclusion of the initiation ceremonies (Kuringal), the 
 
 I "The practice," says Hartland, "of offering the wife or other female dependent to a 
 guest for temporary companionship" is very widespread among various American tribes 
 {op. cit., II, 229). Cf. also Karl Schmidt, Jus Primae Noctis, Freiburg, 1881, passim. 
 On women themselves making the advances, see Hartland, op. cit., II, 129 IT., also the 
 examples gathered by Potter, op. cit., 172 ff., and Zimmer, "Der kulturhlstorische 
 Hintergrund in der altirischen Heldensage," Prussian Academy, 1911, pp. 175 11. In 
 Raoul de Cambrai Bemier Is wooed by Gerin's daughter (vs. 5696): 
 
 " Pren moi a feme, frans chevaliers eslis: 
 Si demorra nostre guere 5, toz dis." 
 
 See Hertz, Parz.', 502, on the TobiasnadUe in Parz., § 203. 
 
 » PauUn Paris. Romans d. I. table ronde, V, 306 ff. ' ZffS., XXXVI (1910). 18. 
 
 * Among other examples that can l>e cited, Crestien's Erec contains a clear instance 
 of the reliance Arthur placed on Gawain: 
 
 "Biaus niCs Gauvains! conselliez m'an 
 Sauve m' enor et ma droiture! 
 Car jo n'ai de la noise cure." (Vss. 308-310.) 
 
 A recent attempt to prove the primitive character of the Syr Percytelle has been made 
 by R. H. Griffith, Sir Perceval of Galles (University of Chicago diss.), 1911. 
 
 * See vss. 510. 1610, 6360 ff. 
 
 * Vs. 1612. The instructions given by Gomemanz, vss. 1610 ff., should be compared 
 with those found in the Ordene de Chetalerie, printed by M6on, Fab., I, 59 ff. 
 
 316 
 
The Sister's Son and the Conte del Graal 27 
 
 youth is not invested with the man's belt until he has been told: 
 (1) to heed the advice of the old men; (2) to share the fruits of the 
 chase with others; (3) not to interfere with the women of the tribe; 
 (4) not to injure his kindred by means of evil magic.^ " The intention 
 of the ceremonies," continues Howitt, "is evidently to make the 
 youths of the tribe worthy members of the community, according to 
 
 their lights Before the novice is permitted to take his place 
 
 in the community, marry, and join in its councils, he must possess 
 those qualifications which will enable him to act for the common 
 welfare." Gillen and Spencer writing^ of the central Australians 
 observe that in the initiations of the natives along the Finke River 
 there is fastened around the initiate's waist a large uliara, "that is, 
 the human hair girdle worn by the men, the girdle being provided by 
 an Oknia of the boy." Previously an Unkulla man had twined round 
 the youth's hair "strands of fur string, until it looked as if his head 
 were enclosed in a tight-fitting skull cap." The Unkulla man is 
 "the brother of the boy's mother," and the fastener of the girdle 
 is the latter's son. During the ceremonies which follow the boy is 
 told that he must be obedient, must never reveal any of the tribal 
 secrets, must not speak unless spoken to. The final rites are "a long 
 series of ceremonies concerned with the totems, and terminating 
 with what may be described as ordeals by fire." The object of the 
 ceremony as a whole is, "firstly, to bring the young men under the 
 control of the old men . . . . ; secondly, to teach them habits of 
 self-restraint and hardihood; and thirdly, to show the younger men 
 who have arrived at mature age, the sacred secrets of the tribe which 
 are concerned with the Churinga^ and the totems with which they 
 are associated." "Every Australian native," they affirm, "so far 
 as is known, has in the normal condition of the tribe to pass through 
 certain ceremonies of initiation before he is admitted to the secrets of 
 the tribe, and is regarded as a fully developed member of it." This 
 shows, especially when taken together with my previous remarks 
 on the Eleusinia,* what a close connection there is between Perceval's 
 entire career and the ceremony at the Grail Castle, if the latter is, 
 
 1 Native Tribes of South-East Australia (Macmillan, 1904), pp. 231 ff. 
 » Native Tribes of Central Australia (Macmillan, 1901), pp. 210-86; 347-73. 
 ' Sacred emblems. * Op. cit. 
 
