Causewayed camps palisade enclosures 2004

Joumal of Nordic Archanlogiml Science 4, pp. 1913 Qoo4)

Causewayed camps, palisade enclosures and central settlements of the Middle Neolithic in Denmark Poul Otto Nielsen The

NationalMueum, Danish Prehistoric

Collections, !:raderihsholms Kønal 12,

DK-1220 Copenhagen Id Denmarh

(poul.otto. [email protected]) Many causewayed camps were rransformed into settlements where various rituals continucd to be performed. Such sectlements aLe among the largest known from thc Danish Middle Neolirhic and are labelled 'central settle, ments' because ofthcir sizc and character as ceremonial centres. The palisade enclosures of the latc Funncl-Beaker Culture are interprered along the same lines. A parallel devclopmcnt towards larger sertlemenrs took place within contemporaneous cuhure groups in Central Europe. Keyuorls

: Funnel-Beaker Cuhure, Pittcd-\Øare Culture, Battle-Axe Culrure,

Middle Neolithic, causewayed camps, palisade cnclosurc, settlement, ritual, culture change

Introduction

there was a causewayed camp there (nos. 4, 5, 18, 19,

Since the first discovery and cxcavarion of a causewayed camp on Danish soil at Sarup, south-west Funen, in 1971-84 (Andersen 1997 ; 1999a: 1999b), causewayed camps have been identified in different parts ofthe country This has happened partly as a result of ncw excavations and partly through the study and re-intcrprctation of reports from earlier excavations. Thc causewayed camps belong to the FunnelBcakcr Culture and were built during a period covering thc latcr part of the Early Neolithic (EN) and first part of the Middle Neolithic (MN), c. 3500-3200 sc. Thus their construction coincided chronologically with thc building of megalithic tombs in Denmark. The late discovery of causewayed camps in Den-

'l'he purpose ofthis survey is to obtain a clearer picture of the coincidence between causcwayed camps and one or more phases of scttlcment during the MN A (c.

mark b1 compari,'on wirh neighbouring counrries in and Central Europe is peculiar and can only be explained by former excavation practices, which allowed only minor parts of even large settlemenr complcxes to be investigated. Attention was directed more

3300 2800 Bc). Did the causewayed camps have a primary function as ritual centres and a secondary function as settlemcnts in the MN? Or were they in fact settlements as wcll as sacred meeting places, in which case, what was the function ofsuch settlements?

towards the recovery ofarteåcts than towards explora-

Later than the causewaycd camps are the palisade

'W'estern

tion ofthe features and layout ofthe

sites.

At six ofthe

20 causewayed camps selected for this survey (listed in theAppendix) small or even large-scale excavation had been undertaken decades before it was realized that

21, 23). These older excavations were carricd out manually, and it was actually the introduction of the mechanical excåvator at archaeological sites, which enabled the uncovering of areas spacious enough for

the inspection of the largc features of causewayed camps, as happened at Sarup in 1971 (Andersen 1997:15 20). The earlier excavations at sites that later rurned out to be causewayed camps wcrc perFormed in order to recover part ofthe rich settlemcnt marerial ofthe Fun-

nel Beaker Culture thar is often found at such

sites.

enclosures from the latcr part of the Funnel-Beaker Culture. These are datcd to the MN AV in Denmark, while in Swcdcn their dates overlap with the time of thc Battle-Axe Culture (Svensson 2002). Unlike the r9

POUL OTTO NIELSEN Table 1. Chronology of the Danish Early and Middle Neolithic.

Years BC

Name of culture and period

(calibratedl

--

2450

--

--

2600 ---

----

2800 --2900 --3000

SingleGrave/

Middle Neolithic

Battle-Axe

B

Culture

t

lt

--

3500 ---

Middle FunnelBeaker Culture

Neolithic

A (MN A) Early

Upper Graves

Per.

