DOI chapter:
Grenzgänger, traders and the last hunter-gatherers of the North European Plain
DOI chapter:Anscher, Theo J. ten; Knippenberg, Sebastiaan: Unexpected dimensions of a Swifterbant settlement at Medel-De Roeskamp (the Netherlands)
DOI Page / Citation link:https://doi.org/10.11588/diglit.66745#0161
160
Unexpected dimensions of a Swifterbant settlement at Medel-De Roeskamp (the Netherlands)
ing to a structural shift in the subsistence economy
(Gehasse 1995). There is an echo of this development
in the ritual depositions. Already in the earliest phase
of the Swifterbant culture pot offerings are attested,
as are offerings of red deer antlers and aurochs skulls
from bulls, whereas the oldest 14C dated depositions
of domestic cattle horns, again from bulls, occurred
much later.1
Residential mobility is considered to be the pre-
vailing settlement system, with hardly any exception
(Raemaekers 1999). Swifterbant sites are usually
situated in gradient rich wetland zones near (small)
streams. The duration of the habitation varied, depend-
ing on the possibilities of the location in question.
Some sites like Bergschenhoek, on a flat shore, are
thought to have been inhabited very shortly: just a few
days or a week at most (Louwe Kooijmans 1985). The
majority of Swifterbant sites, however, are located on
elevated grounds, and seem to have been periodically
used for tens of years to a few centuries, such as for
instance the river dune sites Hardinxveld-Giesendam-
Polderweg, Hardinxveld-Giesendam-De Bruin and
Hoge Vaart-A27 (all belonging to phase Swifterbant
1 [SW1] which is dated between c. 5,000-4,400 cal-
BC), the levee site Swifterbant-S3/S5 and other sites
on levees and sand dunes around Swifterbant, and
the river dune site Hazendonk-1 (all dating to phase
SW2: c. 4,400-3,900 cal BC).2 The discovery of adze
marks pointing to an arable field on a clayey levee at
Swifterbant-S4, and subsequently the identification of
tilled soils in thin sections arguing for arable fields at
other nearby sites as well, ended the discussion about
grain being imported or cultivated locally at wetland
sites in favour of the latter option.3
A few sites were situated on locations that con-
tinued to be easily accessible and inhabitable for a
very long time. They span the complete duration of
the Swifterbant culture and beyond. Hude I (Lower
Saxony, Germany), a palimpsest with a problem-
atic stratigraphy on a higher ground bordering lake
Dummer, is a well known example. At the time the
Hiide ceramics and features were published in full, its
Swifterbant component was heavily underestimated
1 Kroezenga et al. 1991; Prummel / Van der Sanden 1995;
Ufkes 1997; Raemaekers 2019.
2 Louwe Kooijmans 2001a; 2001b; Hogestijn / Peeters
2001; Peeters 2007; De Roever 2004; Raemaekers 1999. -
Phasing based on ceramic characteristics according to Ten
Anscher2012, 127-129.
3 Cappers / Raemaekers 2008; Huisman et al. 2009; Huis-
man / Raemaekers 2014; Schepers 2014.
(Kampffmeyer 1991).4 This is quite understandable
since back then publications on the Swifterbant cul-
ture were scarce, preliminary and based on just a few
sites, all dating to SW2 (e.g. De Roever 1979). It took
another two decades to reach a better understanding
of the width and depth of this culture. Hiide I has a
close western counterpart in Schokland-P14, located
on a large sand ridge on a substrate of boulder clay
bordering the Overijsselse Vecht (Ten Anscher 2012;
2015). In particular, the ceramics are very similar.5
Schokland-P14 has a more reliable chronostratigraphy
though its dating is based on 14C-dates from charred
food crusts on pottery sherds. If the crusts contained
fish remains (whether this is the case is unknown),
the dates might be too old due to the reservoir effect.6
Schokland-P14 at least provides an impression of
the characteristics of the ceramics and worked flints
in subsequent stages that could offer a typological
key for a new interpretation of the Hiide I phasing.
Schokland-P14 has yielded several house plans, argu-
ing for logistic mobility rather than residential mobili-
ty.7 However, it is not clear whether they belong to
SW2 or to the next phase, the Pre-Drouwen phase of
the West group of the Funnelbeaker culture (Trichter-
becherkultur, or TRB). In the Pre-Drouwen phase (c.
3,900-3,400 calBC), the Swifterbant (ceramic) tradi-
tion lived on, alongside new elements that are typical
for the early TRB, partly derived from the Michels-
berg culture, partly indigenous developments (Ten
Anscher 2012; 2015). The view that the Swifterbant
culture could have ignited the TRB North group as
4 Kampffmeyer’s interpretations and suggested periodisation,
relying heavily on the developments within the TRB North
Group, are no longer tenable, also because his periodisation
was not only hampered by the inadequate knowledge of the
Swifterbant culture at the time, but was also affected by his
assigning the same sherds to different phases. See Ten Anscher
2012, 585-589 for an extensive comment on the developmental
model presented in Kampffmeyer 1991. See also Ten Anscher
2015, 349-353. Still, Kampffmeyer’s Anhang B: Tafeln’, con-
taining hundreds of drawings of sherds, many so similar to the
sherds from for instance Schokland-P14, is a treasure of lasting
value.
5 Compare Ten Anscher 2012 fig. 5.1, 5.16, 5.19 and 7.4 with
Kampffmeyer 1991, Anhang B: Tafeln’.
6 It should be noted that the 14C-dates of the Swifterbant oc-
cupation of Medel-De Roeskamp, which on the basis of the ty-
pological characteristics of both its ceramics and its worked
flint must have been by and large contemporaneous with Layer
B from P14-Schokland, tally with the dates from Schokland-P14
and confirm the global validity of the dates proposed for Layers
A-C at Schokland-P14 (Ten Anscher 2012, 58-62).
