The Spear - Digital Documentation Sheds New Light on Early Bronze Age Spear Carvings from Sweden – an Analysis with Some Comparative Examples from Valcamonica, Italy

Ulf Bertilsson

The Spear - Digital documentation sheds new light on Early Bronze Age spear carvings from Sweden – an analysis with some comparative examples from Valcamonica, Italy

Introduction Depictions of metal weapons such as spears, is the case with well-known carvings from swords, and axes are frequent in the Swed- Bohuslän and Uppland. This demonstrates ish Bronze Age rock carvings of Bohuslän, the importance of the spear as the warrior’s Uppland, Östergötland, and Skåne. Most principal weapon in the Scandinavian Early images are non-representational and mostly Bronze Age. This study presents previously portray the object type rather generically. unknown examples of the practice of re- Occasionally, the carvings exhibit details of carving spearheads, which provide unprece- the objects including the shape of the hilt, dented insights into the close link that must ferrule (or butt-spike), and scabbard, so as have existed in Bronze Age Sweden be- to suggest that they are images of real ar- tween real and represented weapons. Some tefacts that existed during the Bronze Age. interesting parallels from rock carvings in It is at times possible to determine precisely Valcamonica in Italy are also presented. the type and chronology of the object depicted. This is the case, for example, with the swords shown on the Ekenberg rock Odin’s spear Gungnir carving in Norrköping (Östergötland), which Fighting and weapons are frequent themes compare to actual Bronze Age swords in the Nordic sagas; their prominence is from the archaeological collections of the due to the notion that the ultimate and Stockholm Historical Museum (Bertilsson 2015; Hildebrand 1869,). It has often been Period I 1750/1700–1500 BC argued that most of the weapons depicted Period II 1500–1300 BC in Swedish rock art are to be dated to the Period III 1300–1100 BC Late Bronze Age, in particular to Period V Period IV 1100–920 BC (c. 920-720 BC; Vogt 2011). However, de- Period V 920–720 BC tailed analysis of the carvings, including Period VI 720–550 BC the digital-technology analysis presented here, indicate that this may not be the case, Table 1. Periods of the Nordic Bronze Age as a number of carvings appear to depict spear types from the Early Bronze Age, 1750 -1100 BC. This includes the spearhead from most glorious way for a warrior to die was the second Valsömagle metal hoard from as a result of single combat or in battle. Denmark, which is dated to period IB of In this way, the fallen would achieve hero the Nordic Bronze Age (Vandkilde 2008). status and be admitted to the company of In some cases, it appears that carved spear- other warriors in Valhall, the hall of Odin heads were updated to represent new types described in the Grimnesmal poem cited over a period of several hundred years, as below. There, the glorious dead were resur-

64 Adoranten 2017 rected the morning after a ritual violent Sun god, was a widespread religious belief banquet: in southern Scandinavia during the Bronze Age (Nordberg 2013, Bertilsson 2014). Until Easy is it to know | for him who to Othin recently, this discourse dominated research Comes and beholds the hall; into Nordic Bronze Age religion, which is Its rafters are spears, | with shields is it roofed, vividly embodied by the story of the sun’s On its benches are breastplates strewn. journey across the sky during day and night (see for example Flemming Kaul’s 1998 Easy is it to know | for him who to Othin book on the subject). Although there is a Comes and beholds the hall; consensus among researchers that the basic There hangs a wolf | by the western door, tenets of this model are correct, over-reli- And o’er it an eagle hovers. ance on it has driven alternative hypotheses about cult and religion in the Scandinavian Bronze Age into the background (Kristian- (The Poetic Edda, Grimnesmal, verse sen 2012; cf. also Nordberg 2013). 9-10; (Bellows, (Trans.). 1923. Available at www.voluspa.org 2008). In an attempt to offer an alternative approach to research in this field, this chapter The weaponry used by the warriors and will shift the focus of the analysis to the gods mentioned in the Nordic sagas encom- rock carvings of Bronze Age weapons, pay- passed swords, axes, and more rarely spears. ing particular attention to images depict- The spear was the weapon of Odin, the ing spears. One of the ideas underlying highest Norse god; he prevailed over the liv- this approach is that the weapons and ing and the dead, and his spear was called fighting scenes displayed on rock carvings Gungnir – meaning “the swaying one” in may express cultic or religious beliefs. This Old Norse. The significance of the spear is chapter will also discuss whether the “war- highlighted in multiple sources including riors and weapons discourse” characteris- the book of Gylfaginning (The Prose Edda, ing the Nordic sagas can be traced back chapter 51), which describes Odin, on the to the Bronze Age and its petroglyphs. To way to the battlefield at Ragnarök, riding explore these problems, I shall analyse a on his horse Sleipner above Einherjar (those number of images depicted on Swedish rock warriors who have died in battle and are carvings, namely the spears engraved on brought to Valhalla by the Valkyries) wear- panels from Bohuslän and Uppland. I will ing a coat of mail and a golden helmet, not apply quantitative methods of analysis armed with his spear Gungnir, which he will to their study, but will instead focus on the use to attack Fenrir the Wolf. Gungnir was, qualitative reading and interpretation of said to be, so well balanced that it could hit selected carvings displaying extraordinarily any target. clear images of spears. These are the panels from Litsleby, Kalleby, and Finntorp (Tanum, The Norse sagas, first written during the Bohuslän), and also from Tuna (Bälinge, Up- Late Iron Age, from 500 AD, bear witness to pland), all of which exhibit spear carvings of the importance of the spear as a weapon of unusual type and quality. the highest status. A common view among scholars of Old Norse religion is that the Some of these images are of notable his- gods appearing in the classic Iron Age my- toric importance; for example, the Litsleby thology would have replaced an earlier be- carving was the first ever-documented rock lief system based on the sun cult, which had carving from Tanum, as shown by an ink dominated the region since the Bronze Age drawing produced by Carl G.G. Hilfelings in (Almgren 1927-28; Nordberg 2013). Several 1792 (Fig. 1) (Bertilsson 2015b). However, strands of archaeological research have Hilfeling’s drawing does not report all the convincingly argued that the cult of the carvings visible on this rock, whose num- sun, and probably also that of a personified ber is now known to exceed 200 images.

