Aivar Kriiska

Published: Coastal Estonia. Recent Advances in Environmental and Cultural History. PACT 51. Rixensart 1996, 373–384.
 
 

Abstract

There are ten known Neolithic settlement sites in the lower course the Narva River within an area of approximately 200 km² (Fig. 1). Narva Joaoru, Riigiküla I–IV, Väiküla, Kudruküla and Lommi I–III. All of these, except Lommi, are located by the banks of the Narva River, but are rather more connected to the seashore of Neolithic times. Archeological excavations have been made at six sites : Lommi III (1940, 1952) (Indreko, 1940; Gurina, 1961), Riigiküla I (1951–52), Riigiküla II (1952), Riigiküla III (1953, 1958) (Gurina, 1955, 1958, 1967), Narva Joaoru (1954, 1957, 1960, 1962–1964) (Jaanits, 1954, 1957, 1960, 1962, 1963, 1964) and Kudruküla (1980–1981) (Jaanits, 1981; Efendijev 1983). In all, about 2000 m² of the settlement area have been studied. The assembled material is vast, including over 30,000 pottery sherds.

The pottery finds are of six basic types: (1) Narva-type pottery, (2) Typical Combed Ware pottery, (3) Late Combed Ware pottery, (4) Corded Ware, (5) Early Textile-impressed Ware. The Corded Ware and early Textile-impressed Ware comprise only an insignificant part (over 70 sherds) of the total. Since this material is less revealing from a technological standpoint, these pottery types will not be considered in this survey and only the Narva-type pottery. Typical Combed Ware pottery and Late Combed Ware pottery will be described. Besides such common denominators as their conical shape and pointed or round base, these types can also be related by their marker's livelihood fishing and hunting. So, if a farming economy is the primary feature of the Neolithic then this pottery should be considered as sub-Neolithic.

The Narva-type pottery has been found at Narva Joaoru, Lommi III and all four Riigiküla settlements. Typical Combed Ware pottery has been encountered at Narva Joaoru, all four of the Riigiküla settlements, Lommi I and III and the Väiküla and Kudruküla settlements. Late Combed Ware pottery has been recovered from the Riigiküla I, II and III, Narva Joaoru, Lommi I and III, Väiküla and Kudruküla settlements.

All three types of pottery were probably produced from varved glacial clays. Varved clays occur at several places in the region of the lower course of the Narva River (Orviku, 1936) and, because of the thin overlying soil, they were easily accessible. At Riigiküla, for example, the varved clays are covered by eolian sand, and during the period of the settlement, the clays were most probably sporadically exposed in places along the coast. M. Sakson (Institute of Geology, Estonian Academy of Sciences) has examined the diatom content of some of the Narva Joaoru, Lommi and Riigiküla pottery sherds. The fact that diatoms were not observed in any of the sherds seems to support heir varved clay origin.

Four separate tempers were mixed into the paste: (1) plant remains, (2) crushed shell fragments, (3) rock debris and (4) chamotte. It seems that the choice of temper did not depend on the characteristics of local paste, but on a pottery tradition, which originates in technological and functional needs developed outside the region under consideration.

Organic material has generally been used in the Narva-type pottery. The temper of most of the sherds has been shell (Fig. 2). According to E. Tavast (Institute of Geology, Estonian Academy of Sciences) shells of the freshwater clam Unio tumidus are mixed into the clay probably along with the clam Anodonta cygnea. The shells are predominantly finely crushed and large fragments are rare. Pottery sherds in which the organic admixture, probably crushed plant remains, has been burned out frequently occur (Fig. 3). The use of this temper is only confirmed by the lightness of the

