The Icelandic annals as historical sources

This art icle was downloaded by: [ Universit et sbibliot eket i Bergen] On: 11 July 2014, At : 05: 25 Publisher: Rout ledge I nform a Lt d Regist ered in England and Wales Regist ered Num ber: 1072954 Regist ered office: Mort im er House, 37- 41 Mort im er St reet , London W1T 3JH, UK Scandinavian Journal of History Publicat ion det ails, including inst ruct ions for aut hors and subscript ion informat ion: ht t p:/ / www.t andfonline.com/ loi/ shis20 The icelandic annals as historical sources Eldbj ørg Haug a a Ospelia 1, Fet sund, 1900, Norway Published online: 23 Jun 2008. To cite this article: Eldbj ørg Haug (1997) The icelandic annals as hist orical sources, Scandinavian Journal of Hist ory, 22:4, 263-274, DOI: 10.1080/ 03468759708579356 To link to this article: ht t p:/ / dx.doi.org/ 10.1080/ 03468759708579356 PLEASE SCROLL DOWN FOR ARTI CLE Taylor & Francis m akes every effort t o ensure t he accuracy of all t he inform at ion ( t he “ Cont ent ” ) cont ained in t he publicat ions on our plat form . However, Taylor & Francis, our agent s, and our licensors m ake no represent at ions or warrant ies what soever as t o t he accuracy, com plet eness, or suit abilit y for any purpose of t he Cont ent . Any opinions and views expressed in t his publicat ion are t he opinions and views of t he aut hors, and are not t he views of or endorsed by Taylor & Francis. The accuracy of t he Cont ent should not be relied upon and should be independent ly verified wit h prim ary sources of inform at ion. Taylor and Francis shall not be liable for any losses, act ions, claim s, proceedings, dem ands, cost s, expenses, dam ages, and ot her liabilit ies what soever or howsoever caused arising direct ly or indirect ly in connect ion wit h, in relat ion t o or arising out of t he use of t he Cont ent . This art icle m ay be used for research, t eaching, and privat e st udy purposes. Any subst ant ial or syst em at ic reproduct ion, redist ribut ion, reselling, loan, sub- licensing, syst em at ic supply, or dist ribut ion in any form t o anyone is expressly forbidden. Term s & Condit ions of access and use can be found at ht t p: / / www.t andfonline.com / page/ t erm s- and- condit ions The Icelandic Annals as Historical Sources Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 Eldbjørg Haug Up until around 1280 there was an abundance of writing on Norwegian history. The historiography consisted mainly of sagas, most of them written by Icelanders. The last saga we know of is about Magnus Lagabøte, but unfortunately only a fragment of it has survived. From the 14th century onwards we find no writing of this kind, but there are narrative sources relating to the history of Norway in the Icelandic Annals. Owing to the lack of sources, the 14th century is a rather obscure period in Norwegian history. The Icelandic Annals therefore hold a central position as sources for the political history of that century. It has been assumed that the real annalistics on Iceland started around 1300. Issues such as the coronation of Magnus Eriksson in Stockholm in 1336, the Black Death in Norway and the church policy of Queen Margaret all have the Icelandic annals as important sources (Haug 1996). Hallvard Magerøy used them to determine the communications between Norway and Iceland (Magerøy 1993). Some historians evidently regard the annals as such a comprehensive source category that they insist that what is not mentioned in them, has never occurred. Edvard Bull has maintained that the annals are the main sources to medieval history, and Knut Dørum took the same position in a debate with me recently (Haug 1995b). I began to question the Icelandic Annals when I needed some of their notices to throw light on Magnus Eriksson's and Queen Margaret's church policy (Haug 1996). At the outset I shared the common opinion that the Annals were contemporary to the events they described. But I soon discovered that an event was not always annotated in the same year in different annals. Some events were even written about twice in the same annal, but often under different years. It is difficult to decide which notice to rely on when there are only two sources - one that states a and one that states b. It is thus necessary to ascertain the years of the events according to the yearbook. It is also necessary to question to what extent the notices were contemporary. Could it be that the annals had not been written year by year in an ongoing process after all? Equally important - did there exist other sources to events annotated in the notices which could be studied as exhibits to the annals? And, if we regard the yearbooks from a holistic point of view, when did they take Eldbjørg Haug, born 1947, cand.philol, PhD. is an advisor in the Ministry of Foreign Affairs. Important published works include Bygdehistorie for Fet fra de eldste tider til ca. 1800 (1980); and Provincia Nidrosiensis i dronning Margretes unions- og maktpolitikk (with an English summary) (1996). Address: Ospelia 1, 1900 Fetsund, Norway. Stand. J. History 22 264 Eldbjørg Haug the form in which they have been left to us? Would it be possible to date them in relation to each other? To answer these questions, I first explain something of the research situation. Then I try to demonstrate source criticism applied to the Icelandic annals. As an example on how to evaluate a yearbook, I use notices from the Lögmann's annals. Finally, the source value of the annals is assessed. Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 1. The characteristics of annals Annals were written by the Romans, but the source class is held to be medieval, starting in the 8th century. It is believed that medieval annalistics started in Britain as notices in the Easter tables. Every monastery had to keep such tables because of the complicated calculation of Easter. It was considered heretical to celebrate Easter at the wrong time. Since some of these yearbooks start with the life of Jesus or the consulate of Caesar, we know that at least some of the events in the annals are neither contemporary to the writer nor immediate, first-hand sources. In the Icelandic Annals we thus find some letters in the margin, which are taken from the chronological system of Bede and tell the date of Easter Sunday for each year. There are many other similarities with English annals as well, and there is reason to believe that the Icelandic annalists started by copying English annals. It is a general opinion that the annals are narratives of a more fragmented and incoherent character than a chronicle or a saga. Being records, they are to a certain extent the opposite of literary sources. The Norwegian historian Ottar Dahl holds this to be significant. "The literary form of a narrative will influence it, and further its prejudice", he says. "Fragmented records of annals will, on the other hand, leave the data without cosmetics." The contents of the annals are sparse. We often find that the description of an event is not written out in complete sentences. In this respect the annals differ both from history writing like the sagas, and from chronicles. As the name indicates, they are written year by year - often starting with the Incarnation. The Icelandic Annals are also considered to be written in an on-going process, although critical examinations of other Nordic annalistics have indicated that all their notices as they appear in the manuscripts are written at a fixed period, not much later than their youngest notices. All the oldest Icelandic annals are based on older yearbooks, but it has been assumed that the last notices in each of them were independent continuations and contemporary information, written down consecutively year by year. One could imagine, then, that each Easter Eve the annalist would go to his manuscript, pontificate on what major events had occurred since last Easter, dip his quill in the ink and write the events down. Then he would put the manuscript away until the following Easter. 1 2 3 Cf. Beda. Translated from Dahl 1967: 45. Bolin 1931; Christiansen 1974. Scand. J. Histoņ 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 The Icelandic Annals as Historical Sources 265 The annals are often written by several generations of different and mostly anonymous authors. The Fragmented Annals have ten different writers, and the Lögmann's Annals have five. Their emphasis is on events, and they give contemporary notices of historical facts which were important to the annalists. Doubts about the annals being on-the-spot accounts led to the next popular belief: As the annalist could not have been present at all the occasions he wrote about, he must have based his information on oral information and perhaps rumour. Carried out to their extreme, the annals may be perceived as the newspapers of the Middle Ages. The place of origin adds a character to the sources, which could be significant in explaining certain characteristics and peculiarities. Annals were mostly written in monasteries and chapter houses, and were intended for internal use. But they were often copied in one place, and then continued somewhere else. The copyist might also add new information to the years already described. It is therefore difficult to create a stemma, as well as to determine each yearbook's place of origin. The Icelandic Annals were mostly written by clerics in monasteries or at bishop Sees, and events which are important from an ecclesiastical point of view are given a relatively large amount of space. On this background the study of annals belongs to the study of medieval historiography. The history writing from the Middle Ages and well up to the 17 th century was a teleological history writing. Building upon Hebrew antecedents, Christianity introduced a new linear notion of time into the Greco-Roman world. The Judeo-Christian time line literally began at one moment and would