Ragnars saga loðbrókar (Ragn)
Skaldic vol. 8; ed. Rory McTurk
Ragnars saga loðbrókar ‘The Saga of Ragnarr Hairy-breeches’ (Ragn), the prose narrative incorporating the forty stanzas edited here, is an Icelandic fornaldarsaga dating in its fullest form from the second half of the thirteenth century or somewhat later (see further below), and dealing with the exploits of its hero, Ragnarr, and his sons in the British Isles and elsewhere. The ostensible reason for Ragnarr’s nickname, loðbrók ‘Hairy-breeches’, is the protective clothing he wears in the saga’s description of his slaying a snake, the increasing size and ferocity of which had posed a threat to Þóra, daughter of Herruðr, whom Ragnarr wins in marriage by this exploit. Introduced as a king’s son, Ragnarr is presented, from an early stage of the saga’s action, as himself a king, with the centre of his power in Denmark. The climax of the saga’s plot is Ragnarr’s death in England in a snake-pit, and the revenge taken by his sons.
It is likely that the saga character Ragnarr loðbrók is a combination of two different historical personages, one of them Reginheri, the leader of the viking attack on Paris in 845, and the other a more shadowy figure whose name came to be attached in the form loðbrók to that of Reginheri/Ragnarr; see McTurk (2011b, 1-14); for a contrary view, see Rowe (2012, 164-6), reviewed by McTurk (2013, 94-5). Of the sons attributed to Ragnarr loðbrók in Ragn, those named Ívarr, Bjǫrn, and Sigurðr are identifiable with ninth-century historical vikings mentioned in contemporary annalistic sources who, chronologically at least, could have been sons of Reginheri (McTurk 2011b, 5-9; 2013, 95-8); while Rǫgnvaldr, a further son according to the saga, may historically have been a viking active in the early tenth century (see Note to Anon Krm 15/6). On the possible association of Reginheri/Ragnarr with the ninth-century poet Bragi inn gamli ‘the Old’ Boddason, the composer of Ragnarsdrápa, see the Introduction to Bragi RdrIII and McTurk (2003).
There are a number of extant written versions of the exploits of Ragnarr and his sons, some of which are incomplete or allusive. One of the reasons for the popularity in Iceland of legends about Ragnarr loðbrók is the fact that several prominent Icelanders of the twelfth and thirteenth centuries claimed descent from him. The first of them known to do so is Ari Þorgilsson, who, in his Íslb, written 1122-32 (ÍF 1, xvii-xviii, 4), gives the earliest recorded instances of the name and nickname Ragnarr and loðbrók in combination, later listing Ragnarr loðbrók among his ancestors (ÍF 1, 27-8). While some subsequent sources use only one or other of these two names, most of them imply awareness of their combined use for the same person. One possible exception here is the runic inscription in Maeshowe on the island of Mainland in Orkney, referred to below. The earliest extended surviving witness to a legend of Ragnarr and his sons in Scandinavian tradition is probably Háttalykill ‘Key to the Metres’ (RvHbreiðm HlIII), composed c. 1150 by the Orcadian jarl Rǫgnvaldr Kali Kolsson and the Icelander Hallr Þórarinsson. This poem, which in its surviving form does not mention the nickname loðbrók, devotes no fewer than twelve stanzas to celebrating Ragnarr, the manner of his death, and his sons (see the Introduction to RvHbreiðm Hl in SkP III). The Maeshowe runic inscription, numbered 23 by Barnes (1994, 178-86) and also dated to the mid-twelfth century (ibid. 39-40), indicates knowledge of a version of the legend in referring to the sons of one Loðbrók (see Note to Ragn 39/4), though with no mention of the name Ragnarr; it may also be associated with Rǫgnvaldr jarl and his entourage. The now lost Skjǫldunga saga, generally dated to c. 1200 (K. Wolf 1993), and accessible only through a sixteenth-century Latin epitome, contained a narrative of Ragnarr loðbrók and his sons, as does Book IX of Saxo Grammaticus’s Gesta Danorum (Saxo 2015 I, ix. 3. 1-4, 4. 1-39, 5. 1-8, 6. 1, pp. 628-69), of about the same date. The anonymous poem Krákumál ‘Speeches of the Crow’ (Anon Krm), preserved within or in conjunction with Ragn, and edited separately in the present volume, also bears witness to the popularity of Ragnarr loðbrók in Old Norse tradition in the period around 1200, the probable date of Krm’s composition; it mentions the nickname but not the name, while clearly assuming knowledge of the latter. In addition, there are three extant Icelandic prosimetrical versions of the narrative of Ragnarr loðbrók, two versions of Ragn, discussed below, and the text known as Ragnars sona þáttr ‘The Tale of the Sons of Ragnarr’ (RagnSon) preserved as part of the early fourteenth-century compilation Hauksbók.
