Ships on Bronzes. A Study in Bronze Age Religion and Iconography (Kaul 1998)

Much unlike many other prehistoric cultures, and especially, Bronze Age cultures in central Europe, the Nordic Bronze age provides a broad number of rich illustrations on metal and stone which, since their discovery, have provoked interpretations regarding their religious or ritual meaning, as they provide a unique window into a remote and somewhat obscure past. For Bronze Age Scandinavia, ships seem to be of particular importance, and this post discusses one of the primary contributions to the interpretation of Bronze Age iconography on Danish razors (with an example in the following piece from Skane).

Razor with 2 folded ship(s) (left, top and right edge, one above the other) with "crew strokes", two circles ("suns") and two bands of S-shaped figures ("extremely stylized sun horses") (Wikimedia)

1. Background

Flemming Kaul's study of ship iconography on such Bronze-Age razor knifes from Danmark represents a seminal example for how the creation of searchable data bases and their statistical evaluation can facilitate, support and guide the traditional qualitative methods of qualitative analysis in the Humanities. In the following, I try to provide an overview over the original study (which has been a basis for numerous subsequent publications not taken into account here), as well as a critical review of the methodology and (possibly) conclude with some remarks regarding alternatives to his interpretation.
Jensen (1993, p.54) gives a nice description of the state of the art regarding the interpretation of Bronze Age iconography prior to Kaul's study (emphasis by me):

Die Axt, die von Pferden gezogene Sonne, das Schiff, der Fisch, das Sonnenzeichen usw, sie alle hatten eine Bedeutung, die in Mythen berichtet worden sein muss und dargestellt in heiligen Zeremonien. Aber eine zusammenhängende Vorstellung dieses Universums von Göttern und Menschen können wir uns nicht machen. Wir können bloß konstatieren, daß die Symbole wieder und wieder gebraucht wurden und ständig in neuen Zusammenhängen. Und wir wissen auch, daß sie in ganz Skandinavien gebraucht wurden, wo sie Ausdruck für ein gemeinsames Bedeutungssystem gewesen sein müssen ...

In order words, by this time, the basic iconographic elements had been identified, but most researchers had basically resignated in their efforts to come to a principled interpretation and focused on descriptive analysis. Kaul (p.11-16, 49-57, 66-72) gives a brief overview over earlier approaches. Broadly speaking, many were based on analogies with presumably related traditions, including the Edda (same cultural sphere, about 1500 years later), the Ugaritic Baal cycle (Syria, roughly contemporary), the Egyptian sun ship (Egypt, roughly contemporary), or religious and iconographic developments in early Greece (introduction of Apollon, bird iconography on geometric and Mycenean ceramics). The fundamental methodological problem of every analogy-based approach is that these analogies involve cultures whose chronological and/or geographical relation with Nordic Bronze Age is remarkably remote, at best. A related school of thoughts discusses ties to reconstructed systems, e.g., continental Celtic mythology (partially reconstructed, neighboring cultural sphere, about 1000 years later), or Dumézil's reconstructed Proto-Indo-European mythology (reconstructed from [mostly] Indo-Iranian, Greek and Germanic sources). While these systems may be closer related, indeed, they are not directly attested and should thus be taken into consideration with great care only.

The first and foremost principle of Kaul's approach is thus to prioritize the immediate Bronze Age evidence: "Instead of looking around for traces of the names of gods and specific myths in time and space to throw light on Bronze-Age religion, it ... will be most fitting to perceive, assess and understand the material and pictorial evidence about Bronze-Age religion for what they are and not by comparing them with other later or distant religious systems." (p.15)

What is remarkable about Jensen's quote from above is that along with the limitations of our current understanding, he points out that "the symbols were used over and over again, and always in new combinations". But repetitive co-occurrence represents an ideal basis for a combinatoric analysis and precisely this is Kaul's innovative methodology: Where earlier approaches had to rely on analogy and/or cautious abductive interpretation, a database was created and statistically evaluated [1].

