INTRODUCTION

The history of the investigation of the Stone Age in Estonia dates back more than 100 years. The beginning of this scientific research is connected with Andreas Constantin Grewingk (1819–1887), Professor of Geology at the University of Tartu in the second half of the 19th century, who also lectured on archaeology (Rõõmusoks 1982, 197). Among his numerous publications, the investigations connected with Kunda Lammasmägi (Grewingk 1882; 1884), and the systematisation and mapping of the Stone Age finds from the Baltic provinces of his days (e.g. Grewingk 1865, 1871) should be mentioned. His diverse knowledge of biology, zoology, and first of all his profession of geologist-mineralogist enabled him to investigate the relations between the Stone Age finds and the natural environment and to identify the rock materials of ancient artefacts and animal bones found from Kunda. The main problem was the lack of finds. In 1865, when Grewingk published his first investigation, only 14 Stone Age finds were known in present-day Estonian territory (Tallgren 1922, 23), and no significant excavations were carried out at Stone Age sites in the 19th century.

After the death of Grewingk, Stone Age investigations remained at an amateur level for a long time. A few articles discussing the Stone Age were written by Richard Hausmann, Professor of History at Tartu University (1842–1918; Hausmann 1904; 1912), whose research mostly centred on the Iron Age, but more influential were the Baltic German amateur archaeologists assembled in the Antiquarian Society of Pärnu. They collected Stone Age artefacts, but also to some extent did research and carried out excavations. Particularly active was a doctor from Vändra, Martin Bolz (1868–1917), who collected 500 ancient artefacts, mostly stone axes and chisels, from Viljandi and Pärnu districts (Kriiska 1997c, 27–28). Bolz documented and mapped his finds thoroughly, and published a catalogue of his collection (Bolz 1914b), and also carried out the excavations of the Stone Age burial site at Kivisaare (Bolz 1914a).

In Pärnu, the consul Friedrich Rambach (1853–1916) and the veterinarian Eduard Glück (1866–1918) were interested in and collected Stone Age finds. The greater part of the collection of Glück, 774 objects made mostly of bone and antler, was gathered in 1904–1905 from the gravel at the bottom of Pärnu River (Kriiska 1997c, 25–27). He also published the greater part of his collection and wrote a survey of the Stone Age (Glück 1906). Rambach's collection consists of more than 500 objects made mostly of bone and antler, gathered from the banks of Pärnu and Reiu rivers (Indreko 1926). Later, and until the 1940s, the collecting for the Pärnu Prehistory Society for Antiquities was continued by the brewer Eduard Bliebernicht (1902–1943). He developed active co-operation with professional archaeologists (Kriiska 1997c, 28) and also published his finds (Bliebernicht 1924).

In the 1920s the number of Stone Age finds increased but was still relatively small and uneven. According to an estimate by Aarne Michaël Tallgren (1885–1945) (1922, 23) nearly 900 stone finds and about 1900 bone and antler artefacts were known at that time. Most of these came from two main sites — Kunda and the lower reaches of the Pärnu River. In his two-volume survey of Estonian prehistory, Tallgren, Professor of Archaeology at the University of Tartu, presented a thorough outline of the Stone Age, discussing artefacts and sites of discovery as well as general cultural and chronological problems (Tallgren 1922).

In the mid-1920s, Richard Indreko (1900–1961) began his career as a researcher of the Stone Age. He was educated at the University of Tartu by Tallgren and Birger Nerman. He initially concentrated his interest on the Mesolithic, studying the area on the northern shore of Lake Võrtsjärv, previously known for stray finds (he excavated the settlement sites of Siimusaare and Moksi), and in 1933–1937 carried out archaeological excavations on Kunda Lammasmägi (Indreko 1936). Indreko, who fled to Sweden during the Second World War, published the results of his research in 1948 in a comprehensive monograph (Indreko 1948b). In it he discussed all known Mesolithic sites and artefacts in Estonia, as well as the problems of the origins of Kunda Culture. In the 1930s Indreko was also interested in the Neolithic period. In 1933–1934 he excavated the cemeteries of the Corded Ware Culture in Ardu (Indreko 1938) and Sope, and published a review on the burials of the Corded Ware Culture in Estonia (Indreko 1935). At the end of the 1930s he turned his attention to Neolithic settlement sites. In 1936–1943 he excavated and carried out survey trips at the settlement sites of Akali, Kroodi, Kullamäe, Lommi, Villa I, Tamula and Undva (Indreko 1948a). The settlement site of Undva was the first archaeologically investigated Stone Age site on the Estonian islands.

Indreko's research into the Neolithic period were continued by Lembit Jaanits, who graduated from the University of Tartu after the war. Jaanits has discussed various problems of the Stone Age: chronology, settlement, artefact typology, ethnic aspects, religion, etc. (Saluäär 2000). He has carried out archaeological excavations on more than ten Mesolithic and Neolithic sites, including Akali, Kroodi, Kullamäe, Kääpa, Narva Joaoru, Riigiküla III, Siimusaare, Tamula, Pulli, Valma, Villa I settlements on the mainland, and the Kõnnu, Loona and Naakamäe settlements on Saaremaa. He is the author of all of the substantial general discussions of the Stone Age published in the post-war years (Jaanits 1959b; Jaanits et al. 1982; Jaanits 1992). Since the end of the 1940s attention was focused on Neolithic settlement and the extensive excavations of sites (Jaanits 1954; 1955; 1959a; 1959b). Especially important is the differentiation of Narva pottery from Late Comb Ware, and their analysis. This was supplemented by the Russian archaeologist Nina Gurina excavating the settlement sites of Riigiküla (I, II, III) in NE Estonia at the beginning of the 1950s, the results of which have been published in a monograph (Gurina 1967).

In the 1960s–1970s the investigation of the Mesolithic period was resumed. The excavations of Narva Joaoru at the beginning of the 1960s, and the Early Mesolithic settlement of Pulli from 1968–1973 and 1975–1976 should be mentioned (Jaanits & Jaanits 1975; 1978). These established the basis for the conclusions about the origins of the Kunda Culture (e.g. Jaanits et al. 1982, 32), as well as its development into the Narva Culture (Jaanits 1970). At the end of the 1960s, Kaarel Jaanits began his research into the Mesolithic period. He has dealt mainly with the settlement sites of the Kunda Culture rich in flint in Central Estonia (the excavations of the settlement sites of Jälevere, Lepakose and Umbusi), and also with the typological analysis of flint artefacts (Jaanits 1973; 1981; 1989; Jaanits & Ilomets 1988, etc.).

While the sites one the Estonian coastal area have retained their importance throughout the history of Stone Age investigations, the Stone Age sites on the islands have been studied only sporadically. Up to the 1980s only the Neolithic, especially Late Neolithic, sites on Saaremaa were known and had been investigated to some extent (Jaanits carried out excavations at the settlement sites of Loona, 1957–1959, Naakamäe, 1958–1959 and 1962, and Kõnnu, 1977–1986). Unfortunately, the material gathered in the excavations has been only briefly discussed. Though the first Mesolithic settlement site on the islands was discovered by Vello Lõugas in 1986 (Pesti & Rikas 1991, 12), and based on the absence of pottery in the trial pit he presumed that the Kõpu I settlement in Hiiumaa, also dated from the Mesolithic period (Lõugas 1982). This opinion has prevailed up to the present day, expressed by Tanel Moora a couple of years ago: We may presume that, being the inhabitants of a forest zone, the people of the Mesolithic Kunda Culture were not yet able to survive on seal hunting, which can be argued in the case of the Neolithic inhabitants of Saaremaa (on the basis of the finds from Kõnnu and Naakamäe settlements) and Hiiumaa (Kõpu) (Moora 1998, 64).

In the 1990s, investigations of the Stone Age have become more extensive with regard to the number of researchers as well as research directions. In addition to archaeologists, zoologists, geologists and geographers have also been interested in the problems of Stone Age settlement and economy. Though the co-operation between archaeologists and other scientists is known throughout the history of Stone Age investigations, the number of researchers and their productivity has increased in recent years. Substantial research of Stone Age fauna has been performed, especially concerning marine mammals and fish (Lõugas 1997). Pollen investigations of lake and bog sediments have helped determine the human impact on the environment in the Stone Age; the finds of pollen of Stone Age cereals are throughout Estonia (Veski 1998; Kriiska 2000, tab. 1 etc.). Several scientific discussions of Stone Age pottery (Kalm 1996, Kalm et al. 1997) and stone finds (Suuroja 1996) have been published.

The search for new Stone Age sites has been intensive. Of the 115 known sites of hunter-gatherer cultures, approximately 80 were discovered in the 1990s, 52 of these during the survey trips organized by the author. The focus has mainly been on the coastal areas and especially the islands, where all the latest excavations of Stone Age sites have taken place (a total of 12 settlement sites). The results obtained enable us to see the settlement history of coastal areas and islands in a different perspective. The archaeological problem of original settlement has received some answers, and the basis has been established for conclusions about economic history, and ¹⁴C datings enable us to revise the absolute chronology of the Estonian Stone Age. These three subjects will form the core of the present publication.
 