Ground Graves

5 Per.

Bottom Graves

Per.

tb le

Bundsø Blandebierg Klintebakke

c

Troldebiem Fuchsberg / Virum

B

Volling

A

Oxie

Neolithic (EN)

places, or were they settlements as well? This paper looks at causewayed camps and palisade enclosures with special focus on the evidence forsettlement that is contemporaneous with or later than the

time ofconstruction ofthese sites. Of the 23 localities chosen for this analysis (see Fig. 1 andAppendix), only a few have been subject to more or less toral excavation (nos. 4, 9). The sites listed here are those about which there is informarion on the date, location in rhe land-

ofthe main constructional features. A

number of the sites mentioned here are unpublished excepr for short notes in Arkeokgishe (Jdgrauninger i Danmarh, in some cases supplemented with information fiom the person in charge ofthe excavation. No attempt is made to enumerate all the possible causewayed camps, and some sites that have not be€n finallyassigned to this category are not included. More possible causewayed camps are mentioned byNiels H. Andersen (1997:267ff .) .

Danish causewayed camps A causewayed camp is defined

site demarcated by one or more parallel rows oflong or short ditches inter-

20

as a

Palisade enclosunes

Passage graves Enlarged dolmens Early dolmens

Causewayed

camps

Eanhen long banows

causewayed camps, rhese enclosures have no interrupted ditches, and they may have multiple palisades. There are many indications ofrituals having been performed, but there is also evidence of massive seitlement acdviries at rhese sites. At least in one case, a palisade enclosure was built on the site of a causewayed camp. - \fere the palisade enclosures sacred meeting

scape and layout

Banows,

st. valby t

Type of enclosure, buildino De.iod

stone cists and flat graves

4-5

(MN B)

Pitte(l.Ware

---- 3't 00 --- 3200 ---- 3300 --

Grave type, buildins period

rupted by causeways and flanked on the inner side by

a

palisade, which may have lorwa rd- pro jecting sections. One feature that is shared by most causewayed camps

is their position on projecting landscape formations. The two overlapping camps Sarup I and II (no. 9) are situated on a plateau surrounded by river valleys on three sides, and 15 ofthe causewayed camps considered here are located in a similar topographical posi-

tion (nos. 1,4,5,6,8,9,10, 11, 13, 14,15,t7,18,20, 21). Only Liselund (no. 12) is described as situated in a flat landscape. The camps may be found on elevated or low ground. steep

-

4rile Bjerggård (no. 20) is located on

hill 84 m above

a

level, Hygind (no. l0), Lønt (no. 22) and Bundsø (no. 23) are to be found on flat, low-lying ground near the shores offiords. The posisea

tion of the ditches may also differ. At Sarup II the ditches form a curv€ ov€r the top of the plateau, and ditches with such a position are also found at Sigersted III, Tielleborg, Vasagård and Mølbjerg (nos. 4, 5, 8, 21), for example. In conrast to thes€, the ditches at Blakbjerg and Bjerggård (nos. l J, 20) encircle rhe top of a hill, while the 250 m-Iong row of ditches a1 Markildegård (no. 6) lies on the slope ofa hill, reaching almost to the bottom ofthe slope. There is a big diflerence in the size ofthe area surrounded by rhe palisade and ditches. The sizes ofthe areas are given in hectares in theAppendix, but the size can only be measured in 500lo ofthe cases, and some of the measurements recorded are only approximate. After the discovery ofthe very large causewayed camps at Liselund (no. 12) and Lokes Hede (no. l3), covering

--CAUSE'iøAYED CAMPS, PALISADE ENCLOSURES

AND CENTRAL SETTLEMENTS Figure 1, Map of the causewayed camps and palisade enclosures listed in the Appendix.

14-5 and" 12-20 hectares respectively, even Sarup I, with its 8.5 hectares, can be regarded only as a site of moderate dimensions.

At seven ofthe causewayed camps listed here there is only one row ofinterrupted ditches, while at 13 there are two parallel rows. Most of the camps with only one row ofditches are located on Zealand. The initial construction of the ditches took place within the EN C (the Fuchsberg/Virum phase) and MN Æa-b. Thus the building of causewayed camps can be dated to the same time interval as the megalithic tombs (Table 1). In three cases, howevet there is no material available for dating the earliest layers deposited in the ditches.