end at another, and it revealed God's purposes. In the Christian schema, the turningpoints of sacred history- the Creation, the Incarnation, the life and death of Jesus, and the prospect of the Last Judgement - set the framework for all historical time. Sacred history gave all of time its meaning. The Christian time schema occupied scholars right into the 17th century. Before turning to the next point, it should be noted that there are annals on Nordic ground other than the Icelandic yearbooks. The Danish chronicle of Zealand is in the form of a yearbook (Christiansen 1974). The Vadstena Diary is another example (Gejrot 1988). Lists of kings also belong to this category (Bolin 1931). Although it has been assumed that annalistics was something the Norwegian clergy did not occupy themselves with, some sources of this kind from Norway are issued in Volume IV of The Old Norwegian Laws. 2. Research The first Nordic historian to introduce modern source criticism based on philology was Gustav Storm. His publication of sources to Norwegian medieval history, with critical commentaries, is impressive. Among his publications we find the already mentioned Volume IV of the Old Norwegian Laws {Norges gamle Love), the ten oldest Icelandic Annals (Storm 1888), and Monumenta Histórica Norwegica (Storm 1880). He was also the first to issue the Chronicle of Hamar, which is possibly a remnant of annalistics at the medieval bishop See of Hamar (Storm 1895, 1890a, 1890b). His Introduction to the Icelandic Annals is of such great value that it may have Scand. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 266 Eldbjørg Haug prevented other historians from taking up the subject. He did not himself, however, consider the theme as exhaustive, as can be seen in his encouragement to other historians to continue the research on the Icelandic Annals (Storm 1888). The next important historian in the critical school was the Danish historian Krislian Erslev, who introduced the dichotomy of narratives and remnants (Erslev 1926/1987). He was succeeded by the Swedish historian Lauritz Wdbull, with his critical examination of the history of the Nordic countries around the first millennium. His dissertation was issued in 1911 (Weibull 1911), and in 1913 Halvdan Koht followed up with his radical criticism of the sagas (Koht 1913). In treating the dichotomy, narrative — remnant, Koht's main point was that the sagas reflect the time in which they were issued. They should therefore be treated as remnants of their issuing situation, rather than as narratives about the events they described. In 1931 Sture Bolin, known among historians and archaeologists for his "History of the Coin", presented a large critical treatise on the oldest Swedish annals (Bolin 1931, 1962). He showed that these annals from the very beginning were learned compilations. When an archaeologist finds an old treasure of coins in the ground, he knows that it could not have been laid there earlier than the year of issue of the earliest coin in the treasure. Bolin found that the Swedish yearbooks were constructed in a similar way. From the beginning they were the result of learned compilations, copied into the annals not much later than their earliest notices, instead of being written into them consecutively, event by event. The annalists were historians - they used the sources and evaluated them. Only one thing was missing - a systematic criticism. After Storm presented his Introduction no critical treatises on the Icelandic annals, similar to Bolin's on the Swedish ones, have appeared. It is true that Sven Axelson held his doctoral dissertation on Sweden in foreign annalistics with a particular view to the Icelandic Annals (Axelson 1955). His intention was, however, to trace the origin of the information on Swedish affairs, not to clarify the origin of each of the yearbooks in the way Bolin had done. If he had discussed that issue, his large body of work would have had a greater value. Tage E. Christiansen emphasized this point in his study of the chronicle of Zealand. The question of the origin situation is crucial to an understanding of the character of the source: The annals are remnants of the past - all sources are remnants. But are they also narratives? Are we working with a contemporary or non-contemporary source? Are our sources first-hand records? 3. What is criticism of sources? If we are going to make use of the narratives of the annals, we have to be certain that what they tell us is true. The information must be reliable. Our point of departure must be to determine die manuscript: Is it a copy of an older text, or is it the work of a historian? Annals should be unfolded from their final entry. The notices should be regarded from the final year's point of view, as an expression of how a compiler would formulate them from the perspective of hindsight. Do we know who wrote the annals? For whom were they written? Why 4 Christiansen 1974. Scand. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 The Icelandic Annals as Historical Sources 267 were they written? This critical position may give us a clue to understanding the unintelligible, and may explain errors and peculiarities that we were unable to explain before. There is also the question of how to arrive at a date for the manuscript. We should not take the year of entry of the last notice as the terminus ante quern at face value. The first thing to do is to read the source in its entirety and look for discrepancies in the notices. The mistakes may give us a clue. Even if the Skalaholt annalist writes for the year 1349 that Pope John XXIII was elected, it is worth while asking whether the notice was written after 1410, when the Council of Pisa elected a pope of that name. It is also reasonable to assume that no one would record the death of a person before he or she was actually dead. If it is possible to prove that the annals were not annotated consecutively year by year, this may explain why the chronology varies from one piece of work to another. Even events which would presumably have occurred in the immediate surroundings of the writer are dated wrongly and described falsely. This is a consequence of the fact that, contrary to what is believed, the annals were not written consecutively year by year. Secondly, we should decide whether our source is biased. Some notices could be seen in this light. When the annalist suddenly presents the cause of an event, we should be on the alert. This is not typical of annalistics. Another form of prejudice could be in the character given to some persons, with their virtues or vices specified. For example, the characterization of Archbishop Nicholas Rusare in the Lögmann's annals: He did not perform any of a bishop's tasks before he died. Or the characterization of Queen Margaret in the Vadstena Diary: She had a very happy life as to mundane matters. The bias of a source must be evaluated according to the time in which it was written, not according to what it describes. The bias of the source is thus important in itself, and may give us a clue as to the time when the evaluation first appeared. It is therefore necessary to form an opinion on the bias of the annals and ask for which historical situation it is typical. Besides mistakes in the manuscripts themselves, we find, of course, information from other manuscripts such as the sagas, which we are able to trace in the annals. This, then, is our third task: to look for information drawn from other written sources. Are there any elements from other manuscripts which can be found in the annals? If the answer is yes, this may disappoint us regarding the establishing of one type of historical fact. It will, however, give us a source to another historical issue. 4. The "Lögmann's" annals I have studied these questions more closely for five Icelandic Annals, which are presumed to have been written in the second half of the 14th century.5 One of these sources is the so-called Lögmann's annals (the Yearbook of the King's local justiciar). Storm assumed that these annals were older than the socalled lost annals from the Northern Coast of Iceland, on which Gottskalk's annals 5 Haug 1996. Scand. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 268 Eldbjørg Haug are based upon, and also older than the Flateyarbók. This assessment depends on how our concepts are clarified. The oldest manuscript of the Lögmann's annals dates from the 14th century, and is written by five different annalists or "hands". This manuscript is considered to be the original one. The yearbook was continued after 1393, as can be seen from a copy, and went on until 1430, but the original last part of it is lost. From this, however, it can be concluded that in its complete form the yearbook was younger than the Flatøy annals, as well as the lost annals from the North Coast. The first author of the annals is the priest and official principal of Hólar, Einar Havlidesson. Thus the annals must have been written close to the Northern bishop see on Iceland. Einar Havlidesson's edition stops in 1362. Einar Havlidesson also wrote Bishop Laurentius's saga, probably after he had finished the annals, as they and other annals are referred to and quoted in several places in the saga. His narrative of the Black Death has been used both by authors of historical novels and by historians as a contemporary source on how the disease spread. I would, however, be reluctant to use this narrative as a source before checking what other contemporary authors had written about the plague. In any case, Einar Havlidesson's narrative should not be regarded as older than the description given in the annals from Skalaholt for the year 1348. When I looked for errors in Einar Havlidesson's annals, I noted the description of the death of the Swedish king Erik Magnusson in 1359. The annalist writes that Erik Magnusson died from poisoning. This is considered to be