Waffen und Graber: typologische und chronologische Studien zu skandinavischen Waffengrabern 520/30 bis 900 n. Chr.

RUNDKVIST, MARTIN

ANNE NORGARD JORGENSEN. Waffen und Graber: typologische und chronologische Studien zu skandinavischen Waffengrabern 520/30 bis 900 n. Chr. (Nordiske Fortidsminder ser. B Vol. 17). 417 pages, 279 b&w and colour figures, 4 tables. 1999. Copenhagen: Kongelige Nordiske Oldskriftselskab; 87-87483-43-2 ISSN 0105-578X hardback DKK500 +VAT, p&p.

The bronze jewellery sequence of Late Germanic Iron Age Scandinavia (6th-8th century AD) has been fairly well understood since the early 20th century, and has been tied down firmly in recent years through computerized typological seriation performed by Karen Hoilund Nielsen and others. The contemporary iron weapon sequence has received much less treatment, and has only been loosely piggy-backed on the jewellery through the work of Birger Nerman and others. This has now changed with the doctoral dissertation of Anne Norgard Jorgensen. She has collected comprehensive typological data from the three Scandinavian areas with abundant Late Germanic weapon burials -- Bornholm, Gotland and southern Norway -- and given them the full treatment of Danish neo-typology. 338 well-preserved and well-documented assemblages have been selected, and the sample is presented with measurements as well as simple line drawings. This data collection alone is a priceless contribution to the field of research.

Sadly, the author's typological method is inconsistent and marred by impressionism. Many undefined terms are used, and the line between type definitions and population descriptions is vague. Although many types are adequately defined, the definitions of others overlap and a `type' can even be presented without any definition whatsoever (lance head L5, p. 94). No statistics, even bivariate ones, are offered to document the morphological distance between populations and thereby justify their delimitation as types. Ideal types seem to have been envisaged without the aid of any rigorous study of the actual variation. This is exemplified by the fact that the length-width proportions crucial to the definitions of the different types of single-edged swords have been calculated as mean values of populations previously delimited by some other means (p. 47, caption to Abb. 11), apparently visual impressions. A unique piece can, seemingly at random, be used to define a single-member type (lance head L8, p. 97), while others are relegated to a rag-bag group (p. 100).

Let us examine the type system for the shield bosses (pp. 77-87). The bosses are divided into those with neck and top disc (SBA), two types with neck but no top disc (SBB and SBC), and Galgenbergian high ones with weakly pronounced neck and a tapered dome (SBD). SBB and SBC cannot be separated morphologically nor on the grounds of the metric definitions. The quoted intervals of the two types' measurements overlap for all measured parameters. The only parameter that seems somewhat to separate the `types' is the width of the brim where the overlap is only 1 mm. Nevertheless, it is clear from the illustrations that the combined SBB&C material shows systematic variation in shape that would be best expressed through proportional definitions, e.g. height through diameter and height above neck through total height. Parallel variation of shape is found in the bosses with top discs, but this variation is entirely subsumed under the single typological element of the disc, which is an additive one in many cases. Thus, only the SBD type is a methodically sound one. The primacy of absolute dimensions and yes/no morphological traits over proportions in these and other type definitions indicate that the author's experience with the single-edged swords has detrimentally influenced her work with other artefact categories. Not all artefacts follow a Kurzsax-Langsax sequence.

Despite the weaknesses of the individual typological systems, the sheer number of artefact categories involved and the interregional correlations nonetheless lend some credibility to the five-phase sequence resulting from the analyses. Single-edged swords, the grips of double-edged swords, shield bosses, lance heads, axe heads, arrowheads, knives, strap mounts and horse bits are all used to seriate the grave assemblages. Due to the continued post-Merovingian use of single-edged swords in Norway, two Viking Period phases up to c. 900 AD are defined for these finds.

Having completed the gigantic typological and absolute chronological work, with the early Swedish boat graves pinpointed in the resulting system, the author turns to sociohistorical interpretations in the final chapter, `Das militarische System'. The models of Merovingian military organization formulated by Heiko Steuer and others are adapted to the Scandinavian context, and illuminating comparisons are made with written sources of the Scandinavian High Middle Ages. This is a condensed version of studies previously published in English (reviewed in ANTIQUITY 74 (2000): 245-6).

The aptly named warren und Graber will become the obvious entry point for future study of Late Germanic Scandinavian weapon graves. Above the level of type definitions the analyses are exemplary. Unfortunately, these higher-level analyses depend on the quality of the groundwork. As the basic typological order imposed upon the finds is to no small degree imaginary, the higher-level results must be treated with caution until the implications of revised groundwork become clear. Such revision can, thanks to the author's painstaking collection work and accessible presentation, be performed at any archaeologist's desk.

MARTIN RUNDKVIST Department of Archaeology University of Stockholm arador@algonet.se