| The Burial Cairns and the Landscape in the Archipelago of Åboland, SW Finland, in the Bronze Age and the Iron Age | ||
|---|---|---|
| Prev | Chapter 9. On Continuity, Settlement, and Subsistence | Next |
The hypothesis concerning the peripheral nature of the archipelago which was introduced in the 1930s has a central position in the research history of the archaeology of the archipelago of Åboland; according to this hypothesis the archipelago was, in view of the agrarian population which is the focal point of archaeology, a remote and barren wild region whose settlement came to an end at the end of the Bronze Age with no resettlement in the Iron Age. The first new settlers came to the archipelago only in the 12th century with the Swedish colonization (see chapter 2). On the mainland coast of South-West Finland the course of events was different: according to the present conception the settlement there was uninterrupted all through the Iron Age (e.g. Salo 1995). Field archaeological investigations have, however, brought forth observations which are not in favour of the hypothesis mentioned above. In the following, the problems associated with the hypothesis, the archaeological evidence of the settlement of Åboland in the Iron Age, and the aspects of pollen analysis and place name studies in reference to the problem of settlement, will be discussed.
The concept according to which the absence of finds in a certain region within a given prehistoric period B implies the disappearance of settlement in that period (B) despite the reported presence of such finds from the preceding period A, occurs over and over again in the research tradition of Finnish settlement archaeology. This concept may be based even on a conspicuous scarcity of finds in that period (B), or on finds which might be associated with the economic exploitation of uninhabited wilds. It has been thought that the principle of caution results in a conditional statement according to which there has been no settlement, if its existence cannot be demonstrated specifically. This condition may, of course be completed with additional statements concerning the field archaeological research situation. In the following, a remote example will be given: In J.R. Aspelin’s days in the 1870s some ten Bronze-Age metal artefacts had been reported from Finland, and by virtue of the scarcity of the finds, the possible settlement of Finland was, in his opinion, at any rate questionable.
The concept described above leads to certain problems if applied indiscriminately. In the first place one must assume generally speaking that the biological survival of man must, for a long period of time, have been the principal rule at least in regions where the natural conditions have allowed the subsistence of man. The population in Finland could not have multiplied in course of millenniums if this assumption did not prevail: according to calculations made by Eino Jutikkala the population of Finland in the Stone Age was 2500–10000 people, and by the Famine Years 1695–1697 the population had risen up to half a million. The growth of the population was not monotonous but the trend of the development must have been ascendant in the long run (Jutikkala 1987). Consequently, also the long-term trend of the settlement must have been mainly ascendant and increasing. Otherwise it is rather difficult to discuss the settlement processes in the prehistoric period and in the Middle Ages (e.g. Orrman 1991 b), or to vindicate conclusions which have built culture-historical bridges across dozens, even hundreds of generations when discussing continuities in material culture or archaic cultural practices, categories, and phenomena, such as beliefs, rituals, and institutions pertinent to the cultural traditions of wildernesses (Sarmela 1984), the linguistic footprints of early iron industries (Salo 1993), simple boat constructions (Itkonen 1941), the use of vessels (Salo 1989), the history of agricultural tools (e.g., Vilkuna 1971), and old etymologies (Häkkinen 1984; Anttonen 1996; Häkkinen 1999).
The disappearance of settlement, its desolation is thus a specific process in prehistory, a phenomenon deviating from the general expansive trend, which should have one or more locally and temporally limited explanations. Case studies from Europe, Africa, the Middle East, and the two continents of America suggest that permanent abandonment of dwelling sites took place only in exceptional economic, demographic, ecological, or regional sociocultural circumstances (Tomka & Stevenson 1993), to say nothing about whole large areas being depopulated.
Differences in excavation or survey methods, or in the coverage of the field research between various regions cannot be considered responsible for such depopulation. Also the intensity of the survey affects strongly what is observed and what remains unobserved; this, on the other hand, is drastically reflected in interpretations and conclusions (Plog et al. 1978). The same applies, naturally, to the effect of survey methods, the abundance, clustering, obtrusiveness, visibility, and accessibility of the archaeological remains (Schiffer et al. 1978) on the result of the field work. Strict caution should be observed in interpretations, if the field research is in its initial phase, or if the finds derive from other operations than field research. The artefacts registered by J.R. Aspelin were obtained from such specific situations in field research. The continuity of settlement through the Bronze Age has no more been questioned for a long time (Salo 1984 a). Generally speaking, the contemporary notion is that there is an obvious continuity of culture and settlement on the coast of Finland from the Subneolithic Stone Age to the Iron Age (Salo 1981a). Being wise afterwards does not, however, justify us to ignore the fact that we are still facing the same problems as Aspelin because there are regions and periods where the registrations of finds are very scarce.
