New research at Rinnukalns, a Neolithic freshwater shell midden in northern Latvia.
Berzins, Valdis ; Brinker, Ute ; Klein, Christina 等
[ILLUSTRATION OMITTED]
Introduction
Shellfish have been a fundamental food resource for human
populations since the Palaeolithic, and shells as waste products of
their exploitation are preserved as archaeological features in coastal,
lacustrine and riverine environments worldwide (Alvarez et al. 2011). In
Europe, the most prominent features of this type are shell middens
created by Late Mesolithic hunter-gatherer societies on the Atlantic
littoral (Milner et al. 2007; Gutierrez-Zugasti et al. 2011). Marine
shellfish were also a fundamental food resource of Mesolithic people on
the south-western margin of the Baltic Sea, where large and famous
middens comparable to those on the Atlantic coast are preserved
(Andersen 2007: 32).
The marine molluscs of the eastern Baltic did not include species
large enough to be important as food, which explains the absence of
prehistoric coastal shell middens in this area. Freshwater mussels did,
however, represent a food resource in the Stone Age. There are thin
layers or small heaps of mussel shells at various inland prehistoric
sites in the circum-Baltic region, and their shells were widely used as
temper in early ceramics. Nevertheless, only a few freshwater shell
middens exist (Noe-Nygaard 1995? 64; Koch 2003? 217-18), and only one
such site is known in the eastern Baltic.
This site is Rinnukalns, on the bank of the River Salaca at its
outlet from Lake Burtnieks in northern Latvia (Figures 1 & 2). The
midden, forming a mound on the riverbank measuring 20 x 30m, was first
excavated in the 1870s, and sporadically re-investigated until the
1940s. After a break of almost 70 years, we have resumed research on
this important site. Excavation in 2011 demonstrated that significant
parts of the midden are still intact, yielding rich assemblages of fish
bone and mollusc shell, as well as herbivore, human and bird bones,
together with artefactual remains.
Research history
In the late nineteenth and early twentieth centuries, Rinnukalns
(formerly also named 'Rinnekalns' or 'Rinnehugel')
was considered one of the most famous Stone Age sites in the eastern
Baltic (Ebert 1913: 507, 1927-1928). It was first recognised as a
prehistoric site by local antiquarian Count Carl Georg Sievers, who
started excavations in 1873 and 1874 (Sievers 1875, 1877). He observed
well-stratified layers of freshwater mussel shells and fish, ash and
charred remains including larger animal bones and fragments of worked
bone and other artefacts (stone, amber and pottery). This stratified
sequence was 0.8-1.1m thick. The underlying sediment was also rich in
shell remains, animal bones, pottery and bone artefacts. Of special
importance were at least four human burials, with some bone and stone
grave goods, which were found in the lower shell deposit under intact
layers of the shell midden (Berzins et al. forthcoming). Sievers
considered these human remains, in contrast to others found in the upper
strata, as Stone Age burials. The midden itself he described as a
residential site of Stone Age hunters, fishers and gatherers.
This interpretation was challenged, and both the anthropogenic
origin of the shell midden and the age of the Stone Age graves were
questioned (Grewingk 1876, 1877). It was after a visit to Rinnukalns by
Rudolf Virchow, who essentially confirmed the observations of Sievers
(Virchow 1877), and the discovery of the Kunda site in present-day
northern Estonia, that the interpretation of Rinnukalns as a Stone Age
site was widely accepted (Grewingk 1884). The age of the presumed Stone
Age graves remained in dispute, because renewed excavation in 1881
uncovered only early modern graves (Sommer 1884). Although the site was
thereafter widely considered exhausted, successful smaller excavations
of undisturbed layers were carried out later by Karl von Lowis of Menar
(1895), Max Ebert (1913) and finally Eduards Sturms in 1943
(unpublished). However, due to the chequered history of the twentieth
century, its research potential was almost forgotten, and Rinnukalns
seemed to be only of interest in relation to research history.
[FIGURE 1 OMITTED]
New investigations
Archaeological investigation resumed in 2009 and 2010. Underwater
archaeological surveys discovered the first finds of animal bones and
bone artefacts on the bed of the River Salaca, directly in front of the
site (Figure 3 & Table 1).These were found lying on the stony
riverbed, now covered by up to 0.5m of mud, along the edge of the river
where the current flows slowly. Further underwater finds were made in
front of a second site, Kaulenkalns, on the riverbank opposite
Rinnukalns. [FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
On the Rinnukalns site itself, hollows were observed that might
have been left from the nineteenth-century trenches, and the question
arose as to whether it really had been completely destroyed, or whether
undisturbed cultural deposits might still be preserved.
In spring 2011,a geophysical survey was carried out at the site.
Geomagnetic prospection using an array of 6 Fluxgate magnetometers (Fa.
Dr. Forster, type Ferrex DLG 4.032.82) revealed a few features of
possible archaeological significance next to the midden. Coring results
supported this preliminary interpretation because the sub-soil was
slightly humic and contained small pieces of charcoal. More importantly,
a ground-penetrating radar survey, using 2-channel measuring equipment
(GSSI, SIR-20) and an antenna frequency of 400MHz, indicated the
existence of layered deposits, offering good prospects for the discovery
of intact midden stratigraphy (Figures 4 & 5).
To verify the geophysical results and assess the state of
preservation, a small-scale excavation was carried out in summer 2011.
