History in the 'political sagas.' (medieval Icelandic sagas)
When Isidore of Seville defined historia as 'narratio rei gestae',(1) he went on to emphasize its origins in eyewitness accounts: 'None of the ancients would write history unless he had been present and seen what he narrated ...'(2) More recent historians are suspicious of eyewitness accounts and aim 'to discover, select, analyse and interpret' the sources they use.(3) According to Beryl Smalley, a 'critical study of the remote past, as distinct from mere compilation of earlier sources, called for tools and equipment which were lacking in the Middle Ages', and Isidore may have performed a useful service by redirecting the energies of mediaeval historians into 'contemporary history'.(4)
We find mediaeval Icelandic historians somewhere along the scale between the ancients, apparently copying out firsthand reports, and the modern historian, critically studying the remote past with the tools of editions, translations and secondary sources. Ari porgilsson (1067/8-1148) was careful to distinguish between eyewitness accounts and secondhand reports; he used (foreign) written sources and had a modern interest in chronology.(5) But while Ari's methods may be recognizable to modern historians, most Icelandic historical narratives are thought to tend towards fabula more than historia, and are correspondingly dismissed by historians for failing accurately to divulge the res gestae.(6) But rather than blaming the mediaeval Icelanders, it might be more productive to make positive use of their 'failure' to become 'modern' historians. Icelandic texts are our main gateway to the medieval Norse past, and for the philologist(7) there is still much to be learned from the approaches to recording and transmitting the past that can be discerned in these texts.
The 'political sagas'
I propose to look at aspects of how mediaeval Icelanders went about the narratio rei gestae in Orkneyinga saga, Jomsvikinga saga and Fcereyinga saga.(8) The grouping together of these three sagas has sometimes been by default, because they do not fit neatly into the categories of Islendingasogur, kings' sagas and the like. But some scholars see positive similarities between them. Melissa Berman classifies them together on thematic grounds, because they all deal with 'issues of political power'.(9) Peter Foote would also see a thematic link, in his case a Christian one, and he emphasizes similarities in their narrative methods.(10) In my view, the most important link between these three sagas is the probable early date of Composition of their first (now lost) versions (c. 1200), as well as their complicated textual history and fragmentary preservation. Such an early date would place these sagas towards the beginning of Icelandic historical writing, and disentangling their intricate textual history can provide insights into the development of attitudes to and methods in Icelandic historiography and saga-writing.
A large part (about two-thirds) of Orkneyinga saga tells of twelfth-century events, and it gets particularly detailed from just before the middle of that century. It is generally believed that the saga was based on traditions from Orkney, even that its author 'had probably met some of the people who figure in his account'.(11) It would thus appear to be a prime candidate for the type of history based on eyewitness reports advocated by Isidore. And indeed there is one clear reference to such a report. In ch. lxxv, after Sveinn Asleifarson's kidnapping of Jarl Pall, we are told that Sveinn returned to Orkney, while Pall remained in Scotland: 'Ok er petta frasogn Sveins um penna atburb' ('And this is what Sveinn said about this event').(12) Sveinn was not only an eyewitness, but also a participant. But the historian who transmitted this was not an uncritical compiler of eyewitness reports, for he goes on immediately to cast doubt on this account by giving an alternative version, 'en pat er sogn sumra manna' ('but the account of some people is...', telling of Pall's sticky end at the hands of his sister Margret and Sveinn). He claims not to know which of these two versions is sannara ('truer'), but points out that Pall was never heard of again, in either Orkney or Scotland, Indeed, he has been exercising caution throughout the episode leading up to the reference to Sveinn's account, for he twice points out a lack of information: 'Ekki er getit orda peira Pals jarls ok Sveins, medan peir foru badir saman' ('There is no record of the conversation between Jarl Pall and Sveinn, while they were both journeying together') and 'Ekki er getit fleiri orda jarls en pessa' ('There is no record of further words of the jarl than these').(13) These are heavy hints to distrust the eyewitness account being presented in such detail.