[1] Kaul describes results of his statistical analysis. He does not, however, report any correlations or significance scores, and indeed, as I show below, many correlations he observed are statistically not significant -- this is an important methodological flaw that may give rise to misinterpretations. A reconstruction based on insignificant observation requires additional support in order to be acepted as anything more substantial than a scientific hypothesis (see Sect. 4).

2. Empirical Analysis

Using a database, Kaul could develop a coherent (though necessarily incomplete) assessment of the Bronze Age universe of men and gods in an empirical fashion on the basis of correlation analysis.

As for the data set considered in this study, Kaul is aiming for an exhaustive analysis of a manageable number of objects and hence focuses solely on bronze objects with ship depictions from the Danish Bronze Age. Ship imagery from Southern Swedish and Norther German bronzes is excluded (for geographic reasons) as well as the rich repertoire of Scandinavian rock carvings (which may include depictions of cultic practices rather than actual myths, cf. p. 49ff., 196f.). Bronze objects without recognizable ships were also excluded (as the ship is taken to be leitmotiv for Bronze Age imagery, cf. p. 84f.). The majority of the remaining images is from razors, and this association between a specific iconography and a type of object indicates that, indeed, a coherent set of beliefs may be covered by this study.

In the analysis of this data set, then, the respective constituents (entites) and their features are identified and discussed, these include the ship itself (p.165-185), its direction (p.185-187), and assotiated motives (p.188-256). An important novel feature considered is the direction of movement, that Kaul motivates from an analysis of the Trundholm Sun Chariot (p.33f). The Sun Chariot features a horse pulling a sun disc, both being put on wheels. It is of high importance here as the ship is frequently accompanied by the sun-horse and the clear difference in ornamentation of both sides of the sun disc that implies a religious or ritual significance of direction: Both sides are ornamented in a similar fashion (hence both sides are meant to be shown), but one being gilded whereas the other is plain bronze. Kaul (p.33-34) concludes that the horse moves from left to right over day (showing the gilded side) but from right to left over night (showing the blank side):

If we look on the side of the sun-disc that is gold-covered, and which must be considered to be a symbol of the radiant sun, we notice that the horse is facing to the right and must be assumed to be "drawing", or moving together with, the sun to the right. This is of course the direction of the travel of the sun seen from the earth, where it happens to go from east to west. If we turn the sun-image round so that we can see the dark, non-gold-covered side of the sun-disc, then the horse is facing left. ... In our physical world, however, the sun never moves to the left; but it might well have been thought to do so in Bronze-Age ideology, namely at night, when it was drawn from west to east under the ground or perhaps also under the sea, in order to begin at the same starting-point at the next morning. ... [I]t should be emphasized that no trace of gold foil was found at the "dark side" of the disc ...

day side (Wikimedia)

night side (Wikimedia)

The sun-horse was deposited around 1400 BC and is thus older than most other bronze objects in the data set. But as Scandinavian ship iconography was intensified by contacts with the Urnfield culture and their "Vogelsonnenbarke" motif (p.277-284), it is possible that the ship acquired its role as carrier of the sun (along with the horse) precisely as a result of this contact and that the earlier, indigenuous idea had been the horse as the sole carrier of the sun. The latter would be in line with other Indo-European (esp. Indo-Iranian, but also later Germanic) conceptions. We may also compare Greek mythology where the sun is drawn by horses at day, but traveling in a bowl on the Oceanos at night -- apparently correlating with the same ideas about the direction of movement.

Regular (day) view on Helios

(Heracles traveling in the) Bowl of Helios

Determining the direction of ships is difficult, given the relatively high degree of abstraction of the Nordic razor engravings. Kaul employs the following criteria:

• "Crew strokes". Lines "on board", sometimes with dots ("heads") on top. Assuming that these represent rowers, they may be slanted towards the direction of moving (p.73,185).
• Keel extension of fore stem. From Mediterranean rams, Kaul concludes that keel extensions, even those not being staight, but pointing upwards, mark the fore stem. This is also confirmed by the crew strokes: Where the crew strokes indicate the direction, the raised keel extensions is always (12/13 observations, one outlier on an atypical piece) the fore stem. Accordingly, this is taken as an indicator, even without crew strokes (p.182,185).