THE CHRONOLOGY OF THE ESTONIAN STONE AGE

Chronology has been one of the central subjects of Estonian Stone Age investigations, which practically none of the researchers since Grewingk has neglected. The chronological systems of the 19th century were rather primitive and were readily abandoned when new material was added. Thus, for example, Grewingk first dated the end of the specific Stone Age in the Eastern Baltic region to the 6th century AD (1871, 49), and later to the 1st century AD (1874, 34).

Tallgren developed a longer-lasting chronology at the beginning of the 1920s (1922). He drew a fundamental distinction between Stone Age cultures based on bone and those based on stone, establishing several chronological stages, using place-names known already as appellatives and connecting them with the phases of the Baltic Sea or megalith constructions: 1) Kunda — the period of Ancylus Lake (6000–5000 BC), 2) Võisiku — the maximum extent of the Litorina Sea (5000–3000 BC), 3) Burtnieki — the dolmen period and 4) Pärnu — the period from dolmens to passage graves. He also established the stages of the Combed Ware Culture and the Boat Axe Culture. Tallgren's system persisted for years, notwithstanding the crushing — though often unfounded — criticism of Julius Ailio (Ailio 1924, 46–50). It has been only slightly adapted; for example the Pärnu stage became associated with the Kunda Culture.

In the second quarter of the 20th century, the connection of archaeological finds with the climatic periods, established with the help of pollen analysis, assumed an important role in Estonian Stone Age chronology. Paul William Thomson (1892–1957), one of the founders of pollen-analysis, played a significant role there, publishing several investigations on this subject (Thomson 1928, 1930) and advising archaeologists (e.g. Indreko 1932). In the first volume of Eesti ajalugu (Estonian History), published in 1936, Harri Moora, the author of the chapter on the Stone Age, distinguished two stages of the Stone Age in Estonia — the Early and Late Stone Age — and dated them to the 6th–4th millennium BC, and 3000–1300 BC respectively. Both stages were in turn divided into two phases. The Early Stone Age comprised 1) the Kunda Culture of the Ancylus Lake period, and 2) the Võisiku Culture of the maximum extent of the Litorina Sea. The Late Stone Age comprised 1) the Combed Ware Culture and 2) the Boat Axe Culture (beginning in 2000 BC) (Moora 1936). The inanity of the differentiaton between cultures based on bone or on stone (or, concealed, Kunda and Võisiku) was demonstrated by Richard Indreko (1948b, 79–81) on the basis of materials found in Kunda. At the beginning of the 1950s Lembit Jaanits developed at chronology of the Neolithic period, mainly on the basis of the stratigraphy of Akali, pollen analyses and analogies from neighbouring countries, especially Finland. He distinguished two phases referring to the type of economy: the hunter-gatherer Neolithic (beginning in the early 3rd millennium BC), and the Neolithic with primitive cattle-breeding and agriculture appearing alongside the hunter-gatherer economy (beginning in the second quarter of the 2nd millennium BC) (Jaanits 1955, 191–192). On the basis of pottery Jaanits distinguished five cultural periods: 1) proto-Combed Ware phase (later Narva pottery) — the first half and middle of the 3rd millennium BC, 2) Typical Combed Ware — the second half of the 3rd millennium BC, 3) Late Combed Ware — the first half of the 2nd millennium BC, 4) Corded Ware — the second quarter of the 2nd millennium BC, and 5) Late Corded Ware, Textile-impressed Pottery and Striated Pottery — beginning in the mid-2nd millennium BC (Jaanits 1954, 364; 1955, 191).

Fig. 1. Chronology of the Estonian Stone Age.

Absolute chronology was revised after the introduction of the radiocarbon method. In 1959, the laboratory for geo-bio-chemistry was established at the Institute of Zoology and Biology of the Academy of Sciences of the Estonian SSR (Lõugas 1988, 20), which has provided the greater part of the ¹⁴C datings of the Estonian Stone Age. By the 1970s, about 20 samples of organic matter, mainly charcoal collected from Estonia's Stone Age sites, had been analysed (Ilves et al. 1974). The absolute chronology developed on the basis of these analyses persisted in broad outline for a couple of decades and has been published in the basic treatments on archaeology (Jaanits et al. 1982; Jaanits 1992). According to the uncalibrated dating of the Pulli settlement site, the Mesolithic began in about 7500 BC. The changes concerning Neolithic datings are minimal, the ¹⁴C datings were mostly considered secondary and simply fit into existing schemes. Within the Neolithic period, Jaanits distinguished 2 sub-periods: the earlier period of hunter-gatherer economy, 3000–2200 years BC, and the later period of hunter-gatherer and farming economy, 2200–1500 years BC (Jaanits et al. 1982, 61). The introduction of pottery, marking the first Neolithic culture — Narva Culture — was dated to about 3000 BC (Jaanits et al. 1982, 62). Though the Typical Combed Ware Culture lacked the ¹⁴C datings definitely connecting it with this culture (two datings from Kääpa were presumably connected with it), its beginning was dated, relying mostly on Latvian material, to about 2500 BC (Ilves et al. 1974, 190; Jaanits et al. 1982, 76). The formation of the Late Combed Ware was dated, on the basis of the samples form Tamula, to the end of the 3rd millennium, and the beginning of the Corded Ware Culture, relying upon the earliest datings of Finnish and Lithuanian material, to 2200 BC (Jaanits & Liiva 1973, 159; Liiva et al. 1974, 190; Jaanits et al. 1982, 102).

The new chronology of the Estonian Stone Age was presented by the author in 1995. It was developed on the basis of fresh research findings and the calibrations of the earlier datings, and it is the foundation of the chronological system presented here. It has been published in full only a couple of times (Kivimäe et al. 1998; Kriiska et al. 1999). Those calibrations were made on the basis of Gordon W. Pearson’s and Minze Stuiver’s ¹⁴C calibration curve (Stuiver & Pearson 1993; Pearson & Stuiver 1993), mostly with the help of the computer program OxCal v2. 18 cub r:4 sd:12 prob[chron]. Now a new program has been developed — CAL40DATA OxCal v2. 18 cub r:4 sd:12 prob[chron] — which corrects the datings quite remarkably, especially concerning the Early Mesolithic (125–150 years), and to a smaller extent also in later periods. The new calibrated chronology has been compiled in this system, and is first published in the present publication (Fig. 1). The comparative datings from the neighbouring countries have also been calibrated in the same program. Hitherto we have 65 more or less reliable Stone Age datings from Estonia, 23 of wich originate from the sites excavated and inspected by the author in the years 1994–1999 (Tab. 1).

Three reliable ¹⁴C datings come from the oldest hitherto known settlement site of Pulli, from the beginning of the Mesolithic: 9620±120 (Hel-2206A), 9600±120 (TA-245) and 9575±115 (TA-176) ¹⁴C years (Raukas et al. 1995, 121). These belong, with at probability of 95.4 %, to the period 9300–8600 cal. BC, which makes the average 8950 cal BC, considering the probability of 68.2 % even 9000 years cal BC. The Mesolithic archaeological complex in the Eastern Baltic bears the common name of Kunda Culture. In Estonia about 50 Mesolithic settlement sites are known at present, and some burials in the Kivisaare cemetery in Central Estonia probably also date from the same time. Naturally, it is not homogenous throughout the 4000 years of existence, but comprises special local and temporal features. The changes within the Mesolithic in the Eastern Baltic are especially accentuated by Latvian and Lithuanian researchers. Tomas Ostrauskas suggests that only the Early Mesolithic phase, of the preboreal climatic period, characterized by arrowheads of the Pulli type, should be included in Kunda Culture (Ostrauskas 2000). Ilze Loze considers the common features between the Late Mesolithic and the subsequent Narva Culture so overwhelming that she uses the name Proto-Narva Culture to refer to both (Loze 1988, Tab. 17). Since the changes seem to follow developmental continuity, the author asserts that these specific traits should be discussed (at least in the present stage of research) among the variations of the Kunda Culture and not separately.
 
 

The introduction of pottery has been considered the feature marking the beginning of the Neolithic period, in Estonia as well as elsewhere in North and East Europe (e.g. Oshibkina 1996), in contrast to regions where the beginning of the Neolithic is connected with the introduction of farming. According to periodisation based on economic changes, we could talk of the Neolithic period in Estonia only in the latest stage of the Stone Age, therefore the preceding stage of the Stone Age containing pottery is sometimes still called the Mesolithic, Sub-Neolithic and Forest-Zone-Neolithic period.

