The Lewis Chessmen on a Fantasy Iceland

The Lewis Chessmen on a Fantasy Iceland

By Morten Lilleøren

In 2010, engineer Gudmundur Thorarinsson, helped in some undefined capacity by his public relations specialist Einar Einarsson, published the seldom-visited view that the Lewis pieces were made in Iceland. The revised version of this article, Are the Isle of Lewis Chessmen Icelandic?, as well as his subsequent publications on this topic, may be found at his website. Thorarinsson’s view was addressed in an article by Dylan Loeb McClain of the New York Times, in his somewhat cryptically entitled, Reopening History of Storied Norse Chessmen). Their notion of an Icelandic origin for the Lewis pieces was given substantial promotion by Thorarinsson and Einarsson at those conferences and symposia they could attend. The idiosyncratic idea was not entirely ignored in academic and specialist circles as well.

I entered this discussion because I was shocked at the poor method employed by Thorarinsson in the pursuit of support for his theory. Accordingly, I published a riposte to his initial article in May, 2011. I entitled this first essay of mine, The Lewis Chessmen were never anywhere near Iceland, and published it on Chess Café. Thorarinsson then published a counter to my questions. Continuing to center his argument on a series of circumstantially-derived and – supported points, he went on to state that, “On behalf of Norway, I am thoroughly disappointed…[and moreover] I meant only to participate in literate discussions and studies” (op. cit, and also ChessBase).

I am Norwegian, but am unsure why Thorarinsson seems determined to cast this debate in nationalistic terms. Indeed, all of his mud-slinging and aspersions seem a bit off point. Instead, as I and others have noted repeatedly, despite his claims of “potency”, Thorarinsson’s well-aired suppositions about the Lewis chessmen are based almost entirely on somewhat flaccid argumentation. He sidesteps acknowledging his disregard for a large number of facts with a coy demurral that, as what he terms “forensics” are inadequate to explain the Lewis pieces, all “conclusions about the Lewis chessmen …are, ultimately, speculative in nature.” This simply is not true: facts are available. What I have put together previously and here is entirely based on the historical record writ large, requiring recourse to archaeology and artifact, saga and law. Thus, this expansion of my earlier argument serves to complete the necessary task of undermining the credibility of Gudmundur Thorarinsson’s specious construction, for not only is his work circumstantial, but it also is seriously flawed. To keep this publication short, scholarly reference appurtenances have been omitted, but will be included in a later work of wider scope. Of course, specific questions regarding sources or other topics may be directed to me: morlille@hotmail.com.

My contention is: Connecting the Lewis chessmen with medieval Iceland is at best a romantic notion entirely undermined by testing this thesis against the known and accepted historical facts.

The old Norwegian settlements

The language in the Norse texts (sagas and poetry) existed before Iceland was settled by the Norsemen. This language was Old Norwegian, or, if you wish, Old West Norse. The relationship between Old Norwegian and Icelandic can be compared with English and American today, despite that the discrepancies between Old Norwegian and Icelandic were less than the ones existing between English and American today. This situation of linguistic divergence lasted until the end of the 14th century.

The Hebrides came under Norwegian control circa the early 9th century, and control of the region was consolidated by 1098. According to the texts many Norwegian kings visited or lived there during the centuries, and it logically and factually follows that settlers/colonists also sailed in the opposite direction. Etymological surveys have suggested relative proportions of Norse and Gaelic farm-names for Lewis to be 80% Old Norwegian and 20% Gaelic. A rather new count shows that out of 126 village names in Lewis, 99 are clearly of Old Norwegian descendant, while nine are of unclear origin. These numbers give us an idea of the Norwegian presence in the past. Nowadays Lewis is the last piece of land facing the vast North Atlantic Ocean. Once, however, it was a main junction along the watercourse highway from Norway to Dublin.

Nothing lasts forever. The colonies in the larger islands were lost first: Dublin (which possibly was the first Norwegian city!) and all other Irish areas finally passed from Norse/Norwegian hands in 1171. The last part of the mainland of Scotland, the Hebrides and Man were transferred to the Scottish king in 1266. Norway itself later became a part of Denmark, and the Danes pawned its holdings in Shetland and the Orkneys to Scotland in 1468.