Adoranten 2017 65 Fig. 1. Carl G.G. Hilfelings’ 1792 ink drawing of the rock carving from Litsleby, later named the “Spear God” and now fre- quently referred to as Odin. (Source: SHFA. Original: Antikvariskt Topografiskt Arkiv, National Heritage Board, Stockholm)

Instead, he focused on the most impress- ity and cost-effectiveness. Such improve- ing image, namely the 238 cm tall, phallic, ments have broadened its uses, which spear-armed male figure, which he depicted now include the digital documentation in a detailed “portrait-like” manner. In con- of rock art. SFM uses photographs of the trast to the overall quality of his drawing, rock surface to calculate a high-resolution, however, his rendering of the spearhead is three-dimensional model of the engravings. not completely accurate, though the length Contrary to traditional two-dimensional is. The modern three-dimensional images methods such as rubbing and tracing, SFM analysed below discuss the differences be- accurately records structure, topography, tween Hilfelings’ ink drawing and the ac- and texture of the rock panel surface tual carving (Fig. 1 and 3). (Bertilsson et al. 2017).

One problematic aspect of SFM is that it Three-dimensional documentation generates extensive amounts of data, es- with Structure For Motion (SFM) pecially when documenting large rock art In recent years, a 3D analytical technique panels; for example, the Aspeberget carv- known as Structure For Motion (SFM) has ing (Tanum) required 1500 images for a full seen marked improvements in its accessibil- 3D documentation (Bertilsson et al. 2014).

66 Adoranten 2017 Depending on the computer system utilised, lines, blurring important details including computing this amount of data can take superimpositions. For example, the Litsleby several hours. Despite the time and effort carving depicts a wheel directly adjacent to involved, SFM does produce highly accurate the so-called “Spear God” discussed below. 3D models of the carvings, showing more The wheel is painted in as complete and details than rubbing and tracing. A pilot shows up so in the textured version of the study conducted in Tanum in 2014 com- 3D file. However, close examination of the pared SFM with other techniques of rock un-textured 3D model shows that the wheel art recording (Bertilsson et al. 2014). The does not have a continuous rim, but one results show that SFM provides a significant that is made from several non-contiguous advantage over traditional methods includ- sections (Fig. 2). This demonstrates that, ing better opportunities for morphological regardless of how experienced the modern identification and typological dating; it also painter of the carving is, their intervention improves the recognition of superimposi- adds another layer of interpretation to the tions and is ultimately less time-consuming recording process. Nevertheless, as SFM than rubbing and tracing (cf. Sevara & highlights very fine details, this drawback Goldhahn 2011, DeReu et al. 2012). In the is mitigated to some degree, in a way that following pages, I shall present examples to enables us to “see through” the paint (Do- illustrate this point. mingo et al. 2013). Therefore, SFM creates better opportunities to record, describe, A challenging factor in the study of rock art and analyse previously unrecognised details is the recent practice of painting carvings that may be of the greatest importance for red in an effort to make them more visible our understanding and interpretation of for the convenience of tourists. The red the rock carvings. paint tends to smooth over the engraved

Fig. 2. 3D model of the Litsleby carving, textured (left) and untextured (right). (Source: SHFA. Photograph: Ellen Meijer)

Adoranten 2017 67 Fig. 3. Structure For Motion (SFM) 3D model of the “Litsleby Spear God” marking the three different spearheads discussed in the text I, II and III. The SFM picture also shows that the arm and hand holding the spear overlie the shaft, and are thus younger than the shaft itself. (Image editing: Catarina Bertilsson. Source: SHFA. Photograph: Ellen Meijer)

The mighty spearman from Litsleby The chronology of the Litsleby panel Our first case study concerns the extremely Before a detailed chronology is proposed large spear-carrier depicted on a rock panel for the carving events documented on the from Litsleby, Tanum. The SFM recording Litsleby panel, a discussion of the overall has revealed hitherto unrecognised details chronology of the panel itself, and of the in the carving, and a careful analysis of the spearman in particular, is necessary. The 3D model of the spear has shown previously spear-carrier has long been considered to unnoticed information of great importance. date to Period V of the Nordic Bronze Age In particular, it was observed that the spear (c.920-720 BC in absolute terms; Maraszek was engraved before the human figure 2015). This proposal is grounded in the because the arm and hand overlie the spear comparison with other large human carv- shaft (Fig. 2 and 3). The spearhead itself ings such as “Skomakaren” (or “The Shoe- is notable because it may have been re- maker”) at Backa, Brastad. This chronology, carved twice after the original engraving, however, is solely based on the style of the considering that the 3D model highlights carving. three different spearheads on the rock surface (Fig. 3, marked I, II and III). Another For a better appraisal of the chronology of noteworthy observation stemming from this figure, attention has to be paid to the the SFM documentation is that the wheel is boats engraved on the panel, since their not only interrupted on its upper and left typology and relative height convey impor- sections but is also overlain by the phallus tant information concerning the carving of the spearman. This means that the phal- sequence and absolute dating of the images lus was carved after the wheel had been (Ling 2008; Kaul 1998; 2003). The earliest drawn. The absolute chronology of the two boat associated with the spear-carrier is a engraving episodes is not easy to establish small vessel overlain by the spear shaft. The but their relative sequence is interesting vessel belongs to the so-called “Nag-type”, nonetheless. which dates to the Late Neolithic, from 2100-1700 BC (Bengtsson 2014). A similar chronology can be suggested for the three-