The first settlement phase with pottery is called, after the settlement sites of the lower reaches of the Narva River, the Narva Culture. It includes practically the whole area of the preceding Kunda Culture — Estonia, Latvia, Northern Lithuania, the northern part of Byelorussia, and a part of Northwest Russia (e.g. Timofeev 1988; Kriiska 1997b). While in Lithuania the tradition lasts throughout the Neolithic period (Girininkas 1994, 259), the Narva Culture in Estonia is mainly an Early Neolithic phenomenon. At present, more than twenty settlement sites of the Narva Culture are known. The majority of these are located on the coasts. Yet, this most likely does not reflect the Stone Age situation but is the result of the better degree of investigation of the coastal areas. The only presumable Early Neolithic burials in Estonia are from the Kõnnu settlement site on Saaremaa and the Narva Joaoru settlement site in Northeast Estonia. The archaeological finds have led to a rather consentient opinion that the Narva Culture developed from the Kunda Culture, evidently without any noteworthy migration (e.g. Jaanits 1970, 86; Zagorska 1993, 114–115). On the basis of pottery, areas with specific features can still be determined there. In the whole culture area these regions are many, and in Estonia alone three local groups can be determined — North Estonia, East Estonia and the islands of West Estonia (Kriiska 1997b, 17).

The latest datings of the pre-pottery Stone Age (Mesolithic) in Estonia are from about 5000 years cal BC. The latest dating from the Võhma I settlement site on Saaremaa, is 5175 cal BC (6245±200 ¹⁴C-years — Ta-2652), and from the Kõpu VII/VIII settlement site on Hiiumaa it is 5120 cal BC (6172±51 ¹⁴C-years — Tln-2024), and from Kunda Lammasmägi and Narva Joaoru even 4950 cal BC (respectively 6015±210 ¹⁴C-years — TA-16, and 6020±120 ¹⁴C-years — TA-17). The oldest reliable dating of the Stone Age with pottery (the Narva Culture) so far comes from the Riigiküla IV settlement site: 6023±95 ¹⁴C-years (Tln-1989), which makes an average of 4950 cal BC, and considering a probability of 68.2 %, perhaps even 4900 cal BC. The latter has been provisionally considered the border of the Mesolithic and the Neolithic periods, although, considering the context of the neighbouring areas, such a date for the beginning of pottery is clearly too late. The earliest dates of the layers containing pottery in the East Latvian Osa and Zvidse settlements are ca. 5500 cal BC; 6533±120 (Ri-272) and 6535±60 (TA-862) ¹⁴C-years respectively (Loze 1988, Tab. 16). Thus the difference with the beginning of typologically very similar pottery would be as much as 600 years. Several explanations are possible, but most likely the phenomenon is not objective but caused by the different level of investigation. Considering the general distribution logic of the craft of pottery-making, the author asserts that the oldest settlement sites with pottery should be discovered in the southern part of Estonia, not on the coast and islands, from where all existing reliable ¹⁴C datings originate. Here we must also consider the possibility that, especially in the earlier phase, earthenware was not taken along on seal-hunting trips. It is hardly likely that the craft of pottery-making did not spread from East Latvia, an area very close to Southeast Estonia both culturally and geographically over some twenty or thirty generations. In the final part of the Narva Culture there are several unanswered questions, and therefore the dating of these processes is hindered. It is not yet clear whether the Narva Culture continued into the Middle Neolithic period. The latest reliable datings of the Narva Culture give an average of 4200–4100 cal BC: Ruhnu II — 4200 cal BC (5400±150 ¹⁴C-years — Le-5628 and 5400±100 ¹⁴C-years — Ta-2716), Kõpu I — 4160 cal BC (5330±90 ¹⁴C-years — TA-493) and Riigiküla XII — 4100 cal BC (5268±58 ¹⁴C-years — Tln-1992).

The beginning of the Middle Neolithic is marked by the development of the Combed Ware Culture. This culture encompassed a rather extensive area around the Baltic Sea. In the north it included Finland up to Rovaniemi, even reaching North Sweden. In the northeast the Combed Ware Culture extended across the greater part of Karelia, in the east the areas of Petersburg and Novgorod, and in the south it reached through Latvia to the Curonian Spit in Lithuania and to the coast of Poland. Compared with the earlier period, great changes took place in material culture. This phenomenon is especially apparent in pottery, but also in flint artefacts, the distribution of amber, etc. The similarity of artefactual materials may indicate that the Combed Ware Culture developed in the whole area over a relatively short period. Since this area was formerly inhabited by several cultures with different local groups, the author is convinced that the Combed Ware Culture was not formed only as a result of local evolution. The distribution of the new culture evidently took place under foreign influence (and, at least on Estonian territory, probably also immigration), as well as with local development. For the Combed Ware Culture, two typological and evidently also temporal groups can be established on the basis of the pottery: 1) Typical Combed Ware Culture and 2) Late Combed Ware Culture. In Estonia, Typical Combed Ware has to date been found in about twenty settlement sites, and several burials (Valma, Naakamäe, Kõnnu, etc.) can also be associated with this period. Unfortunately there are no reliable ¹⁴C datings of the Typical Combed Ware in Estonia. The dating of human bones of a burial from Tamula — 5310±85 ¹⁴C-years (Ua-4828), which gives the average 4150 cal BC (with a probability of 95.4 % 4330–3970 cal BC) — could possibly be connected with the beginning of the phase. If the dating is correct, the correlation between the settlement site and the burial ground of Tamula must be revised, assuming that the burial ground had been established at the site before the settlement, where the Late Combed Ware dominates and Typical Combed Ware is completely absent. Since the period of the Typical Combed Ware has been rather short and relatively contemporaneious in other different areas, we may also apply the datings from neighbouring countries. Numerous ¹⁴C-datings of the Typical Combed Ware Culture have been performed in Finland. Of the 56 datings, the earliest belong to the period 4350–4300 cal BC: Hankasalmi Autioniemi — 5510±170 ¹⁴C-years (Hel-30), Ruotsinpyhtää Holmgård — 5460±150 ¹⁴C-years (Hel-19) and Honkilahti Kolmhaara — 5440±160 ¹⁴C-years (Hel-39) (Pesonen 1999, 200). In Latvia the earliest date of the Typical Combed Ware comes from Zvejnieki cemetery grave no. 206 (Zagorska 1997, 43): 5285±50 ¹⁴C-years (Ua-3543), the average being 4100 cal BC. The beginning of Estonian Typical Combed Ware could be provisionally dated to 4150 cal BC.

Combed Ware has undergone relatively rapid changes. While the rest of material culture resembles the earlier period, in pottery the differences can be observed in the composition of the clay as well as in the ornamentation. In Estonia, several local groups differ on the basis of the Combed Ware. Most marked are the differences between the earthenware of mainland Estonia and of the West Estonian islands. This phenomenon has been called Late Combed Ware, accentuating the evolutionary continuity in the relations with Typical Combed Ware. Late Combed Ware has hitherto been found in more than twenty sites and, compared with the earlier phases, a relatively large number of graves have also been excavated (Tamula, Riigiküla I). The earliest datings of the period are from human and seal bones from the Kudruküla site: 4860±60 ¹⁴C- years (Cams-6266) and 4835±100 ¹⁴C-years (Ua-4827), which give an average of 3650 cal BC. At the same time pottery also changed in Finland, where the Late Combed Ware of Uskela type was formed (Carpelan 1999a, 259). The time of the end of the Combed Ware Culture cannot be determined at the present level of investigations. The latest to date is based on an elk bone found from the settlement site of Villa I — 2000 cal BC (3570±240 ¹⁴C-years — TA-20). It is possible that the tradition of the Combed Ware continued up to the end of the Stone Age, and probably even to the beginning of the Early Bronze Age.

The beginning of the Late Neolithic in Estonia is marked by the development of the Corded Ware Culture (the Boat-axe Culture in earlier literature). This is one of the Corded Ware or Battle Axe cultures that spread across Central, Eastern and Northern Europe in the final stage of the Neolithic. In Estonia a completely new cultural tradition appeared alongside the population of the Late Combed Ware Culture, differing from the former in artefacts and burial customs as well as in settlement pattern and economic system (Jaanits 1992, 47; Kriiska 2000, 71). In Estonia about fifty settlement sites and nearly twenty cemeteries of the Corded Ware Culture are known (Kriiska 2000, 70). These were mostly discovered incidentally during excavations of other sites, and only on a couple of sites were the settlement traces of that period not mixed with the finds of other periods disturbed ploughing. The finds obtained from the excavations have not been numerous — only a few thousand potsherds and a small number of stone tools. The latter includes triangular and heart-shaped arrowheads. It seems that the Corded Ware Culture in Estonia was formed under strong influence from the south, together with new settlers (the theory of moderate migration).