 317 
 
28 William A. Nitze 
 
 as I believe, an initiation. Obviously Nutt was correct in affirming 
 that the questing initiate was always a part of the grail story: the 
 Perceval and the grail stories are essentially one, and any attempt 
 to separate them is, I believe, a mistake. Not that the initiate 
 need always have been called Perceval: Crestien may well have 
 given him the name, which he first mentions in Erec (vs. 1526) — et 
 Percevaus li Galois. But even that is conjectural.^ The essential 
 fact, however, is that the youth's training or education is of a piece 
 with the grail adventure, which is, as we have seen, its culminating 
 point.^ And while other works outside the Arthurian pale, and 
 within it, deal with the sister's son as a recognized personage, none 
 of them treats, as does Crestien's work, his particular problem, per 
 se, as characteristic of tribal life. 
 
 This does not signify that Crestien was aware of the fundamental 
 import of the story he was "setting to rhyme."' He may or may not 
 have been. Probably he grasped the " mystery " as little as we should 
 today. Being a problem in conduct, the story would appeal at once 
 to his scholastic temper. But let us not be misled into thinking that 
 he was primarily bent on being consistent and clear. His Erec and 
 Charrete show that he was not. As Baist says: "Er liebt es, seine 
 Wunder in hellster Beleuchtung hervortreten, aber dann verdam- 
 mern zu lassen." We cannot quarrel with him for doing so — the 
 grammarians afforded him ample justification for respecting tradi- 
 tion, even when it was not understood — and did not Marie de France* 
 on the basis of Priscian extol the ancients for being obscure ? 
 
 Es livres que jadis faiseient 
 assez oscurement diseient 
 par eels ki a venir esteient 
 e ki aprandre les devcient 
 que peiissent gloser la letre 
 e de lur sen le surplus metre. 
 
 » Sco below, p. 31. 
 
 » Baist (Parzival u. der Oral), 19, expresses a somewhat similar view, though without 
 mentioning the initiatory character of the grail ceremony, when he says: "Immerhin 
 wlirde mit der Natur dieser Resto sich die Annahme hesonders gut vertragen, dass in der 
 Vorlage die Weisheitslehren in viel engerer Beziehung zur Handlung standen als bei 
 Chrestien. Dann aber wird es maglich, dass der Oral in seiner ersten Gestalt ohne jede 
 wundcrbareEigenschaft war und nur die Kegel exempliflzieren half, dass unter Umstanden 
 auch Reden Gold sei." 
 
 » Vs. 63 : A rimoier le meillor conte. 
 
 * Lais, ed. Warnke, Prol., vss. 11 ff. ; the passage is a misconstruction of Priscian, 
 7n«(. (beginning). 
 
 318 
 
The Sister's Son and the Conte del Graal 29 
 
 In short, Crestien interpreted his material as far as he was able; when 
 unable, he reproduced it without sacrificing its traditional features. 
 