Out of 415 catalog items, 97 (23%) can be confirmed as sailing to the right ("day") ships, 26 (6%) as sailing to the left ("night") ships. For the catalog, however, only the latter criterion was employed (p.247).

As for the associated symbols, circles are generally interpreted as sun symbols and represent the most frequent motif cooccurring with ships on bronzes (p. 195ff.). Out of 126 objects with sun images, 41 occur with ships sailing to the right, none with ships sailing to the left, thereby confirming their interpretation as day ships (with sun) and night ships (invisible sun), respectively (p.186, 195). Based on Kaul's numbers, I can confirm that the difference is significant (chi², p<.0005).


Sun-horses, i.e., horses pulling a sun symbol, are found 11 times on bronzes with ships (p.200), 10 pulling the sun from left to right, 1 the other direction (p.209). Kaul interprets the majority of them as being not directly on a ship, but before (or, less likely) behind (p.206). Accordingly, they may represent a phase in the cycle of the day travel of the sun, where they take the sun from the ship or bring it there. They never co-occur with night ships (p.186). It is important to note that the correlation with movement to the right, even if exceptionless, is not statistically significantly different from the distribution of day and night ships (Fisher, p>.05). 
The absolute number of sun-horses would be substantially larger if more abstract representations are counted, by which the sun-horse is reduced to an S-shaped symbol, partially associated with a dot (p.209-215). Although still slightly more frequently associated with day ships (86%=30/35, 5 with night ships) than to be expected from the relative number of day ships (79%=97/123), these results are not statistically significantly different from the distribution of day and night ships (chi², p>.05). Remarkably, though, the difference from the distribution of sun-ships is statistically significant (Fisher, p<.005). Kaul's conclusions from this data are thus empirically not well-founded, and in particular, his data on the S-shaped symbol cannot be used to substantiate his claims on the association between sun-horse and day ship.

Similar to sun-horses proper, a mushroom-like symbol of unknown function seems to be exclusively associated with day ships (9 objects with day ships, 2 objects with unrecognizable direction, none with night ships, p. 188). The correlation is exceptionless, but the difference is not statistically significantly different from the distribution of day and night ships (Fisher, p>.05). Kaul (1998, p. 189f.) interpreted it as a cult axe (rotated by 90°) or standart echoing the form of two consolidated ship stems (hence a "folded" ship, p. 194), earlier accounts saw it as a sail (Müller 1921, unlikely), or a sacred tree (Almgren 1927). More recently, Kaul (2014) interpreted it as a symbolic depiction of the travel of the sun, an interpretation that seems to follow Gelling & Davidson's (1969) idea that it may represent the vault of the heavens. Personally, I think that the interpretation as a tree (or another plant) is much more likely than the cult axe, in particular in the light of its similarity with the capital of Ionic columns that also evolved out of stylized trees (an early predecessor can be found at a late Bronze Age cauldron from Cyprus, now at the Neues Museum Berlin, placed between two opposing Cherubim/Lamassu where usually, a tree of life would be expected). In fact, the Germanic (and Proto-Indo-European) concept of a tree as bearer of the universe would be capable to harmonize this idea with the "vault of the heavens" idea. The question remains, though, why this tree as the axis of the universe would be carried around on a ship.

Unlike individualized sun-horses, horse-headed stems occur both with day (80%, 28/35) and night ships (20%, 7/35). This ratio is basically identical to the relative frequency of day (79%, 97/123) and night ships (21%, 26/123), and may simply indicate that the ship has taken over certain qualities from the earlier horse (p.186,200). Their distribution is statistically significantly different from that of sun-ships (Fisher, p<.005). S-shaped fore stems may represent stylized horse-headed stems (p.210) and have a similar distribution, with 82% (55/67) correlations with day ships and 18% (12/67) with night ships (p.186). Again, their distribution is statistically significantly different from that of sun-ships (Fisher, p<.005). Accordingly, there is no empirical basis to claim that horse-headed stems or S-shaped stems are specifically correlated with day ships and/or the sun.