The language was more long-lived. The last “Ostmen” (Norse descendants) in Dublin were mentioned at the end of the 13th century. On the Isle of Man the language disappeared in the 14th century, in Caithness (Scotland) in the 15th century, and in the Hebrides in the 16th century. In the Orkneys the language was still known in the latter part of the 18th century.

The place names Lewis (Ljodhus), the Hebrides (Sudreyar) and Uig bay (Vik) are not Icelandic names. The Norse raids, trade and settlement of these regions started in the late 8th and the beginning of the 9th century, before the settlement in Iceland. The Norwegians settled in various Atlantic islands, and islanders therefore spoke (Old) Norwegian. They did so in Shetland, the Faroes, Orkney, the Hebrides and the Isle of Man. Indeed, Old Norwegian was spoken in the North of Scotland as well as several places in Ireland, including Dublin, the largest Norse colony. So place names were given before Iceland was even settled. Thus, for a crucial section of his Lewis-was-Icelandic theory, Thorarinsson does not even aspire to the post hoc ergo propter hoc fallacy. Indeed, he does not even make historical sense.

The Icelanders and the Walrus Ivory Trade

Colleen Batey finds that there is no archaeological evidence that ivory was worked in Iceland during the Free state in the Medieval age. The newly discovered chess piece from Siglunes in Iceland does not alter this, as it is made of a fish bone, not walrus tusk. The workshops at Skâlholt mentioned by Gudmundur Thorarinsson did not process walrus tusk. More importantly, they dated to approximately 1500: this is 300-350 years after the Lewis chessmen. Thorarinsson further points to a walrus ivory bishop’s crozier as evidence both of ivory working and of religious iconography. While religious iconography must and will be addressed shortly, suffice it here to say that this ivory bishop’s crozier belonged to the Bishop of Gardar in Greenland….no Icelander. Moreover, the skeleton in the grave where the crozier was found has been radiocarbon-dated to 1272*. This artifact and the argument hung on it are both therefore irrelevant to any serious contribution to research regarding the Lewis chessmen.

Prior to 1135, Iceland may well have participated in the walrus ivory trade. However, three of the four ships that were in Greenland around 1130 were Norwegian ships. It should also be noted that the sole ship with an Icelandic crew afterwards sailed to Norway, presumably to sell the cargo from Greenland there, before they made the turn back to Iceland. This Icelandic-crewed ship is the last Icelandic ship that is recorded in any written source visiting Greenland before Iceland became a part of the Norwegian kingdom in 1264. The ship which sank off Hitarnes was not Icelandic. Instead, the ship was wrecked on its way back to Norway from Greenland with cargo destined for Norway. Thus, there is no record of any Icelandic participation in the walrus ivory trade after 1135.

It is a facile assumption that the Icelanders took part in the voyages to the Irish Sea. The simple fact is that they did not. The best testimony about whom the Icelanders used to have connections with is without doubt the Icelandic free state laws. There we find special provisions that apply to people from Norway, Shetland, Faroe, Orkney, Caithness and Greenland. There is, though, no mention whatsoever to peoples from the Hebrides, Man or Ireland.

Gudmundur Arason did indeed travel to the Hebrides. This was, however, due to bad weather; the ship was heading for Norway, where Gudmundur Arason was to be ordained as Bishop of Holar by the archbishop of Nidaros. Beyond the vagaries of travel, the simple fact is that Arason’s ship was a Norwegian one.

In the period 1193-1211, when Pàll Jònsson (the man Gudmundur Thorarinsson assumes was the comissioner of the chessmen) was bishop, there is no record of an Icelandic shipwreck in the vicinity of the Isle of Lewis. More globally, there are not “numerous references” of the sailing of the Icelanders to the Southern Isles. Indeed, there is no record of any Icelandic ship whatsoever – not one – visiting the Hebrides in the period 1150-1260.