68 Adoranten 2017 line boat overlain by the left upper arm. clearly rounded compared with the angular The spearman’s thighs superimpose on a shown in red paint. This interpretation was third boat. This boat is of a type dating to confirmed by crosschecking existing docu- Period II of the Nordic Bronze Age, from mentation obtained via frottage (rubbing). 1500-1300 BC (Ling 2008:105, cf. Kaul 1998 This spearhead shape is consistent with and Fig. 2). Based on these considerations, actual objects belonging to the Valsøma- one could say that the large spearman was gle type, dating to Period IB of the Nordic carved after Period II, when the latest boat Bronze Age, about 1600 BC (Horn 2013; was drawn; however, nothing indicates that Jacob-Friesen 1967; Vandkilde 1996, 2011). it has to be assigned to Period V. A bronze spearhead matching the Litsleby On the panel, but not in direct connec- carving almost perfectly was discovered in tion to the spearman, are other boat carv- Falköping in Västergötland. This is 37 cm ings showing significantly lower chronolo- long overall and has a 9 cm long socket, gies. In particular, two large and heavily which is beautifully decorated with spirals stylized boats found in the northern section forming a running dog motif framed and a of the panel are overall similar to Hjort- continuous band of double arches. Monte- spring-type vessels, which are dated to the lius published drawings of this spearhead early 4th century BC by a Danish bog find (1917: 61, no 917), as did Jacob-Friesen of a wooden boat of the same design (Kaul (1967, Tafel 28:1) and Oldeberg (1974: 303, 2003). If we accept this chronology for the no. 2373; fig. 5a). A direct comparison of three boats carved on the Litsleby panel, the archaeological and carved spearheads this would suggest that multiple carving was made using Photoshop (Fig. 4, right). events took place on the panel over a time Bearing in mind that the drawing of the possibly extending from the Late Neolithic spearhead had to be scaled down for the to the Pre Roman Iron Age, 2100-300 BC comparison to be made, it is noteworthy – a remarkable 1800 years period. The 3D that the Falköping specimen fits almost model also indicates that three different perfectly into the carved spearhead. This spearheads were carved on the panel, with strongly suggests that the Litsleby carv- two shortest in lengthwise succession ap- ing intended to show a Valsømagle type pearing to overlie the longest. This raises spearhead of which that from Falköping is the question as to whether it is actually an excellent example. This interpretation is possible to determine which spearhead was further supported by a similar comparison, carved first. Analysing the spearheads will which the author has made using a photo- help clarify the complex chronology of the graph of the Falköping spearhead obtained panel and of the spear-carrier. from the National Historical Museum (Fig. 4, centre, 5 and 6).

The Litsleby spearheads: A cursory examination of the second carved typology and chronology spearhead based on the SFM 3D model The longest spearhead on the Litsleby panel (Fig. 3, II) indicates that this may depict a has rounded shoulders and a long, slen- Hulterstad-type spearhead. Spearheads of der tip (Fig 3, tip marked I). The image is this kind are shorter and more compact deeply carved and its design is so detailed than those of the Valsømagle type, and as to resemble real metal spearheads from of Period II spearheads in general (Jacob- the Bronze Age, examples of which can be Friesen 1967). The eponymous Hulterstad found in museum collections. Some years spearhead was found in a burial mound ex- ago, the author noticed that the shape of cavated in the 1920s on the island of Öland. the carved spearhead deviated from shape The burial assemblage can be dated to Pe- of the spearhead that was painted in mod- riod III of the Nordic Bronze Age, 1300 – 11 ern times. This is especially apparent in the BC. Further comparisons were carried out bulge at the bottom of the blade, which is to validate the initial hypothesis, using the

Adoranten 2017 69 Fig. 4. Left: Photograph of the bronze spearhead from Falköping next to an SFM 3D model of the Litsleby spearhead carving. Centre and right: drawing of the Falköping spearhead from Montelius (1917: 61, no 917) and superimposition of the photograph and drawing on the 3D model. (Comparative analysis by Christian Horn, Gothenburg University) method discussed above. These comparisons and intended to depict them as realisti- have confirmed that the second spearhead cally as possible. Moreover, the exceptional belongs to the Hulterstad type, while also craftsmanship of the spearheads from firming up the idea that the longest spear- Falköping and the Valsømagle hoard, as head was carved first, and the two shorter well as their unusually rich and artful deco- second (Fig. 3, tip I –III and Figs. 7-9). rations, suggest that these were not ordi- nary weapons, but objects that presumably The near-perfect matches between carved belonged to great warriors and quite pos- and archaeological Bronze Age spearheads sibly prominent members of their communi- discussed above demonstrate that the Lit- ties. This consideration sheds new light on sleby rock carvers must have had access to the Litsleby carvings, which, I argue, depict actual bronze spearheads from their time spearheads of special value and importance.

Fig. 5. Photograph of the bronze spearhead from Falköping; the object belongs to the Valsømagle type, which is dated to Period IB of the Nordic Bronze Age. (Photograh: The Swedish History Museum – SHM)

70 Adoranten 2017 There are, however, several factors making this hypothesis problematic. At Litsleby, there are no images depicting opposing warriors armed with swords, axes, or spears, which are commonly found in many other carvings from Tanum including Bro Utmark, Aspeberget, Vitlycke, and Fossum (Ling & Bertilsson 2016). The Litsleby carving is instead dominated by a single, supernaturally large human figure – the «Spear God» –, which is unparalleled in Scandinavian rock carvings. As pointed out above, the figure is likely to be dated to Period V of the Nordic Bronze Age, 920 -720 BC, by analogy with, the likewise famous human figure «The Shoemaker» from Backa Brastad as well as several other figures of the same type. I have argued above that since the figure›s hand overlies the spear shaft, dated to period I, 1750 – 1500 BC, the large human figure must therefore be later, indicating that the proposed Period V chronology of the supernaturally large human figure may actually be correct; in this case, it could have been simultaneously with the third Fig. 6. The Valsømagle hoard from Denmark; note the spearhead carved in period V, 920 – 720 BC. spearhead on the left and the cast-hilt sword in the centre. This suggests that the first spearhead of the (Aner&Kersten 1976, plate 79-80 no. 1098). Valsømagle type together with the spear shaft was carved much earlier, in the later This proposal is hampered to some extent phase of Period I, 1750 – 1500 BC, and that by the fact that the Litsleby spearheads this original spear was the most important were seemingly redrawn over a period of and prominent figure on the carving be- several hundred years. A possible explana- cause it was on the rock individually with- tion for this is that the carvers belonged to out a human figure. If this assumption is a leading warrior dynasty that dominated true, how could it be explained? In my view, the area during a period of nearly a mil- a possible reason behind could be that this lennium; they would have used the carving spear was a unique weapon of great power and re-carving of the Litsleby panel as a and symbolic importance, justifying its de- social strategy to manifest and consolidate piction in «splendid isolation». This being their power (Kristiansen & Larsson 2005). the case, a possible explanation for its carv-