Two trenches were excavated in the midden itself along the georadar
profile lines, to interpret the georadar profiles (Figure 4). The most
exciting discoveries were made in Trench 1,a 5m-long trench along the
edge of one of the hollows attributed to nineteenth-century excavation,
its eastern side precisely following the line of georadar profile 4.
Although the upper layers of the midden had been partially destroyed by
early modern graves and previous excavation, the lower part of the
midden was intact along the whole length of the trench, and an
unexcavated sequence of midden strata was preserved in part of the
trench (Figures 6 & 7). This corresponds very well to the georadar
image, where the intact layers show up as distinct reflection surfaces
(Figure 5). Evidently, there are areas of intact midden stratigraphy
beneath nineteenth-century spoil heaps, and even in areas already
excavated the base of the midden had not
been reached everywhere.
[FIGURE 5 OMITTED]
[FIGURE 6 OMITTED]
The 2011 excavation showed that the shell midden consisted of
alternating layers of unburnt mussel shell, burnt mussel shell and fish
bone, containing artefacts, animal and human bone as described by
Sievers (1875? 220-22) and Virchow (1877: 398-99). In the best-preserved
part of the profile, there were as many as 13 successive intact midden
layers. Several black burnt patches that might be remains of small
hearths were observed, particularly on top of the fisnbone layers. A
layer of compact, well-decomposed organic matter, also containing
pottery, underlay the basal midden layer in the small area, less than
1[m.sup.2], where the midden was excavated completely.
Artefactual material
Excluding the potsherds, 134 artefacts were recovered in the 2011
excavation. Predominant among these were whole and fragmentary bone and
antler tools: arrows, fragments of spears and harpoons, chisels,
wedge-shaped tools, awls, bodkins, and parts of composite fishhooks
(Figure 8, nos. 1-4). Jewellery consisted of tooth pendants (elk, wild
boar and beaver) and bird-bone pendants (Figure 8, nos. 5-10). Flint
tools and debitage, a stone axe and some grinding stones were also
unearthed.
[FIGURE 7 OMITTED]
The small ceramic assemblage from the organic deposit under the
shell midden is dominated by Typical Comb Ware (Figure 9, no.
1),tempered with crushed granitic rock and decorated with deep pits and
comb impressions. Typical Comb Ware marks the beginning of the Middle
Neolithic in the eastern Baltic region (for a chronological table, see
Larsson & Zagorska 2006: 3). Dates from the nearby Zvejnieki
cemetery place this complex approximately in the period 4460-3800 cal BC
(Zagorska 2006: 99-101, tab. IV). By contrast, the pottery from the
undisturbed midden layers is shell-tempered, with a striated surface,
most commonly decorated with geometric designs of wound cord
impressions, occasionally with shallow pits or incised linear decoration
(Figure 9, nos. 2-4). Comparable to the ceramics published already by
Virchow (1877: pl. XVIII), it fits the general pattern of
'hybrid' late Middle Neolithic pottery in the region, which
displays a mix of typological traits from the Comb and Narva ceramic
traditions (see Berzins 2008: 40).
Most of the other finds can similarly be attributed to the Middle
Neolithic period. However, the site as such was already being used at a
much earlier date. Thus, certain of the fragmentary bone points from
disturbed contexts represent Mesolithic forms. Particularly
characteristic is a fish spear with a triangular cross section,
recovered in Sievers' excavation. Human remains
Besides a larger number of human bones found in layers associated
with nineteenth-century trenches, which probably derive from early
modern burials cut into the midden, disarticulated human bones found in
apparently undisturbed shell midden layers were of special interest. Two
of these were radiocarbon-dated to test whether they were prehistoric.
The first was the right humerus of a neonate, not older than 12 months.
It was found at the boundary between the disturbed part and layer 7, the
uppermost undisturbed layer (Figure 6), but radiocarbon dating (see
below) shows that it is late medieval in date, and clearly reworked
during the earlier excavations.
[FIGURE 8 OMITTED]
The second bone fragment was found significantly lower, in the
undisturbed layer 15 (Figure 6). The expected prehistoric age was
confirmed by radiocarbon dating (see below), providing the first proof
of Stone Age human remains at Rinnukalns, and suggesting that at least
some of Sievers' Stone Age graves may have been correctly
attributed to this period (Berzins et al. forthcoming). The layer 15
fragment is a part of a maxilla, consisting of parts of the right and
the left side of the alveolar process, which fit together at the median
palatine suture. The maxilla belonged to a juvenile aged between 12 and
18. The sex was not determinable.
Fish remains
The original excavations at Rinnukalns produced a considerable list
of mammal, bird and fish species, including wild animals, domesticates
and even marine species (Rutimeyer 1877). However, their stratigraphical
positions are questionable. During the 2011 excavation, in addition to a
relatively small assemblage of mammal and bird bones, thousands of fish
remains were collected (Table 2). So far, 2222 bones (NISP) have been
identified at least to family level, and mostly to species level (Figure
10). These bones are from bulk samples taken from three different midden
layers in which fish remains were clearly visible. All samples were
wet-sieved down to 2mm and fully sorted.