The function of causewayed camps The current interpretation of the function of causewayed camps is based on observations made during the excavation of Sarup I and II (no. 9). The occur-

rence of human bones sade, although small

in the ditches and at the pali-

in number, leads to the hypo-

thesis that interment took place in the ditches. Similar observations have been made at causewayed camps in

'W'estern

and Central Europe (Andersen 1997:307-

09). The place was intended for gatherings and for the

performing of riruals, including temporary inrerment ofthe dead, and parts ofhuman bodies were manipulated in the course ofrituals at different places within the area of the causewayed camp, as were other items found buried in the ground, such as pottery vessels, some ofwhich contained seeds ofcereals, other pottery vessels, flint tools and battle-axes (Andersen 1997 :301-18; 1999 a:296-302).The rituals involving human bones that were enacted at the causewayed camps were part of a burial custom which also included the placement ofskeletal parts in the chambers

of the megalithic tombs. The location of the

cause-

wayed camp in relation to the distribution ofdolmens and passage graves is well illustrated by investigations

PO

UL OTTO NIELSEN

in thc Sarup area (Andersen 1999a292 95). According to Niels H. Andersen, the causcwayed camp was a mccting place for the inhabjtants of small settlements distributed ovel a larger area, who belonged to a supposcdly segmented tribal sociery. The causewayed camp symbolised the uniry of thc many small settlemcnts or territories (Andersen I999a:299). Each of thcm may have had access to their own ditch as a place For temp.rrat1 intermenr. Thi. interpretetion i:.uPporred by the many re-cuttings observed in the ditches and by finds ofhuman bones at the bottoms ofditches bclonging to Sarup I, from thc Fuchsberg Phase (EN C). No human bones were lround in the ditches belonging to Sarup II, from the Klintebakle l'hase (MN

AIb), but burnt bones were found in two out offbur posrholes intended for the substantial wooden upstructure surrounded by a narrow near the edge ofthe plateau inside scmi-circular ditch thc palisade and dirches (Andcrsen I999a:249-50). Somc ofthe funerary ceremonies at Sarup 11 may thus rights belonging to

a

have taken place at certain noticeable features

within

rhc limits of the palisade and ditches. The re-cuttings in thc ditches outside the palisadc, however, seem to havc conrinued fiom Sarup I and II through Sarup III

(MN AID and Sarup IV (MN AllI-lV), which leads u\ (o a\\ume rhar the cu:rom of temlorary ittterment persistcd for quite a long time.

Human bones have also been found in the ditches at other causewayed camps (nos. 4, 10, 11, 15, l7). At three sitcs, Åsum Lnggård, Blakbjerg and Ballegård (nos. 11, 15, 17), they are found in the bottom layers. Only single bones or small collections ofbone material occut and ncver whole skeletons. These findings support the thcory that human bodies were placed at the bottoms of the ditches, and that they were later removed in a decomposed state, during which process minor parts wcre left behind. During the excavation at Bundsø (no.23) hurnan skeletal remains, including skulls, were also found outside the ditches, cmbedded

in the waste from the later settlement (Hoika 1987:150 52). Most often only minor sections of the ditchcs have been excavated, but in a few instances larger parts

of

the bottom layers have been revealed. At Markildcgård (no. 6) there were horizontal layers ofwood and bark which may have served a-s a "floor" on which sherds of

pottery vessels reprcsenting the Virum Phase were found, together with a few bones of oxen and shcep (1?Ø. Sørensen 1995:1 9). At Store Brokhøj (no. 16), a structure interpreted as a pottery kiln filled a large part of one of the ditches, apparently contradicting interprctations that the ditches werc places only lor the per22