Fig. 7. Bronze spearhead of type Hulterstad from Rynge, Scania (left), and Hulterstad, Öland (right), scaled down proportionally to fit the carved spearhead from Litsleby. Both objects match the carving reasonably well but the Hulterstad-type spearhead provides a better fit, suggesting that this was the template used for the re-carving of the Litsleby spearhead. (Com- parative analysis by Christian Horn, Gothenburg University)

Adoranten 2017 71 Fig. 8. Period III spearhead found in a Bronze Age burial mound at Hulterstad, Öland. (Photograph: Swedish History Mu- seum – SHM) ing is that it represents the most important that the objects were perceived as divine, spear of all, Odin’s spear ‘Gungnir’, or its or as gifts from the gods. During the Late Bronze Age precursor. Bronze Age, new mythological ideas may then have been established, which would This readings chime with current interpre- have called for the depiction of a god-like tations of the human figure as Odin, the carrier of metal weapons, along with other, supreme Norse god. However, the spear less exalted figures (e.g. heroes and warri- carving may also be interpreted as an illus- ors) armed with simpler, more schematically tration of a non-anthropomorphic religious depicted spears as well as other weapons. belief dating to the Late Neolithic and Early Seen from this perspective, the spear en- Bronze Age, which focused on weapons gravings may suggest that a ‘weapon cult’ made of new shiny materials: copper and would have emerged in the Late Neolithic bronze. These new materials may have and Early Bronze Age; this would have been thought to radiate so much power evolved into a fully articulated warrior ide-

Fig. 9. Period V spearhead from Nynäsvägen, Kalmar. This provides a possible template for the third spearhead from the Litsleby carving. (Photograph: Swedish Historical Museum – SHM)

72 Adoranten 2017 ology in the Late Bronze Age and Iron Age. tion Imaging (RTI) has namely shown that This ideology seemingly informs the myths the original spearhead from Finntorp was and tales of the Nordic sagas, which narrate redesigned on two occasions, and was also heroic deeds and supernatural feats that carved before the human figure which is may have emerged from a body of mytho- now its bearer (Horn and Potter in press). logical beliefs ultimately dating to the Late Both the carving process and the engraved Bronze Age, 1100 – 550 BC (Bertilsson 1987). motifs display striking similarities with the Litsleby panel. However, both weapon and human figure differ stylistically from the The spear and warrior Litsleby ones as the spear has a curved shaft from Finntorp (Tanum, Bohuslän) and its heads most likely represent flint A question arising at this point is whether points (Fig. 10, 11 and 12). In his discussion the re-design of the spearhead on the of the “pagan” antiquities from Tanum, Litsleby panel is unique. Interestingly, an Montelius (1874-1879: 354 - 358) describes examination of southern Scandinavian rock flint spearheads from Early Bronze Age art shows that this is not the case, for other cairns at Arendal, a village in northern examples of spear carvings and re-carvings Tanum. These objects may date to the end can be identified, although none are identi- of Late Neolithic or the beginning of the cal to the remarkable Litsleby panel. In par- Early Bronze Age, 2100 -1700 BC and are ticular, a rock carving from Finntorp, Tanum thus several hundred years older than the shows a spear with a head looking different spearheads carved on the Litsleby panel. from the one from Litsleby but probably ac- Moreover, the body of the Finntorp spear- complished in a similar way. Recent research man consists of concentric circles represent- carried out using Reflectance Transforma- ing a shield akin to that depicted on the

Fig. 10. Photograph of the central section of the rock carving from Finntorp (Tanum, Bohuslän) showing the great spearman and the so-called Wismar warrior. This human figure overlies a spear with slightly curved shaft and an angular arrow- like head. (Source: SHFA. Photograph: Bertil Almgren)

Adoranten 2017 73 Fig. 11. RTI-image (A) and drawing (B) of the Finntorp spearhead showing three spearheads of different length and design. (Photograph and drawing courtesy of Richard Potter and Christian Horn)

Wismar horn, which, according to Kristian- period albeit the ships could have been sen & Larsson (2005), dates to Period I of carved later than the human figure with the the Nordic Bronze Age, 1750 – 1500 BC (Fig. shield-shaped body. Additionally, the spear 10). This means that the Finntorp spearman shaft with the original tip appears to have could be significantly older than the Litsleby been carved earlier because the human fig- “Spear God”. According to Horn and Pot- ure cuts over it (Fig. 10). ter (in press) based on a recently performed documentation using RTI (Fig. 11), however, Regardless of the chronology of the the Finntorp shield resembles Watenstedt- Finntorp spearman, the early chronology of type shields, for which several comparanda the spear seems supported by the nearby exist in Spain and the British Isles; they are depiction of an anthropomorphic figure dated to Period II-III, which would be con- driving a plough drawn by a pair of oxen. sistent with the chronology of the Wismar Similar engravings are found at other Euro- Horn, now also modified in a later study by pean rock art locales including Monte Bego, Kristiansen (2012. The Wismar Horn shows France, and Valcamonica, Italy (de Lumley et ships with interned bows as well as ships al 2003: e.g. 350, 381 and 381), Anati 1976: with stylized horse-headed bows. Based 74, 84 - 85). Here, they have been dated to on the palaeodating of the Bronze Age the III and early II millennia, 2900 - 1700 BC, shoreline, ships with horse heads were first in the local Chalcolithic and Early Bronze carved in Bohuslän and Uppland around Age (Arcà 2010: 21-23). If we conclude the 1400 BC, in the advanced Period II (Ling above, a possible dating of the spear at 2008; Ling 2013). This type of ship is also Finntorpshällen appears to be the transition displayed on the Finntorp panel, thus sug- phase between Neolithic and Bronze Age, gesting that the carving may date to this around 1800-1700 BC, assuming the hy-