However, only one reliable dating of the Corded Ware Culture has been obtained in Estonia. The analysis of the charcoal sample from the cultural layer of the site of Riigiküla XIV gave the result 3970±100 ¹⁴C-years (Ta-2680), which gives an average of 2500 years cal BC (2900–2100 years cal BC, with a probability of 95.4 %). Typologically, the pottery of the Riigiküla XIV settlement site clearly represents the later, and not the earlier phase of the Corded Ware. When determining the beginning of the culture, we must again rely upon the datings from the neighbouring countries, and the wider context of the Corded Ware/Battle-Axe cultures in Europe.

Eleven ¹⁴C datings of the Corded Ware Culture are known from Latvia (Loze 1992, Tab. 1). The earliest is the dating of peat from the Eini settlement site — 4735±60 ¹⁴C-years (TA-2250), which gives an average of 3500 cal BC. A series of datings come from the settlement sites of Ica and Abora I, the earliest ones being 4420±80 ¹⁴C-years (TA-2248) and 4490±80 ¹⁴C-years (TA-2144) respectively, giving averages of 3125 and 3145 cal BC. Four datings of the Corded Ware Culture are so far known from Finland (Edgren 1992, 92). The earliest come from Vantaa Jönsase and Lieto Kukkarkoski graves, 4520±130 ¹⁴C-years (Hel-1006) and 4320±170 ¹⁴C-years (Hel-831)respectively, which gives an averages of 3250 and 2950 cal BC. The other datings are also relatively early.

The earliest dates of the Battle Axe cultures elsewhere in Central and Eastern Europe also belong to the period of about 200 years before 3000 cal BC, while the Scandinavian ones are a little later. The earliest date of the Baltic Coastal Culture (Rzucewo Culture), distributed troughout Lithuania and on the Polish coast near the Bay of Gdansk, comes from the Nida settlement site (Rimantiene 1989, 176). It is 4460±100 ¹⁴C-years (Vs-632), the calibrated average being 3175 cal BC. In Denmark, the earliest ¹⁴C datings of the Single Grave Culture come from the settlement Langagergåd II (Jensen 1989, 16) and the Engedal grave (Malmros & Tauber 1977, 82), 4270±85 (K-5362) and 4240±90 (K-2501) ¹⁴C-years respectively, the calibrated averages being 2825 and 2800 cal BC. The datings of the Battle-Axe Culture in Sweden are even later (Larsson 1989, 68). The earliest — 4010±115 ¹⁴C-years (U-154) — comes from the settlement site of Kabusa (Larsson 1989, 64) and gives an average of 2550 cal BC. Considering the above datings and the logic of the distribution of the Corded Ware Culture we may assume that the beginning of the Corded Ware Culture in Estonia falls at about 3200 cal BC.

At the present state of knowledge the duration of the culture cannot yet be estimated. It is possible that, as in the case of the Late Combed Ware Culture, it continued up to the end of the Stone Age or even to the beginning of the Bronze Age. Unfortunately the processes of that period have been very poorly investigated in Estonia. The recognised boundary between the Stone and the Bronze Ages, 1500 BC, the date of the earliest known bronze objects in Estonia, has generally been used (Lõugas 1970, 97).
 

THE FORMATION AND GENERAL DEVELOPMENTAL FEATURES OF COASTAL SETTLEMENT IN ESTONIA IN THE STONE AGE

The coastal areas and islands of Estonia formed a specific economic, and in certain respects also cultural unit during the Stone Age. This area is clearly discernible throughout almost the whole Stone Age as an original technocomplex (with the only exception being the Pulli site with its extensive use of imported flint), where quartz dominates as the material for small tools produced by a splitting technique. Among the three basic rocks — quartz, flint and Baltic red quartz-porphyry (typical of the islands) — the share of quartz is over 40 %, mostly even over 80 % (Fig. 2). As with the use of rocks, the basis of the economy was also determined by environmental conditions. Adaptation there made life in coastal areas possible.

Coastal Estonia — an area flooded by different phases of the Baltic Sea in the post-glacial period — was inhabited even in the Early Mesolithic. The oldest known settlement site in Estonia, Pulli (Southwest Estonia, Fig. 3), was evidently a summer habitation site, situated on the Pärnu River only a few kilometres from the Yoldia Sea (Raukas et al. 1995, 121–122; Rõuk & Vuorela 1992, Fig. 140). The second site, Kunda Lammasmägi (North Estonia), was located near Ancylus Lake on a small island within the wetland (Moora 1998, 65). The wide range of datings (8700–4950 cal BC) and the changes in the environment indicate that Lammasmägi had been inhabited repeatedly, although presumably only seasonally, over a long period.

The earliest known traces of habitation on the seashore of mainland Estonia and the islands date from the second half of the Mesolithic. Considering the general development of the Baltic region, it is possible that these areas could have been inhabited somewhat earlier. For instance, the island of Gotland, at a distance of 90 km from the mainland, was inhabited in the Early Mesolithic (Larsson 1997, 14), and the boat equipment from a fishing net and other objects from Antrea, Karelia, sank to the bottom of the strait of that time in about 8500 cal BC — could also be associated with the Kunda culture (¹⁴C datings published by Carpelan 1999b, 160–161). In Estonia only one site, Sõitme I, has been discovered to date. It could have been situated on the coast of Ancylus Lake (Vedru 1998, 62). The formation of the coastal settlement took place mainly at the beginning of the Litorina Sea period. The Narva Joaoru settlement site (Northeast Estonia), not immediately on the seashore but on the riverbank a few kilometres upstream was inhabited in about 6550 cal BC (¹⁴C datings published by Ilves et al. 1974). The settlement sites Vihasoo I and II (North Estonia), and the site near Valge-Risti (West Estonia) belong to the end of the Mesolithic. In Vihasoo, the settlement was located at the ancient mouth of the river, and with the regression of the sea the site was also moved. The Valge-Risti site was located on the shore of a spit in the Litorina Sea. In Southwest Estonia, the settlements of Metsaääre I and II, presumably also from the Mesolithic, are situated in the coastal area, but they are most likely connected with the Reiu River, not with the seashore. The Suurupi and Liikva I–III settlement sites (Northwest Estonia) probably date from the Mesolithic as well (Lang 1996, 420), but their palaeogeographic situation has not yet been established. In a broader sense, the Mesolithic settlements on the ancient shores of Lake Kahala also belong to Coastal Estonia (Vedru 1998).

More numerous are the Neolithic settlement sites of hunters and gatherers on the coast (Fig. 4). On the mainland they were often located on spits forming lagoons. In Riigiküla (Northeast Estonia), fourteen hunter-fishermen’s settlements have been found on a spit of a lagoon of the Litorina Sea (Kriiska 1999). In the same region, Lommi I, II and III, Narva Joaoru, Kudruküla and Väiküla settlement sites are known to have had a maritime economy (Indreko 1948a, 298–299; Kriiska 1995b, 58–60). In North Estonia, Kroodi (Kriiska 1997b) and possibly also Jägala (Spreckelsen 1925) were connected with lagoons. In Southwest Estonia, the Metsaääre III Neolithic site at the midpoint of the Reiu River was also among the coastal settlements. The irrepular coastline with numerous lagoons was favourable for fishing, hunting water-fowl and seal, while the adjacent forests favoured game hunting. This ecological margin effect was skillfully used by the Stone Age people. It afforded the habitants subsistence in a relatively small area, and also a rather settled habitation within the hunting territory.

COLONIZATION OF THE ISLANDS AND GENERAL FEATURES OF DEVELOPMENT IN THE STONE AGE

The oldest traces of settlement on the islands are mainly connected with the transgression of the Litorina Sea and the period following it. Nine Mesolithic settlement sites have been discovered in the NW part of Saaremaa, in the vicinity of the villages Võhma and Pahapilli (Võhma I–VII and Pahapilli I–II). Only the Võhma I settlement site, which was once located on the shore of a small cove, has been archaeologically investigated (Kriiska 1998). The charcoal samples gathered from the fireplaces were dated to 5825–5175 cal BC. Eight Mesolithic sites are known from the Kõpu Peninsula on Hiiumaa (Kõpu II, III, IV/V, VI, VII/VIII, IX, XIV and XVII), of which the Kõpu IV/V and VII/VIII sites on the SE coast of the ancient island have been excavated. On the basis of the samples of charcoal and hazelnut shells the sites have been dated to 5645–5120 cal BC (Kriiska 1996b, 401, 407). In the Late Mesolithic, about 5300 cal BC, Ruhnu Island in the Gulf of Livonia was also seasonally inhabited. Relying upon the finds and geographic location, we may assume that the inhabitants of the islands came from the coastal regions of West Estonia. Unfortunately we know little of the Mesolithic period in that area. Only a part of the finds from the bottom of the Pärnu River and two settlement sites at Metsaääre (I, II) date from the period between the early Mesolithic Pulli settlement site and the site of Valge-Risti, probably contemporaneous with the early settlements on the islands. The islands were probably initially inhabited only temporarily. They offered the possibility of erecting a camp, finding firewood as well as raw material for making tools, and of leaving some supplies behind. Later, but possibly still in the Mesolithic, the settlement on the islands became permanent. The first area of permanent settlement must have been the largest island of this region, which was located in the western and northern part of present-day Saaremaa. The similarity of the finds with those from Hiiumaa and Ruhnu leads to the conclusion that the people who settled on Saaremaa also used other islands and islets of the region for obtaining food.