 The famous grail-question (Cui Van an sert ? or Quel riche home 
 Van an servoit f) is, I believe, such a feature. The riche home is the 
 Grail King; besides if asked, the question will heal the suffering 
 Fisher King, and when not asked, it actually brings harm to the 
 knights and ladies of Arthur's court.^ I thus explained its function^ 
 as ritualistic, and the conclusions reached in the present study bear 
 out this interpretation. To argue, as Heinzel does,^ that this aspect 
 of the ''question" is secondary, seems to me to ignore the raison 
 d^itre of the whole story. Yet those who derive the grail legend from 
 the Byzantine Mass must resort to some such reasoning. Admitting 
 that there is danger here for considerable mutual misunderstanding 
 and that it behooves us to keep an open mind on such intricate 
 matters, nevertheless we should remember that the grail legend 
 tends toward, and not away from, Christianity. The later not the 
 earher versions are the most Christian. Says Martin:^ "Stufen- 
 weise geht die Verchristlichung weiter; immer starker tritt der 
 mystisch-asketische Zug hervor, bis er zuletzt in der torichten 
 Vorstellung gipfelt, dass nicht Perceval, der, wenn auch Mensch, 
 doch vermahlt war, sondern der vollig jungfrauliche Galaad den 
 Gralzauber lost und zwar nachdem er den Gral hat suchen miissen, 
 obschon er selbst im Hause des Grals aufgewachsen ist." Further- 
 more, why should Robert, whose orthodoxy is not his strongest 
 point, nevertheless distinguish the grail ceremony from the Christian 
 mass? ''Nicht das Messopfer," says Heinzel,* "aber etwas dem 
 Messopfer Aehnliches stellt sich der Dichter vor." Or was Robert 
 honestly actuated by the desire to conform to Roman usage by 
 rejecting any contamination from the East? Certainly, the Byzan- 
 tine theory must be at a loss to explain that Robert himself does not 
 so much as mention the bleeding lance. Personally I cannot conceive 
 how then the Byzantine material could be the immediate basis of the 
 grail romances in the twelfth century. What may have happened 
 in Ireland at an earlier date is another question. But even so, 
 ritualism is so general among primitive societies, that the assump- 
 
 1 See, above, p. 5. * Parzival u. Tilurel, II, L. 
 
 ' See Elliott-Studies, 49. * Qp. cit., 87. 
 
 » Op. cit., 185. 
 
 319 
 
30 William A. Nitze 
 
 tion of eastern borrowings on the part of the Celts does not seem to 
 me materially to alter our discussion.^ On the other hand, assuming 
 as we do that Robert is mainly responsible for the Christianization 
 of the grail material, it requires no argument to see that having 
 removed the grail from its primitive setting, and explained it ex post 
 facto on a Christian basis, in the manner somewhat of the enfances 
 of an epic hero, he naturally missed the central motive of the original 
 grail story, represented in the tribal organization. 
 
 If then the "question" is primarily ritualistic, as a part of a 
 pagan ceremonial, and probably involved an explanation of the rite 
 in which the initiate had shared (the so-called "secrets" of the grail), 
 it was instrumental also in establishing the kinship, the "blood-tie," 
 so to speak, which bound the neophyte to the head of the clan, and 
 made him a recognized member of the social group. To use Heinzel's 
 words it was "eine Erkennungsfrage," though we do not share 
 Heinzel's view that it was accidental (zufdllig). A similar procedure 
 is found in stories of the father-and-son combat type. In Sohrab and 
 Rustem, the son says to his father: " One question I desire to ask you, 
 and do you answer that truthfully. Tell me frankly, what is your 
 birth?" Likewise, Hildebrand^ asks his son "wer sein vater waere 
 in der menschen volke, 'oder welches geschlechtes kind du seist.'" 
 Again, we found Conchobar saying to the youthful Cuchulinn: 
 "What is your name and your family?" In this way the question 
 would naturally serve as a method of identification, as a recognition- 
 formula. This has long been observed by scholars. Of all Arthurian 
 stories, that of the grail depends most on the principle of adequate 
 identification. How will the hero of destiny make his presence 
 known ? Why, by a question he will ask. Or, if for some reason, this 
 is not sufficient, by a special seat which he will occupy. Thus we 
 get the Siege Perilous, so similar, as I have pointed out,' to the 
 Irish Lia Fdil or Stone of Destiny, which like the lance and the sword 
 of Lug, and the Caldron of Dagda, was one of the "four jewels" of 
 
 ' On syncretism, frequently overlooked by the reviewers, see my "Fisher King." See 
 the view expressed by Klttredge, Harvard Studies and Notes, VIII: " The mere fact that 
 a story is oriental in its ultimate origin is no reason for refusing to regard it as Celtic if 
 it once made its home among the Celts and came from them, charged with their peculiar 
 genius, to fructify the literature of France and of the world." 
 