sherds (pottery sherds with organic matter are very light their size), their porosity and the imprints cleavage planes of the sherds. In typical Combed Ware pottery, only mineral tempers were used. These as a rule were rock debris from the granite-gneiss group (Fig. 4), in which quarts and feldspar dominate and less frequently fragments of mica schist and conglomerate. The mineral grains are not worn and have sharp edges which confirms that it is a question of crushed rock and not gravel or sand. The size of the grains varies up to a diameter of 1 cm and the density of the grains also varies from vessel to vessel. Chamotte (sherd temper) occurs less often (Fig. 5). Characteristic reddishbrown clay pieces with a structure differing from the remainder of the sherd, have been observed in the Typical Combed Ware pottery of the Lommi III and Narva Joaoru settlements. As for the sherds of the Lommi III settlement, the diameter of the chamotte grains are sometimes up to 0.7–0.8 cm in diameter their maximum size being even up to 1.2 x 0.8 cm. Some chamotte grains contain the same type of mineral grains, as in other pottery sherds. This gives us reason to believe that finely crushed pottery has been mixed into the paste. B. Hulthén has assumed, with analogies of primitive cultures in mind, that the mixing of crushed pottery into the paste is perhaps tied to religious behavior and is not a technological feature (Hulthén, 1985, 335). Maybe this explanation can also be applied to the pottery tempered with chamotte from the sites along the lower course of the Narva River. In any case T.-L. Soininen's hypothesis, that chamotte could also have strayed into the pottery from old, crumbled pottery sherds at the places of productions, is difficult to believe (Soininen, 1990a). In Late Combed Ware pottery, all four types of temper are represented. Organic tempers dominate while mineral tempers are much scarcer and then only along with organic tempers. Grushed plant remains are profusely mixed into the clay, and they have generally been burnt out. S. Hiie (Institute of History, Estonian Academy of Sciences) has found carbonized plant remains in some Kudruküla pottery sherds, including one graminaceous and even individual non-carbonized herbaceous remains. Crushed shell debris is also widely used in the same way as in the Narva-type pottery (Unio tumidus and maybe also Anodonta cygnea, according to E. Tavast). The debris is predominantly fine-grained. In the paste of many vessels, both crushed shell debris and plant temper is present. Rock debris of the granite-gneiss group is occasionally represented and, as mentioned above, only along with organic tempers. The same applies to chamotte which has so far only been observed in the Late Combed Ware pottery of the Lommi III settlement. The quantity of the temper differs from vessel to vessel and it is especially abundant in the pottery of Kudruküla and Väiküla. Rock debris is not present in the pottery of these settlements in contrast to the pottery of other settlements.

P. Rummi (Institute of History, Estonian Academy of Sciences), using the x-ray fluorescence method, determined the chemical composition of 15 Narva-type pottery sherds. 13 Typical Combed Ware ones and one later Combed Ware sherd. Iron-rich clays have been used. The iron (III) oxide (Fe₂O₃) content of the Narva-type pottery is on the average 6.7%, for the Typical Combed Ware pottery 7.8% and the Late Combed Ware pottery 6.8%. The calcium oxide (CaO) content of the Narva-type pottery is surprisingly high, being 7% on average and even as high as 18.5%, This is because of the abundance of crushed shell temper. For the Typical Combed Ware pottery the corresponding value is only 2.1%.

Fig. 2.

Disregarding the few, small bowl-shaped, rounded-base vessels from Narva-type pottery, which have been pressed from a single clay lump, the side walls of all three types have been formed by the coil technique with the base of the clay vessels prepared from a single piece of clay and joined to the parts made by coiling. There are, however, big differences in how the coils were connected and width of the coils between various pottery types. The Narva-type clay vessels are formed using relatively narrow coils. Generally, the width of the coil does not
 

Fig. 3.

Fig. 4.

Fig. 5.

exceed 3 cm. The average width of clay coil of the Narva-type pottery from the Narva Joaoru settlement is about 1.5 cm. There ever occur coils which are only 0.5 cm wide. Clay coils from the Typical Combed Ware pottery in which both contact surfaces of the coils are preserved are a rarity, and the coils are predominantly 3–5 cm wide. In the Late Combed Ware pottery two groups can be distinguished. The clay coils of the Lommi, Narva Joaoru and Riigiküla settlements are 3–5 cm wide, on average, and rarely exceed 5 cm in width. In the pottery from Kudruküla and Väiküla, very wide coils are abundant (8–10 cm, in some cases even wider).

The clay coils have so-called U and N contact surfaces (Fig. 6). In the U type, one of the contact surfaces of the coil is convex and the other is concave. During construction, the lower edge of the coil is thought to have been smoothed round and after some drying a new coil was placed on it, the edge of which becomes curved by the already hardened rim of the previous coil (Fig. 6). There are two variations of the N type:

1. Both lower and upper coils thin-out into sharp edges in opposite directions, forming a wider, inclined contact surface (Fig. 6).

2. Coils with similarly wide contact surfaces, but with convex, not pointed, rims (Fig. 6). This type of connection can be distinguished as a separate type (Soininen, 1990b). The transition from one type of connection to another is not clear-cut. In some cases it is very difficult to determine which group or subgroup a coil belongs to; there also occur coils where one end belongs to one and the other to another sub group (Fig. 6). About 2/3 of the clay coils of Narva-type pottery have the U method of connection, a smaller number comprise the N type plus clay coils with U connection on one end and N one on the other. In the N type sharp-angled forms prevail. Most clay coils in both typical and late Combed Ware pottery have N type contact surfaces, and they usually end with pointed rims, less commonly convex ones. U type clay coils is a unique phenomenon. A few coil edges from typical and Late Combed Ware pottery of the Narva area are undulating. This is a technological method to strengthen the contact surface and it is also used by potters today. The author can confirm that this fact was well-known in the Stone Age, because he has encountered undulating clay coils in the Typical Combed Ware pottery at the Rusko Merttilä Lankila settlement in Finland (KM 27638) and the Early Combed Ware pottery of the Kokemäe Kraviojanka settlement (TYA 116). B. Hulthén has observed from the Ertebölle pottery, that clay coils with N contact surfaces are usually associated with thinner walled pottery than U type coils (Hulthén, 1977). The same is also true in the case of the early Finnish Combed Ware pottery (Soininen, O., 1991) and Kiukais pottery (Soininen, 1990c). This is not the case with the Neolithic pottery of the lower of the Narva River. The majority of the side walls of U contact surface clay coils in Narva-type pottery are 8 mm thick. The average thickness of the side walls of the N contact surface walls of coils formed in typical Combed Ware pottery is 10 mm, and even 12 mm in the Late Combed Ware vessels.