The fish occurs on 9 objects (p.216), with 7 day ships and 2 night ships (p.186). This distribution is not significantly different from the total distribution of day and night ships (Fisher, p>.05), but its difference from the distribution of sun-ships is marginally significant (Fisher, p<.05). Kaul's interpretation is, however, much more specific. Based on a razor where a bird of prey feeds on a fish, he develops the plausible hypothesis that the sun is transferred from the fish to the bird of prey in the presence of the ship (p.217ff.)

Like the fish, the distribution of the snake is not signifiantly differently from the total distribution of day and night ships (Fisher, p>.05), but its difference from the distribution of the sun-ship is marginally significant (Fisher, p<.05): 5 snakes occur with day ships, 2 with night ships (p.186). Unlike the fish which may always be upside-down (and hence, heading the other way), the direction of movement of some snakes can be recognized (in particular of hybrid horse-snakes that also have legs, p. 223). From 6 snakes with recognizable and unambiguous direction, 5 move to the right, 1 moves to the left (p. 221-229). This is neither significantly different from the total distribution of day and night ships nor from the distribution of sun-ships (Fisher, p>.05).

The bird of prey (p.243-245, birds with a down-turned beak) are relatively rare, so that Kaul includes Swedish material into the discussion. Without a quantitative analysis, he argues that many of them stand particularly close to the fish, and one apparently is about to devour the fish. The interpretation, mentioned above already is that the bird of prey may be following the fish in the presence of the ship.

Aquatic birds (p.242-243, birds with up-turned beak) however, are considered static participants of the cycle of the sun. Kaul interprets them as being fixed on standarts or the ship's stems rather than being integrated in the "narrative", reflecting problems to come to a coherent interpretation.

As for the very rare instances of clearly recognizable human figures (p. 248-256), Kaul assumes that these may represent a relatively late stage in Bronze Age mythology development, and may mark the advent of a anthropomorph divinities. As these actually co-occur with crew strokes, they may be larger-than-life representations of human beings, but qualitatively different from ordinary crew members. A remarkable feature is that heads can be augmented with halos (if the ship moves to the right) or with helmets (once, if the ship moves to the left), but never with ordinary human features such as a nose. Given the scarcity and heterogeneity of the data, an association with solar divinities, or the celestial twins (dioskuroi), remains, however, speculative.

Out of these observations, but including additional criteria such as the relative position of the sun(s) in the images, or the sequential order and orientation of symbols, Kaul then constructs the daily cycle of the sun (p.262-265), basically by arranging selected razors into a coherent sequence and assigning them tentative temporal interpretations.

3. Interpretative Analysis

Before describing these in detail, I want, however, to point out what Kaul's numbers and my significance analysis have shown so far: We could clearly identify two different patterns of distributions, one being associated with an exclusive movement of the ship to the right (circles, possibly [exceptionless, but insignificant] sun-horse, "mushroom", humans with halo), and interpreted as being related with the movement of the sun at day, and one distributional pattern that is insensitve to the division between left- and rightward movement, resp. day and night ships (horse-headed stem, S-shaped stem, fish, snake, possibly [no data] crew strokes and humans with helmets [singleton]). In addition, narrative elements indicate that the fish may be followed by a bird of prey. Both classes show statistically significant differences, so that we can infer a basic differentiation between celestial ("day only") and general ("day or night") aspects as shown below:

Kaul's quantitative data does not provide more information than in the diagram above. In particular, we have no structured information about co-occurrences between, say, fish/snake and sun-horse. Only such data would allow us to identify entities that we can locate more precisely in the lower or middle part of the diagram. In principle, this analysis would be possible on the basis of the second volume (catalog), but it would require a full retrodigitalization of the database.