Once the reference to Sveinn has alerted us to the fact that the narrative is following an eyewitness account, it becomes clear that the whole story of the kidnapping (chs. lxxiv-lxxvi) is told from at least two different points of view: first Sveinn's, then that of the supporters of Jarl Pall.(14) The effect is to remind the reader that even eyewitness accounts can conflict, and shows a sophisticated and subtle questioning of the evidence. Once we are made aware of this approach, we begin to appreciate the way in which the saga is based on eyewitness accounts. Thus, some eyewitness accounts that operate at the diegetic level (i.e. as events within the story) may also have extradiegetic significance (i.e. derive from reports used as sources by the author/narrator). There are two examples in ch. lxxvi: Borgarr's report of having seen Sveinn's kidnapping journey in both directions, and Sigurdr of Westness's account to his friends of his meeting with Hakon karl when the latter brought the news of Pall's kidnapping.(15) Particularly in chs lxxiv-lxxvi, the historian's critical approach is expressed in the narrative devices employed. Close literary analysis reveals this critical presentation of the evidence of eyewitness accounts.
Despite Isidore, mediaeval historians did try to write history even where they lacked eyewitness accounts. And although Orkneyinga saga is at its most detailed when dealing with recent history, the critical attitude is also apparent when it is dealing with events from the remoter past, not based on firsthand reports. Just as competing eyewitness accounts are balanced in chs. lxxiv-lxxvi, so other traditions, of various types, are juggled throughout the saga. Metatextual references to alternative or additional information, often using a formula such as 'sumir menn segja' ('some people say'), can be found in chs. xx, xxix, xlii and l. In ch. xlii, one of the alternative versions of the place of death of Erlingr Erlendsson is supported by a reference to Snorri Sturluson. Besides such unspecified traditions, references to named texts give
the appearance of being to written works, even if we cannot identify them with surviving texts. Such references include a 'saga Magnuss konungs' and an 'aevi Noregskonunga' in ch. xxi, and a saga of Erlingr skakki in ch. lxxxix.(16) There are also two references to poems which are not quoted, a drapa about Hakon Palsson in ch. xliii and a kvaedi about Magnus and Hakon in ch. xlvi.(17)
Then there are the many skaldic verses quoted throughout the saga. Like the reports of events, these can operate at the diegetic level (i.e. they can themselves be narrated as events) or at the extradiegetic level (i.e. they can be the reports on which the narrative is based). A few examples should make these distinctions clear.(18)
The majority of skaldic poems in Orkneyinga saga are cited as speech acts that are an integral part of the events being narrated. For example, when Rognvaldr kali is shipwrecked in Shetland, 'spurdu menn at um ferdir hans' ('people asked about his journey') and, in reply, 'jarl kvad visu' ('the jarl spoke a verse') and his stanza describing the shipwreck is quoted (ch. lxxxv).(19) The verses in the second half of the saga (from ch. lviii onwards) are almost always introduced by the verb kvad as a part of the narrative in this way, and most of them are spoken by Rognvaldr himself. Even when Rognvaldr's men have captured a ship off Sardinia (ch. lxxxviii) and there is some discussion about the exact sequence of events and who was the first to board, the speculation takes place entirely at the diegetic level. Some of Rognvaldr's men thought that 'pat vaeri omerkiligt, at peir hefdi eigi allir eina sogn fra peim stortidendum' ('it would be foolish if they all had different accounts of these great events'), and so the doubt is banished at the level of the story by Rognvaldr in a verse in which he declares Audunn to have been the first to board.(20)
The verses in the earlier part of the saga, however, referring to events much further back in time, are presented as the reports on which the narrative is based. These reports can have a peripheral function in relation to the rest of the narrative, serving to confirm or amplify information that does not seem entirely dependent on the verse, as when, in ch. xix, a verse by Ottarr svarti is quoted to illustrate Rognvaldr Brusason's relationship with King Olafr. Mostly, however, the verses function as eyewitness reports, either by implication or explicitly. Seven of the eighteen verses attributed to Arnorr jarlaskald in the saga contain the first person pronoun ek or possessive adjective minn, including several that make explicit his presence at the events being described. Thus, of a battle on the Isle of Skye, Arnorr says: 'gorla sak at gindi / grar ulfr of na sorum' ('I saw clearly that the grey wolf gaped over the wounded corpse'); of the battle of Raudabjorg, Arnorr said: 'Hvarn tveggja sak hoggva / hird a Pettlandsfirdi, / vor prifusk mein at meiri, / minn audgjafa sina' ('I saw both of my benefactors cut down each other's troop in the Pentland Firth, our hurt became even greater').(21) Many of the remaining verses can be associated with those which reproduce Arnorr's eyewitness report of events: in ch. xx two verses (one containing an ek, one not) are quoted after the introduction 'pessa getr Arnorr i porfinnsdrapu' ('Arnorr tells of this in Porfinnsdrapa'),(22) and most of the others must originally be stanzas from this poem. A verse preserved only in Uppsala, Universitetsbibliotek, MS 702 emphasizes the poet's visual and aural impressions of Porfinnr's battles in England: 'hornablastr' ('the blowing of horns'), 'hristisk hugsterks jofurs merki' ('the banner of the resolute prince was shaken'), 'vigljost' ('light enough for battle'), 'skulfu jarn' ('iron [weapons] shivered').(23) While Arnorr saw and heard the events he is describing, his eyewitness report is not reproduced purely mechanically. His Porfinnsdrapa is dismembered and plundered for historical information; it has been 'selected, analysed and interpreted'. Extracts from Arnorr's poems are almost invariably introduced with a formula such as 'sva segir Arnorr' ('as Arnorr says'), which emphasizes their function - to support the statements in the narrative.