Ragn is preserved in two major redactions: Y, preserved complete in NKS 1824 b 4° (1824b), dated to c. 1400, and X, preserved fragmentarily in the late fifteenth-century ms. AM 147 4° (147). In the X redaction, Anon Krm is placed in the mouth of Ragnarr as he dies in the snake-pit; in 1824b, on the other hand, this poem appears as an appendix to Y. In neither 147 nor 1824b is Krm preserved complete; see further the Introduction to Krm in the present volume. In 1824b, presumably reflecting the Y redactor’s intention, Ragn occurs immediately after the text of Vǫls, thus presenting the former as a sequel to the latter. This may also have been the case with the X redaction, although the present state of 147 makes it difficult to be sure: 147 contains no text of Vǫls, but a retrospective reference in its text of Ragn to the meeting of Sigurðr and Brynhildr on a mountain (a major event in the extant Vǫls) suggests that the X redaction of Ragn may originally have been preceded by a text of Vǫls (see Ragn 1906-8, 180-1, cf. 46-7; cf. also McTurk 1977, 569). The existence of a third version of Ragn may be tentatively deduced from the combined evidence of X and RagnSon, a mainly prose compilation preserved in the hand of Haukr Erlendsson in AM 544 4° (Hb), which forms part of the compilation Hauksbók, of which Haukr (d. 1334) was the chief scribe. In RagnSon (Hb 1892-6, 459) reference is made to what is told í sǫgu Ragnars konungs ‘in the saga of King Ragnarr’ of Ragnarr’s marriage to Þóra and his subsequent conquests; in its account of Ragnarr’s fight with the snake RagnSon differs markedly from Y (the relevant part of X is now illegible in 147), and in its account of the revenge taken for Ragnarr’s death by his sons is much less close to Y than to X, which at this point can be read with relative clarity (see Ragn 1906-8, xcii, 167-8, 193‑4).
There is some uncertainty about the relationship between the RagnSon version of the legend of Ragnarr loðbrók and those of X and Y. McTurk (1991a, 56; 2013, 98-9), following Einar Ólafur Sveinsson (1936, 194-5) and Bjarni Guðnason (1969, 37), has proposed that RagnSon reflects a lost, ‘older’ version of Ragn, while Rowe (2009, 353-60; 2012, 22, 228-36) has argued to the contrary that the extant RagnSon was ‘an attempt to “improve” the extant Ragnars saga, sometimes by shortening it and sometimes by making it closer to Skjǫldunga saga’s version of the story’ and was originally written c. 1300, very possibly by Haukr Erlendsson. By ‘the extant Ragnars saga’ Rowe (2012, 207) clearly means what is here called the Y-redaction. This redaction shows the influence of Þiðr (see Bjarni Guðnason 1969, 31 and McTurk 1977, 578-82), which is now thought to have been written 1230-40 (Mundt 1973, 353-6), or even earlier (Andersson 1986, 364-6), and a reference to what seems to be this same redaction of Ragn appears in HálfdEyst, dating from c. 1350 (see Schröder 1917, 94; cf. McTurk 1977, 582-3; Naumann 1993, 26). This would allow for a dating of Y to the period c. 1250-1350. Rowe’s view just outlined, however, if accepted, would narrow the terminus ante quem for Y down to c. 1300. There is even less certainty about the dating of X and the saga of Ragnarr referred to in RagnSon. Bjarni Guðnason’s view (1969, 37) that X, like Y, shows the influence of Þiðr (see however McTurk 1977, 572-8, and 583 n. 128), and that Y (as McTurk 1975, 61-4, agrees) is largely derived from X, would suggest, if accepted, a date for X of c. 1250. As for the saga of Ragnarr referred to in RagnSon, Rowe, as shown