Instead of continued statistical analysis, Kaul now begins to develop his model, but again: Beyond the information in the diagram above, this is an interpretation not substantiated with statistical (empirical) data. From a logical point of view, the following discussion is an abductive interpretation, i.e., based on plausibility considerations, and hence, not logically provable.

Based on (initially) 12 selected razors with particularly "narrative" character and partial overlaps, Kaul reconstructs the following sequence of events and augments them with a tentative temporal interpretation:

Journey of the sun travelling over the sky (Nationalmuseet, slightly updated from Kaul 1998, p. 262)

Kaul (1998) explains:

Our journey with the sun on its cyclical daily journey (...) begins at the bottom left [10] with (...) a ship sailing to the left, followed by a fish swimming to the left. ... [H]igher up [1] (...) the sun is being pulled upwards and to the right by a fish at sunrise from a night-ship to a day-ship sailing to the right ... The fish is allowed to travel on board the ship [2] (...), is devoured by a bird of prey [3] ... The sun horse, in this case two sun-horses, are about to pull the sun from the ship with the aid of a cord [4] (...), and finally at the sun's highest point at noon, (...) a sun-horse has just collected the sun from the ship [5] ... Now in the afternoon, the sun is beginning to sink in the sky, and the sun-horse lands with the sun on the ship [6] ... Some time after the sun-horse has landed, the sun is taken over by the snake from the afternoon-ship [7] (...), and below can be seen yet another snake in front of this ship presumably "concealing" the sun in its spiral-curls [8] ... [A]t sunset, a bird-head's ship can be seen [omitted in the revised diagram above] (...), pulling the sun down. Finally, (...) two ships can be seen sailing to the left, the one above the other [9] ... The wheel has come full circle.

(quoted from p.262, references to catalog numbers replaced by references to the numbers in the diagram in square brackets)

One should add at this point that Kaul's analysis was subsequently extended, e.g., by clarifying his interpretation of the "crew strokes": Kaul (2005) suggested "that also the souls of the dead had an important role in this divine play, being paddlers of the sun-ship or sun-ships that every day and night travelled through all the spheres."

4. Critical Discussion

Kaul's interpretation given above implicitly presupposes that:

1. the majority of razors cover different aspects of a single coherent narrative, 
2. this narrative pertains to the daily voyage of the sun,
3. the razors used for the reconstruction provide well-defined, but partially overlapping episodes in this narrative, 
4. the sequential order of symbols implies a chronological order, and
5. the relative position of the circles allows us to infer an absolute temporal interpretation.

These assumptions are plausible, but out of the archeological record, they are unprovable. Even though appealing, this is methodologically problematic: If any of these assumptions turns out to be false, the entire reconstruction is obsolete. Personally, I am particularly suspicious with respect to assumption 2 (it could be another natural cycle correlated with the sun, e.g., that of the year), assumption 3 (if it is a daily cycle: why should the narrative on a razor be limited to a single episode in the narrative instead of selecting the main stations of the entire cycle), and assumption 5 (the relative position may be enforced by external constraints, e.g., the shape of the medium).

Even though the approach and its result seem plausible, the selection process, and especially the association of certain stations with specific times (beyond day and night) is a highly subjective undertaking. This may be a methodological issue, in particular with respect to episodes without obvious overlap. To support his analysis, a explicit evaluation is thus required in order to assess the quality of this reconstruction beyond mere plausibility considerations. Such an evaluation is, however, not performed in a systematic way.
As a criterion for intrinsic evaluation one may consider counting the number of bronzes that are support this reconstruction (i.e., all representations with sufficient episodic content to be assigned to one or several points in Kaul's reconstruction), and the number of bronzes that do not contradict this reconstruction (e.g., all representations of ships without additional symbols associated with them). Kaul (1998) does not provide these numbers, but we may note a few exceptions which may not refer to this cycle. In particular, this includes all razors with representations of humans (they are not part of his reconstruction). However, even with these numbers provided, such an evaluation remains questionable in its significance, as is not performed against an independent evaluation set, but only against the source material from which the reconstruction was built. It is possible to find additional illustrations from Northern Germany, Sweden and Norway (and some of these are covered in his discussions of the different motifs), but to compile a corresponding catalog requires several years of research -- comparable to the time Kaul took to prepare this publication (1993-1998) -- and is thus beyond reach. Subsequent extensions of Kaul's interpretation to the analysis of other objects provide such an intrinsic evaluation to a limited extend (qualitative interpretation in archeology seems to lack the concept of a baseline that this interpretation could be compared with and measured against). I remember several exhibits in different museums that may be used to support his interpretation and which were not mentioned in his discussions (e.g., at the Neues Museum Berlin), but such unsystematical observations provide merely episodic evidence, at best. At the moment, evaluation thus has to rely on extrinsic information, i.e., additional support from sources other than Nordic Bronze Age iconography. Such an evaluation, however, is beyond the scope of Kaul's monograph and his primary area of expertise.