74 Adoranten 2017 Fig. 12. Flint spearheads of different types that could have provided templates for the re-carved spearheads from Finntorp. All objects shown here are dated to the Late Neolithic / Early Bronze Age, 2100 – 1750 BC (cf. Montelius 1917). (Photo- graphs: Swedish History Museum – SHM) pothesis that spearheads are in flint. After spearhead and a long, narrow shaft (Fig. that, the human figure with circular body 13). The analysis of the 3D model of the and the horse headed ships has been added scene obtained using SFM shows that this during period II, 1500-1300 BC. The plough spearhead resembles the largest spearhead scene appears to supports the early dating. from Litsleby. The comparison was made using the same methodology applied to the Litsleby spearhead carvings including their The waving spear proportional downscaling. To carry out the from Kalleby (Tanum, Bohuslän) comparison, a Långaröd-type spearhead Another relevant example has been dis- from Scania was first superimposed on the covered on a rock art panel from Kalleby Kalleby carving, but its shape proved to be showing three human figures playing more slender than the carved spearhead on bronze lures and two additional, larger the lower part of the blade (Fig. 14, left). A figures above them on the panel. The first Valsømagle-type spearhead was then super- is horned and carries a scabbard with an imposed on it, and this matched the carving artfully designed winged chape (i.e. the extremely well (Fig. 14, right). As for the metal point of the scabbard) dating to the spear shaft, it could be argued that it was Period V of the Nordic Bronze Age based intentionally carved as thin and undulat- on comparisons with archaeological objects ing to illustrate the ability of the spear to (cf. Coffey 1913, Fig. 67). The second, and sway, perhaps as a reference to Odin’s spear larger, human figure holds a hammer-like Gungnir, or «The Swaying One». weapon in his raised right hand. To the left of this figure, an upstanding spear has been carved comprising a well-designed

Adoranten 2017 75 Fig. 13. 3D model of the rock carving from Kalleby (Tanum, Bohuslän) showing an axe bearer and a spear with a slender, waving shaft. (Source: SHFA: Photo: Ellen Meijer)

Fig. 14. From left to right: The Valsømagle-type spearhead from Falköping; the same spearhead scaled down and The spear outlines superimposed on the carved spearhead from Kalleby; from Tuna (Bälinge, Uppland) the carved spearhead, and a spearhead of the Långaröd type from Scania superimposed on the carved spearhead. Another example of the of the spear’s sig- (Comparative analysis by Christian Horn, Gothenburg nificance during the Nordic Bronze Age, University). is the spear carvings from Tuna (Bälinge, Uppland). On this panel a number of natural striations exist, to which spearheads have artfully been added, if in a rather coarse manner (Janson 1960) (Fig. 15). It is noteworthy that two of the spears appear to be held by previously unrecognised human figures with raised arms (Fig. 16). The spearheads are simply outlined and show a relatively wide base and rounded shoulders, which overall give them a roughly trian- gular appearance (Kjellén 1976; Kjellén &

76 Adoranten 2017 Fig. 15. The spear carvings from Tuna (Bälinge, Uppland) (Photo: Harald Faith-Ell, after Kjellén 1976).

Hyenstrand 1977); however, they lack fer- of the panel are reminiscent of Early Bronze rules or any other features that could aid Age objects, with the one sitting at the comparison with archaeological objects. It bottom resembling Valsømagle-type spear- is, therefore, difficult to identify the kind heads from Period IB, ca.1600 BC of the of spearhead that the carvers intended to Nordic Bronze Age (Vandkilde 1996). depict with any degree of certainty. As for the two spearheads to the left of Modern recording using SFM has not yet the panel, the bottom one has a rather been attained for the Tuna carvings. The short shaft and a head shaped differently sole documentation available consists of from the others, which may be taken to black-and-white photographs dating to represent a Late Bronze Age weapon of the late 1940s, as well as rubbings made by the kind schematically depicted on a rock Dietrich Evers in the 1970s. Due to heavy art panel from Åby (Tossene, Bohuslän). It overgrowth of lichens, the rock face would has been suggested that this might be the now require extensive cleaning prior to doc- depiction of a real spearhead found in a umentation with SFM or any other method. nearby tomb there (Goldhahn 2014). The Although the weapons are highly stylized, current spear at the Tuna carving seems the four spearheads engraved to the right to be superimposed on one human figure

Fig. 16. Rubbing of the central section of the Tuna panel with the spearheads. (Source: SHFA. Rubbing: Dietrich Evers)