above, sees this saga as identical with Y and hence, from her perspective, as dating from the second half of the thirteenth century. She is here arguing against Bjarni Guðnason (loc. cit.), who, following Einar Ólafur Sveinsson (also loc. cit.), saw the saga in question as a precursor of the X and Y redactions, and as complete by c. 1230, the birth date of Randalín Filippusdóttir, a member of the Icelandic Oddaverjar family, which claimed descent from Ragnarr loðbrók (see ÍF 35, lv n. 6). The name Randalín, unique as applied to a historical person, occurs in RagnSon as an additional name for Ragnarr’s wife Áslaug (as it also does in Y; X is here illegible), and Bjarni argued that its adoption by the Oddaverjar was influenced by an early version of Ragn in which Áslaug was so named. In the present edition it is simply assumed that behind RagnSon there lies a lost version of Ragn of uncertain date, resembling and differing from the X and Y redactions in ways which can be deduced from a comparison of RagnSon’s narrative with those of X and Y.
Ragn also has analogues in certain Danish, Norwegian, and Faroese ballads. The ballads in question are those edited (in DgF I) as DgF 24 (in Danish) and (in Landstad 1853) as Landstad XI (in Norwegian), both dealing with a serpent-slaying episode, and preserved from the seventeenth and nineteenth centuries respectively; the Faroese Ragnars kvæði, dealing with Ragnar’s marriages to Tóra (cf. Þóra) and Ásla (cf. Áslaug), and preserved from the eighteenth century (Djurhuus and Matras 1951-63, 215-43); and the two Danish ballads, edited as DgF 22 and 23, both about a princess brought up as a herdswoman before marrying a king, and both preserved from the sixteenth century (McTurk 1991a, 51‑93).
All known versions of the legend of Ragnarr loðbrók drew selectively on what must originally have been verse and prose oral traditions. In earlier, lost versions of the extant saga texts Ragnarr seems to have been presented, as he is in X and Y, as marrying twice, first to Þóra, the daughter of Herruðr, and then, after Þóra’s death, to Áslaug, the daughter of Sigurðr and Brynhildr. Krm concentrates on his battles, and mentions both Þóra (Anon Krm 1/5) and Áslaug (Anon Krm 26/3), the former as the wife he won by his serpent-slaying and the latter as the mother of the sons who will avenge him; whereas Saxo concentrates largely on his marriages and sons, reporting that Regnerus Lothbrog (sic) had three wives, the second of whom, Thora (cf. Þóra), he won by slaying two serpents, but among whom Áslaug is not included. (The absence of Áslaug, the daughter of Sigurðr and Brynhildr, from Saxo’s account would seem to suggest that this figure was unknown to Saxo, at least in connection with Ragnarr loðbrók; for a contrary view, however, see Rowe 2012, 100‑5.)
The basic saga narrative appears to have described, in the following order, Ragnarr’s winning of Þóra by slaying a snake; his marriage after Þóra’s death to Áslaug, the daughter of Sigurðr and Brynhildr; the deaths in Sweden of his two sons by Þóra, namely Eiríkr and Agnarr, and the resulting revenge mission led by Áslaug and her own sons by Ragnarr, namely Ívarr, Bjǫrn, Hvítserkr and Sigurðr; Ragnarr’s death in England in King Ella’s (OE Ælle) snake-pit after his abortive invasion of that country; his sons’ revenge, involving the founding of York by Ívarr and the cutting of a blood-eagle on Ælle’s back; and various further enterprises of his sons, including their conquest of Lúna (Luni) in Italy and their proposed invasion of Rome.