The assemblage is dominated by several cyprinid species, which are
archaeozoologically difficult to distinguish (Cyprinidae, 49.0% NISP,
26.5% MNI); seven species are definitely proven. Perch (Perca
fluviatilis, 26.0% NISP, 26.5% MNI) and eel (Anguilla amuilla, 20.7%
NISP, 27 MNI) are the next most frequent. Of minor importance were pike
(Esox lucius, 2.3% NISP,11.4% MNI) and pike-perch (Sander lucioperca,
1.8% NISP, 6.1% MNI). Other species are recorded only by single remains
(cf. Figure 10). Comparison with the results of monitoring programmes
(Birzaks et al. 2011) shows that the modern fish community is very
similar to that of the Stone Age, confirming that the local ecosystem
has been extraordinarily stable for more than 5000 years.
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
Eel bones are significantly more frequent in the upper layers than
in the basal ones and the average length of eel vertebrae increases from
3.2mm (n = 6) in the bottom layer, to 5.1mm (n = 38) in the middle
layers, and 5.6mm (n = 249) at the top of the midden. This represents an
increase in the average length of the eels caught, from c. 370mm to c.
65?mm (Figure 11). The increasing length is a species-specific
phenomenon; cyprinids, pike or perch do not show the same development.
The estimated average length of perch, for instance, varies between 190
and 240mm in all three samples. Clearly, there was a conscious attempt
to catch large eels. This probably indicates a specialised fishing of a
migratory species, which has not previously been recorded from the
eastern Baltic Neolithic (Lougas 1996: tab. 2).
[FIGURE 11 OMITTED]
Mollusc remains
The mollusc fauna from the 2011 excavation (Rudzite et al. 2012) is
overwhelmingly dominated by swollen river mussel, Unio tumidus. Other
large edible mussels, Unio pictorum (painters mussel) and Anodonta sp.,
were also present, as well as the much smaller gastropod (snail) species
Viviparus contectus and Valvata piscinalis. Many whole mussel valves
were found, with damage only to the thin edges and loss of the coloured
surface layer (periostracum). Signs of deliberate breakage or pounding
are absent, although these may have been present on the edges. Since
Anodonta tend to be relatively poorly preserved, their shells being
naturally thinner, they may have constituted a higher proportion of the
gathered assemblage. U. tumidus will also inhabit locations with a slow
current, whereas the other mussel species are typical of plant-rich,
warm water with a muddy bottom, which suggests that they were collected
from the lake margins, rather than the river; mesotrophic lakes in this
area typically have muddy bottoms.
In the land snail assemblage from the intact midden strata, two
habitats are represented: robust pillar (Cochlicopa nitens) and lovely
vallonia (Valloniapulchella) are characteristic of calcareous wetlands,
whereas moss chrysalis (Pupilla muscorum) and ribbed vallonia (Vallonia
costata) are also calciphiles, but need dry, stony meadows with sunny
habitats. Typical forest species are absent, even though the majority of
Latvia's land snails are forest species. Snail habitat preferences
suggest residence on the midden itself, which was probably not covered
by trees, and in meadow-like areas close to the lakeshore.
Radiocarbon dating and stable isotopes
Twelve samples from Trench 1 of the 2011 excavation were submitted
for radiocarbon dating (Table 1).Most were bones of terrestrial
herbivores (aurochs, wild boar and red deer). Three charcoal samples,
two human bones and one bird bone were also submitted. Two bones from
the organic deposit below the midden failed, owing to a lack of
collagen.
Two of the dated bones were still in articulation and must have
been deposited before their soft tissues had completely decayed. The
lack of mixing between fishbone-rich and shell-rich layers also suggests
that these layers are intact, so we may assume that most bone and
charcoal samples are contemporary with the layer in which they were
found. Dating of the disarticulated human maxilla confirmed that it came
from a disturbed prehistoric burial. The neonate humerus found at the
interface between layer 7 and one of the older archaeological trenches
dated no earlier than cal AD 1420-1480 (95% confidence).
The remaining midden samples date to the later fourth millennium,
while charcoal from layer 20, the organic deposit under the midden,
dates to the mid sixth millennium cal BC (Table 1).The layer 15 charcoal
fragment is apparently older than the charcoal from layer 17, and must
be reworked from older sediments; layer 15 also contained the
redeposited maxilla. Otherwise, the results fit the stratigraphic
sequence of layers 17 to 7 (Figure 12). JL he disarticulated aurochs
bone in layer 6 is slightly older than the rest of the midden samples,
but it is stratigraphically later than the medieval human bone in layer
7, and may be reworked. The undisturbed midden layers excavated in 2011
may have been deposited over only a few years: the remaining results
from terrestrial samples are statistically consistent with a single
date, probably within a decade or two of 3350 cal BC. The layer 6
aurochs and the stray find dated by KLA-43697 to 3090-2900 cal BC (Table
1) indicate that the site was used over a longer period.
Collagen extracted for radiocarbon dating was subsampled for stable
isotope analysis. The layer 15 bird bone, from a red-necked grebe
(Podicepsgrisegena, which feeds mainly on small fish), gave isotopic
values suggesting a diet based on marine or estuarine species (Figure 13
& Table 1),consistent with its radiocarbon age, which implies a
dietary reservoir effect of c. 250-300 radiocarbon years. Data from
modern mollusc and water, and mid-Holocene plant macrofossil samples
indicate that the local freshwater reservoir effect was probably c.
800-900 radiocarbon years (Meadows et al. 2014). This bird therefore
died before it could acquire a more local isotopic signature.