forming of rituals. In seeking an explanation for this feature, we might assume that pottery was fired in this particular place in order to meet the dcmand for vessels used during the lunerary ceremonies. Svend Nielsen has suggcsted that pottery production may in fact have been one of the activities carried out at causewayed camps. If so, it may help explain the uniformity of pottery shapes and decoration over large arcas during the IIN C and early MN A. The causewayed camps, being meeting places for a large number ofpeople performing rituals and celebrating social events, served as centres of communication (S. Nielsen 1999:I33). Thc above interpretation of the function of the causcwayed camps belonging to thc Funnel-Beaker Culture in South Scandinavia as ritual meeting places finds support in observations from cxcavations ofsites of thc same nature within other culture groups. Most

comparable

with the

causcwayed camps

Scandinavian ones are of the Michelsbcrg Culture,

the be-

oftheir large size and difference in layout. Finds of human bones and whole skeletons in the ditches, wherc re-cuttings have been observcd just as in the Danish ditches, indicate a similar lunction (Alrdersen 1997:184-203). When larger parts ofthe inner areas of Michclsberg causewayed camps have been examcause

ined, the features and finds do not reflect ordinary settlement activities. In particular, the contenrs ofpits in-

dicate dcliberate deposition of material. The causewayed camps ofthe Michelsberg Culturc are therefore looked upon as central places for rituals (Bertemes 1991).

Settlement occupation at the causewayed camPs In spite of clear evidence of rituals having been performed at thc Danish causewayed camps, not all ofthe material found there indicates ritual behaviour. Tiaces of senlement activities have also been found at most of the sites, indicating either short-rerm or long-tcrm occuparion. First we have to consider that the presence of a large number ofpeople engaged in the construction ofa causewayed camp resulted in a certain quantity of refuse. The crew working with the timber uscd for building the palisades needed effective tools, which had to be repaired from time to time and were finally discarded, so that wc may expect to find flint waste and maybe left-overs from regular tool production. Fecding the labour force would require accessories such as querns, fireplaces, clay vessels, etc., resulting in refusc like that lound at ordinary settlement sites.'We would also expect the people attending ceremonies year after

CAUSE\(AYED CAMPS, PALTSADE ENCLOSURES AND CENTRAL SETTLEMBNTS

Table 2. Causewayed camps, indications

ofsetdement

F,N C

MNAI

ab

MNAII

MNAIII

MNAV

Pitted-Ware

1. Skævinge

4. Sigersced

III

5. TrelJeborg

6. Markildegård 8. Vasagård

10. 1

Hygind

l. Åsum Enggård

12. Liselund 13. Lokes Hede 14. Ginnerup 15. Blakbjerg 16. Store Brokhøj 17. Ballegård 18. Voldbæk 19.

Toftum

20. Bjerggård 27. Mølh1erg 22. Lønr 23. Bundsø

year at the causewayed camps to leave acertain amount

ofgarbage behind. From the sheer quantity of settlement material it is some times possibl€ to determin€ whetherwe are deal-

ingwith temporary occupadon in connection with the construction of the site or its ritual use, or whether there was more continuous occupation. The most obvious evidence oflong-term occupation occurs ar sires where thick sediments with mixed s€ttlement debris cover not only the interior ofthe causewayed camp but also the surrounding ditches, thus sealing these features and extending beyond the confines of the original site. At many causewayed camps, however, no such occupation layers have suryived, due to disturbalces and erosion caused by modern agriculture. Thus settlement material may be present only in the

fill ofthe ditches and in pits dug into the subsoil. In such cases long-term occupation can only be deduced

from the quantiry and character ofthe arteåcts deposited, Furthermore, the presence ofcertain categories of objects such as harvesting implements should be seen as evidence of permanent habitation.

Vith

due respect to the ambiguiry of the evidence, we will now try to look at the indications for settlement occupation during successive stages at the causewayed camps lised in the Appendix (summarised in Tåble 2).