false. Its provenance is Hbellus de Magno Erici Regis, which was written while Magnus Eriksson was the prisoner of King Albrecht of Sweden, i.e. between 1365 and 1371. The tendency of the Libel is to justify the Swedish aristocracy for supporting Albrecht of Mecklenburg as Swedish king while dethroning Magnus Eriksson. The anonymous author used material from St. Birgitta's revelations, but the poisoning of King Erik is an original accusation. The libel also accused Queen Blanche of being a poisoner. The annals did not include this accusation. Einar Havlidesson must have sensed the prejudice of his exhibit and therefore omitted any accusation against the king's mother. The notice of King Erik's poisoning indicates that Einar Havlideson wrote this after the Libel had been written between 1365 and 1370. He seems to have written the annals in one process, and not consecutively, year by year. This indicates 1365 as terminus post quern, and Einar's death in September 1393 as terminus ante quern. The next four writers of the Lögmann's annals are anonymous. The 2nd hand started with notices for 1362, and continued until 1380. One piece of information could give an indication to the provenance. For the year 1364 the annalist states that King Erik Magnusson and Lady Blanche were betrayed at the court. This notice may also have been constructed from the Libel, although not as precisely as the other one. One might therefore suspect that the 2nd hand was familiar with the serious accusations of the libel, but had not read them himself. He knew that Queen Blanche was dead, and also that her son was dead. He had heard about the wedding in Copenhagen between Margaret, King Valdemar's daughter, and King Håkon VI of Norway, but did not have any exact 6 Kraft 1927; Andersson 1928; Christiansen 1974. Scand. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 The Icelandic Annals as Historical Sources 269 information. Tage E. Christiansen has pointed to the Libel as the source of The Chronicle of Zeeland concerning its narrative of the wedding in Copenhagen and Prince Christopher's death at that time. The Icelandic 2nd hand might have mixed these rumours without having read any of the manuscripts himself. He knew that Queen Blanche and her son, King Erik, were dead, but was not capable of evaluating his information, either chronologically or factually. My hypothesis is that the 2nd hand started his work some time after 1380, which is the last year of his notices. The notices of the 3rd hand for the years 1379 and 1380 do not give any clues to indicate provenance. The 4th hand is of greater interest. This "hand" has written notices for the years 1380-1383 and 1388-1392, many of them relating to the church policy of Queen Margaret. Here is the narrative of the provision of the Dane Nicholas Rusare to the Arch See of Nidaros, contrary to the election of the chapter. Some errors give us a clue to the date of authorship. The canonization of St. Birgitta is mentioned for the year 1390, whereas she was actually canonized in 1392, so the notice cannot be older than that. For 1391 it mentions the deaths of some well-known Norwegian nobles: Håkon Johnson, Håkon Stumpe and Gaute Eiriksson. Their deaths cannot have been written into the annals before these men had actually passed away. Gaute Eiriksson lived longest of all. He was alive in 1412, but mentioned as dead in 1413, his death thus being the terminus post quern for the writings of the annalist. And since we have already established that these annals were continued after 1392, a possible hypothesis would be that the 3rd, 4th and 5th "hands" all belong to that part of the yearbook which is now known only as a copy, and is often referred to as the New Annals. All the errors in the chronology indicate that the notices must have been written down a relatively long time after the events had occurred. The 5th hand in the Lögmann's annals fills in the gaps from 1384 to 1387, 1388 and 1389. It is this "hand" who has written the famous aftermath of Archbishop Nicholas Rusare: He did not perform any of a bishop's duties while being in Norway, therefore he did not perform any consecrations and did not confirm any children. The 5th annalist also made a mistake in his chronology. From the fact that he wrote his notices in the gaps left by the 4th hand, we conclude that they were written after that writer had stopped. We have already demonstrated that 1413 was terminus post quern for number 4. But since the manuscript ends in 1393, we cannot tell from palaeographical indications exactly when annalist No. 4 put his quill down. We may, however, get somewhat further by studying the annals' place of origin. Einar Havlidesson was a priest on the Northern Coast of Iceland, close to the bishop See of Hólar. According to Storm, his successor should be traced in the same circles. But from about 1380, or with the 3rd hand, the manuscript must have been continued at Skalaholt. Its continuation, only known through our copy, has so much information on Bishop Vilkin of Skalaholt, that someone close to him probably wrote those notices. My hypothesis is that this annalist is identical with the 3rd hand. It is difficult to say exactly when he stopped — he may have continued for some years after Vilkin's death. But from 1403 and up to 1420 another ecclesiastical character dominates the text - the next bishop at Skalaholt, Arni Olafsson. My next hypothesis, therefore, is that a person close to this bishop Scand. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 270 Eldbjørg Haug continued the annalistics, and that he is identical with the 4th hand. Since this hand began his writing after 1413,1 am inclined to believe that he started after the death of the bishop in 1420. If my other hypotheses are correct, who is the last annalist, who is identical with the 5th hand? Let us take a look at the dramatic events at Skalaholt after Bishop Arni's death. The vacancy lasted until 1426, when he was succeeded by John Gereksson Lodehat. The new bishop had been dismissed as archbishop of Uppsala in 1421, owing to misconduct. Before becoming archbishop, he had been Chancellor to Erik the Pomeranian. From the narrative of the Icelandic Annals we learn that the infamous archbishop and bishop ended his days in 1433, when rebellious Icelanders drowned him. The rebellion was linked with a proposal to Margaret Vigfusdatter, daughter of the Icelandic "hirdstjore" and sister of Ivar Vigfusson. The suitor was the bishop's son, but Margaret turned down the proposal, and her brother supported her. As a revenge their house was set on fire. Ivar perished, while Margaret survived. She took an oath that she would only marry a man who would avenge her brother. Elements of this dramatic narrative echo throughout the Icelandic family sagas. There is little else to read about John Gereksson in the last parts of the new annals. However, they do not present a positive image of conditions on Iceland. The church at Skalaholt is reported to be in a bad state, as it has been without a bishop since the death of Arni Olafsson. The official principal is old and almost blind. The new bishop arrives from England, bringing with him a great many Danes who are of no use. Only two priests are mentioned by name, probably because they collected tithes for the bishop in great haste, to take to England to be sold. These are not events that one would expect to be written at the bishop See, in close proximity to the new bishop. I would rather suspect that the annalist belonged to the same circle that drowned the bishop in July 1433. If we assume that the compilation of the last part of the annals took place at the bishop see after the drowning of the bishop, we arrive at the time of the Council of Basle, where the drowning ofjohn Gereksson was treated. On a more homely base, the Dala rebellion raged from 1434 to 1436. In Norway, Amund Sigurdsson Bolt rebelled in 1436. And on Iceland, the English behaved like pirates towards the inhabitants. If we see this political climate in connection with the 5th hand of the annals, new light is thrown on the notices from the 1380s. The Norwegians and Icelanders were in conflict with Danish bailiffs and English merchants and pirates. Since Nicholas Rusare was also a foreigner, actually a German, although according to the annals a Dane, this could be used against him in the actual situation of conflict with foreigners. The annalist's rather biased summing up of the archbishop office of Nicholas Rusare probably reflects the political situation in Iceland at the time that it was written. My hypothesis concerning the last part of the yearbook, which is only preserved as a copy, cannot be considered as proven. What is established, though, is that the yearbook was not finished until the middle of the 1430s. 5. The Icelandic Annals as historical sources The examples from the Lögmann's annals demonstrated that they were secondary Scaad. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 The Icelandic Annals as Historical Sources 271 sources, not contemporary ones, and to a certain extent based on rumours of doubtful origin. Moreover, the origin of their information is confined to Bergen and the Western Coast of Norway. The information from Oslo and the Eastern part of Norway is very limited. We cannot consider the annals to be a source of such richness that every significant event would be noted in them. As already mentioned, it has been maintained that the Icelandic Annals are the most important sources to Norwegian political history in the 14th century. Furthermore, it has been maintained that since the Norwegian coronation of Erik the Pomeranian in 1392 was not mentioned in these sources, the coronation did not take place. In other words, the historian introduces an argument from silence. This leads to the last issue to be discussed: What is the source value of the Icelandic Annals? Are they so-called "complete"? Are