[FIGURE 12 OMITTED]
Isotope results (enriched [delta][sup.15]N, depleted
[delta][sup.13]C; Table 1) from the prehistoric maxilla suggest that
this individual consumed a good deal of freshwater fish and shellfish;
the values are at the aquatic end of the range of human bone isotope
results from Zvejnieki (cf. Eriksson 2006). The medieval neonate had a
more terrestrial diet; the high [delta][sup.15]N value is probably a
nursing effect, not the result of fish consumption. The [delta][sup.13]C
results do not indicate significant consumption of plants using the C4
photosynthetic pathway (e.g. millet), or of marine resources. Results
from herbivore bones are consistent with expectations for wild animals
from a mid-Holocene forest habitat.
Rinnukalns and its significance in eastern Baltic prehistory
Geophysical survey and test excavation have shown that parts of the
Rinnukalns shell midden are still preserved in situ, with an undisturbed
stratigraphy at least 0.8m thick. The newly excavated area of the midden
represents a short phase of the eastern Baltic late Middle Neolithic. An
early Middle Neolithic (Typical Comb Ware) layer exists under the shell
midden, and the radiocarbon date from layer 20 and some typical bone
tools from the nineteenth-century excavation indicate Mesolithic
occupation as well. Small assemblages of pottery, lithics, bone
artefacts and tooth ornaments have been recovered from secure contexts,
as well as the first confirmed prehistoric human remains at this site.
[FIGURE 13 OMITTED]
Preservation of mollusc shells and fish bones is exceptional. This
site currently represents the best source of information on the
prehistoric use of freshwater mussels in the Baltic region. A wide range
of fish species has also been identified, and there are indications that
the species composition and even the dimensions of fish vary between
midden layers, although cyprinids, eel and perch predominate in all the
sampled contexts.
In the eastern Baltic, as in neighbouring areas, the term
'Neolithic has generally been retained as a designation for the
ceramic final stage of the Stone Age. Thus, the Mesolithic and Neolithic
simply correspond to 'aceramic' and 'ceramic',
respectively. Currently, various lines of evidence point to the Middle
Neolithic as the period of incipient agriculture in the eastern Baltic
(for more details see Berzins 2008: 371-73, 403-404). This period may be
viewed, following Zvelebil (1996: 329), as representing a prolonged
'substitution phase', during which wild food resources
continued to provide the main subsistence basis.
The settlement of this period at Rinnukalns maintains the focus on
rich concentrations of aquatic resources associated with the major lake
basins. The finds from the old excavations, the artefactual evidence
pointing to a range of domestic activities, as well as the burials,
indicate that Rinnukalns was much more than a specialised
fishing/mollusc-gathering station. Rather, it may be classed along with
the settlement complex at Zvejnieki and similar sites in other lake
basins in the region as a multi-purpose residential site in a location
offering excellent access to a range of wild food resources--with the
difference that during the Middle Neolithic this particular spot came to
be used for the mass dumping of mussel shells.
The thick shell accumulation sets Rinnukalns apart from other
Middle Neolithic sites in the eastern Baltic. On the other hand, in
terms of site location, economic basis and artefact typology, it fits
the regional pattern. In all of these respects, we may identify
particularly close parallels with the approximately contemporaneous
settlement of Tamula in south-eastern Estonia (Jaanits 1984). Similarly
situated by a lake outlet, Tamula has produced very similar bone tools
and ceramics, and a faunal assemblage consisting primarily of wild
mammal, bird and fish bone, supplemented with a small quantity of
remains from domestic livestock. A further parallel is the occurrence of
burials within the settlement area (Jaanits 1957). Especially
significant is grave XIX, which not only gave an uncalibrated
radiocarbon date and a [delta][sup.13]C value (Kriiska et al. 2007: tab.
1:lab. no. Hela-1337; [sup.14]C-year: 4925 [+ or -] 25 BP;
[delta][sup.13]C: -25.0) almost identical to that of the human maxilla
from layer 15, but also contained some bird figurines of bone (Kriiska
et al. 2007: fig.10, AI 4118: 2501, 2502) comparable to finds from
Rinnukalns (Ebert 1927-1928: pi. 25n).
The widespread use of mussel shell as temper in ceramics suggests
that freshwater mussels were in fact a significant element of diet
across the eastern Baltic region. Sites with mussel shells are quite
exceptional, however. Apart from Rinnukalns, evidence for large-scale
consumption of shellfish has only been found at Narva Riigikula I in
northeastern Estonia (Indreko 1964? 68-79; Gurina 1967: 22,159,
figs.10,12 & 14). Mussel shells are also reported from Neolithic
layers at other sites along the River Narva (Indreko 1964: 66; Kriiska
1996: fig. 3).
The general absence of shell layers at Neolithic sites might be
explained by a combination of factors: the re-use of shell for pottery
temper, non-preservation due to soil acidity, and/or destruction in
historical times as a source of agricultural lime. On the other hand, it
is questionable whether the shell from freshwater mussels collected
primarily for food could have completely disappeared as tempering
material; and post-depositional factors do not satisfactorily account
for the absence of shell layers at stratified sites with well-preserved
antler and bone.
In the neighbouring southern Baltic, Unio sp. freshwater mussels
were used not only as food but also for the production of ornaments in
the Brzesc Kujawski group of the Lengyel culture, which represents the
northernmost tradition of the Central European Early Neolitnic. At
Brzesc Kujawski sites, freshwater mollusc shells have been found as
production or consumption waste in various Kinds of pits, including a
special type of pit with a regular bell-shaped profile that could have
been used for the live storage of shellfish (Grygiel & Bogucki
1981:17;1986: 130). Similarly, pits with large quantities of Unio sp.
shells (Krysiak & Lasota 1973) are known from the settlement of
Zawichost-Podgorze in south-eastern Poland.