Adoranten 2017 77 and held by another figure with a horned Foppe di Nadro, Cemmo and Corni Fres- head. The spear above this scene has an chi, are well known and well documented extension added to the shaft that appears (Anati 1976, Sansoni et al 2003). The weap- to be held by a third human figure, while ons depicted there are daggers of differ- its head differs from the others in that its ent kinds, including halberds. Here there shoulders are straighter and a short crossbar are similarities with some rock carvings in has been depicted where one would expect Scandinavia, for example, with Madsebacke to see a socket. Such a type of spearhead in Bornholm, where a halberd have been is unknown in Scandinavia and could be depicted (Bertilsson 2014), with Järrestad later than the four spears to the right of the where some palstaves have been depicted panel (see above). Objects of this kind are (Bertilsson, C. 20115) and with Kasen Ryk in called “barbed spearheads”, and date to Tanum with a few daggers (Fig. 17 below). 900-800 BC in the British Isles (Davis 2015, Nos. 1301 and 1370). Pending a more de- Actually, there is a specific carving in Valca- tailed documentation using modern meth- monica that shows similarities to the Tuna ods, it can preliminary be suggested that at panel, namely at Dos Costapeta where pol- least four, and possibly five, of the spears ished and expanded striations, on otherwise were engraved on the Tuna panel before smooth rocks, at Dos Costapeta have been any of the human figures were added. This spearheaded during the Bronze Age (Fig. is, thus, another example illustrating our 18 and 19, Sansoni et al 1999). This dem- hypothesis that «in the beginning there was onstrates that similar types of weapons as the spear». in Scandinavia were depicted in the Alpine area. In Valcamonica, the carved arms are without carriers seam to appear from Cop- Spears in Valcamonica, Italy per Stone Age or Late Neolithic until Early The question is whether this habit of depict- Bronze Age, from approx. 2500 till approx. ing individual weapons was distinctive to 1200 B.C., while the first warriors do not oc- Sweden or if it also occurred in other rock cur until Middle or Late Bronze age. It have carving areas of Europe at this time? One been suggested that e.g. halberds could such area of particular interest in this con- have represented deities and their “quali- text is the rock carvings area in Valcamon- ties” (like thunder and seasonal changes), ica, northern Italy, between which there has but also to be depicted to represent weap- been demonstrated similarities between ons for the offerings (Fig. 20). In a second different carvings motifs (Sansoni 2015). phase, the gods might also have been That weapons without carriers are depicted depicted as human-likes carrying weapons, on rock carvings there, including Valtellina, but always with the possibility to depict

Fig. 17. Dagger depicted on rock carving at Ryk Kasen in Tanum, Sweden. Source: SHFA. Photo: Bertil Almgren.

78 Adoranten 2017 Fig. 18. Engraved Bronze Age spearheads at Dos Costapeta in Valcamonica. The spearheads have been placed on natural ice striations that have been expanded. Courtesy and photo: Umberto Sansoni. Fig. 19. Another engraved Bronze Age spearhead at Dos Costapeta in Valcamonica where the spearheads have been placed on an expanded natural striation. The spear being held by an antropomorphs that existed before it. Courtesy and photo: Umberto Sansoni.

Adoranten 2017 79 Rock art is an expression of the social value of the objects being engraved, and it may signify a desire to possess the objects. Al- though warrior scenes can be complex and the message that the carver(s) is trying to convey is not always immediately evident, it is still possible to suggest that carvings do not merely depict mundane representations of everyday objects. Instead, the act of carving objects on rock surfaces may indicate that they held special significance in cult practices. If we examine carving practices from an historical perspective, we may conclude that the objects displayed on Bronze Age rock carvings reflect core ideological values of the society that expressed them, such as weapon cult, ideas of warfare, and perhaps the glorification of conflict and violence. Importantly, such values were still dominant in the Viking Age (Bertils- son 1987). It can perhaps be suggested that this ideology became institutionalized over time, and informed both social hierarchies and religious beliefs to which all members of early Scandinavian society would have been exposed from birth. Such ideas would Fig. 20. Engraved spears at Luine, Valcamonica, Italy with have been reaffirmed through recurrent so- spearheads and axe-blades (the upper and longer possibly cial action and the recounting of mythologi- a halberd) of Early Bronze Age types (cf. Montelius 1895, cal narratives. Pl. 3:14, 9:2 and 3:18, 34:19) (Depiction with bi-chromatic technique by Emmanuel Anati, Anati 1982) If we accept that the rock art reflects a new ideology introduced in southern Scandi- them only with reference to specific weap- navia in the Early Bronze Age, and further ons (Sansoni 2015). developed with the establishment of personified deities during the Late Bronze Age, we must also presume that society would Concluding remarks have felt the need to affirm the new ideo- The four case studies from Sweden dis- logical order most acutely during the initial cussed here constitute remarkable examples phase, and also at later, significant junctures of the life histories of ideologically and main the evolution of the belief system. This terially important Bronze Age spearheads, may be the reason behind the abundance and of the complex, long-lived biographies of images of weapons and warriors on the of their rock carvings (Kopytoff 1984, Gos- rock carvings, which mark the Early Bronze den & Marshall 1999, Melheim 2013, Ling Age beginnings and Late Bronze Age de- & Bertilsson 2016). The comparison made velopments of a cultic tradition spanning with Valcamonica clearly demonstrates that the best part of two millennia. I argue that similar depictions appear also there indicat- it is this tradition that gradually morphed ing the existence of a shared ideology and into the mythological narratives related in pantheon in the Early Bronze Age. More the Nordic sagas – narratives that remained detailed analysis focussing on that issue will influential long after the practice of carving be performed to further demonstrate this. weapons on rock panels had ceased to exist.