Items that appear to have been introduced in the X redaction include an account of Ragnarr’s first meeting with Áslaug, after Þóra’s death, at Spangereid in Norway, where she is living as a herdswoman under the name of Kráka; the mention of another son of Ragnarr by Áslaug, namely Rǫgnvaldr, and an account of the latter’s heroic death; accounts of the slaying by Ívarr of three magical cows, two of them killed in the conflict in which Rǫgnvaldr falls, the third despatched as part of the process of avenging the deaths of Þóra’s sons in Sweden; and an account of the conquest by Ragnarr’s sons of Vífilsborg (Avenches), which they attack with fire and catapults. The fragmentary and largely illegible text of 147 makes it difficult to determine the course of events at all stages of X, but it is likely that the circumstances leading up to the deaths in Sweden of Þóra’s sons are described there more in the manner of Y (preserved in 1824b) than in that of RagnSon (cf. the Context to Ragn 11); and that in X, as in Y, the Lúna-Rome episode, presented as occurring before Ragnarr’s death (rather than after it, as in RagnSon), is more elaborate than in the basic narrative, involving the decision by Ragnarr’s sons not to invade Rome after being told by an old man in Lúna that, in walking there from Rome, he has all but worn out two pairs of iron shoes. Krm also seems to have been added in the X redaction; on its apparent relegation to the status of appendix by the Y redactor, see the Introduction to Krm.
A very likely addition by the Y redactor is the introductory account that forms the first chapter of Y (ch. 2 of which leads into the account of Ragnarr’s serpent-slaying and winning of Þóra), about how Heimir, Brynhildr’s foster-father, brought Áslaug as a three-year-old child to Spangereid in a harp, and how Áslaug was found by the local farmer and his wife after they had killed Heimir for his riches, and was brought up by them under the name of Kráka. It is however conceivable that an earlier version of this account already existed in association with X, as part of a lost redaction of Vǫls to which X may originally have formed a sequel (see above and cf. McTurk 1977, 568-70, 584‑5).
In the Y redaction it is London, whereas in RagnSon it is York, that Ívarr founds in connection with the avenging of his father. While the account of the revenge in X is closer to RagnSon’s account than to Y’s, the state of 147 is such that it cannot be said whether or how the motif of the founding of a city featured in X. The last three chapters (18-20) of Y may be regarded with reasonable confidence as additions to the Y redaction, since they follow its account of the revenge taken on Ælle by Ragnarr’s sons, whereas X appears to come to an end just after its account of this event (see Ragn 1906-8, lxxxvi and 193-4), and RagnSon’s narrative and Y’s diverge increasingly after their different accounts of it, as Rowe (2012, 232-3) has shown. Chapter 18 of Y describes the deaths of Ragnarr’s sons Hvítserkr and Ívarr, and mentions some of the descendants of Sigurðr and Bjǫrn; ch. 19 tells how two warriors meet at a royal funeral feast, and after an exchange of verses recognise one another as former followers of Bjǫrn and Ragnarr; and ch. 20 relates how a trémaðr ‘wooden man’, discovered by the followers of one Ǫgmundr on the Danish island of Samsø, claims in verse to have been placed there and treated as a cult-object by the sons of someone with a name corresponding closely to Ragnarr’s nickname, loðbrók.
From the evidence of RagnSon in Hb it seems clear that the stanzas numbered here Ragn 11 and 13 (spoken by Eiríkr when at the point of death in Sweden), and those numbered Ragn 18-22 (spoken by Áslaug and her sons Sigurðr, Bjǫrn, Hvítserkr and Ívarr, one stanza each, in that order, when planning revenge for the deaths of Eiríkr and Agnarr) were present in the basic narrative. Also present there, to judge from RagnSon and X, was the half-stanza in tøglag metre edited as Sigv Knútdr 1I, quoted from Sigvatr Þórðarson’s Knútsdrápa in connection with the account in the prose of the cutting of the eagle on Ælle’s back. This is not found in Y. Possibly also present in the basic narrative, though by no means certainly, was the stanza edited in the present volume as KrákÁsl Lv 11 (RagnSon 1); this appears in RagnSon, where it is spoken by Áslaug, according to the prose, on receiving the news of the death of her son Sigurðr (nicknamed ormr-í-auga ‘Snake-in-eye’). It is not found in Y, however; nor does it seem to have been present in X. These stanzas, i.e. Ragn 11, 13, 18-22, Sigv Knútdr 1I and KrákÁsl Lv 11, are the only ones that occur in RagnSon.