The explanation for this is a simple one. Whilst ocean-going vessels were necessary to travel to both Greenland and the Hebrides, Icelanders did not have such ships anymore. It is true that during the settlement, they had many ships. These boats, though, did not last long: experts estimate that these ships lasted roughly 20-30 years. Many of the islands settled during this period were not forested. Even in Iceland a limited access to wood disappeared long before the Lewis chessmen were made. This made it impossible to repair and replace ships as necessary to maintain an oceangoing fleet. Helgi Thorlaksson states, “At the end of the eleventh century the Icelanders possessed a fleet of oceangoing vessels, but in the second half of the twelfth century they were very few and around 1200 there were none. All transportation over the ocean was in the hands of the Norwegians and men from the Orkney and Shetland islands.” Njörður Njarðvik went on to explain that,

In conclusion: There is no evidence that there was an Icelandic presence in the Hebrides at the time the Lewis chessmen were made. Further, there is no record of any Icelandic ship visiting the Hebrides in the period 1150-1260, while during this time the Norwegians were in firm control of the region. The Norsemen had been there for centuries, and maintained a significant presence.

The Clonard Queen

“The Clonard Queen” is a drawing of a chess piece similar to the queens in the Lewis sets. She has the same characteristics and design, and hence quite possibly is of the same origin. This queen was found in an Irish bog 14 years earlier than the Lewis Chessmen were found. This very significant find indicates as well that other chess pieces had been transported (and lost!) in the same region as the Lewis pieces. This strongly suggests that the Lewis chessmen came to Lewis in a regular, quotidian sort of way. As has been established, however, the connection between Iceland and Ireland was minimal or non-existent. Hence the very existence of this piece is difficult to explain in conjunction with Gudmundur Thorarinsson’s assumptions.  Moreover, some idiosyncratic series of events leading to Icelanders nonetheless being in the region so as to lose or hide the Lewis pieces is undermined by the finding of the Clonard Queen.

     

The Old Norwegian Language and the Bishop at the Chessboard

 Gudmundur Thorarinsson’s main argument hinges on his claim that the word bishop for the chess piece in the Old Norse language was only used in one of many areas sharing the language. In other words, despite a shared language, only Iceland but not all of Scandinavia had this usage. Linguists usually assume just the opposite of this, believing instead that if a word occurs in a language, and that word denominates an activity known elsewhere within the area where this language is spoken, the word/denomination is used in the same capacity by all contemporaneous practitioners of the language. Thorarinsson flies in the face of practice by professionals and experts with his linguistic privileging. Moreover, Iceland was a part of the Kingdom of Norway when the bishop-mentioning text in question was written.

Archaeological finds demonstrate that chess was known in Norway as early (if not earlier) than the first half of the 12th century. However, Gudmundur Thorarinsson claims that Norwegians never used the word bishop, and always used the word “loeper”. If accepted, this idea of Thorarinsson’s produces an unsupportable chain of suppositions: Norway must have been the first country using this word, and the Norwegians used this word centuries before anyone else. Hence, even if all the surrounding countries used the word “bishop”, Norway used the word “loeper”.  Furthermore, even if parts of the same kingdom (viz., the Faroese, or Iceland) used the word “bishop”, the mainland used “loeper”. Indeed, the argument extends, the Norwegians clung to this word even when Norway became a part of Denmark…despite that the Danes used the word “bishop”.  All of this is simply untenable.

In the original version of his article, Thorarinsson wrote, “The Lewis Chessmen are the only chess pieces that include bishops with crozier and miters and full ceremonial clothing.”  To challenge this claim, I presented six images of other Medieval chess pieces depicting a bishop. These pieces are classified as chess pieces by a number of researchers and Gudmundur Thorarinsson has so far been unable to present evidence to refute their opinion. Instead, he silently elided his statement in a later version of his article to read, “The Lewis chessmen are also to my knowledge the first known chess pieces that include bishops...”.  Even this watered-down rephrasing is not, however, in accordance with the archaeological facts.