80 Adoranten 2017 It so seems that the cult of weapons dis- at Sacred Texts: Sagas and Legends: The Po- cussed in this chapter preceded the materi- etic Edda.) alization of the gods wielding them – gods Bengtsson, L. 2014. Rock Art Ships as a that, if we accept the chronology proposed Method for Dating. In Counterpoint: es- above, only began to appear in the Late says in Archaeology and Heritage Studies in Bronze Age, hundreds of years after the in- Honour of Professor Kristian Kristiansen. S. troduction of bronze spears in Scandinavia. Bergerbrant & S. Sabatini, eds.. BAR Interna- It is also possible that ideas of a supernatu- tional Series 2508, 2013 (pp. 253 – 258). ral spear-wielder may have formed in this early period, only to coalesce into the ‘clas- Bertilsson, C. 2015. Front cover photo. In sic’ image of Odin as the spear-god par ex- Picturing the Bronze Age. Swedish Rock Art: cellance from the Late Bronze Age onwards. Volume 3. Skoglund, P., Ling. J & Bertilsson, U. eds. Oxbow Books. Oxford & Phildelphia. Ulf Bertilsson Bertilsson, U. 1987. The Rock Carvings of Svenskt Hällristnings Forsknings Arkiv - Northern Bohuslan. Spatial structures and Swedish Rock Art Research Archives, social symbols. Stockholm Studies in Archae- University of Gothenburg ology 7. Stockholm. www.shfa.se Bertilsson, U. 2015a. Examples of ap- [email protected] plication of modern digital techniques [email protected] and methods - Structure for motion (SFM) +46702515096 and Multiview Stereo (MVS) for Three-dimensional Documentation of rock carvings in Tanum creating new opportunities for interpretation and dating. In Prospects for References prehistoric Rock Art research. F. Troletti ed. Almgren, O. 1927. Hallristningar och XXVI Valcamonica Symposium 2015 (pp.57- kultbruk. Bidrag till belysning av de nor- 62). Capo di Ponte: Edizione del Centro, diska bronsaldersristningarnas innebord. 57-62. Kungl. Vitterhets Historie och Antikvitets Bertilsson, U. 2015b. From folk oddities Akademien, Handlingar 35. Stockholm. and remarkable relics to scientific substra- Anati, E. 1976. Evolution and Style in tum: 135 years of changing perceptions Camunian Rock Art. An Inquiry into the For- on the rock carvings in Tanum, northern mation of European Civilization. Archivi 6. Bohuslan, Sweden. In Picturing the Bronze Ediszioni del Centro. Capo di Ponte. Age. Skoglund, P., Ling, J.. and Bertilsson, Anati, E. 1982. Luine collina Sacre. U. eds. Swedish Rock Art Series: Volume Edizione del Centro. Capo di Ponte. 3(pp.5-20). Oxbow Books. Oxford and Phila- Aner, E., Kersten, K. 1976. Die Funde der delphia. älteren Bronzezeit des nordischen Kreises in Bertilsson, U., Bertilsson, C., Meijer, E. Dänemark, Schleswig-Holstein und Nieder- 2014. Dokumentation av Världsarvets Häll- sachsen. 2. Holbæk, Sorø und Præstø amter. ristningar med 3D teknik – Structure from København und Neumünster. Motion. Etapp 1. Pilotstudien på Aspeber- Arcà, A. 2010. Comparisons between the get 2014. Rapport Svenskt Hällristnings For- most ancient phases of the two main rock sknings Arkiv. 106s. Göteborg. art poles, Mount Bego and Valcamonica. L Bertilsson, U., Ling, J., Bertilsson, C., ´arte rupestre delle Alpi/Alpine rock art. (pp. Potter, R., Horn, C. 2017. The Kivik tomb 21-23). Capo di Ponte. - Bredarör enters into the digital arena - Bellows, Henry Adams. (Trans.). documented with OLS, SFM and RTI. In New 1923. The Poetic Edda: Translated from Perspectives on the Bronze Age. (Berger- the Icelandic with an Introduction and brant, S. & Wessman, A. eds. :289 -305.Pro- Notes. New York: American-Scandinavian ceedings from the 13th Nordic Bronze Age Foundation. Reprinted Lewiston, NY: Edwin Symposium in Gothenburg, Sweden in 2015. Mellon Press. ISBN 0-88946-783-8. (Available Coffey, G. 1913. The Bronze Age in Ire-

Adoranten 2017 81 land. Hodges, Figgis & Co. Dublin. Johansson, S-O. & Magnusson, J. 2004. Davis, R. (2015). The Late Bronze Age Developing new technique for 3D docu- spearheads of Britain. Steiner Verlag. Stutt- mentation of rock art. In The Valcamonica gart. symposiums 2001 and 2002. Bertilsson, U. & De Lumley, H. et colaborateurs. 2003. McDermott, L. eds..: pp.125- 133. Riksantik- Gravures protohistoriques et historiques de varieämbetet. Stockholm. la region du mont Bego. Tome 14. Tende, Kopytoff, I. 1984. The Cultural Biography Alpes-Maritimes. Sevteur des Merveilles. of things: Commoditization as Process. In Zone du Grand Capelet. Zone XII. Groupes I The social life of things: commodities in à VI. Sarle Édisud. Aix-en-Provence. cultural perspective. Appadurai, A. ed. Cam- De Reu, J., Plets, G., Verhoeven, G., De bridge Cambridge Univ. Press. Smedt, P., Bats, M., Cherretté, B., De Clercq, Kaul, F. 1998. Ship on bronzes. A study in W. 2012. Towards a three-dimensional cost- Bronze Age religion and iconography. PNM effective registration of the archaeological Studies in Archaeology and History 3:1/2. heritage. Journal of Archaeological Science, København. 40(2),(pp.1108–112)1. doi: 10.1016/j. jas. Kaul, F. 2003. The Hjortspring find. In 2012.08.040 Hjortspring – a Pre-Roman Iron Age War- Domingo, I., Villaverde, V., López- ship in Context pp. Crumlin Pedersen, O. Montalvo, E., Lerma, J. L., & Cabrelles, and Trakadas, A. eds. p. 141-186. Roskilde, M. 2013. Latest developments in rock art Ships and Boats of the North, vol. 5. recording: towards an integral documenta- Kjellén, E. 1976. Upplands hällristningar. tion of Levantine rock art sites combining The Rock Carvings of Uppland. Kungl. Vit- 2D and 3D recording techniques. Journal terhets Historie och Antikvitets Akademien. of Ar chaeological Science, 40(4), (pp.1879– Almqvist & Wiksell International. Stock- 1889). doi: 10.1016/j.jas.2012.11.024 holm. Goldhahn, J. 2014. Engraved biographies Kjellén, E., Hyenstrand, Å.1977. Hällrist- – rock art and life-histories of Bronze Age ningar och bronsålderssamhälle i sydvästra objects. Current Swedish Archaeology 22, Uppland. Upplands Fornminnesförenings (pp.97–136). Tidskrift 49. Uppsala. Gosden, C. & Marshall, Y. 1999. The cul- Kristiansen, K., Larsson, T. B. 2005. The tural biography of objects. World Archaeol- Rise of Bronze Age Society. Travels, Trans- ogy 31(pp.169-178.) missons and Transformations. Cambridge Hildebrand, B. E. 1869. Till hvilken University Press. Cambridge. tid och hvilket folk bora de Svenska Kristiansen, K. 2012. Rock art and reli- Hallristningarne hanforas? AntiqvariskTid- gion – the sun journey and the Indo-Euro- skrift för Sverige 2(pp. 417-432). pean mythology and Bronze Age rock art. Horn, C. 2013. Weapons, fighters and Adoranten Vol. 2012 p. 69-84. combat: spears and swords in Early Bronze Ling, J. 2008. Elevated rock art – Towards Age Scandinavia, Danish Journal of Archae- a maritime understanding of Bronze Age ology, DOI: 10.1080/21662282.2013.838832 rock art in northern Bohuslan, Sweden. GO- Horn, C. & Potter, R. (in press). Trans- TARC Serie B. Gothenburg Archaeological forming the Rocks: Time and Rock Art in Thesis 49. Bohuslän, Sweden, European Journal of Ling, J. 2013. Rock Art and Seascapes in Archaeology. Uppland. Swedish Rock Art Series, Volume Janson, S. 1960. Hällristningen vid Tuna 1. Oxbow Books. Oxford and Philadelphia. i Bälinge socken. TOR. Uppsala universitet. Ling, Johan & Cornell, Per. 2015. Krieger Uppsala. und Felskunst im bronzezeitlichen Skan- Jantzen, D. & Terberger, T. 2011. Gewalt- dinavien. In Krieg Eine Archalogische Spu- samer Tod im Tollensetal vor 3200 Jahren. rensuche. Meller. H and Schezik, M.eds. p. In: Archaäologi in Deutschland. Nr. 4 (pp. 265-268.Lande- samt fur Denkmalplege und 6-11), Theiss, Stuttgart 2011, ISSN 0176- Archaologie Sachsen – Anhalt. Landesmu- 8522, seum fur Vorgeschichte. Halle (Saale).