Personally, I am enormously impressed by the scope and the results of Kaul's catalog and in-depth study, and its apparently enthusiastic reception by the community confirms that it represented an enormous methodological step ahead in the field. But as a researcher with computer science background, I also see some methodological shortcomings, most noteably

1. the insufficient mathematical formalization of the statistical evaluation (including the complete lack of significance and correlation tests) and
2. the lack of extrinsic evaluation.

As for the first point, I performed some of the significance tests using Kaul's data and could confirm the significance of some of his conclusions, but the scores also indicate the speculative nature of others. Given the limited amount of source data, the applicability of these methods is limited, but a correlation study investigating, e.g., the cooccurrence between sun and fish/snake should be possible and would be an important first step.
For the second point, extrinsic evaluation, one would need to confirm some of the conclusions of Kaul that go beyond objective quantitative measurements from the comparison with related cultures, iconographies and religions. As mentioned above when criticizing earlier analogy-based approaches, this is problematic because of considerable chronological and/or geographical gaps between the Nordic Bronze Ages and comparable material. To support the plausibility of this reconstruction, four main strategies can be employed:

1. confirming selected aspects with replication studies on different data sets from related cultures,
2. confirming conservation of selected aspects in descending or from preceding cultures,
3. confirming selected parallels with contemporary cultures, or
4. confirming selected parallels in other, unrelated cultures with similar socio-economic, political and/or ideological systems.

It should be noted that each of these strategies should be applied as they tackle different aspects of the model, i.e., iconographic parallels (1), preserved cultural heritage across different socio-economic, political and ideological systems (2), flow of ideas anong contemporary cultures on a comparable level of development, and cultural universals (4). In addition, it is necessary for (1)-(3) to identify and to confirm paths of transmission. Otherwise, any confirmed parallel may be attributed to an insignificant chance resemblance, or, even worse, confirm the author's expectations from prior knowledge that were implicitly guiding the development of the reconstruction.
An important difference between the suggested extrinsic evaluation and earlier, analogy-based approaches is that here, the reconstruction is developed before any explicit comparison with other mythological systems is performed. The basis of reconstruction is the archeological-iconographic record alone, and reconstruction and evaluation should be performed by independent teams.
I developed some ideas on extrinsic evaluation, and actually gained some insights, already, but that's far beyond the scope of this blog ;)

References
Jørgen Jensen (1993), Führer durch das Nationalmuseum. Dänische Vorzeit. Nationalmuseum Kopenhagen.
Flemming Kaul (1998), Ships on Bronzes. A Study in Bronze Age Religion and Iconography. Studies in Archeology & History 3(1&2), PNM (Publications from the National Museum), Copenhagen.
Flemming Kaul (2005), Bronze Age tripartite cosmologies, Prähistorische Zeitschrift 80(2):135–148.
Flemming Kaul (2014), Vom ewigen Kreislauf der Sonne, AiD (Archäologie in Deutschland) 2/2014

Nationalmuseet Kopenhagen: Sonnenwagen von Trundholm

Im August 2011 (ja, schon eine Weile her) habe ich Kopenhagen besucht und bei dieser Gelegenheit auch das Nationalmuseum, eine großartige und umfassende Sammlung zur dänischen Vergangenheit, so umfangreich allerdings, dass ich mich auf die Abteilung für Ur- und Frühgeschichte beschränkten musste. Die wichtigsten Gründe für diesen Besuch waren der Sonnenwagen von Trundholm und (natürlich) der Gundestrup-Kessel, aber der überragende Umfang der Sammlung hat mich wirklich überrascht.