Piece d'Echec en Ivoire, from the Collection of Jean-Jacques Marquet de Vasselot

The private collection of Jean-Jaques Marquet, curator of the Louvre, contained an artifact of striking pertinence to this discussion: a bishop carved of walrus ivory. The ivory itself has been radiocarbon dated within the usual 95% probability to 770-990 CE.  Also, note that the bishop’s miter is worn facing sideways, to borrow James Robinson’s description, rather that frontally. This is in accordance both with contemporaneous changes in miter orientation and the earliest European chess piece designs. The reverse elevation of carnopy/piece reveals the old arabic abstract shape of the alfil piece, complete with “horns”.

Thus here we see an artifact that is strongly likely to predate the Lewis pieces that is connected with both early European and Arabic chess set design and a crozier-bearing, sideways-miter-wearing bishop.

     

The Written Sources

Thorarinsson’s original argument regarding use of the word bishop was, "The word 'bishop' for a chess piece is only used in two languages, Icelandic and English."  This was amended in his next publication to the rhetorical query, “The question arises, when was it first used for a chess piece? That is the main issue here.” He then goes on somewhat ponderously to answer, “At the time the chessmen were made, this word in relationship to chess was only used in Iceland… It is safe to assume therefore that the word bishop as a chess piece is much older in Icelandic than English. This suggests that the chess term may have originated in Iceland."

Gudmundur Thorarinsson writes expansively of  Old Norse “manuscripts” and “written sources” mentioning the chess bishop. This also is simply incorrect. There is only one manuscript in the Old Norse language, Màgus Saga Jarls, that refers to the “bishop".  The oldest known version of the Màgus Saga Jarls dates to 1300-1325. The Lewis chessmen date to the period 1150-1200.  This chronological gap of 100-175 years renders the Màgus Saga Jarls, like the Bishop of Gardar’s crozier, quite irrelevant to our discussion. Moreover, H. J. R. Murray himself cites two Latin texts circa 1200-1250 mentioning the bishop on the chessboard. Hence, even if the (allegedly) Icelandic manuscript or manuscripts referred to by Gudmundur Thorarinsson were pertinent here, they are predated by the texts identified by Murray.

Furthermore, several scholars (Cederskiöld, Halvorsen,Glauser, Kalinke) argue that this particular text is possibly or probably of Norwegian origin. There is no scholar who has argued unequivocally that this text has its origin in Iceland.  Thus even if the surviving version were Icelandic, it would seem reasonable to believe it to be a copy. Hence, the Màgus Saga Jarls, be it Norse or Icelandic, is of unclear provenance. It therefore cannot be mobilized to identify Iceland rather than Norway in use of the word “bishop” for a chess piece.

Finally, Thorarinsson makes much of his point about the word "bishopsmate" as part of his chain of evidence so as to argue for early use of the word bishop. He points to this “bishopsmate” because he wishes us to accept that if two words are joined to form a new one, each of the original two words is older. Prima facie, this is a sensible enough argument, but this second foray into linguistics is even more ill-conceived than his first, for it rests on false premises. In the oldest extant version of the text (parchment AM 580 B, characterized by Cederskiöld as "almost a hundred years older than any of the others") the term actually appeared as “biskups mat”. Two words. Not one word. In this matter, we were misinformed by Gudmundur Thorarinsson from the beginning.

The Icelandic Church's Opposition to Chess and Berzerkers

One of Gudmundur Thorarinsson’s main arguments against Trondheim as the place where the chessmen were made was the Church's opposition to such a project. As will be discussed shortly, this has little relevance as an argument against Trondheim. Possible clerical or legal opposition to games writ large is, though, a valid refutation of Thorarinsson’s idiosyncratic notion that the bishop of Skalholt commissioned the Lewis pieces. Nonetheless, Thorarinsson repeatedly emphasizes his idea that the Church opposed chess, while he vigorously promotes the peculiar and contradictory idea that the Lewis pieces were specifically commissioned by the bishop.  However, the Icelandic codex of law, Gragas, states unequivocally that,

Simply put, Iceland was not a European safe haven for game players who wanted to place stakes on their play. These laws were not shared by any other Nordic country at the time the Lewis chessmen were made. Even though chess as it is played today and perhaps chess as it was played then is not generally conceived of as a game of stakes, the Icelandic environment seems somewhat less friendly than elsewhere, where games pieces were regarded as commodities purveyed by merchants, in response to a robust demand.  Moreover, if the Lewis pieces were bespoken ones (a whole cloth supposition on Thorarinsson’s part), for an Icelandic bishop to be the patron of such work seems extremely unlikely.