The firing of clay vessels varies. The sherds of Narva-type pottery are predominantly grey and in colour and occasionally black or yellowish-red. This leads us to believe that the vessels were usually fired in a reducing environment, although deviations existed. The firing temperature was obviously not very high, although taking into consideration the complete burning out of organic matter, the temperatures probably exceeded 700°C. Sherds of Typical Combed Ware pottery are mostly light red or with yellowish brown tones, and there are even bright red sherds. The vessels have been fired in an oxidizing environment at a fairly high temperature, possibly even much as 900°C. The Late Combed Ware pottery can be divided, according to construction technique and temper, into two groups. The vessel sherds of the Lommi I and III, Narva Joaoru and Riigiküla I, II and III settlements are mostly of lighter colour and the plant remains added to the paste have been burned away.

The vessels have been fired in an oxidizing environment and very strongly, as in the Typical Combed Ware pottery. The Late Combed Ware pottery of the Kudruküla and Väiküla settlements occasionally have very different firings. The color of the sherds varies from black to yellowish-red, which suggests the lack of a well defined firing environment. In many vessels the organic matter has not been entirely burnt away, so the firing temperature has usually not exceeded 700°C.

The earliest pottery type of the lower course of the Narva River is, as is the case for the whole of Estonia, the so-called Narva pottery. The question of the origin of this pottery is not yet entirely solved. Primarily on the basis of the ornamental elements of the Narva pottery in the Narva area, L. Jaanits found parallels with the younger Early Combed Ware pottery (style I : 2) (Jaanits, 1965; 1985). Technologically, these pottery types differ. In the I : 2 style mineral tempers, rock debris and sand (Edgren, 1984; Soininen, O., 1991) have been mixed into the paste and more than just the Narva-type of clay coils with N contact surfaces occur. If the pottery of this region does indeed only have its beginnings at the turn of the third and fourth milleniums BC, as suggests by the stratigraphy of the Narva Joaoru settlement (where the last Mesolithic stratum is still dated in the fourth millenium BC, ¹⁴C dates 5300±250 BP (TA-7) and 5820±200 BP (TA-33)) (Ilves et al., 1974). This means that there is a gap of almost 1500 years in the pottery of the Early Narva-type (Early Narva-type pottery is ¹⁴C-dated according to Latvian settlements 6450±250 BP (MGU-1008), 5510±70 (TA-1799) (Loze & Liiva, 1989). In the region studied, there are no sherds which would reflect the development of local pottery or of some apprenticeship period. It seems the pottery has arrived there through already well-developed traditions and this allows perhaps to surmise that the pottery came to the area (The Kunda-Narva culture) as an inside innovation.

An altogether different type of pottery tradition is exhibited by the Typical Combed Ware pottery. One can not believe that this kind of dramatic change could have occurred from the basis of the Narva-type of pottery. Any kind of find which could point to continuity in the development of the pottery types is lacking. The production technology of the pottery seems to demonstrate that the question is the arrival of new settlers. A continuity from both of the earlier pottery types can be seen in the Late Combed Ware period. The Late Combed Ware period of the Narva Joaoru, Lommi I and III and Riigiküla I, II and III settlements are special transitional forms, showing the continuation of a series of Typical Combed Ware pottery forms. However, it is likely that the organic matter temper of the clay content to be came predominant, probably due to the influence of Narva pottery. In the case of the pottery of Väiküla and Kudruküla, it is a question of an already developed pottery type, known in other parts of Estonia, especially in the Tamula settlement.

ACKNOWLEDGEMENTS

The manuscript was translated from Estonian to English by Karl S. Altan, Helsinki and revised by Hille and Valdar Jaanusson, Stockholm.

REFERENCES

EDGREN, T., 1984, Kivikausi. Suomen historia I. Espoo. 1984. 32 p.

EFENDIJEV, E. F., 1983, Raskopki v zapadnom Prinarovje, in Arheologitsheskije otkrõtija 1981 goda. Sbornik. Moskva.