One explanation for these two approaches to verses as evidence and verses as speech acts is that the verses in the latter part of the saga, cited as speech acts, were too recent and had not been in the tradition long enough to achieve their place in the 'historical perspective' which was the prerequisite for Norse history-writing.(24) But, as we have seen, even recent events could be subjected to critical scrutiny in the saga's prose narrative, and the Sardinian episode shows an awareness of the historical gap between events as they happen and a later account of them to an audience. Many of Rognvaldr's verses are playful and have much entertainment value, especially in their prose contexts.(25) The Rognvaldr episodes in Orkneyinga saga already show a tendency towards the fictionalizing process by which skaldic verses became an integral part of saga entertainment.
The version of Jomsvikinga saga found in Copenhagen, Arnamagnaeanske Institut, MS 510 4to contains a number of verses not preserved in other manuscripts of the saga. In ch. xxxix a description of Hakon jarl and his sons just before the battle of Hjorungavagr ends with the information 'peir styrdu allir skipum. Sva sagdi pordr Kolbeinsson, er hann orti um Eirik' ('they all captained ships, as pordr Kolbeinsson said when he composed a poem about Eirikr'), supported by two stanzas describing the Norwegian fleet on its way to the battle.(26) In chs. xlv-xlvi (describing the course of the battle after Hakon's sacrifice to Porgerdr Hordatroll), nine whole and two half stanzas by Tindr Hallkelsson are quoted. In ch. xlv, the occasion is Hakon's casting-off of his armour: 'Hakon jarl bardisk sva djarfliga, at um sidir steypti hann af ser brynjunni fyrir sakir hita ok erfidis; sva segir Tindr' ('Jarl Hakon fought so boldly, that eventually he threw off his mail-coat because of the heat and the toil; so Tindr says').(27) This is followed by three verses, in two of which this action is described.(28) Then the statement that twenty-five of the Jomsvikings' ships were cleared introduces five and half stanzas of conventional battle description, the first of which includes the relevant numeral.(29) But first the compiler felt moved to stress the importance of the source of this hard fact: 'pat segir Tindr Hallkelsson i flokki peim, er hann orti um Jomsvikinga, og heyrir sva par til, at hann var par sjalfr' ('Tindr Hallkelsson says that in the poem which he made about the Jomsvikings, and it is relevant to this that he was there himself').(30) Indeed two of teh following verses (although it is hard to be sure, for they are quite mangled) contain the first person pronoun, although not in a context which would necessarily indicate that the poet was present at the battle. In ch. xlvi the description of Bui jumping overboard with two chests of gold is followed by one and a half stanzas referring to his watery end.
These are the only verses that are cited as historical reports in Jomsvikinga saga. All the other verses in AM 510 4to (in chs. xlii, xlvi-xlvii, xlix and liii) are presented as diegetic events, i.e. speech acts in the narrative. Some of these verses are also cited in one or more of the other versions of the saga and are presented in the same way there, as the speech of the characters.