In the north merman lowlands and Denmark, freshwater molluscs are
very sparsely represented at sites of the fifth and fourth millennia BC,
contrasting with the plentiful evidence of marine mollusc consumption.
From late Ertebolle contexts there are only scattered examples of
freshwater mollusc shells used as beads or pendants (e.g. Lubke 2009:
560). So far, the only evidence for actual consumption of freshwater
mussels comes from the small inland sites within the extensive Amose bog
area on Sjaelland, Denmark. Small heaps or middens of the shells are
reported from several sites, generally belonging to the Funnel Beaker
culture (Noe-Nygaard 1983, 1995: 64; Koch 1998:147, 2003; Fischer &
Heinemeier 2003). They occur in definite connection with traces of food
preparation involving fire, namely charcoal concentrations and burnt
shell or fish remains (Noe-Nygaard 1983: 137)--very similar to
Rinnukalns.
The sporadic evidence for freshwater mollusc consumption across the
Baltic region presents an enigma. On current evidence it is difficult to
say whether mussels represented a secondary resource that obtained
dietary importance only in certain periods when the exploitation of
primary resources was in some way restricted, or whether shellfish
consumption was actually a more widespread dietary practice than is
indicated by the preserved evidence. Certainly, a much clearer
understanding of freshwater shellfish as a component in the annual cycle
of food acquisition is required, and in this regard Rinnukalns offers
exciting possibilities. Already at the time of the 2011 excavation, it
was thought that the sequence of distinct layers of fish bone, unburnt
shell and burnt shell might represent a recurring seasonal pattern. The
radiocarbon results, demonstrating that this sequence probably
accumulated over only a few years, appear to fit this interpretation and
suggest that each of the midden layers is a seasonal deposit reflecting
intensive exploitation of particular resources. Thus the Rinnukalns
midden offers a very fine-grained picture of seasonality in resource
exploitation, which will be a focus of future research.
Moreover, research will address the place of Rinnukalns in the Lake
Burtnieks regional settlement system. The relationsmp between Rinnukalns
and Zvejnieki is still unclear, but several indicators suggest that
Rinnukalns was not simply a fishing station for the Zvejnieki
settlement. The Zvejnieki graveyard demonstrates that Lake Burtnieks was
a preferred occupation area throughout the Stone Age. Further research
on both sites may clarify which resources were used at what time in the
area, when Rinnukalns became an important component of the settlement
system, and why freshwater mussels became extraordinarily significant at
the end of the eastern Baltic Middle Neolithic.
Acknowledgements
The research described in this paper has been undertaken within the
framework of the strategic research theme 'Man and
Environment' of the Centre for Baltic and Scandinavian Archaeology;
and within projects 09.1539 and 276/2012 funded by the Latvian Council
of Science, implemented at the Institute of Latvian History at the
University of Latvia.
Field research at Rinnukalns in 2011 and analysis of excavated
material was undertaken with support from the State Culture Capital
Foundation of Latvia, the Centre for Baltic and Scandinavian
Archaeology, and the Baltic-German University Liaison Office with funds
from the German Academic Exchange Service (DAAD). Gunita Zarina
(Institute of Latvian History) identified the human bone. Prof. Carl
Heron and Andy Gledhill (Bradford University) undertook the stable
isotope measurements. Prof. Oliver Nelle (Kiel University) identified
charcoals for radiocarbon dating. Maris Rudzitis (University of Latvia)
assistea in the interpretation of mollusc remains. We would like to
thank Prof. Aivar Kriiska (Tartu), Mari Torv (Schleswig) and two
anonymous reviewers for helpful comments on an earlier draft of the
text.
References
ALVAREZ, M., I.B. GODINO, A. BALBO & M. MADELLA. 2011. Shell
middens as archives of past environments, human dispersal and
specialized resource management. Quaternary International 239? 1-7.
http://dx.doi.Org/10.10l6/j.quaint. 2010.10.025
ANDERSEN, S.H. 2007. Shell middens ('Kokkenmoddinger') in
Danish prehistory as a reflection of the marine environment, in N.
Milner, O. E. Craig & G.N. Bailey (ed.) Shell middens in Atlantic
Europe: 31-45. Oxford: Oxbow.
BERZINS, V. 2008. Sarnate: living by a coastal lake during the East
Baltic Neoliwic (Acta Universitatis Ouluensis B Humaniora 86). Oulu:
Oulu University Press. Available at: http://herkules.oulu.
fi/isbn9789514289415/isbn9789514289415.pdf (accessed 26 June 2013).
BERZINS, V., U. BRINKER, H. LUBKE, J. MEADOWS & I. ZAGORSKA.
Forthcoming. New research on the human burials of Rinnukalns, Latvia, in
Mesolithic burials--rites, symbols and social organization of early
postglacial communities. Halle: Congresses of the State Museum for
Prehistory Halle.
BIRZAKS, J., E. ALEKSEJEVS & M. STRUGIS. 2011. Occurrence and
distribution of fish in rivers of Latvia. Proceedings of the Latvian
Academy of Sciences Section B 65? 57-66.
BRONK RAMSEY, C. 2009. Bayesian analysis of radiocarbon dates.
Radiocarbon 51:337-60.
EBERT, M. 1913. Die baltischen Provinzen Kurland, Livland, Estland.