Hochspannend in diesem Zusammenhang war insbesondere die Präsentation des Sonnenwagens von Trundholm.

Sonnenwagen von Trundholm, Tagesseite (Wikimedia)

Der Name "Sonnenwagen" ist zunächst etwas irreführend, denn tatsächlich ist es kein Wagen, sondern die auf Räder montierte Darstellung eines Pferdes, das die Sonnenscheibe zieht. Im weiteren Sinne ist die Bezeichnung zwar nicht unrichtig, aber sie weckt völlig falsche Assoziationen mit etwa dem Sonnenwagen des Helios: Auch bei den Griechen wurde die Sonne von Pferden über den Tageshimmel gezogen, jedoch anthropomorph umgedeutet und (wie ein Mensch) in einem Streitwagen.

Abgesehen von der einmaligen Möglichkeit, ein so berühmtes Stück einmal aus der Nähe zu sehen, hat mich aber überrascht, wie es in einen weiteren ikonographischen Kontext eingebunden wurde, was die Teil-Rekonstruktion einer komplexen Sonnenmythologie ermöglichte. Konkret liegen der Präsentation dort Forschungsarbeiten von Flemming Kaul zugrunde, der den Sonnenwagen in Zusammenhang mit spätbronzezeitlichen Darstellungen von Sonnenpferden und -schiffen brachte, die sich v.a. auf nordischen Rasiermessern finden. Im Ergebnis steht die Rekonstruktion einer Sonnenfahrt, bei der die Sonne zu verschiedenen Tageszeiten von unterschiedliche Tieren und weiteren Objekten begleitet wird.
Im ersten Moment hat mich diese Interpretation sehr beeindruckt, aber nicht vollends überzeugt. Das System scheint in sich plausibel und auf einer soliden empirischen Basis (und nicht etwa durch Analogieschlüsse) entwickelt worden zu sein, aber die Interpretation ist zu detailliert und zu "glatt", um nicht Skeptizismus hervorzurufen -- bei mir jedenfalls. Skeptizismus vor allem auch deswegen, weil rein gar nichts in der historisch überlieferten Mythologie oder deren Nachklängen in späteren Märchen und Riten auf eine so starke religiöse Fixierung auf die Tagesreise der Sonne hindeutet. Diese Diskrepanz empfinde ich nach wie vor ausgesprochen problematisch. Es ist zwar keineswegs zu erwarten, dass die Überlieferungen der historischen Zeit tatsächlich die Mythologie und Religion repräsentieren, die einer gut anderthalb tausend Jahre älteren Ikonographie zugrundeliegen, aber es wäre verwunderlich, wenn überhaupt nichts davon geblieben wäre. Tatsächlich gibt es Gegenbeispiele, die zeigen, dass über ideologische, politische, soziale und religiöse Umbrüche hinweg bemerkenswerte Kontinuitäten bestehen, die in Mitteleuropa durchaus religiöse und mythologische Fragmente über fast 2000 Jahre hinweg bewahrt haben [1]. Selbstverständlich handelt es sich dabei jeweils nur um ausgewählte Teilaspekte und die Kriterien für diese Auswahl sind nur ansatzweise zu erfassen: Die in [1] genannten Beispiele berühren kosmologische Aspekte (Aufenthalt der Totenseelen und deren Wiederkehr in der Wilden Jagd) sowie Bezüge zum Erntezyklus (Erntesegen an Wodan), die bis in die Neuzeit bewahrt wurden -- ungeachtet offensichtlicher Widersprüche zur Kirchenlehre. Zumindest in diesen Punkten wäre dann jedoch wohl auch ein Fortleben spätbronzezeitlicher Traditionen (ungeachtet eventueller Diskrepanzen zu den vorherrschenden Ideen der Eisenzeit und der frühen historischen Zeit) zu erwarten. Wie dieser Widerspruch aufzulösen ist, war die große Frage, mit der ich die Ausstellung verließ, und mit der ich mich im folgenden ein bisschen intensiver beschäftigt habe: Ich habe mir Kauls (1998) "Ships on Bronzes" zugelegt, um seine Analyse und Methodik nachzuvollziehen, weitere Museen besucht, viel gelesen, ethnographische Parallelen herangezogen, und so die Zeit erlaubt, werde ich versuchen, einige Ideen dazu in diesen Blog zu setzen.
All dies nun hatte mich so beschäftigt, dass ich die großartigen Funde späterer Zeiten, wozu v.a. das Hjortspring-Boot und der Gundestrup-Kessel gehören, leider nur noch überfliegen konnte. Ich muss wieder kommen ;)
Ein Manko meines damaligen Besuchs will ich aber auch nicht verschweigen: Es gab seinerzeit keinen Museumsführer zur Ur- und Frühgeschichte. Das könnte sich mittlerweile geändert haben, denn damals war der offensichtliche Grund die in den letzten 10 Jahren zuvor erfolgte Umgestaltung um den rekonstruierten Sonnenmythos herum ... den älteren Führer (Jensen 1993) habe ich später antiquarisch erworben, einen neuen Führer gab es bis dato noch nicht.

