Navigation sidebar:



  1. Strahlenwälle
    1. Expansion of settlement area
    2. Visualization
    3. Schimmelhain
    4. Defences
    5. Access and Restriction
    6. Hinterer Eulenkopf
    7. Relation to water and cultic significance
  2. Visibility
    1. The ramparts t,u, and v
    2. Settlement
      1. The South
      2. The West
      3. The East
      4. The North
      5. A visitor's impression
  3. Footnotes


The research questions, which were posed in chapter Methodology mentioned different explanations for the use of the Strahlenwälle. Through generating and exploring the model it was tried to gain some insight into their use. Naturally this is a visual and subjective experience of the model, and may answer some of the questions better than others, it might also pose new problems which can only be answered by further research, in particular by excavations.

Expansion of settlement area

Schlott speculated that the Strahlenwälle could be the beginning of the attempt to add further living area to the oppidum. I would argue against that, on grounds of missing parallels. Usually if living area was added towards oppida it took place in a more or less “circular” way (for example Závist or the Heidetränkoppidum, also the three concentric rings of the Dünsberg itself might be taken as an example) and not by building many ray-like extensions extruding from the site. Also the evidence on site is speaking against this thesis. Where we find more or less enclosed areas (for example the area between Strahlenwällen J, L, M, N and banks f and e1, or the area enclosed by banks m, a, W and A) then they provide no evidence of occupation. We also have areas of little or no settlement activity inside the oppida (see section Platforms), and the question remains whether these areas, which are already enclosed by the banks would not have been occupied first, before additions were constructed.

We can also witness that building activity did extend outside the fortification of the hill, especially to the west but also in the south and near the Grinchesweiher (see figure Platforms and Ramparts). In the west we find no Strahlenwälle associated with the occupation (if we are not taking ramparts R and S as the start for extending the settlement area). In the south the platforms lie either not close enough to the Strahlenwälle to be associated with them, or they lie outside their boundaries, so that they are separated from the oppidum by the Strahlenwälle.

Platforms can be found inside and outside the annex, which encloses the Grinchesweiher, but Strahlenwälle C, F and G do not seem to represent an attempt to enclose the outlying settlements.


Through the model it was possible to first get a general impression of the features on the hill, and then to chose suitable locations for stills and animations, in order to investigate the points of interest. The Viewpoint media model and the possibility to easily manipulate it made it possible to keep an overview over the whole of the hill while the details were investigated.

While watching the panoramas and the animations, which show the Strahlenwälle from outside the oppidum, I had the impression that some of the Strahlenwälle did seem to guide the observer visually to some of the gates, but were obstructing others. A sum up of my impressions can be seen in figure accessibility. The dark areas denote the points from where the access to the gates seems obstructed, while the light grey areas show the positions from where an observer seems to be guided towards one of the gates. In case of the light grey area between Strahlenwällen D and C the passage through the gap left by the two banks seems to be “inviting”, even though this is not a real gate.

Especially the area around the Grinchesweiher brings further questions. Dehn proposed that the ramparts around the Grinchesweiher seemed to be built in two phases (see section Strahlenwälle). When viewing the ramparts from the northeast (see Banks_DEC.avi on the CD), this seems very likely, as Strahlenwall D with its lower L-shaped part seems to form a straight line with H, but it is also observable that the longer part of D, which runs parallel to E, seems to form an “entrance”, by curving in into the direction of E and I. Is this maybe part of an original entrance to the area of the Grinchesweiher, which was in use at the time when only the first phase of the ramparts existed? Going to the south (Banks_FDIJ.avi) the picture is very different. It seems like the longer part of bank D, with its curving end, blocked the entrance between the ramparts D and E. And the L-shaped part of D appears to have exactly the same function: blocking off the access to D and E. From this position it does not seem clear any more whether there were really two phases to the banks enclosing the Grinchesweiher.


As can be seen in figure accessibility, it seems that access appeared to be restricted when approaching the oppidum from the south east and the north, but the settlement seemed approachable when coming from the southwest and the northeast. The only entrance which seems openly accessible from the southeast is gate 6, because we find a gap in the line of J in front of it. Studying Reeh's (2001) map (see figure Reeh) it is obvious that exactly in the region of the gap springs are frequent2, thus the area might have been naturally closed off, assuming that it was a marshy ground, which on the other hand would restrict access to gate 6.