AM 510 4to is a late (sixteenth-century) manuscript of Jomsvikinga saga, the main merit of which is that it appears to have made use of a lost version of the saga which was also a source for both Fagrskinna and Heimskringla.(31) Since the verses by pordr Kolbeinsson are also quoted in Fagrskinna and some of the verses by Tindr are, as we have seen, quoted in Heimskringla, it is likely that these particular sections of AM 510 4to represent this lost version of the saga. They are therefore worth more careful consideration.
In ch. xxxix, the verses by Tindr are introduced by a fairly detailed description of Hakon jarl's fleet, as seen by the Jomsvikings: var hann eigi einskipa ok eigi med .ii., heldr varu meir en .ccc.; pat varu snekkjur ok skeidur ok kaupskip ok hvert fljotanda far, er jarl fekk til peira, pau er ha varu bordi, ok oll varu skipin baedi hladin af monnum ok vapnum ok grjoti.(32) (he had not one ship, and not two, but rather there were more than 300; there were sailing-ships and warships and merchant ships and every high-sided seaworthy vessel the jarl could get them, and all the ships were laden with men and weapons and stones).
This could be based on the verses: the first stanza contains the words snekkja, knorr and skeid, and the second talks of hava stafna. There is no equivalent passage in the other versions of the saga. For instance, Stockholm, Kungliga Biblioteket, MS Perg. 4to no. 7 does not describe Hakon's fleet at all,(33) while Copenhagen, Arnamagnaeanske Institut, MS 292 4to, after noting the Jomsvikings' perception that the inlet was covered with (Hakon's) ships, tells what the Norwegians saw: 'pa sa peir jarlarnir, Hakon og synir hans, hvar peir eru komnir Jomsvikingar' ('then the jarls, Hakon and his sons, saw that the Jomsvikings had come').(34)
Tindr's verses in ch. xlv of AM 510 4to also follow a shift of perspective towards Hakon in the prose narrative. After a colourful description of the trollwomen fighting on his side, the reprehensible devilishness of it all is somewhat undermined by the emphasis on Hakon's bravery in fighting: pat er sagt, at peir Hakon jarl gordu hardar atlogur ok bordusk nu djarfliga. Nu er pess getit, at Hakon jarl bardisk sva djarfliga, at um sidir steypti hann af ser brynjunni fyrir sakir hita ok erfidis.(35) (It is said that Jarl Hakon and his men made fierce attacks and now fought boldly. It is also said that Jarl Hakon fought so boldly that he eventually threw off his mail-coat because of the heat and the toil.)
The only counterpart to this in the other versions is a general comment (at a different point in the narrative) that it was so warm that many men took their clothes (but not their armour) off.(36)
The vignette of Bui leaping overboard with two chests of gold is told with gusto in all versions of the story, but in AM 510 4to the clear statement of Tindr's verse that he was forced overboard somewhat undermines the preceding heroic description: er pat mal manna, at engi einn hafi meiri kappi verit i lidi peira Jomsvikinga, en Bui digri; hafdi hann ok sva margan mann drepit i bardaganum, at pat kunni enginn madr at telja.(37) (people have said that no one was a greater champion in the Jomsvikings' troop than Bui the Stout; he had indeed killed more men in the battle than anyone could count).
This unevenness in chs. xxxix and xlv-xlvi arises from teh incomplete integration of sources which basically concentrate on the Hladajarls (Tindr's and pordr's poems in praise of Hakon and his son Eirikr respectively) into a text that is otherwise primarily interested in the deeds of the Jomsvikings.(38) Norman Blake called Jomsvikinga saga 'pure fantasy' and 'the end product of many years of literary accretion.'(39) Mostof all in the version he edited (MS perg. 4to no. 7), but also in the others, this process resulted in a highly fictional and entertaining narrative, with the Jomsvikings as heroes, continuing into Bjarni Kolbeinsson's Jomsvikinga drapa with its glorification of Vagn Akason. But what AM 510 4to demonstrates is that, at an earlier stage in the development of the saga, parts of it were attempts at historical narrative, based on the testimony of skaldic verses, at least some of which were thought to be eyewitness reports.