Fußnoten

[1] Ein Beispiel ist die bis ins frühe 20.Jh. im gesamten deutschsprachigen Raum überlieferte Wilden Jagd, deren Anführer regional noch als Wodan (Waur, Wode, Muote usw.) bekannt war und sich so unmittelbar an die germanische Mythologie anschließt. Nicht nur grundlegende mythologische Vorstellungen wurden dabei bewahrt, sondern auch komplexe Ernterituale, vgl. z.B. Grimm (1875, S. 127-130) zu niederdeutschen Kornsegen -- immerhin fast tausend Jahre nach der Christianisierung und beinahe vierhundert nach der Reformation. Andere Vorstellungen, nach denen die Wilde Jagd von einer Frau (thüringisch-hessisch Holle, süddeutsch Berchta) geleitet wird, die auch die Seelen von Kindern aufnimmt, scheinen viel weniger mit der in der Edda verarbeiteten spätgermanischen Mythologie vereinbar und reflektieren möglicherweise noch älteres Gedankengut, vielleicht zu verbinden mit der spätantik dokumentierten Göttin Hludana (Grimm 1875, S.211f.), und falls so, dann womöglich über mindestens 1700 Jahre hinweg bewahrt. Falls wir mit Grimm annehmen, dass Hludana mit der altnordischen Hlodyn nicht nur sprachlich verwandt sein sollte, sondern beide auf gemeinsames Vorbild zurückgehen, dann sind die mit ihr verbundenen Ideen bis in protogermanische Zeit, also etwa die Mitte des 1.Jt.v.u.Z. zurückzuprojizieren.

Literatur
Jacob Grimm (1875), Deutsche Mythologie, Bd. 1, 4.Auflage, Berlin.

Jørgen Jensen (1993), Führer durch das Nationalmuseum. Dänische Vorzeit. Nationalmuseum Kopenhagen.

Flemming Kaul (1998), Ships on Bronzes. A Study in Bronze Age Religion and Iconography. Studies in Archeology & History 3(1&2), PNM (Publications from the National Museum), Copenhagen.

Anm. zur externen Links: Wikipedia-Inhalte haben selbstredend keinen wissenschaftlichen Anspruch, sind aber hier aufgrund ihrer vermutlich bleibenden Verfügbarkeit die am ehesten geeignete Ressource, um Hintergrundinformation einzubinden ;)