It is clear from 147 that the stanzas numbered here Ragn 1 (spoken by Ragnarr on killing Þóra’s serpent) and Ragn 2 and 3b (spoken by Kráka-Áslaug on first meeting Ragnarr), if not Ragn 3a, spoken by Ragnarr on that occasion (see Rloð Lv 2 (Ragn 3a), Note to [All]) were present in X. It is likely also, though the state of 147 does not allow this to be confirmed, that Ragn 4 (spoken by Ragnarr when offering Þóra’s shift to Kráka-Áslaug) and 5 (spoken by Kráka-Áslaug when declining the offer) were present in X. Clearly present in X were Ragn 6 (spoken by Kráka-Áslaug on first sleeping with Ragnarr), 7 (spoken by Bjǫrn on his and his brothers’ victory at Hvítabœr) and 8-10 (spoken by Ragnarr on the birth of his son Sigurðr ormr-í-auga). It is likely that, of the four stanzas spoken in Y by Eiríkr at the point of death in Sweden (Ragn 11-14), Ragn 11 and 13, at least, both present in RagnSon (see above), were present in X; neither Ragn 12 nor Ragn 14, on the other hand, is present in RagnSon and the state of 147 does not permit any firm statement as to the presence (or otherwise) of any of these four stanzas in X. It is clear that the following thirteen stanzas (15-27) were in X: Ragn 15 and 16 (spoken respectively by Áslaug and by messengers from Sweden), 17 (spoken by Áslaug), 18-22 (spoken by Áslaug and her four sons, one stanza each, in the same order as in RagnSon), 23 and 24 (spoken by Ragnarr before invading England), 25 (spoken by Áslaug when offering him a protective shirt) and 26-7 (spoken by Ragnarr in the snake-pit). In X Krm (spoken by Ragnarr in the snake-pit) appears to have preceded st. 26, separated from it by a few lines of prose.
In Y, as already noted, Krm appears as an appendix to Ragn, rather than as an integral part of it, as in X. Stanzas presumed to have been added in the Y-redaction are as follows: Ragn 28 and 29 (spoken by Bjǫrn on returning with his brothers to Denmark from the south), and Ragn 30-40, which appear in the chapters (18-20) added in the Y-redaction (see above). Ragn 30 and 31 (in ch. 18), are spoken by Áslaug, here named Randalín, on the death of her son Hvítserkr; Ragn 32-7 (in ch. 19) are spoken by the two warriors, one after the other (one stanza each), who recognise each other as former followers of Ragnarr and Bjǫrn; and Ragn 38-40 (in ch. 20) are spoken by the trémaðr ‘wooden man’, discovered by the followers of Ǫgmundr on the island of Samsø.
In his edition of Vǫls and Ragn (Ragn 1906-8), Magnus Olsen gives a diplomatic text of those two sagas as preserved in 1824b (pp. 1-175) and of what he could read of the Ragn text in 147 (Ragn 1906-8, 176-94), followed by a critical edition of the forty Ragn stanzas (Ragn 1906-8, 195-222). For the information given above about the stanzas in X (preserved fragmentarily in 147), the present editor is deeply indebted to Soffía Guðný Guðmundsdóttir. Whereas Olsen (Ragn 1906-8, 176-213), could identify traces of only sixteen of these stanzas, i.e. Ragn 1, 6-10, 15-16, 18-20, and 22-6, Soffía has not only added to Olsen’s readings of those stanzas, but has also identified traces of five more stanzas, namely Ragn 2, 3, 17, 21 and 27.
An attempt at dating sts 1-37 of Ragn, leaving aside KrákÁsl Lv 11 (RagnSon 1), may begin with a reference to Snorri Sturluson’s description and illustration in Háttatal 54 (SnSt Ht 54III) of the metrical variant of dróttkvætt named in one ms. Ragnars háttr ‘Ragnarr’s metre’, in which, says Snorri, Ragnarr loðbrók composed poetry (see SnE 2007, 24-5; cf. Faulkes 1987, 198). For the characteristics of this metre, see Context to SnSt Ht 54III. Snorri’s information shows that, by the time of the composition of Ht (1222-3, see SnE 2007, ix-x), the name of Ragnarr loðbrók had come to be associated with a particular type of relatively loosely composed skaldic verse (cf. SnSt Ht 54III, Note to [All]).