Like Orkneyinga saga and Jomsvikinga saga, Fareyinga saga is preserved in Flateyjarbok (but not in complete form).(40) It is also associated with other expanded versions of the sagas of Olafr Tryggvason and St Olafr. But there are no traces in it of the critical historical approach that is prominent in Orkneyinga saga and detectable in Jomsvikinga saga. The only verse quoted is Prandr's kredda (ch. lvi). The only allusion to the traditions on which the saga was based is the reference to two Icelandic(?) informants and Ari porgilsson (ch. xxvii), but this has been explained by Olafur Halldorsson as a device on the part of the author to make his saga more plausible.(41)
What Fareyinga saga does demonstrate is the way in which techniques deriving from the historical interest in eyewitness reports could be transformed into effective literary devices, and could be used to write fabula rather than historia.
Peter Foote has discussed in some detail the way the saga has of 'presenting events and circumstances without comment and as they appeared to people who were partly or totally misled by them.'(42) An example of this is the scene in chs. ii-iii in which we, along with Harekr, believe he is giving money to his brother Sigurdr, only to discover that it was in fact someone (we never find out who) impersonating Sigurdr. Narratologists call this device 'focalization', a term which serves to distinguish 'between the question who is the character whose point of view orients the narrative perspective? and the very different question who is the narrator?.'(43) This term is useful in discussing Icelandic texts, which are invariably third-person narratives with an omniscient narrator and a panoramic narrative perspective, but where occasionally the narrative perspective is limited to the perceptions of an individual character. This is done quite regularly in Fareyinga saga, often in a scene leading up to an attack of some kind. Thus, in ch. vi Hafgrimr goes (with six men and his wife) to get help from his father-in-law, Snaeulfr, and the audience is made to share their puzzlement as they arrive at the farm, where visitors are clearly expected, but there is no one to be seen: ok er peir komu at eyjunni, sa peir ekki manna uti a baenum ok ekki uti a eyjunni. Ganga nu upp til baejarins ok inn i husin ok verda ekki vid menn varir. Til stofu ganga pau ok er par sett upp bord ok baedi a matr ok drykkr, en vid menn verda pau ekki vor. petta potti peim undaregt.(44) (and when they reached the island, they saw no one outside by the farm or on the island. Now they go up to the farm and in the buildings but can see no one. They go into the main building where tables are set up with both food and drink, but they can see no one. They thought this very strange.)
The next morning they see Snaeulfr and his household arrive in a boat, having made their point that they do not wish to be involved in the quarrel with Brestir and Beinir. In the next chapter we are on Little Dimun with Brestir and Beinir, watching the arrival of 'three ships laden with men and weapons, with twelve men on each ship' whom they go on to recognize as Hafgrimr, prandr and their followers. There are similar uses of focalization in chs. xxxiv, xlv, xlviii and lv; and in ch. xxiii the temple of porgerdr Hordabrudr is revealed through Sigmundr Brestisson's eyes.
In Fareyinga saga this device increases suspense, mystery, enjoyment, but does not appear to have any function beyond a ludic literary one. It is quite different from Orkneyinga saga, where, in chs. lxxiv--lxxvi, shifting focalization gives us the same event from two different points of view: in chs. lxxiv--lxxv, the kidnapping of Jarl Pall is presented with a strong tendency to focalize through Sveinn Asleifarson, the kidnapper, leading up to the reference to 'Sveinn's account of this event', discussed above; in ch. lxxvi, the kidnapping is revealed as it appeared to Sigurdr of Westness and his men, with contributions by other eyewitnesses, such as Borgarr of Geitaberg. There is no suspense in telling the same story twice, but in Orkneyinga saga it is done for historical reasons, to give a comprehensive and critical narratio rei gestae.
The total artistic control of the narrator is further revealed in chs. xiv--xvi of Fareyinga saga. In this, Ulfr, the Norwegian foster-father of Sigmundr and porir, tells them the life story of someone called porkell. It is only at the end of this narrative that Ulfr reveals that he is, in fact, porkell, by a sudden switch to the first person pronoun: ok er stund lidr ferr hann burt or hellinum ok til baejar pess er Poralfr bondi hafdi att ok tekr nu Ragnhildi i burt i annan tima ok raedsk nu a fjoll ok eydimerkr; 'ok her nem ek stadar', sagdi hann, 'sem nu hefi ek bygd mina setta, ok her hefi ek verit sidan ok vid Ragnhildr atjan vetr ...'(45) (and when some time had passed, he left the cave and went to the farm that had been Poralfr's and took Ragnhildr away for a second time and went to live in the mountains and wilderness; 'and I ended up here,' he said, 'where I now have my abode, and Ragnhildr and I have been here for the eighteen years since...')