Nevertheless it is remarkable that some areas look more accessible after a barrier had been crossed, for example once arrived in the Schimmelhain3, Strahlenwälle L and M seem to focus the observer on the entrance to gate 7 (see Banks_LM.avi), they (L and M) appear almost like a Zangentor leading up to another Zangentor (gate 7). This focus on gate 7 is one of the interesting features of the Schimmelhain. It is understandable why Reeh (p. 100) thinks that this area might have been used as a corral. But other uses would be possible: for example the site could have been used as a refugium for the surrounding villages. It is fortified with ramparts, and allows an easy access to the oppidum via gate 7, if there should be further need of evacuation. Noteworthy in this context is the fact that Strahlenwall M would have the ditch (and palisade) on the wrong side of the fortification: it is shielding the area in front of gate 7 from the Schimmelhain and not vice versa.

Another possibility could be something similar to a marketplace, which is connected to the oppidum via gates 6 and 7. Access could not only be gained from gate 7, but also from the gaps between ramparts M and N, or J and N. When choosing either of these entrances to the Schimmelhain the visitor is forced to walk past the ramparts and palisades for some time.

We could also be looking at two phases of additional “fortification”. In figure Accessibility it seems like the access from the southeast was supposed to be restricted. Strahlenwall M would be facing in this direction, but might have been superseded by Strahlenwall J in its use, so that we are not looking at an enclosed area, but on two different lines of fortification.

These are only suggestions, which spring to mind while watching the animations for the Strahlenwälle in this area, the real use of this enclosure can only be verified through excavation.


Generally I think that the Strahlenwälle were not really suitable for defence purposes, because attackers could easily walk around them. The only exceptions might be banks D, E, H and I, which are securing the Grinchesweiher, and also T, which encloses the area of the Kleinen Dünsberg in a parallel line to l4. It is also remarkable how at the same time the Strahlenwall T guides the view towards gate 13, if the spectator is already close enough to the gate, otherwise the view might be blocked by Strahlenwall U.

Access and Restriction

The characteristic of blocking entrances physically, or only from view, is notable at several other places on the Dünsberg. Gate 1a seems to be blocked by Strahlenwall W especially in association with bank V (see Banks_VWA.avi), while gate 1 lies open to the visitor, its accessibility is even enhanced by the presence of Strahlenwall A.

The small rampart P is blocking the view towards gate 7, when approaching it from the southwest, on the other hand gates 8 and 9 seem to be easily accessible (see Banks_hgP.avi)5.

Approaching from the west, gate 10 is widely visible and Strahlenwall R seems to lead towards it, while the view of gate 11 is blocked by Strahlenwälle R and S. Only when the observer is getting close to the gate by walking around the tip of Strahlenwälle R and S the entrance becomes visible (see view_towards_gate11.avi).

One might wonder, whether it is deliberate that one of the entrances seems to be blocked when two gates lie quite close to each other (e.g. gates 1 and 1a, 8 and 9, 10 and 11). Was maybe one of them dedicated to entering the site and one for exiting it, while visual aids helped to communicate which one was which?

Hinterer Eulenkopf

It is remarkable that both of the Strahlenwälle near the Hintere Eulenkopf are facing to the west6, so that they are not really enclosing the Hintere Eulenkopf but rather shield the area next to gate 11. Could it be that we encounter once more two phases of the same structure? First Strahlenwall S could have been built, and later R followed because it was desirable to enclose the Hinteren Eulenkopf as well (because of its ores or because of the springs in this area?). Otherwise the two ramparts would make little sense. In the way they were built they form a dead end (see Banks_SR.avi), and would not offer much of an escape towards the oppidum. Also generally the access to the area enclosed by the two banks is restricted, and with that the access to the springs there.

Relation to water and cultic significance

It seems apparent that there is a connection between Strahlenwälle and sources of water. 14 of the Strahlenwälle can be associated with springs or streams, while 9 of them cannot, or the connection seems questionable (see also \ref{subsec:springs). But the way in which these features relate to the sources of water is very different. The Grinchesweiher and the Schulborn are incorporated into the settlement. F, K and V restrict the access to streams from different areas (F and K from the northeast and V from the east). Also S and R delimit the springs near the Hinteren Eulenkopf. One of the possible explanations would be that the water was in some way assigned to different fields or meadows, and the banks were a means of separation between them. Further boundaries could have been formed by hedges etc. On the other hand it seems a great effort to separate fields by massive ramparts.

The other explanation could be related to cult. The seclusion of sacred areas by means of boundaries is common to Celtic ritual sites (Filip, 1970: 63; Webster, 1995: 458-459) \nocite{Webster-1995, Filip-1970. But the way the boundary is set up is very unusual, because normally the sanctuary would be not only enclosed from one side, but kept more secluded, by walling it off on four sides, as for example the “Belgic sanctuaries” (Webster, 1995), the sanctuaries in Manching (Sievers, 1999) or in Libenice (Filip, 1970). Not only are these sites more secluded from the surrounding area, but also the banks are the focus of ritual activity (Webster, 1995). The partly excavated Strahlenwall M has not provided any finds in its ditches, but it is also not associated with water. As we have no other excavation data concerning the banks F, K and V, it is hard to determine whether they could have been of cult related use, and further research would be necessary.