Prahistorische Zeitschrift 5: 498-559.
--1927-1928. Rinnekalns, in M. Ebert (ed.) Reallexikon der
Vorgeschichte 11:145-47. Berlin: Walter de Gruyter.
ERIKSSON, G. 2006. Stable isotope analysis of human and faunal
remains from Zvejnieki, in L. Larsson & I. Zagorska (ed.) Back to
the origin. New research in the Mesolithic-Neolithic Zvejnieki cemetery
and environment, northern Latvia (Acta Archaeologica Lundensia series in
8[degrees] 52):183-215. Stockholm: Almqvist & Wiksell.
FISCHER, A. & J. HEINEMEIER. 2003. Freshwater reservoir effect
in [sup.14]C dates of food residue on pottery. Radiocarbon 45: 449-66.
GREWINGK, C.1876. Der Kauler- und der Rinne-Kaln am Burtnecksee in
Livland. Sitzungsberichte Dorpater Naturforscher-Gesellschaft 4: 206-25.
--1877. Zur Abwehr. Sitzungsberichte Dorpater
Naturforscher-Gesellschaft A\ 534-39.
--1884. Die Neolithischen Bewohner von Kunda in Estland und deren
Nachbarn. Dorpat (Tartu): Mattiesen.
GRYGIEL, R. & P.I. BOGUCKI. 1981. Early Neolithic sites at
Brzesc Kujawski, Poland: preliminary report on the 1976-1979
excavations. Journal of Field Archaeology 8: 9-27.
http://dx.doi.org/10.1179/009346981791505201
--1986. Early Neolithic sites at Brzesc Kujawski, Poland:
preliminary report on the 1980-1984 excavations. Journal of Field
Archaeology 13: 121-37. http://dx.doi.org/10.1179/009346986791535762
GURINA, N.N.1967. Iz istorii drevnihpiemen zapadnyh oblastej SSSR
(po materialam narvskoj ekspedicii) [On the history of ancient tribes of
the western regions of the USSR (based on the material of the Narva
expedition)]. Leningrad: Nauka.
GUTIERREZ-ZUGASTI, I., S.H. ANDERSEN, A.C. ARAUJO, C. DUPON, N.
MILNER & A.M. MONGE-SOARES. 2011. Shell midden research in Atlantic
Europe: state of the art, research problems and perspectives for the
future. Quaternary Internationally'. 70-85.
http://dx.doi.Org/10.10l6/j.quairit.2011.02.031
INDREKO, R. 1964. Mesolithische und Fruhneolithische Kulturen in
Osteuropa und Westsibirien (Antikvariska Serien 13). Stockholm: Almqvist
& Wiksell.
JAANITS, L.1957. Neue Graberfunde auf dem spatneolitischen
Wohnplatz Tamula in Estland, in Meinander (ed.) Studia neolitica in
honorem Aarne Ayrapaa. 80-100. Helsinki: Suomen muinaismuistoyhdistys.
--1984. Die kennzeichnenden Zuge der Siedlung Tamula. Iskos 4:
183-93.
KOCH, E.1998. Neolithic bog pots from Zealand, Mon, Lolland and
Falster (Nordiske Fortidsminder Serie B 16). Copenhagen: Det Kongelige
Nordiske Oldskxiftselskab.
--2003. Stenalderbonder pa sommerjagt. Nationalmuseets Arbejdsmark
2003: 209-29.
KRIISKA, A. 1996. Archaeological excavations on the Neolithic site
of Riigikula IV. Proceedings of the Estonian Academy of sciences.
Humanities and Social Sciences 45(4):410-19.
KRIISKA, A., L. LOUGAS, M. LOHMUS, K. MANNERMAA & K. JOHANSON.
2007. New AMS dates from Estonian Stone Age burial sites. Estonian
Journal of Archaeology 11(2): 83-121.
KRYSIAK, K. & A. LASOTA. 1973. Szczatki zwierzece z wykopu IV
na neolitycznym stanowisku w Zawichoscie. Wiadomosci Archeologiczne 38:
83-89.
LARSSON, L. & I. ZAGORSKA (ed.). 2006. Back to the origin. New
research in the Mesolithic-Neolithic Zvejnieki cemetery and environment,
northern Latvia (Acta Archaeologica Lundensia series in 8[degrees] 52).
Stockholm: Almqvist & Wiksell.
LOUGAS, L.1996. Stone Age fishing strategies in Estonia. What did
they depend on? Archaeofauna 5: 101-109.
LOWIS OF MENAR, K. VON. 1895. Bericht uber Ausgrabungen am
Burtneckschen See und am Rinnekaln. Sitzungsberichte Gesellschaft fur
Geschichte und Alterthumskunde der Ostseeprovinzen Russlands 1895:
77-84.
LUBKE, H. 2009. Hunter and fisher in a changing world.
Investigations on submerged Stone Age sites off the Baltic coast of
Mecklenburg-Vorpommern, Germany, in S. McCartan, R. Schulting, G. Warren
& P. Woodman (ed.) Mesolithic horizons. Papers presented at the
Seventh International Conference on the Mesolithic in Europe, Belfast
2005? 556-63. Oxford: Oxbow.
MEADOWS, J., H. LUBKE, I. ZAGORSKA, V. BERZINS, A. CERINA & I.