 ' See Das midebrandslied, ed. AJ. Vollmer and K. Hofmann, 11. 
 
 » Ellioll- Studies, 42. 
 
 320 
 
TpE Sister's Son and the Conte del Graal 31 
 
 the Tuatha De Danaan, and therefore the most accessible prototype 
 for a poet dealing with the grail. In at least one of the grail 
 romances, however, the hero establishes his identity by "opening" 
 a tomb, the top of which rises to his touch. This happens in the 
 Perlesvaus,^ and it is known thereby that Perceval is the son of the 
 Veve Dame and also li miaudres chevaliers del monde. 
 
 Now, the Cymric hearth was the symbol of family ownership and 
 inheritance. 
 
 The right of the son on succession was to uncover the hearth of his father 
 or ancestor. The legal term for the recovery by an ejected son of his patri- 
 mony was dadenhudd, or the uncovering again of the parental hearth. The 
 term was a graphic one. The fire-back-stone, set up against the central pillar 
 of the hut supporting the roof (pentow2;aen = head-fire-stone), was a memorial 
 or witness of land and homestead {tir a thyle), because it bore the 7nark of the 
 kindred upon it. And the covering and uncovering of the fire had a picturesque 
 significance.^ 
 
 It took four generations of occupancy to establish a claim to a 
 Welsh family hearth, which thus became, ''in a very literal sense, the 
 focus of the rights of the kindred."^ If, then, we consider that m. 
 Crestien, where the tribal concept is strongest, the grail ceremony 
 takes place about the central fireplace of the Grail Castle, and 
 that the word which Crestien uses, cheminal, probably points to a 
 primitive hearthstone^ (Wolfram's fiverrame of marble), the Welsh 
 custom is doubly interesting, especially as a ceremonial fire is not a 
 phenomenon confined to the Welsh. Still, Crestien does not give us 
 the slightest inkling that the hearth at the Grail Castle bore the 
 mark of the kindred. At the same time, it is after his visit to the 
 Grail Castle that Perceval mysteriously divines his real name,* and 
 our evidence has shown that ultimately the grail ceremony was 
 the means of instating him in his inheritance. For want of further 
 material we must leave the problem at this pomt without coming to 
 a fixed conclusion.^ 
 
 > Pot., I, 179. 2 Seebohm, op. cit., 82. ' Ibid., 83. 
 
 * Elliott- Studies, I, 30 ff. ^ See above, p. 28. 
 
 ' The problem indicated here might repay further investigation. Note, above all, what 
 Van Gennep has to say on the "rites de denomination " in liis Rites de passages, pp. 88 ff.; 
 also chap, vion the "rites d'initiation," where a simmiary is made of the various initiations 
 known up to 1909 — a chapter which corroborates our conclusions on the initiatory char- 
 acter of the grail material. On names the following points may be noted in addition: 
 
 In the Australian Kuringal ceremonies (cf. Howitt, op. cit., 526-641) among the 
 tilings which the novice leams from his Kabosisthe Budjon or totem name. "These 
 
 321 
 
32 William A. Nitze 
 
 But the material adduced has, I hope, thrown some light on the 
 means by which the tribal hero may have won official recognition. 
 
 And now, in conclusion, the name irarr^p given to the tribal 
 god in the ancient mysteries seems to me paralleled by the riche 
 home (the Grail King; according to Nutt, the Mikado of the myth), 
 the [Celtic] tribal ancestor, the religious and juridical head of the 
 clan, the maternal uncle of the sister's son. And, taken all in all, 
 does not the Conte del Graal, even after the adaptation it underwent 
 to meet the demands of twelfth-century society, still furnish an 
 example, more primitive than anything handed down elsewhere in 
 Arthurian literature, of the tribal myth in the sense in which Van 
 Gennep^ has recently defined that dubious word ? 
 