At the beginning of the Litorina Sea period, changes in settlement and economy occurred also elsewhere in the Baltic region. The first traces of settlement from several Finnish (ancient) islands date from that period, e.g. Kemiö (Asplund 1997, 218), Vantaa Jönsas and Kilteri (Nuñez 1978, 7, Fig. 2; Purhonen & Ruonavaara 1994, 91) and Åland (Nuñez & Gustavsson 1995, 223).

11 settlement sites of Neolithic hunter-gatherers in Estonia are known from the islands. Four more settlements are situated on ancient islands presently located on the mainland. On Saaremaa, four Neolithic settlement sites — Undva, Naakamäe, Loona (Jaanits 1965, 28–33) and Kõnnu (Jaanits 1979) are known to have been situated on the seashore. On Hiiumaa, seven Neolithic settlement sites are known from the Kõpu peninsula (Kõpu I, X, XI, XII, XIII, XV and XVI) (partly published in Kriiska 1995a; Lõugas et al. 1996; Kriiska & Lõugas 1999), and at least two of the six Stone Age settlements of Ruhnu Island belong to the Neolithic. In West Estonia, Kaseküla (Kriiska et al. 1998) and Rõuste settlement sites were situated on small islands near the mainland. In Southwest Estonia, Lemmetsa II and Malda settlements were situated on small islets forming a lagoon near the mainland. Later, when the uplift of the land had closed the connection to the sea and a relict lake was formed, the settlement was shifted to the mouth of the river flowing out of the lake (Lemmetsa I settlement site).

In many respects, the oldest investigated settlement sites of Northwest Saaremaa, Kõpu peninsula (Hiiumaa) and Ruhnu are very similar. It was customary in all areas to use local rocks for making tools. The dominant raw material was quartz and white, local grey and beige flint of inferior quality. In processing the stones, a splitting technique was often used. Only 1–2 % of the tools, mostly scrapers, were more thoroughly processed. The small number of chopping tools is also common in these areas. Only a few fragments or blanks of chisels and axes were found (Kriiska 1998, 19). The bone material and the small variety of artefacts at the early sites on the Baltic islands have been interpreted as indicators of seasonality (Moora & Lõugas 1995, 479; Nuñez 1996, 27; Kriiska 1996d, 3). It is neverless possible that seasonality on the islands does not mean — from the end of the Mesolithic — repeated arrivals and returns to the mainland, but probably the moving of a community (communities?) permanently living on the islands, within their hunting territories which may have comprised several islands and islets. This may be indicated by the singularity of the culture group on Saaremaa, Hiiumaa and Ruhnu at the beginning of the Neolithic. Since it differs considerably from the local groups of the Narva Culture on the mainland, it must have been formed to some extent separately. It is very likely that the regional peculiarities of pottery reflect the cultural differences that had already formed during the Mesolithic (Kriiska 1997, 17). The materials found at the Early Neolithic site of Kõnnu (Saaremaa) also indicates more permanent presence. Here, in contrast to the other early habitation sites on the islands, finds of chopping tools are numerous — nearly 500 stone axes and chisels (Kriiska 1998, 19).
 

MARITIME ECONOMY

It is clear that the resources depended on the natural environment of the Stone Age communities. Thus the choice of settlement sites reflects the type of economy, and the type of ancient landscape helps to determine the nutritional base. The animal bones found from the sites provide even more information about the Stone Age economy. Presuming that the share of the identified bone fragments reflects the past economic significance of the species (Söderholm & Ukkonen 1999, 47), several regional and temporal peculiarities can be observed in the Stone Age hunting and fishing economy.

A drastic change in economic and settlement strategies was brought about by the spread of agriculture. Though possibly earlier known on Estonian territory to some extent (Lang 1995), the spread of agriculture as the economy altering everyday life and settlement pattern was associated here with the Corded Ware Culture. From that period date the numerous pieces of evidence clearly indicating the spread of agriculture. Bones of goat, sheep, pig and cattle, as well as artefacts made of them, have been found in graves (Jaanits 1992, 48). A fragment of Corded Ware pottery with a charred grain of barley on its surface was found in the settlement of Iru (North Estonia) (Jaanits 1992, 49). Even more significant are the pollen diagrams of bog and lake sediments. Pollen of cultivated plants of the Stone Age, from the period 3900–1600 cal BC, has been found at 13 locations on the coasts of mainland Estonia and on the islands, and also inland (Kriiska 2000, Tab. 1). Wheat, oats and barley were cultivated.

The settlement pattern of the Corded Ware culture differed from that of the hunter-gatherer society of the Stone Age. The choice of dwelling sites is based on quite different criteria than in other Estonian Stone Age cultures. This is especially striking in Coastal Estonia where the settlement of that period is not located on the seashore but a short distance away. In North Estonia it is concentrated on the North Estonian Glint (Lang 1996, Fig. 101, 120; Lang & Konsa 1998). In several places the areas inhabited earlier, now far from the seashore, have been re-inhabited. The settlement sites of Võhma I (Saaremaa) (Kriiska 1998, 18) and Siimusaare (Central Estonia) (Jaanits 1959b, 161) reveal traces of Mesolithic settlements, and in Kõpu (Hiiumaa) as well as Riigiküla on the lower reaches of the Narva River, new settlements have been established at Early Neolithic sites (Kriiska 2000, 72). Small settlements and burial grounds seem to indicate the formation of communities of single families, typical of (at least) the farmers of Coastal Estonia in the Bronze Age and the Roman Iron Age (Lang 1995, 136; Kriiska 2000, 74). Unfortunately it is impossible to determine the share of grain and the meat of domestic animals in the diet of the people, compared to the products of gathering, hunting and fishing. Relying upon the Late Neolithic and Early Bronze Age bone material from Latvia and Lithuania, the share of domestic animals was not very large (Ostrauskas 1998). Never the less, agriculture determined the choice of settlement sites.

Some changes affecting settlement pattern had also occurred earlier. In the Baltic Stone Age, a considerable shift in the hunter-gatherer economy took place due to the climatic changes of the end of the glacial period. The warming of the climate caused changes in fauna and flora and reindeer as the main game was replaced by other species.

The glacier regressed from the Estonian mainland in about 11,000 cal BC and evidently the natural conditions already rendered human settlement possible (Moora 1998, 54). The nearest camp sites of reindeer hunters known to date have been discovered on the banks of the Daugava River in Latvia (Zagorska 1999), but it would be prematiere not to consider the possibility of hunting trips extending to Estonian areas as well.

The Early Mesolithic traces of human activities at the Pulli site are clearly connected with forests and the hunting of wild animals. Such an economy, lasting throughout the Stone Age, could be called an inland economy. It was based on hunting wild animals, fishing on inland bodies of water and gathering. At the present state of research we may say that one of the central areas of Mesolithic inland settlement was in Central Estonia, on the northern shore of Lake Võrtsjärv and on the banks of the Navesti River. Twelve sites are known from these areas (Fig. 3): Tamme, Jälevere, Moksi, Siimusaare, Umbusi (Jaanits 1981) and Lalsi I–IV, Laeva I–II and Leie (Kiristaja et al. 1998, 214–216). Several Mesolithic settlement sites are also known from the banks of the Emajõgi River.

Inland Neolithic settlement sites are known from the banks and shores of several rivers and lakes in Eastern Estonia (Akali, Kullamägi), Central Estonia (Valma) and Southeast Estonia (Kääpa, Tamula, Vagula, Villa I, II and Väike-Rõsna) (Jaanits 1959a; 1959b; 1976; Kiristaja et al. 1998, 222, 227) (Fig. 4). The main game throughout the Stone Age were elk and beaver, wild boar (in the osteological material of Valma outnumbering even elk, and ranging second in Akali), and auroch (Fig. 5). An extensive variety of fish is represented in the bone material, but pike, perch, common bream and vimba bream dominate (Lõugas 1997, Tab. 2).

In the coastal areas a different type of economy developed, with the hunting of marine mammals playing an important rolle. The time of its emergence is not yet clear, but in the Late Mesolithic since the beginning of the Litorina Sea period, it is clearly discernible in the archaeological and especially in the palaeozoological material. Two trends can be observed in the formation of coastal settlement and economy: 1) specialised seal hunting and 2) the hunting of marine and wild animals. Both of these trends survived for a long time and disappeared only with the spread of agriculture.