The Strahlenwälle leading up to springs (A, B, U, J) may also be interpreted as being cult related, by guiding a passage from the gates or from certain ramparts towards the springs. Unfortunately we cannot say whether these springs were the focus of cult related attention. Indeed it is not attested that the Celts used springs as sanctuaries, and Webster (p. 449-450) points out that springs as sanctuaries in Celtic areas came only into use after the Roman conquest. On the other hand we know that springs were used by the Germans (Stjernquist, 1970), but then we should find sacrificial gifts and other indications of cult in or close to these springs. As already noted in section Springs, several of the springs and also the great basins of the Schulborn and the Grinchesweiher were excavated, but no trace of cult related remains was found, even though the conditions were good enough to preserve organic materials.

Taking together all the facts I think that it can be said that a relation of the Strahlenwälle to cult is rather questionable at the moment, but the picture could be changed by further excavations of these sites.


Having elaborated the perception of the Strahlenwälle from outside the oppidum, it is time to have a look at the features and the visibility from inside the site.

In the folder “Views-Strahlenwaelle” on the CD views from the top of different banks inside the oppidum towards the Strahlenwälle have been collected. These images do not only provide views of the Strahlenwälle, but also show how much of the interior of the oppidum was visible from different ramparts.

From the top of the hill the Strahlenwälle are usually quite well visible. The views offer little new insights into the layout of the banks, only two facts should be mentioned.

In the image from_q_to_LMNOP.jpg the passage between J and N seems almost like a gate, while in the same image Strahlenwälle L and M also appear as a gate leading up to gate 7.

The view from_p_to_Grinches.jpg shows again that H and D seem to join up, and could have formed a single defensive line (see also section Visualization).

The ramparts t,u, and v

An interesting view is offered by view_from_o_to_w.jpg and From_w_to_tuv.jpg. Both show how the banks t, u and v are acting as a barrier between the flat area in the northeast and the massive fortification on top of the hill. View_from_o_to_w.jpg also shows that both of the entrances (gate 21 and 22) to the topmost wall are visible, but only to a limited degree.

It could be assumed that banks t, u, and v were contemporary with the topmost rampart. In this case the construction might be paralleled by what we can see at the oppidum of Závist (Motykova et al., 1991: fig.1; Drda, 1997: fig. 4), where two banks in front of gate D block off a flat area leading up to the oppidum.

After the construction of the middling rampart, banks t, u and v might have become redundant as obstacles against rapprochement.


The stills are also helpful in the investigation of the settlement areas inside and outside the oppidum. Already from plan Platforms and Ramparts it is apparent that some areas are more densely settled than others (see also Platforms).

From the image Houses_in_the_east.jpg it is clear that even though this is the most densely populated area on the plan, the settlement density is not very high. One obvious reason for this is the steepness of the slope, which permits the building of houses only in intervals on platforms. The view may also be biased by the fact that the houses on the platforms are very small and that just one house was placed on each platform, even though some of the platforms had been big enough to supply enough space for several buildings. Still, the picture presented by this image is not what we would expect to be “urban” in a classical, medieval or modern sense.

Other areas of high settlement densities are to the south and to the west of the oppidum.

The South

A good overview of the southern part is given in From_x_to_LMNOP.jpg and from_q_to_LMNOP.jpg. The area seems closely populated, but there are marked exceptions: the area north of rampart O and in front of gate 7 as well as the Schimmelhain are completely free of occupation.

The houses between the gap left by Strahlenwälle O and N, are hardly visible from inside the oppidum. At the end of the sequence of Banks_JN.avi the extent of the settlement between banks O and N can be observed. The houses seem to be situated in a very vulnerable position7, and only the top ones are visible from the topmost rampart of the hill. The question is also why the area between this settlement cluster and the bottommost rampart are kept free of buildings. Was there some kind of taboo to building there or was the situation not favourable enough for settlement? The latter would be rather strange considering the amount of houses situated further north and south of this empty area.

The West

To the west we also have densely populated areas, where the buildings spill over the ramparts into the lower lying ground. The ground is remarkably steep in this region (see walkthrough.avi).

An interesting situation is given at gates 10 and 11, near the Strahlenwälle R and S (see From_r_to_RS.jpg). The entrance area inside the oppidum towards gate 10 is completely free from buildings, while the platforms start to cluster in the direction of gate 11 and on the outside of bank k. Is the area at gate 10 kept free because it is one of the main arrival areas (see section Access and Restriction)? Maybe we could also postulate that gate 10 is for general access 8, and gate 11 is used by the population of the oppidum for minor tasks, and does not bear a great amount of traffic9. This might also explain the visually guiding function of the Strahlenwälle, which lead the arriving visitors directly to the right entrance (gate 10) to the oppidum.

