RUNIC INSCRIPTIONS AS A PRIMARY HISTORICAL SOURCE

by

Hugh R. Whinfrey

The purpose of this paper is to discuss the major issues underlying the use of Germanic runic inscriptions as a primary historical source. The stage will first be set with some general runological background. Then orthographic issues will be covered. Next, issues other than decipherment, and focused on runestones, are examined. And then finally some comments will be directed towards the enigma of runestones in northwestern Russian history.

The first scholars who made systematic surveys of runic monuments were Johan Bureus (d. 1652) in Sweden and Ole Worm (d. 1654) in Denmark and Norway.¹ The first use of runic monuments as historical sources was published in 1893 by the Danish scholar Ludvig Wimmer.² References to the work of Worm and Wimmer permeate the modern runologic literature.

Runic inscriptions appear both on large stones and on portable objects. The runestones are the focus of most of the scholarly works. Published catalogs of runestones are readily available and often include photographs and detailed scholarly analysis of their inscriptions. Runic inscriptions have been found in all areas known to have been extensively inhabited by Germanic tribes, and in other areas where a Germanic presence is strongly suspected.

The runic alphabet of the Germanic world is called the 'futhark', which is simply the first six letters of this alphabet pronounced in sequence ('th' is an interdental fricative). The individual letters are referred to as 'runes'. There is wide disagreement among runologists about when and where the futhark originated. Speculation ranges all the way back to the bronze age (ca. 1500 BC). The general supposition is that the runic script entered the Germanic world in the first several centuries after Christ as a result of commercial interactions with the Roman Empire. The oldest known runestones are on the Scandinavian peninsula and date from ca.300-400 AD. The oldest known example of the script itself has been dated as ca. 200 AD.

Moltke³ gives a detailed comparative analysis of the futhark with the major variants of the Phoenician, Greek, Etruscan and Latin alphabets and arrives at the conclusion that the Latin alphabet of circa 0 A.D. was the basis for the futhark. Ludvig Wimmer also believed that the futhark was based on Latin letters. The Greek and Etruscan alphabets have been taken seriously by scholars in the past as possible sources for the futhark. The latest 'Greek' theory was published in 1988.⁴

Moltke believes that borrowed Latin letters, with some entirely Germanic additions, were modified so all purely horizontal and curved lines were replaced by slanted straight lines. He argues that the forms of the various runes were arrived at by the fact that they were intended to be chiseled into wood.⁵ The purely horizontal lines are more difficult to read on wood than slanted lines, and the absence of curved segments made the work of chiseling them easier. As additional evidence, he cites the recent unearthing from an ancient quay in Bergen of hundreds of runic inscriptions on wood containing everything from love letters to bills of lading.

There are three major chronological variants of the futhark under the classification scheme defined by Moltke⁶: first period (ca. 0 to ca.650 AD), second period (ca.650-ca.1050), and third period (ca. 1050 to ca. 1400). The Latin script had completely replaced the futhark by the end of the third period.

The first period futhark consisted of 24 runes, reading left to right, and a mirrored variant reading right to left. It was sometimes written using a combination of the two, as a plough goes back and forth over a field. This period, referred to as the 'older futhark' by Pritsak, is generally identified with the Kylver stone (ca. 400) found in Gotland, which has the earliest known sequential listing of the 24 runes inscribed on it. There exist several other relatively minor variants from the Kylver stone format. The most significant major variant of the first period is an Anglo-Frisian futhark of 31 letters (ca. 500 AD).

The second period futhark consisted of 16 letters and is considered to have originated just before the dawn of the Viking Age. Pritsak refers to this period as the 'younger futhark'. It exists in two major variants - the Danish and the Swedish-Norwegian. There is only one known example, from South Jutland, that reads from right to left. In this period some of the runes have curved segments. Moltke takes this as evidence that metal engraving and stone carving were probably commonplace uses of this script.⁷

In the third period, the middle ages, the alphabet is properly called a futhork, due to a change in pronunciation of the fourth letter. It had 27 runes, with numerous variants for some of them, and was based on the second period Swedish-Norwegian script. Numerous inscriptions from this period exist on stones used to construct churches, where they appear to be the systematic marks of workmen rather than graffiti.

The major problem confronting a runologist is usually the deciphering of the text of the inscription. Inscriptions dating from the transitional time between the first and second periods are somewhat problematic simply due to a lack of a coherent standard orthography. There are also the obvious hurdles of imperfect preservation such as missing pieces or defacement.

There are also some less obvious hurdles that can enter into the deciphering process. Moltke cites an example where an engraving done by an illiterate metalworker has been concluded to be the reason for an unknown character in an inscription.⁸ Moltke also makes a very cogent point that transliterations from the runic alphabet do not necessarily reflect the spoken language at the time, just as a modern written language often reflects pronunciation from a bygone era.⁹

On an orthographic level, there are several hurdles that can hinder the deciphering process. There is a variant of the second period Swedish-Norwegian futhark, called 'stemless', that omits the main part of each letter, leaving just the secondary parts to indicate the sound. It appears to be a type of shorthand. It was also common practice from even the oldest runic inscriptions to save space by writing two, and later even three, dissimilar runes on top of each other. Each futhark also had minor variants in some of the individual runes which further complicates decipherment.