they of such a nature that we may take the liberty of using silence as an argument? An argument from silence is a negative statement of existence. A narrative is being rejected because no other witnesses tell about the event. It is claimed that this phenomenon was of such a character that it had to leave traces, and that these traces would consequently have been conserved and known. Concerning the coronation of Erik the Pomeranian and the use of argumentum e silentio in that connection, my conclusion is that, since we have a diploma that tells about his Norwegian coronation, we cannot use the silence of the Icelandic Annals to prove that it did not occur. Although some of the narratives of the Icelandic Annals are doubtful or false, I consider the annals to be an interesting source category. Furthermore, some of the information found there must have had a written exhibit — missives, diplomas, as well as other material which might be considered as annals themselves. Obituaries may have been one of them. When a rich person died, he or she often donated the means to a church or a church institution to sing masses for them. Sometimes the masses were supposed to be sung for all eternity at every anniversary of the death or funeral. It was necessary to keep records of such masses in the actual church institution. The records could be kept in separate books, in calendars, anniversaries, missals or in annals. When a notice in the annals records the day of a person's death, we may rest assured that the day is correct. The year, however may be more doubtful, as demonstrated by the untimely death of Gaute Eiriksson. In the Middle Ages all trade from the Norwegian provinces overseas went via Bergen. Bergen was thus the "capital" of Iceland after it had ceased to be the capital of Norway. Most of the communication with Norway went to or through Bergen. So the oral and some of the written exhibits for the annals must have been gathered here. Another important location in Norway concerning the gathering of information for the annalists was Nidaros. It was the duty of every suffragan in the church province to visit the Arch See once a year, either personally or by deputy. The oath of obedience from the bishops included a promise of such visits.8 The obligation to visit the archbishop was maintained, at least up to the last years of the 15th century, but the bishops from Iceland did not travel personally every year. 7 8 Haug 1995. DN VIII 74; Hamre, col. 513. Scand. J. History 22 Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 272 Eldbjørg Haug There are, however, several notices in the annals about the bishops leaving the island with their followers. Their destination was Norway, their mission in most cases to visit the archbishop. It is likely that these visits resulted in both oral and written information about events which could later be written into the annals. Although we cannot always trust the information in the annals, the fact that an event is written down at least gives us a source concerning the diffusion of information. Obituaries in the annals, and also information about bishop successions, are often of such a character that even when we lack other sources, they can be assumed to be correct. This is of course most likely for events on Iceland, but events recorded for Bergen or Nidaros may also be true. The problem is usually the dating, particularly of the year, as the annals cannot be considered to be contemporary sources in a strict sense. Before I finish, let me point to another area of interest - the history of mentality. Events which we would not consider significant nowadays, the annalists found to be so important that they took notes about them. Often it is impossible to decide whether or not the event is true. One of Queen Margaret's dreams is recounted in the Fragmented Annals. Although it is not significant as such, it gives an interesting glimpse of a way of thinking and imagining. ° From Heimskringla we know the dreams of Queen Ragnhild. King Sverre used his dreams consciously in his political propaganda. Dreams have always had a great symbolic value, not only in the Nordic Middle Ages. The annalistic notice of Queen Margaret's dream was probably meant to throw light on her life and work. However, it tells us more about the writer and his world of thoughts than about Queen Margaret. 