GURINA, N. N., 1955, Novõje neolititsheskie pamjatniki v Vostotshnoi Estonii, in Muistsed asulad ja linnused: Arheoloogiline kogumik, 1. Tallinn.

GURINA, N. N., 1958, Die archäologische Forschungen in Ost-Karelien und im Leningrader Bezirk in den Jahren 1948–1957, in Finskt Museum. 65, Helsingfors.

GURINA, N. N., 1961, Drevnjaja istorija Severo-zapada evropeiskoi tshasti SSSR, in Materialõ i issledovanija po arheologii SSSR, 87, Moskva–Leningrad.

GURINA, N. N., 1967, Iz istorii drevnih plemen zapadnõh oblastei SSSR (po materialam narvskoi ekspeditsii), in Materialõ i issledovanija po arheologii SSSR, 144, Leningrad.

HULTHÉN, B., 1977, On Ceramic Technology during the Scanian Neolithic and Bronze Age (Thesis and Papers in North-European Archaeology. 6). Stockholm. 34 p.

HULTHÉN, B.,1985, Temper Variations in Ancient Ceramics. Technological or Cultural Origin?, in Iskos. 5. Helsinki. p. 329–337.

ILVES, E., LIIVA, A. AND PUNNING, J.-M. 1974, Radiouglerodnõi metod i ego primenenije v tshetvertitshnoi geologii i arheologii Estonii. Tallinn. 74 p.

INDREKO, R., 1940, Kaevamised 2. VII 40. Ekaterina Hämelaineni (Jaan Saare) tl. maal Narvataguse, Narva vld. Lommi kruusaaugus, Notika jõe ääres, mis jookseb Luuga jõkke. (Manuscript in the Institute of History, Estonian Academy of Sciences).

JAANITS, K., 1981, Raskopki v Muuksi i Kudrukjula. Arheologitsheskije otkrõtija 1980 goda. Sbornik. Moskva.

JAANITS, L., 1954, Aruanne arheoloogilistest proovikaevamistest muistsel asulal Narva linnas Joaorus 8.–25. juunini 1954. a. (Manuscript in the Institute of History, Estonian Academy of Sciences).

JAANITS, L., 1957, Aruanne arheoloogilistest proovikaevamistest asulakohal Narva linnas Joaorus 10.–28. juunini 1957. a. (Manuscript in the Institute of History, Estonian Academy of Sciences).

JAANITS, L., 1960, Aruanne arheoloogilistest kaevamistest asulakohal Narva linnas Joaorus 6.–29. juulini 1960. (Manuscript in the Institute of History, Estonian Academy of Sciences).

JAANITS, L., 1962, Aruanne arheoloogilistest kaevamistest asulakohal Narva linnas Joaorus 2.–30. augustini 1962. a. (Manuscript in the Institute of History, Estonian Academy of Sciences).

JAANITS, L., 1963, Aruanne arheoloogilistest kaevamistest asulakohal Narva linnas Joaorus 28. juunist 5. augustini 1963. a. (Manuscript in the Institute of History, Estonian Academy of Sciences).

JAANITS, L., 1964, Aruanne arheoloogilistest kaevamistest asulakohal Narva linnas Joaorus 30. juunist 1. augustini 1964. a. (Manuscript in the Institute of Hiatory, Estonian Academy of Sciences).

JAANITS, L., 1965, Über die Ergebnisse der Steinzeitforschung in Sovjetestland, in Finskt Museum, 72. 1965. Helsingfors.

JAANITS, L., 1985, Nekotorõje voprosõ izutshenija kamennogo veka v Pribaltike, in Novoje v arheologi Pribaltki i sosednih territorii. Tallinn.

LOZE, I. A. AND LIIVA, A. A., 1989, Hronologija rannego neolita vostotshnoi Pribaltiki (po materialam poselenii), in Latvijas PSR Zinatnu Academijas Vestis. 2. 66 p.

ORVIKU, K., 1936, Geoloogilisi märkmeid geoloogilis-geograafiliselt õppeekskursioonilt Narva ja selle ümbrusesse 10.–14. juuni 1936, in Eesti Loodus, 4. Tartu, p. 149–157.

SOININEN, O., 1991, Kokemäen Kraviojankankaan keramiikka-ainesto. Varhaiskampakeramiikan typologinen analyysi. Pro Gradu. (Manuscript at the University of Turku, Departement of archaeology).

SOININEN, T.-L., 1990a, Turun Niuskalan Kotirinteen kiukaiskeramiikan valmistustekniikasta ja sen yhteydest keramiikan nykytilan, Pro Gradu. Turku University, p. 31. (Manuscript at the University of Turku, Department of archaeology).

Contents  Previous article  Next article