The reason for the formation of the maritime economy is disputable. One hypothesis from Finland relates it to a sort of ecological restraint occurring as a result of the significant decrease in the elk population (Siiriäinen 1982, 18). The Estonian material does not allow one to draw such conclusions. The author would instead agree with another Finnish researcher who suggests that the sea became more productive (Nuñez 1996, 24). The Litorina Sea was more brackish and evidently also more favourable for an increase of biomass in the sea and in particular also for the seal, increasing their populations.

In mainland Estonia the economy was strongly affected by the jointed coastline rich in lagoons. As already indicated, coastal settlements are known from several regions of Estonia, the lower reaches of the Narva River being the most thoroughly investigated of these. The first archaeological excavations took place there in as early as 1931, and excavations and survey trips have continued intermittently up to the present (for a survey of the research history see Kriiska 1996c). In addition to archaeological finds, rich osteological material has been collected there, giving a good overview of the fauna of Northeast Estonia in that period and enabling the drawing of conclusions about areas with analogous natural conditions. The osteological material indicates that different animals were hunted, elk, auroch and wild boar being the most commonly represented (Fig. 6). The relative significance of wild boar increased during the period of the Comb Ware Cultures, when the lagoon at Riigiküla had become marshy and the settlement had shifted to the banks of the Narva River in the eastern part of the former spit. Considerable changes in the relative significance of different species have taken place on the site of Narva Joaoru. In the Mesolithic (layers II and III) the elk dominates over the beaver and wild boar, while in the Early Neolithic (layer I) their share had been practically equal. A specific trait of the settlements of the Narva region is the great share of dog bones, and dog meat was evidently also been used for food. In the material of the Riigiküla IV settlement of the Narva culture, 35 % of the bone finds belong to pine marten, which might indicate the hunting of fur animals in winter (Kriiska 1999, 177; Lõugas 1999).

Seal has been an important game item on the lower reaches of the Narva River as well as elsewhere in coastal areas. Their relative significance in the material of Riigiküla sites was 12–24.5 %, at Kudruküla even 57.8 % (Fig. 6). Such proportions of seal bones at the latter site area is comparable with those of the islands of West Estonia for their economic base. In the early phase of the Neolithic, ringed seal and grey seal were hunted; since the Middle Neolithic, the harp seal appears in the bone material. The latter has been found on the mainland in Kudruküla and Riigiküla and from the bottom of the Pärnu River together with Stone Age bone artefacts (Lõugas 1997, Tab. 3). A new species of marine mammals appearing in the Neolithic was the only species of Cetacea in the Baltic, the porpoise, the bones of which have been found in the material from Kudruküla (Lõugas 1997, Tab. 3), Riigiküla I, III and IV settlements (Lõugas 1999, Tab. 1).
 

Fishing and fowling also played an important role in the economy. Of fish, pike dominated in coastal settlements too, but next in importance was pikeperch instead of perch (Lõugas 1997, Tab. 2; 1999, Tab. 1). Unfortunately only a few bird bones found from coastal settlements have so far been identified. Relying upon the bone finds, it has been determined that of waterfowl various ducks, the whooper swan, common scoter, etc. were hunted; of forest birds, the black grouse and capercailli were represented (Glück 1906, 275–276; Lõugas 1999, Tab. 1).

As already mentioned, the earliest traces of settlement on the islands date from the Late Mesolithic period. It is possible that the more remote islands were discovered only during long seal-hunting trips, at a time when hunting for marine mammals became an important source of livelihood for the communities of the coastal settlements of mainland Estonia. In any case, the hunting of marine animals was one of the main subsistence activities on the islands. In all the Mesolithic and Early Neolithic settlements investigated to date the proportion share of seal bones is remarkably large. In the Mesolithic settlements (Võhma I, Kõpu IV/V and VII/VIII), all the bones identified by species belong to seals (Fig. 7). In the Kõpu I settlement (Hiiumaa) of the Narva Culture, apart from the seal bones only four fragments of a hedgehog’s mandile have been found. This is very interesting since the hedgehog is unable to populate the islands isolated by the sea, so they must have been brought along by people. It is possible, of course, that they were taken along as a food supply but, relying upon ethnological parallels, we may hypothesise that the hedgehog bones possessed a religious rather than economic importance (Lõugas et al. 1996, 206–207). Seal bones were predominant (93.8 %) also in the osteological material of the Kõnnu site (Saaremaa) of the Narva Culture. But there, stray bones of elk, beaver, wild boar, pine marten and fox were also found (Lõugas 1997, Appendix II, A.). Small amounts of birds’ bones— eider, long-tailed duck, cormorant, merganser, goldeneye (Kriiska & Lõugas 1999, Tab. 6) and some fish bones — pike, roach, perch, cod and turbot (Lõugas 1997, Tab. 2) have also been found. Fish bones are so rare that they have come from seals’ bowels. Fowling may have occupieda realatively important rolle in the early economy of the islands, as is the case with the gathering of birds’ eggs. In connection with the early habitation of Åland, an interesting phenomenon of the Baltic Sea has been discussed: polynia — reccurent spaces of open water in the midst of ice, often used for living and feeding places by fowl and marine animals. Such places offered Stone Age people possibilities for successful hunting (Nuñez 1996, 29–32). Hazelnut shells (Kriiska 1995, 413) have been found at the Kõpu I settlement site — one of the few traces of gathering from Estonian Stone Age settlement sites.

The presumable period of residence at the early sites of Võhma and Kõpu was early spring —  the best time for hunting ringed seal and grey seal. In February and March these species migrate, even today, to the present territorial waters of Estonia to give birth to their young, and become easy targets. For the ringed seal puppying takes place mainly at the end of February, and for the grey seal in March (Aul et al. 1957, 268–269). Although both species can be hunted to some extent all year round, early spring has always been the best and most productive time for hunting seals (Kalits 1963, 136; Art 1988, 13). The direct evidence hunting during the puppying period is the bone of a ringed seal of at most a week old found at the Kõpu I site (Moora & Lõugas 1995, 479). The close stone settings of the fire-pits, accumulating heat and keeping the dwellings warm for longer, might also indicate the habitation of the site in the cold season.

In the second half of the Neolithic the hunting economy of the islands had undergone considerable changes (Fig. 8). The main game was still seal with its proportion among mammal bones remaining 72.3–94 %. In addition to the grey seal and the ringed seal, harp seal and porpoise also appear among the game. Of the bone finds from Naakamäe identified by species, 67 % belong to harp seal. The respective percent in Loona is even as high as 92.4 %. There has presumably not been a local population of harp seal in the Baltic (hitherto no finds permit a contrary assumption). Evidently the very difficult ice conditions and the favourable feeding base in the Litorina Sea during the sub-boreal climatic period forced them to migrate from the Arctic Ocean to the Baltic Sea in autumn and early winter for feeding (Lepiksaar 1964; Lõugas 1998). At strategic points the catching of such animals, migrating in great numbers, must have been relatively easy. It is possible that harp seal was sporadically caught with nets at this time (Lõugas 1998).
 

The importance of salt-water fish was increasing. From the bone material of the settlement of the Late Combed Ware Culture in Kaseküla, 11 fish species could be identified, since the limestone soil had preserved the bones in excellent condition (Kriiska et al. 1998, Tab. 3). Among them flounder, perch, cod, eel and pike prevailed. On distant islands, the predominant fish was cod. This was especially predominant (97 %) in the rich bone material (ca 10,000 fish bones) of the Loona settlement site. Among the few fish bones of the Naakamäe site, the bones of common sturgeon and turbot should be mentioned (Lõugas 1997, Tab. 2). The finds of deep-water cod indicate active fishing on the open seas. From the settlement of Loona (Saaremaa), a relatively large number of wild boar bones (20.8 %) were found, and from the settlement of Kõpu XI (Hiiumaa) and the Kaseküla settlement which was then located on a small island near the mainland, eel bones were found. Among the numerous pig bones from Loona, 30 pieces have roused suspicion that they belonged to half-domesticated young animals (Paaver 1965, 440).

Considering that the best time for hunting grey seal is early spring, that harp seals could be hunted only in autumn and early winter, and that porpoise appears in the Baltic Sea mostly in summer and autumn, we can presume that the above-mentioned Late Neolithic sites were inhabited in at least those periods, but possibly year-round. The archaeological materials found, containing more specific tools and more tools with evidence of secondary processing, also indicated more stable and permanent settlement.

ACKNOWLEDGMENTS

The author wishes to express gratitude for collaboration to the archaeologist Ulla Saluäär (Museum of Pärnu), palaeozoologist Dr. Lembi Lõugas (Institute of History), geographer Dr. Toomas Kokovkin (Island and coast research centre Arhipelaag), artist Jana Ratas (Institute of History) and all the participants in my different archaeological expeditions. For various help I would like to express my gratitude to the archaeologist Prof. Valter Lang (University of Tartu), the archaeologist Prof. Ari Siiriäinen (University of Helsinki), the geologist Prof. Volli Kalm (University of Tartu), the geologist Urve Miller (University of Stockholm), the Master of Geology Juho Kirs (University of Tartu), translators Liis Soon (Institute of History), Triinu Mets-Sõmermaa (Estonian Institute of Humanities), Are Tsirk, Alexander Harding and many other kind friends who have not refused their help. I owe special thanks to my teacher, archaeologist Dr. Lembit Jaanits, and the team of the present electronic publication: webmaster Krista Sutt and archivists Katrin Martsik and Anu Lepp.