 Si [une] .... histoire est accompagn^e d'une pantomime repr^sentant 
 les personnages et les phases successives du theme; si les acteurs qui 
 repr^sentent les h^ros sent revetus d'une quality sacr6e ou si ce sont des 
 magiciens .... si certains personnages se retrouvent dans d'autres r^cits 
 ou c6r6monies ou Ton explique la formation du monde, le renouveau de la 
 v^g^tation, les origines de la tribu; si toute cette representation n'a lieu qu' 
 k un moment solennel de la jom-n^e, n'a pour spectateurs que des hommes 
 adultes et initios aux mysteres, et qui se sont soumis d'abord a des rites de 
 purification, qui enfin seraient punis de mort s'il racontaient aux fenunes, 
 aux enfants et aux Strangers ce qu'ils ont vu — dans ce cas, I'histoire consid^r^e 
 n'est plus un simple conte: c'est une partie essentielle du systeme religieux, 
 un drame sacr^, un mythe. 
 
 William A. Nitze 
 
 The University of Chicago 
 
 names are not much used, and a person does not know much of the Budjans of others. 
 It is the personal name which is used, not the Budjan. The personal name is a tribal one 
 given to an individual in childhood, and the use of the totem name is avoided, lest an 
 enemy might get hold of it and do him an injury by evil magic." In some tribes it is 
 the totem name wliich Is used and the other which is kept secret. Mr. Potter (op. cil., 
 211), adduces the practice of certain Indians of calling each other brother, sister, father, 
 etc., in order to avoid any danger of allowing others to know what their real name is. 
 Cuchulinn is named after Culann's hound, and therefore cannot eat the flesh of his name- 
 sake (cf. Stokes, Revue cellique, III [1882), 176); and Ciwalchmei (the French Gawain) 
 according to Rhps, Arthurian Legend, 168, resolves itself into Gwalch-mei, "the Hawk 
 of the month of May." Gawain is always ready to tell his name (see Hist. Liu., XXX, 
 37 ff.); in the Conte del Graal (vss. 5583 ff.) he says: 
 
 "Sire Gauvains sui apelez 
 Onciues mes nons ne fu celcz 
 An leu ou il me fust requis, 
 N'onquos anoorcs no Ic dis 
 S'aincois demandoz ne me fu." 
 
 On naming after an ancestor or divining names, see Hartland, op. cil., I, 211 ff 
 
 > La formation des ligendea (Paris. 1910), pp. 306-307. 
 
 322 
 
H0il7\9n 
 
 HOME USE 
 
 CIRCULATION DEPARTMENT 
 
 MAIN LIBRARY 
 
 This book is due on the last date stamped below. ^ 
 
 1 month loans may be renewed by calling 642-3405. 
 6-month loans may be recharged by bringing books 
 
 to Circulation Desk. 
 Renewals and recharges may be made 4 days prior q,. 
 
 to due date. 
 ALL BOOKS ARE SUBJECT TO RECALL 7 DAYS 
 
 AFTER DATE CHECKED OUT. 
 
 ^PR 9.5 19751 9 '—i 
 
 [fe_pi^^^UT5 
 
 ■) 
 
 LD21 — A-40m-12,'74 General Library 
 
 (S2700I.) University of California 
 
 Berkeley 
 
 L 
 
 a^^SPf***^ TT .Genial Library 
 
 (CSTffi . 1 „ > ,7,; University of California 
 
 Berkeley 
 
RETURN CIRCULATION DEPARTMENT 
 
 TO— #► 202 Main Library 
 
 LOAN PERIOD 1 
 HOME USE 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 
 Renowols and Rechorges may be made 4 days prior to the due date. 
 
 Books may be Renewed by calling 642-3405. 
 
 DUE AS STAMPED BELOW 
 
 ftOV 2 1992 
 
 AUTO DISC CIRC J 
 "WoTlggg 
 
 IN 30 '94 
 
 FORM NO. DD6 
 
 UNIVERSITY OF CALIFORNIA BERKELEY 
 BERKELEY, CA 94720