Gudmundur Thorarinsson refers passim to Icelandic texts about berserkers, but here as elsewhere he somehow overlooks contemporaneous legal codices, which are generally regarded as being among the best written resources for serious scholarship of this era. As concerns berserkers, No. 7 of the medieval Icelandic laws (the chapter about the Church) states, “If a man goes into a berserk frenzy, the penalty is lesser outlawry, and the same penalty applies to the men who are present unless they restrain him…”.  These laws date to circa 1122-1133, which means that berserkers were outlawed in Iceland at the time the Lewis chessmen were made. Thus, the bishop of Skalholt would have been very brave or very foolish to defy outright a law against berserkers by commissioning a set of games pieces depicting precisely such.

Finally, Thorarinsson’s arguments against Trondheim because there are bishops, not archbishops at the board, are besides the point.  Let us bear in mind that at the time of the advent of the Gregorian reform movement in the church all over Europe, one of the claims was the clergy’s independence of the kings. Thus if someone had launched a game where no less than four archbishops were present, ready to sacrifice themselves for their respective kings, it would seem like that this would have been received by others than the clergy as an intended provocation.

With similar disdain for historical probability, Thorarinsson goes on airily that, “One might imagine that the [Icelandic] bishops…thought it fitting that the men standing closest to the royal couple should be bishops.”  The actual fact is that Iceland was not a kingdom and it had no formal aristocracy at the time the Lewis chessmen were made. Indeed, Iceland had no kings, no queens, only two bishops, no knights and no regular army for the hròkrs/warriors to join. Icelandic society, in contrast to most of western Europe, thus is one of the least likely places in all of the medieval world to give the chess pieces their modern identities on what has been termed “the allegorical chessboard”.

Minor Errata

Last, I would like to use this opportunity to briefly correct some of the less egregious of Thorarinsson’s mistakes, omissions, and conclusions so as to further emphasize that Thorarinsson trades in speculation at best rather than fact.

1. The bishops’ “[m]iters changed again around the year 1200, so the chessmen are unlikely to be much younger than that.”  As reference for this, he cites “Robinson, The Lewis Chessmen; Stratford, The Lewis Chessmen and the enigma of the hoard. “ First of all, the miter did not change again before the 14th century. Secondly, none of the cited authors claimed otherwise.

2. “Some scholars claim that the Icelander Sæmundur the Learned was the first citizen of the Nordic countries to be educated in a university in France.  He studied there for many years and probably paid for his education with precious artifacts from Iceland.”  As universities did not exist in either France or Europe itself at that time, this statement is an anachronism unlikely have been made by Thorarinsson’s purported “scholars”.  While general information batted about on the Internet is that Sæmundur, established at Oddi as of 1078, studied at what would by 1150 become the Sorbonne, modern scholarship tends more to the idea that Sæmundr fróði studied instead in Franconia. Methods of payment for his education are certainly even more speculative rather than “probably… with precious artifacts from Iceland”.

3. “According to historians the ‘Church politic’ in Trondheim was clear. The church should be peaceful and not participate in war or violence.”  Here I wish only to point out that Thorarinsson makes this claim in regards to the time when the crusades were reaching their peak; a time when bishop Absalon, possibly Denmark’s greatest warlord of all times, was at the height of his powers; and a time when the Norwegian church was one of the major forces in the Norwegian civil war (1130 to 1240).  Any historian making such a dubious claim should be identified. As usual, Thorarinsson fails to do so despite the appeals to authority he repeatedly makes with such grandiose ostentions to “historians”.