Ea rty Neolirhir C. the Furhsberg and Virum pbases, c. 3500-3300 nc

Setlement material fiom EN C that must have accu-

mulated shordy after the construction

of

the

causewayed camps has been found at five of the sites

(nos. 1, 6, 9, 12, 19). At Sarup

I

(no. 9), seftlement

refuse was recorded over a large part of the space be-

hind the palisade, while features containing evidence ofritual depositions were found on the periphery near the palisade and on the sloping ground bordering the interior area. No settlement material from this period was found in the ditches, but therc \.i€re 25 pits containing settlement refuse. Niels H. Andersen suggests that the refirse in the pits was derived from the ritual use of the site, or alternatively, that waste was transported from settlements elsewhere to be deposited at

23

P()I'I

.)TT{)

NTFT SFN

the causewayed camp (Andersen 1999a:276 77). A certain numbel of the postholes recorded at the site

mains to be seen how much of rhc rich matcrial of finds from MNAI at Lønt (no. 22) can be attributcd to

rnay have belonged to Sarup I, although no house sites could be identified. At Liselund (no. l2), settlemer.rt material from.EN C ur.lound in pir.,and layer.wirh r\J\le!o!eringpal

settlerDenr activities.

ofthe inside

area ofthe causewayed camp, and settlement material fiom the same period was found in both the bottom and top layers ofthe ditches. -t'his indicates quite extensive occupation shortly after the construction ofthe site. At lbftum (no. l9), settlement material was also lound embedded in the layers within the ditches, lrom top to bottom.

Early Neolithic c.

C

Middle Neolithic AI,

.3500 .3100 uc

There are two instances in which the date ofthe earliest activity at a causewayed camp is reported as EN C MN AI, which means that the material either spans the two periods or cannot be dated more precisely. Both sites belc,ng to the largest ofthe causewayed camps. At Lokes Hede (no. 13) some ofthe pits contained burnt clay, possibly the remains ofovens, while at Blakbjerg

(no. 15) part ofthe inner

area

with occupation layers,

pits and postholes were excavated in a search for house remains, which escaped detection here as at other sites. 'l he material from the inner space ofthe site and finds

from rhe Jirch fill marerial be.rr *itn... to exren\i\e, regular settlement activity at the causewayed camp in rhc rimc Lrlk'wing it. ..rn.truit i.,n.

Middle Neolithic AIa-b, c. 3300-3I00 ec At Sarup II (no. 9) there were 77 pits containing settlement waste, which could be dated to MN AIb, the Klintebaldce Phase. Some ofthe pits were interpreted as having been used for storage, which should indicate occupation of longer duration (Andersen 1999a:291). But as in the previous phase, settlement material was not mixed wirh the fill ofthe ditches. In the only ditch excavated at Åsum Enggård (no. 11) both the bottom and top laycrs containcd finds lrom MN AI, and thc top laycr also containcd accumulatcd scttlcmcnt waitc. Substantial occupation must havc takcn placc hcrc subscqucnt to the building of thc causcwaycd camp. Ar Mølbjerg (no. 21) sherds of pottery from MN AI were forurd during minor excavation ofan occupation layer, and somc wcrc also collected from thc surfacc. Sherds lrrom MN AI wcre found in the ditches ar Voldbæk (no. 18) and Bjerggård (no. 20), but this does not necessarily indicate occupation during that period. It re-

2.4

Middle Neolithic AII, c. 3100 3000 rc: Settlement activity during MN AII has been recordcd at six of the causewayed camps (nos. 4, 5, 9, 10, 17, 22). At Sarup (no. 9) there was an extensive settlemcnt covering c.4 hectares during thc Sarup III phasc. Fivc features contained burnt clay and may bc intcrprctcd as the remains of ovens. Bcsidcs thcsc, thcrc wcrc quetns, hammer-stones, flint wastc and traccs of pottery production. Rows ofpostholcs probably indicarc house remains.'I'he layers in the ditches contained substantial settlement waste from MN AII (Andersen 1999a:310 311). At Sigersted IIl (no. 4) settlement activiry during MN AII covered an area drat extended beyond the ditches of the causewayed camp. As ar Sarup III, the upper layers of thc ditchcs wcrc mixcd with waste from the settlement. Building activiry had left numerous postholes, but no clcar housc plans could be identified. Several pits contained wastc, burnt daub and querns. At Lønt (no. 22) a rich occupation layer also indicates substantial settlement ar the site during this period.