Another problem of methodology in the interpretation of real or seeming shortcomings in the find material concerns the regional scope. Its significance is illustrated by the debate concerning the development of the settlement of southern Ostrobothnia in 800–1200 AD. For a long time the only prevailing concept was that the Iron-Age settlement which is represented by cemeteries and graves with abundant finds until the 7th and the 8th centuries, began to vanish from this area in the 9th century, and Ostrobothnia was desolated (e.g. Meinander 1977). Sporadic finds have been reported from this area from the Late Iron Age but according to the traditional opinion they do not indicate a continuity of settlement; instead, they have been interpreted as being left there by hunters and other travellers in the wilds who came from the more southern province of Satakunta. The present settlement of the region did not start until the beginning of the Middle Ages with the colonization from Sweden. The research team of Evert Baudou joined together archaeological excavations and pollen analyses during their field research in southern Ostrobothnia in the years 1986–1991. Attention was paid not only to central settlement regions with abundant finds but also to regions exposed by rapid shore displacement with very few, if any, registered finds from the Iron Age. According to the reports given by the research team (Baudou et al. 1991; Baudou 1991), the results of the investigations indicated that the discontinuity of settlement was only local, not regional. The conditions of cultivation and cattle raising in the region changed in the Late Iron Age because shore displacement in the flat and even landscape of Ostrobothnia (e.g. Jones 1988) resulted in grazing lands turning into marshy ground. For this reason the settlement moved closer to the sea shore where shore meadows still provided pasture land for the cattle.
Critical views have been presented about the interpretation of the pollen analyses from southern Ostrobothnia (see Orrman 1994 a; Orrman 1999) but, nevertheless, the investigations conducted by Baudou’s team demonstrate very clearly the significance of independent methods and of a sufficiently extensive local scale. If the area surveyed in questions about the continuity, enlargement, and structure of the settlement is rather small – the banks of a river or the shore of a Finnish lake and the like – one easily ends up with a situation in which the conclusions have to be based on a very limited number of finds, occasionally even on one single find. In such a case the possibility of studying the settlement as a regional process adapting to gradual changes like shore displacement etc. is wasted on one hand, and, on the other hand, every new find will change ostensibly the image of the settlement of the region.
The presumption that a certain region is uninhabited, does not thus necessarily turn out to be the simplest and most cautious null hypothesis because it must be altered every now and then with increasing find material. To assume that every region should be regarded as potentially inhabited if such a notion is not challenged by obvious evidence, seems to be more in accord with the nature of the field archaeological material. Such ”obvious evidence” might purport, for instance, natural conditions which make human settlement impossible, natural catastrophes, or an absolute absence of evidence of any settlement, in spite of systematically and purposefully conducted field archaeological investigations[1]. The latter assumption – ”it is there, if only we can find it” – is, with increasing find material, more sustainable than the assumption of desolation or depopulation, but it deprives, naturally, the finds of the fascination which is inextricably linked with finding something new and unexpected.
A dwelling site is a primary indication of settlement in an archaeological material. As Bronze-Age and Iron-Age dwelling sites are usually difficult to trace, the location and the development of settlement have been traditionally investigated by means of the location and distribution of graves and cemeteries. Due to the obtrusiveness of cemeteries and graves they are often more easily discovered than the remains of buildings and dwelling sites which have been covered by earth. Graves have, at least in some cases, been ravaged more seldom than dwellings by construction work and haulage of gravel. As the graves and the cemeteries are often located in the vicinity of known dwelling sites, the graves – particularly those from the Iron Age – are regarded as reliable indications of sedentary settlement even where no dwelling sites have been registered. Indirectly deduced settlement has been called cemeterial settlement (Fi. kalmistollinen asutus, Salo 1995; Salo 1999). The development of settled areas, cultural regions, and barter trade in Finland Proper and Häme (SW and S Finland) in the Late Iron Age, as described by Kivikoski, is an example of writing prehistory on the basis of cemeterial settlement (Kivikoski 1961: 290–293).