The investigations were financed by: the Estonian Science Foundation (grant no. 1022, 2254 and 3332), the Estonian Fund for Furthering Culture, the EU/PACT project (ERBICIPACT 93-0152), the University of Tartu, the Estonian Academy of Science, the Museum of Narva, the Hiiumaa Centre for Biosphere Reserve of the West Estonian Archipelago, the Institute of History, the Fishing Industry of Pärnu and the newspaper Maaleht.
 

REFERENCES

Ailio, J. 1924. Katsaus Viron myöhäisimpiin arkeologisiin tutkimuksiin. — Suomen Museo, XXX, 1923. Helsinki, 41–58.

Art, E. 1988. Hülged ja hülgepüük. — Hülgepüük. Hülgepüügi meenutusi möödunud aegadest. Stockholm, 5–15.

Asplund, H. 1997. Kemiön suurpitäjän esihistoria. — Kemiön suurpitäjän historia, I. Tammisaari, 213–286.

Aul, J., Ling, H. & Paaver, K. 1957. Eesti NSV imetajad. Tallinn.

Bliebernicht, E. G. 1924. Neue Funde aus dem Pernauflusse. Fundbericht von den J. 1920–1922. — Suomen Muinaismuistoyhdistyksen Aikakauskirja, XXXIV:2, Helsinki, 3–19.

Bolz, M. 1914b. Neolitchische Steingeräte aus dem Pernau-Fellinische Kreise und dessen Umgebung. — Sitzunsberichte der Altertumforschenden Gesellschaft zu Pernau, 7. Pernau, I–CXVI.

Carpelan, C. 1999a. Käännekohtia Suomen esihistoriassa aikavälillä 5100...1000 eKr. — Pohjan poluilla. Suomalaisten juuret nykytutkimuksen mukaan. Bidrag till kännedom av Finlands natur och folk, 153. Helsinki, 249–280.

Carpelan, C. 1999b. On the Postglacial Colonisation of Eastern Fennoscandia. — Dig it all. Papers dedicated to Ari Siiriäinen. Helsinki, 151–171.

Edgren, T. 1992. Den förhistoriska tiden. — Finlands historia, 1. Ekinäs, 9–270.

Girininkas, A. 1994. Baltu kulturos ištakos. Vilnius.

Glück, E. 1906. Über neolithische Funde in der Pärnu und die Urbewohner der Pernau-Gegend. — Sitzunsberichte der Altertumforschenden Gesellschaft zu Pernau, 1903–1905, 4. Pernau, 259–318.

Grewingk, C. 1865. Das Steinalter der Ostseeprovinzen Liv-, Est- und Kurland und einiger angrenzeden Landstriche. Schriften der Gelehrten Estnischen Gesellschaft zu Dorpat, IV. Dorpat.

Grewingk, C. 1871. Zur Kenntniss der in Liv-, Est-, Kurland und einigen Nachbargegenden aufgefundenen Steinwerkzeuge heidnischer Vorzeit. — Verhhandlungen der Gelehrten Estnischen Gesellschaft zu Dorpat, VII:1. Dorpat, 1–56.

Grewingk, C. 1874. Zur Archäologie des Balticum und Russlands. — Archiv für Antropologie, VII:1. Braunschweig, 1–54.

Grewingk, C. 1882. Geologie und Archaeologie des Mergellagers von Kunda in Estland. Archiv für die Naturkunde Liv-, Est- und Kurland, I, IX:1. Dorpat.

Grewingk, C. 1884. Die Neolithischen Bewohner von Kunda in Estland und deren Nachbarn. — Verhhandlungen der Gelehrten Estnischen Gesellschaft zu Dorpat, XII. Dorpat, 1–88.

Hausmann, R. 1904. Ueber Gräber aus der Steinzeit im Ostbaltikum. Grabfunde in Woisek und Kölljala. — Sitzungsberichte der Gelehrten Estnischen Gesellschaft, 1903. Jurjew (Dorpat), 71–81.

Hausmann, R. 1912. Das Steinzeitgrab von Karlowa bei Dorpat. — Sitzungsberichte der Gelehrten Estnischen Gesellschaft, 1911. Jurjew-Dorpat, 60–66.

Indreko, R. 1926. Die Rambachsche Sammlung. — Sitzungsberichte der Altertumforschenden Gesellschaft zu Pernau, 1914–1925, 8. Pernau, 283–344.

Indreko, R. 1932. Kiviaja võrgujäänuste leid Narvas. — Eesti Rahva Muuseumi Aastaraamat, VII, 1931. Tartu 1932.

Indreko, R. 1935. Sépultures néolithiques en Estonie — Õpetatud Eesti Seltsi Aastaraamat, 1933. Tartu, 202–223.

Indreko, R. 1936. Vorläufige Bemerkungen über die Kunda-Funde. — Sitzungsberichte der Gelehrten Estnischen Gesellschaft, 1934. Tartu, 225–298.

Indreko, R. 1938. Ein Hockergrab in Ardu, Ksp. Kose. — Litterarum Societas Esthonica 1838–1938 Liber Saecularis. ÕES Toimetised, XXX. Tartu, 185–200.

Indreko, R. 1948a. Bemerkungen über die wichtigsten steinzeitlichen Funde in Estland in den Jahren 1937–1943. — Antikvariska Studier, III. Kungliga Vitterhets Historie och Antivitets Akademiens Handlingar. Stockholm, 291–315.

Indreko, R. 1948b. Die mittlere Steinzeit in Estland, Mit einer Übersicht die Geologie des Kunda-Sees von K. Orviku. Kungliga Vitterhets Historie och Antikvitets Akademiens Handlingar. Stockholm.

Jaanits, K. 1973. Eesti mesoliitilised tulekivist nooleotsikud. — Eesti NSV Teaduste Akadeemia Toimetised, 22:3, 306–314.

Jaanits, K. 1981. Die mesolithischen Siedlungsplätze mit Feuersteininventar in Estland. — Mesolithikum in Europa, 2. 2. Intern Symp., Potsdam, 3. bis 8. Apr. 1978. Veröffentlichungen des Museums für Ur- und Frühgeschichte, 15. Berlin, 389–399.

Jaanits, K. & Ilomets, M. 1988. Umbusi mesoliitilise asula vanusest ja kohast Eesti keskmise kiviaja kronoloogias. — Loodusteaduslikke meetodeid Eesti arheoloogias. Artiklite kogumik. Tallinn, 54–64.

Jaanits, L. & Jaanits, K. 1975. Frühmesolithische Siedlung in Pulli. — Eesti NSV Teaduste Akadeemia Toimetised. Ühiskonnateadused, 24:1, 64–70.

Jaanits, L. & Jaanits, K. 1978. Ausgrabungen der frühmesolithischen Siedlung von Pulli. — Eesti NSV Teaduste Akadeemia Toimetised. Ühiskonnateadused, 27:1, 56–63.

Jaanits, L. 1954. Neoliitilised ja varase metalliaja asulad Emajõe suudmealal. — Eesti NSV Teaduste Akadeemia Toimetised, 3:3, 350–366.

Jaanits, L. 1955. Neoliitilised asulad Eesti NSV territooriumil. — Muistsed asulad ja linnused. Arheoloogiline kogumik. Tallinn, 176–201.

Jaanits, L. 1965. Über die Ergebnisse der Steinzeitforschung in Sowjetestland. — Finskt Museum, LXXII. Helsinki, 5–46.

Jaanits, L. 1970. Kultuuri arengupidevusest Eestis üleminekul keskmiselt nooremale kiviajale. — Studia archaeologica in memoriam Harri Moora. Tallinn, 81–87.

Jaanits, L. 1979. Die neolithische Siedlung Kõnnu auf der Insel Saaremaa. — Eesti NSV Teaduste Akadeemia Toimetised. Ühiskonnateadused, 28:4, 1979, 363–367.

Jaanits, L. 1992. Põllumajanduse eelduste kujunemine. — Eesti talurahva ajalugu. Tallinn, 42–56.

Jaanits, L., Laul, S., Lõugas, V. & Tõnisson, E. 1982. Eesti esiajalugu. Tallinn.

Jaanits, L. & Liiva, A. 1973. Eesti vanima ajaloo kronoloogia ja radiosüsinikumeetod. — Eesti NSV Teaduste Akadeemia aastail 1965–1972. Tallinn, 157–161.