In general, the second period futhark, the Viking Age, sacrificed general legibility for ease of construction. The streamlining of the form of the characters and the lesser number of characters created ambiguities on a semantic level - there could often be several interpretations in Old Norse. Successful deciphering of a second period inscription hence requires that some attention be paid to context, which is only made more difficult by the relative brevity of the inscriptions. Occasionally a dot was used within a rune to create a variant which alleviated some of the semantic ambiguities, however the dots had no meaning with certain runes and yet were still used with them as ornamentation.

Even with a correct reading of the orthography, there are still further complications that can occur. The runes were sometimes written in code for reasons of commercial secrecy or magical/religious reasons. There exists a fair amount of literature devoted to deciphering these. Finally there is the problem of determining what a deciphered word means in Old Norse. Problems caused by obscure words are continuing sources of dispute among scholars.

The runestones of the Viking Age were erected for a number of proposes. Pritsak notes among them the commemoration of civic events, marking of assembly sites, monuments to judicial decisions, and frequently monuments to travellers abroad, especially those who perished while away.¹⁰ Even when the runes are considered to be deciphered, the text of the inscription is usually no longer than a paragraph or so and contains information that is often mundane and personal in character. They do however frequently reference place names, personal names, and both circumstances and objects relevant to the culture. These references are the main body of data that the runestones provide to the historian.

Use of this data is a dance with various degrees of uncertainty. The runestones are often the only available contemporary written documentation of Viking activities, hence they cannot be ignored as supporting evidence and do carry some weight as refutational evidence.

Questions of the authenticity of specific runestones are generally passed over in silence, perhaps rightly so in many cases where knowledge of their existence predates the scholarly attempts to decode them. Yet it certainly remains conceivable that a hoaxster could chisel a runic inscription into a slab of stone. Normal debunking techniques based on Carbon 14 dating do not offer a solution in these cases, as the inscription consists of a systematic lack of material on the face of an inorganic stone. Weathering conditions have enough variables that a situation can easily be concocted to explain a relatively fresh-looking stone. In brief, they are vulnerable to hoaxsters. Perhaps the only remedy is to be particularly cautious about isolated inscriptions that contain implications running counter to generally accepted notions of the Viking Age.

Since runestones were not explicitly dated, dating of a runestone is at the very center of the dance with uncertainty. Supporting archaeological evidence from the site, often non-existent or unexcavated, is required for any truly scientific precision. Art history is commonly invoked when the stones also contain decorative carvings. The form of the runes themselves belie the era to which the inscription belongs, however a finer delineation can only come by cross-referencing with other sources. Perhaps of more interest to the historian is a dating of the information given by the inscription on the runestone. The erection of a monument is not necessarily simultaneous with the event commemorated. Sagas and skaldic poetry from the Viking Age often reference the same events and people as a runestone. Sometimes events with known dates are referenced in the runic inscription. The point is that the process of dating necessarily brings into play other types of historical sources, and some of these sources are not broad generalizations but specifics. These specifics in turn also serve as evidence to the same end as the runic inscription itself. This methodology precludes the runestones from being used in any serious sense as stand-alone historical documentation.

The most tenuous aspect of using the runestones as historical evidence is taking the absence or scarcity of them as supporting evidence to a hypothesis. Considering the millennium or so since their construction, many have been doubtlessly lost forever. A crude estimate made with liberally unrealistic assumptions concerning early English runic inscriptions yields a guess that at most one per cent of the objects actually inscribed are known to scholars today.¹¹ The runestones however were intentionally conspicuous, large and heavy, and altogether more difficult to lose than the predominantly portable objects involved in the preceding estimate. However, runestones have not always been treated with respect as historical objects. Pritsak notes that in the middle of the thirteenth century, runestones were sometimes regarded as building material in Sweden.¹² Moreover, Pritsak asserts that there was a lack of suitable stones in eastern Europe and that the local population probably did not treat foreign runestones with any great respect.¹³

Probably the greatest enigma surrounding the runestones is their relative scarcity in northwestern Russia. Scandinavian activity in the area is well-evidenced by other sources. That stones were scarce and wood consequently used instead is a plausible explanation. Yet unfortunately it pales beside the fact that from Novgorod, where wood was preserved in the ground, only two runic inscriptions have been recovered - one on the humerus of a pig and the other on a fragment of a cow's rib.¹⁴ The riddle remains unsolved.

In summary, runology is a somewhat exotic field. The decipherment process lies close to the boundary between an objective science and a subjective art. The dating process is in a similar, but slightly more objective, position. Despite all the uncertainty surrounding them, they constitute valuable pieces of evidence in issues where evidence is not always particularly abundant. As a historical source, they can not only help answer questions, but pose them also. In truth, it seems they pose more questions than they answer, which by itself is enough to make them valuable to historians.

Seattle, December 1993.

Endnotes

Bibliography