OZOLA. 2014. Potential freshwater reservoir effects in a Neolithic shell
midden at Rinnukalns, Latvia. Radiocarbon 56: 823-32.
http://dx.doi.org/10.2458/56.16950
MILNER, N., O.E. CRAIG & G.N. BAILEY. 2007. Shell middens in
Atlantic Europe. Oxford: Oxbow.
NOE-NYGAARD, N.1983. The importance of aquatic resources to
Mesolithic man in inland sites in Denmark, in C. Grigson & J.
Clutton-Brock (ed.) Animals and archaeology 2: shell middens, fishes and
birds (British Archaeological Reports international series 183): 125-42.
Oxford: Archaeopress.
--1995. Ecological, sedimentary and geochemical evolution of the
late-glacial to postglacial Amose lacustrine basin, Denmark (Fossils
& Strata 37). Copenhagen, Oslo & Stockholm: Scandinavian
University Press.
REIMER, P.J., M.G.L. BAILLIE, E. BARD, A. BAYLISS, J.W. BECK, P.G.
BLACKWELL, C. BRONK RAMSEY, C.E. BUCK, G.S. BURR, R.L. EDWARDS, M.
FRIEDRICH, P.M. GROOTES, T.P. GUILDERSON, I. HAJDAS, T.J. HEATON, A.G.
HOGG, K.A. HUGHEN, K.F. KAISER, B. KROMER, G. MCCORMAC, S. MANNING, R.W.
REIMER, S. REMMELE, D.A. RICHARDS, J.R. SOUTHON, S. TALAMO, F.W. TAYLOR,
C.S.M. TURNEY, J. VAN DER PLICHT & C.E. WEYHENMEYER. 2009. IntCal09
and Marine09 radiocarbon age calibration curves, 0-50,000 years cal BP.
Radiocarbon 51:111 1-50.
RUDZITE, M., I. ZAGORSKA, H. LOBKE & V. BERZINS. 2012. Molluscs
from a new archaeological excavation at Rinnukalns, Latvia. Schriften
zur Malakozoologie 27: 49-56.
RUTIMEYER, L. 1877. Ueber die Thierreste des Rinnekalns.
Sitzungsberichte Dorpater Naturforscher-Gesellschaft 4: 534-39.
SIEVERS, C.G. 1875. Ein normannisches SchifFsgrab bei Ronneburg und
die Ausgrabung des Rinnehugels. Verhandlungen der Berliner Gesellschaft
jur Anthropologie, Ethnologie und Urgeschichte 1875 (Zeitschrift jur
Ethnologie 7)? 214-25.
--1877. Ueber die wahrscheinliche Entstehung des Rinnehugels.
Sitzumsberichte Dorpater Naturforscher-Gesellschaft A: 406-17.
SOMMER, A. 1884. Der Rinne-Kalns und seine Bedeutung fur die
Anthropologie Livlands. Archiv fur die Naturkunde Liv-, Ehst- und
Kurlands. Serie II IX: 345-99.
VIRCHOW, R.1877. Bericht iiber eine archaologische Reise nach
Livland. Verhandlungen der Berliner Gesellschaft fur Anthropologie,
Ethnologie und Urgeschichte 1877 (Zeitschrift fur Ethnologie 9):
365-437.
ZAGORSKA, I. 2006. Radiocarbon chronology of the Zvejnieki burials,
in L. Larsson & I. Aagorska (ed.) Back to the origin. New research
in the Mesolithic-Neolithic Zvejnieki cemetery and environment, northern
Latvia (Acta Archaeologica Lundensia series in 8[degrees] 52): 91-113.
Stockholm: Almqvist & Wiksell.
ZVELEBIL, M. 1996. The agricultural frontier and the transition to
farming in the circum-Baltic region, in D.R. Harris (ed.) The origins
and spread of agriculture an dpastoralism in Eurasia-. 323-45. London:
University College London Press.
Received: 6 August 2013; Accepted: 13 November 2013; Revised: 5
February 2014
Valdis Berzins (1), Ute Brinker (2), Christina Klein (3), Harald
Lubke (4), John Meadows (4), Mudlte Rudzlte (5), Ulrich Schmolcke (4),
Harald Stumpel (3) & Ilga Zagorska (1)
(1) Institute of Latvian History, University of Latvia, Kalpaka
bulvaris 4, Riga 1050, Latvia
(2) State Archaeology, State Authority for Culture and Preservation
of Monuments Mecklenburg-Vorpommern, Domhof 4/5, Schwerin 19055, Germany
(3) Department of Geophysics, Institute of Geosciences,
Christian-Albrechts-University Kiel, Otto-Hahn-Platz 1, Kiel 24118,
Germany
(4) Centre for Baltic and Scandinavian Archaeology,
Schleswig-Holstein State Museums Foundation, Schlossinsel 1, Schleswig
24837, Germany
(5) Museum of Zoology, University of Latvia, Kronvalda bulvaris 4,
Riga 1586, Latvia
Table 1. Radiocarbon results from the underwater survey 2010 and the
Trench 1 midden sequence 2011, calibrated using IntCal09 (Reimer et
al. 2009) and OxCal v4.2 (Bronk Ramsey 2009).
Laboratory
Findspot/layer number Identification
Salaca riverbed
Kaulenkalns bank KIA-43696 antler handle, red deer (Cervus
elaphus)
Rinnukalns bank KIA-43697 cattle/aurochs (cf. Bos sp.)