7 The Overijsselse Vecht (German: Vechte) originates in Ger-
man Rhine-Westphalia.
Unexpected dimensions of a Swifterbant settlement at Medel-De Roeskamp (the Netherlands)
ing to a structural shift in the subsistence economy
(Gehasse 1995). There is an echo of this development
in the ritual depositions. Already in the earliest phase
of the Swifterbant culture pot offerings are attested,
as are offerings of red deer antlers and aurochs skulls
from bulls, whereas the oldest 14C dated depositions
of domestic cattle horns, again from bulls, occurred
much later.1
Residential mobility is considered to be the pre-
vailing settlement system, with hardly any exception
(Raemaekers 1999). Swifterbant sites are usually
situated in gradient rich wetland zones near (small)
streams. The duration of the habitation varied, depend-
ing on the possibilities of the location in question.
Some sites like Bergschenhoek, on a flat shore, are
thought to have been inhabited very shortly: just a few
days or a week at most (Louwe Kooijmans 1985). The
majority of Swifterbant sites, however, are located on
elevated grounds, and seem to have been periodically
used for tens of years to a few centuries, such as for
instance the river dune sites Hardinxveld-Giesendam-
Polderweg, Hardinxveld-Giesendam-De Bruin and
Hoge Vaart-A27 (all belonging to phase Swifterbant
1 [SW1] which is dated between c. 5,000-4,400 cal-
BC), the levee site Swifterbant-S3/S5 and other sites
on levees and sand dunes around Swifterbant, and
the river dune site Hazendonk-1 (all dating to phase
SW2: c. 4,400-3,900 cal BC).2 The discovery of adze
marks pointing to an arable field on a clayey levee at
Swifterbant-S4, and subsequently the identification of
tilled soils in thin sections arguing for arable fields at
other nearby sites as well, ended the discussion about
grain being imported or cultivated locally at wetland
sites in favour of the latter option.3
A few sites were situated on locations that con-
tinued to be easily accessible and inhabitable for a
very long time. They span the complete duration of
the Swifterbant culture and beyond. Hude I (Lower
Saxony, Germany), a palimpsest with a problem-
atic stratigraphy on a higher ground bordering lake
Dummer, is a well known example. At the time the
Hiide ceramics and features were published in full, its
Swifterbant component was heavily underestimated
1 Kroezenga et al. 1991; Prummel / Van der Sanden 1995;
Ufkes 1997; Raemaekers 2019.
2 Louwe Kooijmans 2001a; 2001b; Hogestijn / Peeters
2001; Peeters 2007; De Roever 2004; Raemaekers 1999. -
Phasing based on ceramic characteristics according to Ten
Anscher2012, 127-129.
3 Cappers / Raemaekers 2008; Huisman et al. 2009; Huis-
man / Raemaekers 2014; Schepers 2014.
(Kampffmeyer 1991).4 This is quite understandable
since back then publications on the Swifterbant cul-
ture were scarce, preliminary and based on just a few
sites, all dating to SW2 (e.g. De Roever 1979). It took
another two decades to reach a better understanding
of the width and depth of this culture. Hiide I has a
close western counterpart in Schokland-P14, located
on a large sand ridge on a substrate of boulder clay
bordering the Overijsselse Vecht (Ten Anscher 2012;
2015). In particular, the ceramics are very similar.5
Schokland-P14 has a more reliable chronostratigraphy
though its dating is based on 14C-dates from charred
food crusts on pottery sherds. If the crusts contained
fish remains (whether this is the case is unknown),
the dates might be too old due to the reservoir effect.6
Schokland-P14 at least provides an impression of
the characteristics of the ceramics and worked flints
in subsequent stages that could offer a typological
key for a new interpretation of the Hiide I phasing.
Schokland-P14 has yielded several house plans, argu-
ing for logistic mobility rather than residential mobili-
ty.7 However, it is not clear whether they belong to
SW2 or to the next phase, the Pre-Drouwen phase of
the West group of the Funnelbeaker culture (Trichter-
becherkultur, or TRB). In the Pre-Drouwen phase (c.
3,900-3,400 calBC), the Swifterbant (ceramic) tradi-
tion lived on, alongside new elements that are typical
for the early TRB, partly derived from the Michels-
berg culture, partly indigenous developments (Ten
Anscher 2012; 2015). The view that the Swifterbant
culture could have ignited the TRB North group as
4 Kampffmeyer’s interpretations and suggested periodisation,
relying heavily on the developments within the TRB North
Group, are no longer tenable, also because his periodisation
was not only hampered by the inadequate knowledge of the
Swifterbant culture at the time, but was also affected by his
assigning the same sherds to different phases. See Ten Anscher
2012, 585-589 for an extensive comment on the developmental
model presented in Kampffmeyer 1991. See also Ten Anscher
2015, 349-353. Still, Kampffmeyer’s Anhang B: Tafeln’, con-
taining hundreds of drawings of sherds, many so similar to the
sherds from for instance Schokland-P14, is a treasure of lasting
value.
5 Compare Ten Anscher 2012 fig. 5.1, 5.16, 5.19 and 7.4 with
Kampffmeyer 1991, Anhang B: Tafeln’.
6 It should be noted that the 14C-dates of the Swifterbant oc-
cupation of Medel-De Roeskamp, which on the basis of the ty-
pological characteristics of both its ceramics and its worked
flint must have been by and large contemporaneous with Layer
B from P14-Schokland, tally with the dates from Schokland-P14
and confirm the global validity of the dates proposed for Layers
A-C at Schokland-P14 (Ten Anscher 2012, 58-62).
7 The Overijsselse Vecht (German: Vechte) originates in Ger-
man Rhine-Westphalia.