Jensen, J. A. 1989. Langagergård II — udgravningsresultat. Fund af en amfora fra jysk enkeltgravskultur. — Stidsyxekultur i Sydskandinavien. Rapport från det andra nordiska symposiet om Stridsyxetid i Sydskandinavien 31.X–2.XI 1988. University of Lund Institute of Archaeology Report series, 36, 13–17.

Kalits, V. 1963. Kihnlaste elatusalad XIX sajandi keskpaigast tänapäevani. Tartu, 1963. (Manuscript in the Library of University of Tartu.)

Kalm, V. 1996. X-ray diffraction analysis of Neolithic ceramics: Examples from the Narva area, South-Eastern Coast of the Gulf of Finland. — Coastal Estonia. Recent Advances in Environmental and Cultural History. PACT 51. Rixensart, 385–396.

Kalm, V., Kriiska, A. & Aruväli, J. 1997. Mineralogical analysis applied in provenance studies of Estonian Neolithic pottery. — Proceedings of the Estonian Academy of Sciences. Geology, 46:1, 16–34.

Kiristaja, A., Tvauri, A. & Vindi, A. 1998. 1997. aasta arheoloogilised inspektsioonid. — Arheoloogilised välitööd Eestis 1997. Tallinn, 210–236.

Kivimäe, J., Kriiska, A. Põltsam, I. & Vunk, A. 1998. Merelinn Pärnu. Pärnu.

Kriiska, A. 1995a. Archäologische Ausgrabungen auf dem Standort der ehemaligen Steinzeitsiedlung Kõpu I (Ristipõllu). — Proceedings of the Estonian Academy of Science. Humanities and social sciences, 44:4, 410–416.

Kriiska, A. 1995b. Narva jõe alamjooksu ala neoliitiline keraamika. — Eesti arheoloogia historiograafilisi, teoreetilisi ja kultuuriajaloolisi aspekte. Muinasaja teadus, 3. Tallinn, 54–115.

Kriiska, A. 1996a. Archaeological excavations on the neolithic site of Riigiküla IV. — Proceedings of the Estonian Academy of Science. Humanities and social sciences, 45:4, 410–419.

Kriiska, A. 1996b. Archaeological studies on the Kõpu Peninsula. — Proceedings of the Estonian Academy of Science. Humanities and social sciences, 45:4, 398–409.

Kriiska, A. 1996c. Stone Age Settlements in the Lower Reaches of the Narva River, North-Eastern Estonia. — Coastal Estonia. Recent Advances in Environmental and Cultural History. PACT 51. Rixensart, 359–369.

Kriiska, A. 1996d. Viron rannikkoalueen asutus ja pyyntikulttuurin erikoistuminen kivikaudella. — Muinaistutkija, 4, 1–6.

Kriiska, A. 1997a. Excavations of the Stone Age Site at Vihasoo III. — Arheoloogilised välitööd Eestis 1996. Stilus, 7, 19–28.

Kriiska, A. 1997b. Kroodi ja Vihasoo III asula Eesti varaneoliitiliste kultuurirühmade kontekstis. — Eesti Arheoloogia Ajakiri, 1, 7–25.

Kriiska, A. 1997c. Pärnu muinasuurijad ja muinasteadus. — 100 aastat Pärnu Muinasuurimise Seltsi. Artiklite kogumik. Pärnumaa ajalugu, 1. Pärnu, 18–29.

Kriiska, A. 1998. Mesoliitilised asustusjäljed Loode-Saaremaal. — Ajalooline Ajakiri, 1 (100), 13–22.

Kriiska, A. 1999. Formation and Development of the Stone Age Settlement at Riigiküla, Northeastern Estonia. — Environmental and Cultural History of the Eastern Baltic Region. PACT, 57. Rixensart, 173–183.

Kriiska, A. 2000. Corded Ware Culture Sites in North-Eastern Estonia. — De temporibus antiquissimis ad honorem Lembit Jaanits. Muinasaja teadus, 8. Tallinn, 59–79.

Kriiska, A., Jonuks, T. & Kraas, P. 1999. Eesti muinasesemed. [http://brutus.fil.ut.ee/klass/andaluusia]. 05.05.2000.

Kriiska, A., Lõugas, L. 1999. Late Mesolithic and Early Neolithic Seasonal Settlement at Kõpu, Hiiumaa Island, Estonia. — Environmental and Cultural History of the Eastern Baltic Region. PACT 57. Rixensart, 157–172.

Kriiska, A, Lõugas, L. & Saluäär, U. 1998. Archaeological excavations of the Stone Age settlement site and ruin of the stone cist grave of the Early Metal Age in Kaseküla. — Arheoloogilised välitööd Eestis 1997. Tallinn, 30–43.

Lang, V. 1995. Varane maaviljelus ja maaviljelusühiskond Eestis: ääremärkusi mõningate arengutendentside kohta. — Eesti arheoloogia historiograafilisi, teoreetilisi ja kultuuriajaloolisi aspekte. Muinasaja teadus, 3. Tallinn, 116–181.

Lang, V. 1996. Muistne Rävala. Muistised, kronoloogia ja maaviljelusliku asustuse kujunemine Loode-Eestis, eriti Pirita jõe alamjooksu piirkonnas. Muinasaja teadus, 4. Tallinn, 1996.

Lang, V. & Konsa, M. 1998. Two Late Neolithic to Early Iron Age settlement sites at Ilumäe, North-Estonia. — Archaeological field works in Estonia 1997. Tallinn, 65–77.

Larsson, L. 1989. Boplatser, bebyggelse och bygder. Stridsyxekultur i södra Skåne. — Stidsyxekultur i Sydskandinavien. Rapport från det andra nordiska symposiet om Stridsyxetid i Sydskandinavien 31.X–2.XI 1988. University of Lund Institute of Archaeology Report series, 36, 53–76.

Larsson, L. 1997. Coastal Settlement during the Mesolithic and Neolithic Period in the Southernmost Part of Sweden. — The Built Environment of Coast Areas during the Stone Age. The Baltic Sea-Coastal Landscape Seminar. Session No. 1. A Symposium at the Centenary of Archaeological Excavations at Rzucewo (Rzucewo-Gdansk) 4–9 October 1994. Gdansk, 12–22.

Lepiksaar, J. 1964. Subfossile Robbenfunde von der Schwedischen Westküste. — Zeitschr. f. Säugetierk, 29 (5). Hamburg, 257–266.

Loze, I. 1992. Corded Pottery Culture in Latvia. — Praehistorica, XIX. Scnurkeramik symposium 1990 – Univerzita Kralova. Praha, 313–320.

Lõugas, L. 1994. Subfossil vertebrate fauna of Asva site, Saaremaa. Mammals. — Stilus, 5. Tallinn, 71–93.

Lõugas, L. 1997. Post-Glacial development of vertebrate fauna in Estonian water bodies. A palaeozoological study. Dissertationes Biologicae Universitatis Tartuensis, 32. Tartu.

Lõugas, L. 1998. Postglacial Invasions of the Harp Seal (Pagophilus groenlandicus) into the Baltic Sea. — Proceedings of the Latvian Academy of Science. Section B, 52, ½, 63–69.

Lõugas, L. 1999. Animal Remains from the Neolithic Riigiküla Sites, Northeastern Estonia. — Environmental and Cultural History of the Eastern Baltic Region. PACT, 57. Rixensart, 185–190.

Lõugas, L., Kriiska, A. & Moora, H. 1996. Coastal Adaptation and Marine Exploitation of the Island Hiiumaa, Estonia, During the Stone Age with Special Emphasis on the Kõpu I Site. — Landscape and Life. Studies in honour of Urve Miller. PACT 50. Rixensart, 197–211.

Lõugas, L., Lidén, K. & Nelson, E. 1996. Resource utilisation along the Estonian coast during the Stone Age. — Coastal Estonia. Recent Advances in Environmental and Cultural History. PACT 51. Rixensart, P. 359–369.

Lõugas, V. 1970. Eesti varane metalliaeg (II a.-tuh. keskpaigast e.m.a. – 1. sajandini m.a.j.). Dissertatsioon ajalooteaduste kandidaadi kraadi taotlemiseks. Tallinn. (Manuscript in the Library of Institute of History.)

Lõugas, V. 1982. Beiträge zur Vorgeschichte des Westarchipels Estland. — Eesti NSV Teaduste Akadeemia Toimetised. Ühiskonnateadused, 31:4, 372–376.

Lõugas, V. 1988. Loodusteaduslike meetodite kasutamisest Eesti arheoloogias. — Loodusteaduslikke meetodeid Eesti arheoloogias. Artiklite kogumik. Tallinn, 13–24.

Malmros, C. & Tauber, H. 1977. Kulstof-14 dateringer af dansk engeltgravskultur. — Aarbørger for nordisk Oldkyndighed og Historie 1975. København, 78–95.

Moora, H. 1936. Kiviaeg. — Eesti ajalugu, I. Tartu, 10–62.