4. “[T]he Norwegians know little about their history before 1200, except what was written in Iceland by Icelanders.”  This is an odd statement. I will focus on the last part, concerning the written sources to Norwegian history.  First we have the Norwegian medieval laws, while the medieval Norwegian diplomas are contained in 22 volumes with close to 20000 documents. The three oldest extant history books about Norway were all written in Norway,  by Norwegians. In addition there are several shorter texts about history written by Norwegians.

5. Finally, it is necessary to address the abuse of quotes by old scholars. Thorarinsson points to Madden and Murray because they concluded that the use of “hròkr” (warrior) indicated that the chessmen were from Iceland. They thought this because they thought that only Iceland used this term to denominate the rook: both of these men were unaware of the Faroese language, which still has the same expression. Moreover, they were unaware of the fact that this term was used all over Scandinavia in the Medieval era. It is simply inappropriate to mine these fine scholars for misleading and outdated quotes in such a way.

Conclusion

One of the main deficiencies with Gudmundur Thorarinsson’s work about the Lewis pieces is the sheer number of incorrect historical facts. The text is marred with simple faults. Hence, the historical handicraft is not well done. Moreover, a number of important and known facts indicating conclusions contrary to his own are omitted, despite that some of these facts singlehandedly undo his entire thesis. Additionally, Thorarinsson employs questionable methods, referring to “scholars” and “historians” without naming them, to “manuscripts” without providing any references, to “artifacts” without providing images or even stating where they might be located. When he does name a written source, he does not cite page numbers. Indeed, approximately 25% of his references as a whole are interviews which cannot be verified. Thus, a reader has to take Gudmundur Thorarinsson’s statements at face value.  As demonstrated here, such faith in a work riddled with mistakes, omissions, unsupported assertions, misused sources, and questionable conclusions is a risky business indeed.

Thanks for help from Kristjan Sander (Estonia) and historian Milissa Ellison (USA), who also served as  professional editor for English usage and argumentation. I also wish to thank chess collector Ole Drønen (Norway) for his critical comments. The author, Morten Lilleøren, is university-trained in history. In the chess world he is best known as an ICCF Grandmaster and was a member of the Norwegian National Team that won the 15th Correspondence Chess Olympiad. Morten Lilleøren, November 2011-11-05, Oslo, Norway

Previous articles on the Lewis Chessmen

Lewis Chessmen Symposium at Skálholt, Iceland 01.08.2011 – In the nineteenth century a collection of chessmen, handcrafted in the twelfth century, was found on the Isle of Lewis, and have become the most famous chess set in the world. This year archaeologists discovered a new chess piece in Iceland, and the controversy surrounding the exact origins of the Lewis Chessmen takes a new twist. In August you can attend a symposium on the subject.

On the origins of the Lewis Chessmen – A reply 31.03.2011 – Gudmundur. G. Thórarinsson, the author behind the main material in The Enigma of the Lewis Chessmen, a discussion on the origins of the world's most famous chess set, was dismayed at the belligerent tone in an article seeking to refute his hypotheses regarding its Icelandic roots. He replies here and adds 'the potency of these arguments necessitates formidable counter arguments'.

Norwegian-Icelandic war over the Lewis Chessmen? 05.03.2011 – Last year we published an article on the famous Lewis Chessmen. The two Icelandic authors claimed that the pieces were carved in their country. "With few reservations parts of the chess world have adopted this theory," writes Morten Lilleøren of Norway, who finds the article "filled with faults and oversights." Lilleøren sets out to correct the facts and prove: the Lewis Chessmen are from Norway.

The enigma of the Lewis chessmen 11.09.2010 – In 1831, in Edinburgh, Scotland, a collection of chessmen found on the Isle of Lewis was displayed for the first time. These 12th century handcrafted pieces made from walrus tusks and whale teeth have since become iconic examples of our lasting love for wargames. Their origins, however, is one of theory and controversy. Here is an illustrated article on the world's most famous chess set.

Copyright Lilleøren/ChessBase