Stanzas 1-37 of Ragn frequently depart from the strict rules of dróttkvætt. Not a single one of them conforms exactly to the strict dróttkvætt pattern; on the other hand, not a single one of them conforms exactly to the Ragnars háttr as Snorri illustrates it. Hendingar (including skothendingar) occur in them, but only sporadically, and by no means always in the lines where they might be expected. The head-stave does sometimes fall in the second metrical position (as in Ragnars háttr), but this is by no means the general rule. All the indications are that these stanzas were composed relatively late, at a time when adherence to the strict rules of dróttkvætt was falling into decline.
There are some cases of discrepancy between stanzas and their prose contexts, discussed in the Notes. None, however, give clear indications of whether the stanzas are older than their prose contexts. Thus one has little alternative but to follow Finnur Jónsson (LH II, 142) in concluding that the stanzas are relatively young and probably not older than the thirteenth century. A further consideration is that, as Finnur (LH II, 142), indicates, the stanzas of Ragn (especially sts 8-10) show a clear awareness of the idea that Ragnarr’s wife Áslaug was the daughter of Sigurðr Fáfnisbani and Brynhildr Buðladóttir. Krm 26/3 mentions Áslaug by name, but not her parentage (cf. McTurk 1991a, 147), and Saxo (d. c. 1220) makes no mention of either. Her name and paternity, though little more, may have been mentioned in the lost Skjǫldunga saga (Bjarni Guðnason 1969, 32 and n. 15), dated to c. 1200. It would appear from the prose of RagnSon in Hb (Hb 1892-6, 459), where Áslaug plays a significant part in the action, that Sigurðr and Brynhildr were presented as her parents in Haukr’s source text, where, however, there is no certainty that sts 8-10 were present. It seems that this idea of her parentage did not become firmly established until some time into the thirteenth century. It should finally be mentioned that Finnur Jónsson (LH II, 142, 153-4) argued for the influence of Krm 28 and 24/5-8 on the wording of Ragn 26. This is certainly possible chronologically if the Ragn stanza is thought to date from the thirteenth century, since Krm was composed probably not much later than c. 1200 (see the Introduction to Krm in the present edition), and the verbal resemblances are indeed striking. It is perhaps safest to assume, however, that Krm and the Ragn stanza here go back to a common source.
As for the three stanzas in fornyrðislag (Ragn 38-40) spoken by the trémaðr in the final chapter (20) of the Y-redaction, it seems that the first of these, st. 38, did not belong originally with the other two (for a contrary view, see Poole 1991, 20-2). As argued in st. 38, second Note to [All], it is likely that this stanza was either borrowed from Hálf into the Y-text of Ragn some time in the fourteenth century, possibly by the scribe of 1824b, or, alternatively, was borrowed by the Y-redactor from an unknown source. If so, its history can be traced back no further than to the second half of the thirteenth century, the time (in all likelihood) of the Y-redactor, though it may well be older than that. As for sts 39-40, which seem to have belonged together from the beginning, it has been argued that they are ‘relatively old’ (see st. 39, Note to [All]), apparently on the basis that in st. 40/4, as it has traditionally been read, a form of the name Loðbrók occurs (in the phrase synir Loðbrókar ‘sons of Loðbrók’), uncoupled with the name Ragnarr, thus pointing to a time before the application of the names Ragnarr and Loðbrók to the same person, the first known instance of which occurs in Ari Þorgilsson’s Íslb, written 1122-32, as noted above. It has also been argued (by Gutenbrunner 1937, 141-3 and Olsen 1962c, 21-5) that these two stanzas, in which the trémaðr speaks of himself as a cult-object, weather-beaten and unprotected by flesh or clothing, have been influenced by Hávm 49 and 50, which speak respectively of clothes being given to two trémenn ‘wooden men’ and of a withering tree unprotected by bark or foliage.