Within Ulfr's narrative, the fiction that he is narrating a story about a third person is deliberately maintained. Thus Ulfr says (ch. xv): 'er pat sogn manna at porkell yrdi banamadr hans' ('people say that porkell was his killer'), referring to his alter ego in the third person and pretending not to know something he, as porkell, must have known.(46) In narratological terms, we have Ulfr, a diegetic character in the primary narrative, acting as a diegetic narrator of the secondary narrative. In this secondary narrative, the hypodiegetic character porkell is revealed to be the same person as the diegetic narrator/character Ulfr, bringing us neatly back to the primary narrative. Ulfr/porkell's total control over his narrative, revealing only what we need to know as we need to know it, mirrors that of the narrator of Faereyinga saga.(47)
Even in its present form, Orkneyinga saga reveals an attempt at critical historical narrative. Most of the saga is preserved only in Flateyjarbok, yet many of the textual elements that betray the historian behind the saga are preserved only in the fragmentary mananuscripts. Thus, cross-references which reveal a consciousness of the text as constructed text -- such as 'sem fyrr segir' ('as it says earlier') or 'sem adr var ritat' ('as was written above') -- are rare in Flateyjarbok, but relatively common in the fragmentary manuscripts.(48) Similarly, none of the three references to prose sources in chs. xxi and lxxxix is found in Flateyjarbok. Clearly, there were good reasons for the compiler(s) of Flateyjarbok to excise such metatextual references from the saga, but in doing so they have partially obscured the careful, consciously historical attitude displayed in earlier versions of the saga. Some of the episodes involving Rognvaldr show the tendency to fictionalize even what must have been recent eyewitness reports.
Jomsvikinga saga exists in several other versions besides Flateyjarbok, and it is correspondingly difficult to arrive at a sense of the historical attitudes that may or may not have been displayed in the lost earliest version(s). The surviving versions show little historiographical awareness. Despite the availability of skaldic verse as a source, the stories about the Jomsvikings are treated as fiction rather than as a narratio rei gestae, with the verses generally a part of the fiction. But if some of AM 510 4to does reflect a lost earlier version of the saga, then it may not always have been so. The fictionalizing of the saga of the Jomsvikings seems to have taken place during the course of literary transmission, particularly with the increasing emphasis on the colourful exploits of the Jomsvikings themselves. Despite the availability of alternative sources of information concentrating on the Hladajarls, the extant versions of the saga generally prefer to ignore them.
With Fareyinga saga, there is little, if any, evidence that there ever was an earlier, more 'historical' version of the saga. Indeed, there is precious little evidence that even the main events of the saga ever happened, so it can hardly be called a narratio rei gestae. The saga was clearly written, even in its earliest version, primarily for enjoyment and entertainment, or for whatever other reasons literary texts are composed. But even Fareyinga saga can be shown to have made use of narrative techniques that arose out of the attempt at analytical historical writing. Its interest in visualizing key scenes is akin to the emphasis on eyewitness accounts which led historical writers to value the testimony of skaldic poetry, for instance. Ulfr/porkell's narrative shows a sophisticated awareness of how a teller can manipulate the tale, an awareness that arose in a tradition used to comparing different versions of tales. In Fareyinga saga, we can see how history is 'fictionalized'.
If these three sagas illustrate Icelandic approaches to historical narrative, we can also see in them the future development of Icelandic narrative art without a historical purpose. The mediaeval Icelanders did not succeed in developing a 'critical study of the remote past' that would satisfy a modern historian, although they do seem at least to have made the attempt, and to have been on the right lines. But the historical impulse was sidetracked into narrative for its own sake, giving us the fictions of the Islendingasogur (and even to some extent Snorri Sturluson).(49)
Critics may continue to discuss whether the 'political sagas' have thematic concerns in common and whether they therefore form a sub-genre. But perhaps more importantly, they provide a neat demonstration of the development of Icelandic fictional narrative, along with a sense of the road not taken.(50)
JUDITH JESCH Dept of English Studies, University of Nottingham
(1)Isidori Hispalensis Episcopi Etymologiarum sive Originum, ed. by W. M. Lindsay, 2 vols., Scriptorum Classicorum Bibliotheca Oxoniensis (Oxford, 1911), I, xli.