Excavation 2011, Trench 1
Layer 6 KIA-45798 cattle/aurochs (cf. Bos sp.),
R humerus
Layer 7 KIA-45799 wild boar (Sus scrofa), phalanx II
Layer 7 KIA-45800 human (Homo sapiens) y neonate,
< 1 year; R humerus
Layer 9 KIA-45801 cattle/aurochs (Bos sp.), R metatarsal
Layer 10 KIA-45802 cattle/aurochs (Bos sp.), L proximal
tibia, unfused epiphysis
Layer 15 KIA-45803 human (Homo sapiens), juvenile,
12-18; R maxilla
Layer 15 KIA-45804 charcoal, Alnus sp.
Layer 15 KIA-45805 Podicipedidae, cf. red-necked grebe
(Podiceps grisegena), humerus
Layer 17 KIA-45809 charcoal, Corylus sp.
Layer 19 KIA-45806 red deer [Cervus elaphus), L radial
carpal
Layer 20 KIA-45807 bone, indet. (cf. medium-large
mammal)
Layer 20 KIA-45808 charcoal, indet. bark
[[delta].sup.15]N [[delta].sup.13]N
Laboratory ([per thousand]) ([per thousand])
Findspot/layer number ([double dagger]) ([double dagger])
Salaca riverbed
Kaulenkalns bank KIA-43696 -23.8
Rinnukalns bank KIA-43697 5.34 -23.22
Excavation 2011, Trench 1
Layer 6 KIA-45798 6.26 -22.61
Layer 7 KIA-45799 5.12 -22.46
Layer 7 KIA-45800 12.55 -20.44
Layer 9 KIA-45801 6.42 -23.08
Layer 10 KIA-45802 6.35 -22.84
Layer 15 KIA-45803 11.98 -24.85
Layer 15 KIA-45804
Layer 15 KIA-45805 10.88 ?17.09
Layer 17 KIA-45809
Layer 19 KIA-45806
Layer 20 KIA-45807
Layer 20 KIA-45808
Laboratory Conventional Calibrated date
Findspot/layer number 14C age (BP) (95% confidence)
Salaca riverbed
Kaulenkalns bank KIA-43696 7105 [+ or -] 30 6030-5910 cal BC
Rinnukalns bank KIA-43697 4360 [+ or -] 25 3090-2900 cal BC
Excavation 2011, Trench 1
Layer 6 KIA-45798 4725 [+ or -] 30 3640-3370 cal BC
Layer 7 KIA-45799 4540 [+ or -] 25 3370-3100 cal BC
Layer 7 KIA-45800 440 [+ or -] 25 cal AD 1420-1470 *
Layer 9 KIA-45801 4620 [+ or -] 35 3510-3340 cal BC
Layer 10 KIA-45802 4575 [+ or -] 35 3500-3120 cal BC
Layer 15 KIA-45803 4935 [+ or -] 25 3780-3650 cal BC *
Layer 15 KIA-45804 4755 [+ or -] 30 3640-3380 cal BC
Layer 15 KIA-45805 4840 [+ or -] 25 #
Layer 17 KIA-45809 4560 [+ or -] 20 3370-3130 cal BC
Layer 19 KIA-45806 no yield
Layer 20 KIA-45807 no yield
Layer 20 KIA-45808 6440 [+ or -] 30 5480-5320 cal BC
([double dagger]) stable isotopes were measured using Isotope Ratio
Mass Spectrometry on a Thermo Flash 1112 Elemental Analyser coupled
to a Thermo Delta plus XL mass spectrometer.
Typicd measurement errors of [+ or -] 0.2 [per thousand] are quoted
for both [[delta].sup.15]N and [[delta].sup.13]C in both samples and
standards. Duplicate results were obtained, and agree within the
quoted errors; mean values are shown.
* maximum age--would be more recent if diet incorporated significant
amounts of fish.
# cannot be calibrated because of dietary reservoir effects.
Table 2. Rinnukalns, 2011 excavation. List of identified animals from
undisturbed contexts. NISP: Number of Identified Specimens.
Fish NISP (n) NISP (%)
Cyprinids 1047 49.0
(roach, Rutilus rutilus 21)
(vimba, Vimba vimba 8)
(bream, Abramis brama 8)
(ide, Leuciscus idus 2)
(nase, Chondrostoma nasus 1)
(chub, Squalius ephalus 1)
Perch (Perea fluviatilis) 577 26.0
Eel (Anguilla anguilla) 460 20.7
Pike (Esox lucius) 52 2.3
Pike-perch (Sander lucioperca) 39 1.8
Burbot (Lota lota) 3 0.1
Herring (Clupea harengus) 2 0.1
Salmonids (Salmo sp.) 1 <0.1
Total 2222 100
Fish Mammal NISP (n)
Cyprinids Beaver (Castorfiber) 5
(roach, Rutilus rutilus Cattle/aurochs (?wsp.) 3
(vimba, Vimba vimba Elk (Alces alces) 3
(bream, Abramis brama Red deer (Cervus elaphus) 1
(ide, Leuciscus idus Wild boar (Sus scrofa) 2
(nase, Chondrostoma nasus 14
(chub, Squalius ephalus
Perch (Perea fluviatilis)
Eel (Anguilla anguilla)
Pike (Esox lucius)
Pike-perch (Sander lucioperca) NISP
Burbot (Lota lota) Bird (n)
Herring (Clupea harengus) Red-necked grebe 4
Salmonids (Salmo sp.) (Podiceps grisegena)
Total 4