In the present edition, 1824b is taken as the main ms. for each of the forty stanzas of Ragn, with variant readings supplied from 147 and Hb for the stanzas and parts of stanzas they preserve. Readings are also given for sts 1, 3 (a and b), 4, and 5 from AM 762 4°ˣ (762ˣ), which preserves those four stanzas, each preceded by a short introduction in Danish and followed by a Latin translation and commentary, in the hand of Magnús Ólafsson of Laufás (d. 1636), who describes them as collected, along with other items, Ex Antiqvitatibus ‘from ancient sources’ (see Kålund 1888-94, II, 183). For Ragn 38, which is also preserved in Hálf, readings are given from the main ms. of that saga, GKS 2845 4° (2845). These five mss were used by Finnur Jónsson in Skj; only the first three were used by Magnus Olsen in his critical edition of Ragn 1-40 (Ragn 1906-8, 195-222). These are the only two previous editions in which sustained use is made of 147. Thanks to the work of Soffía Guðný Guðmundsdóttir, the readings from 147 in the present edition are greater in number, and are shown for more stanzas, than in either of them. Ms. 1824b is the obvious choice as the main ms. for each stanza, since its prose text of Ragn provides the only complete and, for the most part, coherent context for all forty stanzas.
Mss of Ragn containing all or some of the Ragn stanzas but not used in the present edition because found on inspection to derive in all probability from 1824b, directly or indirectly, are as follows: Borgarnes: (Héraðskjalasafn Borgarfjarðar): Sagnahandrit Jóhannesar Jónssonarˣ; Cambridge, MA: Harvard Houghton Icel 32ˣ; Dublin: TCD 993ˣ; London: BLAdd 11160 Iˣ, BLAdd 24969ˣ; Reykjavík (Landsbókasafn Íslands): ÍB 76 4°ˣ, JS 8 folˣ, JS 12 folˣ, Lbs 170 folˣ, Lbs 272 folˣ, Lbs 361 4°ˣ, Lbs 824 4°ˣ, Lbs 841 4°ˣ, Lbs 1061 4°ˣ, Lbs 1246 4°ˣ, Lbs 1487 4°ˣ, Lbs 1491 4°ˣ, Lbs 2341 4°ˣ, Lbs 4661 4°ˣ; Stockholm (Kungliga Biblioteket):Holm papp 95 folˣ, Holm papp 38 4°ˣ; Uppsala (Universitetsbiblioteket): Ihre 77ˣ, R 703ˣ, Westin 89ˣ. There are a further twenty-two paper mss of Ragn, yet to be inspected by the present editor, which are here assumed to derive ultimately from 1824b. Also assumed to derive from 1824b are six paper mss containing one or more of the Ragn stanzas independently of the prose text; all but two of these six have been inspected by the present editor.
It is further assumed, for the purposes of the present edition, that the mss of RagnSon other than Hb (see above) descend directly or indirectly from Hb. These are: Cambridge MA: Harvard Houghton Icel. 32ˣ; Copenhagen: AM 307 4°ˣ; Dublin: TCD 1016ˣ; London: BL Add 11131ˣ; Oslo: UB 246 folˣ; Paris: BSG 3714ˣ; and Reykjavík (Landsbókasafn): JS 19 folˣ; Lbs 2796 4°ˣ, Lbs 3795 8°ˣ.
In the first edition of Ragn (Björner 1737), the text, given on each page with a facing Swedish translation and a Latin translation at the foot of the page, has been treated meget vilkaarlig ‘very cavalierly’ (Ragn 1906-8, xi). Close reading shows that it derives ultimately from 1824b. Furthermore, it omits altogether sts 20-4, 26-9 and 31, as well as ll. 2-6 of st. 12, and l. 14 of st. 36, and substitutes for sts 26-7 the twenty-nine stanzas of Anon Krm, the latter’s text based apparently on that of Worm (1651, see the Introduction to Krm). Von der Hagen’s edition of Ragn (1814), based on Björner’s, gives no translation, but otherwise follows Björner in all respects. Neither is used in the present edition. Von der Hagen’s German translation of Ragn (1828), on the other hand, does not have these omissions, nor does it include Krm; it is no doubt influenced by Rafn’s Danish translation of Ragn (1822), which is based on 1824b. The editio princeps of all forty stanzas of Ragn is in fact that of Rafn (1822, 177-246), published in the same volume as his Danish translation, with variant readings and commentary for each stanza. It is not used in the present edition, however, since his text of the Ragn stanzas in FSN (1, 235-99), which is used here, is based on a relatively close reading of 1824b (FSN 1, xvii), and makes use for the first time of 762ˣ.