(2)Translation in Beryl Smalley, Historians in the Middle Ages (London, 1974), p. 24. Isidore's Latin is: 'Apud veteres enim nemo conscribebat historiam, nisi is qui interfuisset, et ea quae conscribenda essent vidisset.'
(3)Smalley, Historians, p. 24.
(4)Ibid., p. 25.
(5)See Jonas Kristjansson, Eddas and Sagas: Iceland's Medieval Literature (Reykjavik, 1988), pp. 120-3.
(6)See Diana Whaley, Heimskringla: an Introduction (London, 1991), pp. 115-19; Sverre Bagge, Society and Politics in Snorri Sturluson's 'Heimskringla' (Berkeley, Calif., 1991), p. 10. Bagge represents a newish trend which attempts to elide the difference between mediaeval and modern historians, not by restoring the credibility of the mediaeval ones, but by emphasizing that even the modern 'historian's reconstruction of the past resembles a work of fiction to a considerable degree' (p. 58), thus shifting the focus to the study of Heimskringla as a reflection of its time rather than of the past it claims to describe.
(7)I like to use this word in the broad (dare I say interdisciplinary?) sense discussed in Den filologiske vitenskap, ed. by Odd Einar Haugen and Einar Thomassen (Oslo, 1990).
(8)For brief notes on these sagas, see Kristjansson, Eddas and Sagas, pp. 164-6.
(9)Melissa Berman, 'The political sagas', Scandinavian Studies, LVII (1985), 113-29 (pp. 113, 125-6).
(10)Peter Foote, 'Observations on Orkneyinga saga', in St Magnus Cathedral and Orkney's Twelfth-Century Renaissance, ed. by Barbara E. Crawford (Aberdeen, 1988), pp. 192-207 (pp. 192-5).
(11)Ibid., p. 195.
(12)Orkneyinga saga, ed. by Sigurdur Nordal (Copenhagen, 1913-16), p. 189. This edition is cited because of its full variant apparatus, but I have normalized the orthography. All translations from Old Norse are my own.
(13)Ibid., pp. 188-9.
(14)See my 'Narrating Orkneyinga saga', Scandinavian Studies, LXIV (1992), 336-55, for a more detailed study of the narrative devices (primarily that which narratologists call 'focalization') used to achieve this dual perspective.
(15)Orkneyinga saga, ed. Nordal, pp. 190, 193.
(16)Ibid., pp. 57, 260.
(17)Ibid., pp. 108, 111.
(18)See also on this Sigurdur Nordal, Snorri Sturluson (Reykjavik, 1973), p. 139; Bjarni Einarsson, 'On the role of verse in saga-literature', Medieval Scandinavia, VII (1974), 118-25 (p. 119).
(19)Orkneyinga saga, ed. Nordal, p. 218.
(20)Ibid., p. 251.
(21)Ibid., pp. 61, 71. Other examples of the first-person pronoun are at pp. 46, 48, 51, 70, 86.
(22)Ibid., pp. 46-7.
(23)Ibid., p. 65 n.
(24)See Bjarne Fidjestol, 'Sogekvaede', in Deutsch-nordische Begegnungen: 9. Arbeitstagung der Skandinavisten des deutschen Sprachgebiets 1989 in Svendborg, ed. by Kurt Braunmuller and Mogens Brondsted (Odense, 1991), pp. 57-76 (p. 58).
(25)On intellectual games in twelfth-century Orkney, see Paul Bibire, '"Few know an earl in fishing-clothes"', in Essays in Shetland History: Heidursrit to T. M. Y. Manson, ed. by Barbara E. Crawford (Lerwick, 1984), pp. 82-98 (p. 97); and Ole Bruhn's forthcoming article, 'Earl Rognvald and the rise of saga literature', in The Viking Age in Caithness, Orkney and the North Atlantic, ed. by Colleen E. Batey et al. (Edinburgh, 1993).