The East

The same phenomenon concerning a free entrance area is observable at gate 4, which is leading towards the Grinchesweiher (see from_p_to_Grinches.jpg and From_w_to_Grinches.jpg). Here a shallow depression in the landscape is left completely free from occupation, while the houses cluster on the steeper hill behind it. This feature cannot be explained by incoming traffic from outside the oppidum, as there is no real entrance to the Grinchesweiher, besides the ones coming from the oppidum. Still gate 4 would be the main access to the Grinchesweiher, as the direct way from gate 5 towards the basin is obstructed by Strahlenwall H. Being one of the main sources of water for the settlement it was certainly necessary to keep the entrance area to this basin free.

The view offered by From_w_to_Grinches.jpg shows not only the empty area in front of gate 4, but also that the areas closest to the walls show no settlement features10. Also the area between the topmost and the middling rampart appears to lack of buildings. %reason?

On image From_o_to_b.jpg it is observable that the houses on the hill are standing in rows, this can also be seen in from_p_to_JK.jpg, but usually the development appears rather “random”. Plan Platforms and Ramparts shows that there seem to be more platforms, which might be built in lines, but this is not reflected by the stills made from the model.

Whether streets or paths along which the buildings are situated could be deduced from this “patterning” has to remain speculation, even though it is obvious that some tracks or ways, between the houses must have existed.

The North

The north is generally sparsely settled. Near gate 1 the occupational area stops abruptly leaving the rather flat area between gate 1 and gate 2 completely empty (see from_o_to_VWA.jpg, From_w_to_AB.jpg and From_o_to_AB.jpg). The question arises, whether this area was truly not settled, or whether the settlement remains were not detected. The surface in this region appears to be less steep than other regions with high settlement density. Did the platforms between gates 1 and 2 escape Reeh's attention, because they were not as clearly visible as on the steeper slopes11? Or were the traces of platforms destroyed through forestry12? Maybe the area in the north was not adequate for settlement or was left open as pasture, refugium etc. Only excavations in this region could shed new light onto these questions.

A visitor's impression

The ramparts in general are obstructing the view inside the oppidum, this becomes apparent in the animation “walkthrough”.avi but also while viewing some of the stills, for example From_y_to_TU.jpg, where the whole of the occupation between the middling and the bottommost rampart is concealed by the middling rampart.

On the other hand the fortifications are widely visible from outside the oppidum13, and give the onlooker an impressive view of the site with its multiple ramparts, which become more massive the further up the hill they are14. The multitude of houses and the vastness of the site must have added to the impression of a strong centre.

The Dünsberg is still a landmark in the region (see section Topography) where it is widely visible over kilometers. Imagining it stripped of the forest and with the clearly visible fortifications and houses, which can be seen in the model, the view of it must have been even more imposing to the population and visitors of the area than it is today.

It is still hard to imagine what the site must have looked like in the past with different buildings, busy inhabitants, cattle and other animals, but I think that the model might give some indication of what we could expect from the ramparts and the settlement density.

1 This area is called “Schimmelhain” by Reeh.

2 This is an indication for Reeh to take this area for a corral.

3 The area enclosed by J, N, M, f and e.

4 Bank_T.avi shows this nicely, especially how the Strahlenwall follows the isolines in this area, while l runs further on the top of the hill, with a steep slope in between them (see also From_s_to_TU.jpg).

5 The Strahlenwall Q did not find its way into the model, because it was not included on the map by the FH-Frankfurt, if it would be in place the view of gate 9 might have been obstructed as well.

6 That is the direction to which their ditch (and palisade?) is directed.

7 They are placed far away from the oppidum, on a steep slope, which is completely exposed to the southeast. Research in the nature of this settlement would be interesting to determine why the houses were built there.

8 For example trade etc. which is related to people coming from outside the oppidum, which need to access the settlement with carts, bring in cattle etc. Activities which will need a lot of space in the entrance area.

9 Reeh (p. 105) has argued in similar lines, i.e. that the two gates had a different purpose.

10 The reason for this might be that the ramps, which are not modelled, occupied this area, which was thus not available for settlement. Also easy access to the fortifications via the ramps would have been necessary.

11 This seems to be unlikely, because Reeh (p. 65) could identify platforms on the flat area between ramparts t, u, v and p (in 1933 around 80 of them were still visible), where they are hardly noticeable. He also identified some platforms on the Kleinen Dünsberg, where the surface is also very flat.

12 The area around gate 8 is vastly disturbed by trackways.

13 For this refer in particular to the Viewpoint Media model.

14 For example view Banks_hgP.avi, Banks_VWA.avi, Bank_N.avi and view_towards_gate11.avi.