Olsen (Ragn 1906-8) and Finnur Jónsson in Skj AII and BII are the only two previous editors to make sustained use of 147. Olsen is the first to do so in print. In addition to his critical edition of the Ragn stanzas, Olsen (Ragn 1906-8, 176-94) prints all he could read of the prose and verse of the 147 text of Ragn (which includes much of Krm as well as Ragn 26 and, it now emerges, Ragn 27). He notes (Ragn 1906-8, lxxxiv) that Guðbrandur Vigfússon was the first to identify a text of Ragn different from that in 1824b in a palimpsest section of 147, fols 93-111, in which the Ragn text is obscured partly by rubbing and partly by legislative pronouncements written over it in a hand dating from c. 1580. Guðbrandur refers briefly to this in the Prolegomenon to his edition of Sturlunga saga (Stu 1878, I, cxcv). Ms. 147 does not appear to have been used by Guðbrandur and Powell, however, in their edition of the Ragn stanzas (and of Krm) in CPB II, 339-53, which is nevertheless one of the editions used here.
Olsen (Ragn 1906-8, lxxxix) notes that Sophus Bugge (1833-1907) had made a careful transcript of what he could read of the text of Ragn in 147, and had placed this at his disposal, and that he, Olsen, after making his own transcript independently of Bugge’s, had found considerable differences between the two transcripts. Bugge’s transcript, which has never been published, had earlier been made available, in part at least, to Storm (1877, 477 n. 1; 1878, 124 n. 1), and to Edzardi (1855-80, III, i-ii, xli), both of whom refer to the text of Ragn in 147 as having recently come to light; cf. also Storm (1877, 429 n. 4; 1878, 83 n. 5). Edzardi (1855-80, III, i-iii, xlv-viii, 256, 329-31) makes considerable use of it in his revised version of von der Hagen’s German translation (1828) of Ragn. Edzardi’s revision of von der Hagen, and Paul Herrmann’s German translation (1923b) of Ragn, have been found helpful in preparing the present edition.
Other editions and translations used in the present edition include Valdimar Ásmundarson’s second edition of Ragn (Ragn 1891), helpful for its incorporation in the verse passages of readings suggested by Jón Þorkelsson (1822-1904) and listed in the preface to Valdimar’s first edition (1885-9, I, ix-xi); Eskeland’s second edition of Ragn (Ragn 1944), with a normalised text based on Olsen’s edition of the 1824b text and a facing translation into Modern Norwegian (nynorsk); and Schlauch’s English translation (1930, 183-256). The editions and translations of Ragn so far mentioned, with the exception of Eskeland’s second edition but including his first (published 1914), are listed in Halldór Hermannsson (1912, 3-4, 34-5; 1937, 48, 60-1). The more recent editions of Ragn used in the present edition are: Guðni Jónsson’s edition in normalised spelling in FSGJ I, 219-85 (cf. xxiv-vi), based ultimately on Olsen’s edition of the 1824b text, but departing from it occasionally in its treatment of the stanzas; Örnólfur Thorsson’s edition (Ragn 1985) of the 1824b text in Modern Icelandic spelling; and Ebel’s edition (Ragn 2003) of the 1824b text, in normalised spelling but including diplomatic texts of the Ragn stanzas as preserved in 1824b and Hb. The Spanish, French, and English translations of Ragn by Ibáñez Lluch (1998), Renaud (2005), and Waggoner (2009) respectively (the first of these now in a second edition, Ibáñez Lluch 2014) have also been found helpful. There is also an Italian translation by Meli (1998).