(26)Jomsvikinga saga, ed. by Carl af Petersens (Lund, 1879), p. 69. This edition is cited in normalized orthography throughout.
(27)Ibid., p. 81.
(28)In Heimskringla, quite logically, only the 1 1/2 stanzas mentioning Hakon's discarding of his armour are quoted: Olafs saga Tryggvasonar, in Snorri Sturluson: Heimskringla, ed. by Bjarni Adalbjarnarson, 3 vols., Islenzk Fornrit, 26-8 (Reykjavik, 1941-51), I, 281-2.
(29)Again, Heimskringla, I, 286, quotes only the stanza that mentions the number of ships.
(30)Jomsvikinga saga, ed. af Petersens, p. 82.
(31)See Jomsvikinga saga, ed. by Olafur Halldorsson (Reykjavik, 1969), pp. 11-12, 14-15.
(32)Jomsvikinga saga, ed. af Petersens, p. 69.
(33)The Saga of the Jomsvikings, ed. by N. F. Blake (London, 1962), p. 32.
(34)Jomsvikinga saga, ed. Halldorsson, p. 176.
(35)Jomsvikinga saga, ed. af Petersens, p. 81.
(36)Jomsvikinga saga, ed. Halldorsson, p. 182; The Saga of the Jomsvikings, ed. Blake, p. 35.
(37)Jomsvikinga saga, ed. af Petersens, p. 85. Although the stanza is obscure, the relevant statement that 'fyrir bord ... at ganga ... Bua kendu' is clear enough.
(38)Although AM 510 4to refers to a 'flokkr' which 'Tindr orti um Jomsvikinga,' the frequent references to the jarl in the verses make it clear that the stanzas quoted are from one poem about Jarl Hakon. (On Tindr's and pordr's poems, see Bjarne Fidjestol, Det norrone fyrstediktet (Ovre Ervik, 1982), pp. 24, 102, 115-17).
(39)The Saga of the Jomsvikings, ed. Blake, p. vii.
(40)Faereyinga saga, ed. by Olafur Halldorsson (Reykjavik, 1987), quoted throughout in normalized orthography.
(41)Ibid., p. clxiv.
(42)Peter Foote, 'On the saga of the Faroe islanders', in Aurvandilsta: Norse Studies, ed. by Michael Barnes et al. (Odense, 1984), pp. 165-87 (p. 177).
(43)Gerard Genette, Narrative Discourse (Ithaca, NY, 1980), p. 186.
(44)Faereyinga saga, ed. Halldorsson, pp. 11-12.
(45)Ibid., pp. 35-6.
(46)Ibid., p. 35.
(47)When Faereyinga saga was excerpted into the version of the saga of Olafr Tryggvason in Copenhagen, Arnamagnaeanske Institut, MS 62 fol., this narrative device was abandoned and Ulfr begins by saying 'ek heiti rettu nafni porkell' ('my real name is porkell'): see Faereyinga saga, ed. Halldorsson, p. 32 n.
(48)There are only two in Flateyjarbok: see Orkneyinga saga, 14/18 and 179/12; but there are six in the other manuscripts: 43/7-8, 45/20, 55/4, 139/5-6, 279/3-4, 279/5-6.
(49)For a similar conclusion reached by somewhat different means, see Sverrir Tomasson, '"Soguljod, skrok, had": Snorri Sturluson's attitude to poetry', in Ur Dolum til Dala: Gudbrandur Vigfusson Centenary Essays, ed. by Rory McTurk and Andrew Wawn, Leeds Texts and Monographs, n.s. 11 (Leeds, 1989), pp. 317-27. This paper has also been published in an Icelandic version as 'Soguljod - skrok - had: Vidhorf Snorra Sturlusonar til kvedskapar', Skaldskaparmal, I (1990), 255-63.
(50)For a demonstration of the road taken in the transformation of firsthand reports into an 'episk sagaform', see Preben Meulengracht Sorensen, 'Historiefortaelleren Sturla pordarson', in Sturlustefna: Radstefna haldin a sjo alda artid Sturlu pordarsonar, ed. by Gudrun Asa Grimsdottir and Jonas Kristjansson (Reykjavik, 1988), pp. 112-26.