82 Adoranten 2017 Ling, J., Bertilsson, U. (2016). Biography Qawasmi, J., Alshawabkeh, Y. & Remondino, of the Fossum panel. Adoranten Vol. 2017. F. eds. (pp.251-266). Amman, Jordan, 13-15 Maraszek, R. 2015. Kampf- und Jagd- March 2011.CSAAR Press. waffen: Lanzen und Speere im bronzezeitli- Vandkilde, H. 2011. Bronze Age Warfare chen Europa. In Krieg - eine archäologische in Temperate Europe. In Sozialarchäologis- Spurensuche. Meller, H. & Schezik, M. eds. che Perspektiven: Gesellschaftlicher Wandel p. 289-293. Landesamt für Denkmalpflege 5000-1500 v. Chr. zwischen Atlantik und und Archäologie Sachsen-Anhalt. Landes- Kaukasus. Hansen, S & Müller, J.eds. Vol. museum für Vorgeschichte. Halle (Saale). Archäologie in Eurasien 24 DAI & Darmstadt Melheim, L. 2013. An epos carved in (pp. 365-380) : Verlag Philipp von Zabern. stone: three heroes, one giant, and a cosmic Vogt, D. 2011. Rock Carvings in Østfold task. Counterpoint: essays in archaeology and Bohuslän, South Scandinavia. An inter- and heritage studies in honour of professor pretation of Political and Economic Land- Kristian Kristiansen. p. 273-282. Berger- scapes. The Institute for comparative Re- brant, S. & Sabatini, S. eds. BAR Interna- search in Human Culture. Novus Press. Oslo. tional Series 2508, 2013. Ancient origins. 2013-2015. Reconstruct- Montelius, O. 1874-1879. Bohuslänska ing the of humanity’s past. http://www. fornsaker från Hednatiden. Bidrag till ancient-origins.net/ancient-places-europe/ kännedom av Göteborgs och Bohusläns alken-enge-and-buried-army-002120 Fornminnen och Historia. Första Bandet. Svenskt Hallristnings Forsknings Arkiv: p.321-422. Stockholm. www.shfa.se Montelius, O. 1895. La Civilisation Primi- tive en Italie. Depuis L’Introduction des Me- taux. Premiere Partie. Italie Septentrionale. Texte et Planches. Imprimerie Royale. Stock- holm. Montelius, O. 1917. Minnen från vår forntid. I Stenåldern och Bronsåldern. Stock- holm. Nordberg, A. 2013. Fornnordisk religions- forskning mellan teori och empiri. Kulten av anfäder, solen och vegetationsandar I idéhistorisk belysning. Acta Academiae Re- giae Gustavi Adolphi CXXV1. Uppsala. Sansoni, U. 2015. Alpine and Scandina- vian rock art in the Bronze Age: a common cultural matrix in a web of continental in- fluences. In Picturing the Bronze Age. Swed- ish Rock Art: Volume 3. Skoglund, P., Ling. J & Bertilsson, U. eds. p.129-141. Oxbow Books. Oxford & Phildelphia. Sansoni, U., Gavaldo, S. and Gastaldi, C. 1999. Simboli sulla roccia. L’arte rupestre della Valtellina centrale dale armi del Bronzo aisegni Cristiani. Edizione del Cen- tro. Capo di Ponte. Sevara, C. & Goldhahn, J. 2011. Image- based Modelling of the Present Past: Building 3DModels of Archaeological En- vironments from Digital Photographs. In Conference Proceedings from Digital Media and its application in Cultural Heritage. Al-

Adoranten 2017 83