Middle Neolithic AIII, c. 3000-2900 nc 'fhere was settlement activity during this period at six of the causewayed camps (nos. 4, 5, 8, 9, 10, 23). Sarup IV (no. !) has been dated to MN AIII-I\{ and finds bclonging to this phasc covcr about 4.5 hectares and indicate settlement activitics. Pits with wastc were not as numerous during this period as during Sarup III, but they were larger (Andcrsen 1999a:315,31920). Ar Sigenred ll I (no. 4) thcrc wa. a .crrain amou n t of settlement material from MN AIII. which accumulated as a continuarion from thc MN AII-sctrlcmcnt, while at Bundsø (no. 23) thc causcwaycd camp is covered by material that originates from a settlement dur-

ing MN AIII.

Middb Neolithic AV

2900 2700 ac Scrtlcmcnr activitics during MN AV have been c.

re-

corded at ten ofthe causcwaycd camps (nos.4, 5,6, 8,

9, i0, 18, 20,21.,23). At the thrcc sitcs whcrc finds from this pcriod prcdominatc, thcy indicatc cxtcnsive, long-term occupation (nos. 6, 8, 20), whilc at six sires the find materjal is more limited (nos.4, 5,9, 10, 18,

23). The distribution of finds lrom MN AV at

T CAU5F,!?'AYI]D CAMP5. PAI,ISADE DNCI,OSI'RES AND CT]NTRAI, 5I,TTI,EMtrNTS

Vasagård Øst (no. 8) extends beyond the ditches ofthe causewayed camp and reaches two parallel palisades at

the foot of the hill, surrounding an area of about 2.5 hectares, and settlement debris from MN AV also covers a large area at Markildegår'd 0.ro. 6) and tsjerggåLd (no. 20), the estimate at the latter being about 5 hecrares. At Sarup, occupation during the Sarup V phase was probably brief. Settlement deposits were left in the upper levels ofthe ditches, but no new re-cuttings took

At Damsbo Mark, about 3 kilometres from the Sarup site, settlement remains frorn rhe MN AV have been recorded over an area ofabout 8 hectares, indicat-

place.

ing that a new location had been chosen for the ma.jor settlement ofthe region during this period (Andersen 1999a:324).

Pitted-r[/are Cuhure (PWC), c. 2800-2600

BC

I)uring the later part of MN AV the Kattegat region with an economy based

was influcnced by the P\X/O,

on terrestrial as well as marine resources and adopting a new flint technology and the characteristic elements of P-WO pottery decoration (Rasmussen 1991). At the newly found causewayed camp at Cinnerup (no. 14) a large P1VC seftlemenr has left deep layers containing fill and cover the upper parts the ditches ofthe causewayed camp. Only limited excavation has taken place at this site so far, and we therefc,re have to wait for a description ofthe stratigraphy ofthe ditches waste which

and a date for the construction of the causewayed camp. C)bservarions made at Ginnerup so far nevetthclcss provide evidence ofthe great size ofsome ofthe P\(/C settlcmcnts in l)cnmark, confirming what has already bccn dcmonstratcd at Kainsbakke, only a few kilomctrcs from Ciinnerup (Rasmussen 1991). The gcncral tcndcncy for large, permanently occupied settlements by thc cnd of thc lunnel-Beaker (lulture is also pcrccptiblc in thc arcas affected by Pr{/(l influence. Ar Sråvic in thc wcstcrn part of Scania, settlement material from MN AV, combined with flinr tools from thc P\flC], has been found in the fill ofthe interrupted ditches bclonging to a causewayed camp, stretching over a distance of250 m (Larsson I !82). As it appeals from thjs short survcy, scftlcmcnt activities on a major scale took place ar thc causewaycd camps as early as EN C and MN AI, be ing most clcarJy demonstrated by thc cxcavations at Liselund (no. 12), Lol