6. Conclusion I have maintained that the Icelandic annals should be studied as historical remnants before we use them as narratives. We date the annals, we establish the place in which they were written, we identify the anonymous writers; in short — we establish the situation of origin. I have demonstrated that one of the annals is not written consecutively. All the Icelandic annals I have studied so far, give the same conclusion. And I have proved the annals to be an interesting source category, although more interesting as remnants than as narratives. My final remarks are about the benefit of studying the annals. When studying them as remnants we perform a historiographical research, or a literary criticism. Since the event's imprints cannot be properly interpreted unless we first put them back into the cultural system which they influenced at the time, it is necessary to bear in mind everything that is known about this culture in order to evaluate the accounts that have come down to us. The study of the annals is a hermeneutical process of knowledge, which gradually reveals to us other aspects of medieval culture and society which were formerly hidden. The annals are interesting sources in the study of the medieval mentality. The characteristics of each of them tell us what the anonymous annalist considered 9 10 Cf. Otto 1933. Cf. Cormack 1996: 185-209; Hamre 1996: 505. Scand. J. History 22 The Icelandic Annals as Historical Sources 273 important. In the words of Georges Duby, we seek to observe "the impact that the imaginary and oblivion have on information, the insidious penetration of the marvellous, of the legendary, and, in the course of a sequence of commemorations, the fate of a memory in the midst of a changing set of mental representations." Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 Bibliography Andersson, I. (1928). Källstudier till Sveriges historia 1230-1436. Inhemska berättande källor jämte Libellus Magnipolensis. Lund. Axelson, S. (1955). Sverige i utländsk analistik 900-1400 med särskild hänsyn til de isländska annalerna. Uppsala. Beda Venerabilis (1990). [Historia ecchsiastica gentis Anglorum. Norwegian] Anglernes kirkes historie, translated by E. Schjøth. Aschehoug, Oslo. Bolin, S. (1931). "Om Nordens äldsta historieforskning. Studier över dess metodik och källvärde", in Lund Universitets årsskrift N. F. Avd. 1, 27:3. Lund. Bolin, S. (1962). Ur penningens historia. Aldus/Bonnier, Stockholm. Bull, Edv. (1920). Annaler. Aschehougs Konversasjonsleksikon, vol. 1. Christiansen, T. E. (1974). "Yngre Sjællandske Krønikes sidste år", Scandia, Lund. Cormack, M. (1996). "Visions, Demons and Gender in the Sagas of Icelandic Saints", Collegium Medievak, Vol. 7:185-209. Diplomatarium Norvegicum (1849-), Vols. I-XXII. Duby, G. (1990). [Le dimanche de Bouvines. English] The Legend of Bouvines, translated by C. Tihanyi. Polity Press, Cambridge. Dahl, O . (1967). Orunntrekk i historieforskningens metodelęre. Universitetsforlaget, Oslo. Erslev, K. (1926/1987). Historisk teknik. Den historiske undersųgelse fremstillet i sine grundlinier. 2nd ed. Copenhagen. Gejrot, C. (1988). Diarium Vadstenense. The Memorial Book of Vadstena Abbey. A Critical Edition with an Introduction by Claes Gejrot. Studia Latina Stockholmiensia, Vol. 33, Stockholm. Hamre, A.-M. (1991). "Til Rødebokens Genesis", (Norsk) Historisk tidsskrift, Vol. 70. H a m r e , L. "Edsformular, Norge", Kulturhistorisk leksikon for nordisk middelalder og vikingtid, Vol. III, col. 513. Hamre, L. (1996). "Eldbjørg Haug: Provincia Nidrosiensis i dronning Margretes unions- og maktpolitikk", (Norsk) Historisk tidsskrift, Vol. 75 (4):497-508. Haug, E. (1995a). "Erik av Pommerns norske kroning", (Norsk) Historisk tidsskrift, Vol. 74,: 1-21. Haug, E. (1995b). "Erik av Pommerns norske kroning nok en gang. Svar til Knut Dørum og Erik Opsahl", (Norsk) Historisk tidsskrift, 74 (4):492-508. H a u g , E. (1996). Provincia Nidrosiensis i dronning Margretes unions- og maktpolitikk, Vol. 13, Skrifter. Trondheim: Historisk institutt, NTNU. Koht, H. (1913). "Sagaernes opfatning av vor gamle historie. Foredrag i Den norske historiske forening 24de november 1913", (Norsk) Historisk tidsskrift, 5.R.IL379-396. Christiania. Kraft, S. (1927). "En pamflett mot Magnus Erikssom i dess idépolitiska och litterära miljö", (Svensk) Historisk tidskrift, Vol. 47. 11 Duby 1990: 8. Scand. J. History 22 274 Eldbjørg Haug Magerøy, H. (1993). Soga om austmenn: nordmenn som siglde til Island og Grųnland i mellomalderen. Det norske videnskaps-akademi, Skrifter 2, Hist.-filos. klasse. Ny serie, Vol. 19. Samlaget, Oslo. Storm, G. (ed.) Norges gamle Love, Vol. IV. Downloaded by [Universitetsbiblioteket i Bergen] at 05:25 11 July 2014 O t t o , A. (1933). Liber Daticus Roskildensis. Anniversarieliste. Roskilde Gavebog og Domkapitlets Storm, G. (1880). "Indledning", Monumenta Historica Norvegię, edited by G. Storm, pp. I-LXIV. Christiania. Storm, G. (1888). "Forord", Islandske Annaler indtil 1578, edited by G. Storm, pp. I LXXXIV. Christiania. Storm, G. (1890a), "Om det gamle Hamar og den gamle 'Hamars Beskrivelse' fra 1553", (Norsk) Historisk tidsskrift, 3.R..I.111-140. Christiania. Storm, G. (1890b), "Om de hamarske Krøniker", Historisk tidsskrift, 3.R.I-.277-308. Christiania. Storm, G. (1895). "Fortale", in Historisk-topografiske Skrifter om Norge og norske Landsdele, forfattede i Norge i det 16de Aarhundrede, edited by G. Storm. Christiania. Weibull, L. (1911). Kritiska undersökningar i Nordens historia omkring år 1000. Gleerup, Lund. Scand. J. History 22