It would, however, be unwise to conclude inversely that the absence of cemeteries or graves is a direct indication of the absence or desolation of settlement. Matti Huurre expresses this notion as follows:
Focusing research exclusively on cemeteries involves certain risks. As a cemetery can be regarded as a distinct implication of Iron-Age settlement, it may lead easily to the concept that the absence of cemeteries is an indication of the absence of settlement. One may, however, pose the question why stable settlement could not have been possible even without a certain type of cemetery. In fact, cemeteries demonstrate only a certain type of burial which, again, is associated with religion and beliefs. Thus the increasing prevalence of cemeteries reflects the distribution of new religious concepts which need not have anything to do with the spreading of settlement. It has been frequently emphasized that one might believe that Finland had turned into a totally desolate country by the advent of Christianity if one relied on cemeterial finds only (Huurre 1979: 141).
The presence of graves and cemeteries is thus not a necessary but often a sufficient condition for the conclusion that the finds indicate a sedentary settlement. The historical-geographical context has been of significance for various researchers in the interpretations of cemeterial settlement. A 12th-century grave with a cremated female body was discovered in Suomussalmi in northeastern Finland. In the otherwise sparse Iron-Age find milieu this grave was interpreted as an isolated interment in the wilds, and not as an indication of sedentary settlement (Huurre 1973 b; cf. Huurre 1992). In western Finland, in Satakunta, 14 cairn cemeteries have been regarded as indications of sedentary settlement from the Late Roman Iron Age, and this has resulted in inferring changes in the assumed central regions of settlement (Salo 1999: 8). The arguments of these interpretations are differentiated by the difference between the remote northern taiga and the fertile south-western agrarian region on one hand, and by the frequencies of the graves on the other. In both cases it is, however, a question of interments in river valleys. Where does the line run within archaeological finds between sedentary settlement, mobile residential settlement, and desolate wilderness? On the mainland coast of Finland Proper there are more than 600 registered Iron-Age graves and cemeteries which have undoubtedly been interpreted as indications of sedentary agrarian settlement. But what about the more than 200 Iron-Age cairns in Åboland at the distance of 10–70 km from the mainland coast towards the archipelago; are they wilderness interments or graves within sedentary settlement?
The basic assumption of cemeterial settlement, the location of Metal-Age graves and dwellings close to each other, seems to be valid in several contexts. With the help of graves we are able to outline the regions which undoubtedly have been inhabited but burial sites cannot throw much light on the location of dwelling sites or on desolate regions. Even a small number of registered graves may be an indication of sedentary settlement but the less and fewer the archaeological indications are, the more cautious one should be. The graves contain, however, also other intimations of prehistoric communities than of their location: they may provide knowledge of beliefs, rituals, the relationship between man and nature, subsistence strategies, gender, and the biological qualities of the deceased. When these traits contain local contexts, connections with the ecological settings of the grave sites, it is not too audacious to regard the graves as culturally significant monuments of locally anchored people whose dwelling sites were somewhere in the vicinity.
For decades it has been a common conception that the Iron Age archipelago of Åboland was a desolate wilderness, exploited mainly for long-distance utilization only and thus resembling the wilderness regions of the inland of Finland. However, in the imagery of the Finnish national prehistory writing the archipelago has never acquired the same prominence as the wooded inland regions. Derek Fewster has characterized the early 20th-century national construction of the Finnish Iron Age with its heroic spirit of the Kalevala, the conquest of wild uninhabited forests, and the tribal organization into semi-state communities whose military power was based on hillforts. In this scenery the worst enemies of Finns were the Vikings (Fewster 1999). In the (pre)historical archtypes of Finnishness the maritime archipelago seems to have been an alien region and therefore left outside of the mainstream of research. The archipelago was the pillaging nest of the Vikings, there was no network of hillforts as there was in the interior and on the south-western coast of the country. And the starting point of the settlement in the archipelago, the Bronze Age with its predominantly Scandinavian tokens – Tallgren used the term the ”Scandinavian” Bronze Age in Finland (Tallgren 1937) – was not a very important part of prehistory in the construction of Finnishness because early prehistorical populations were not regarded as predecessors of today’s Finns.
The urban bourgeoisie discovered the fashion of spending summer in the archipelagos of Sweden and Finland in the late 19th century. For them the archipelago was a more homelike equivalent to an alpine or wilderness milieu, it was an aesthetically appreciated untamed virgin territory. There was no contradiction between this milieu and its inhabitants, the islanders who made their humble living by doing hard work (Eklund 1985; Montin 1996). The literature concerning the archipelago, and the belles lettres with ethnographic characteristics in particular (see e.g. Nygren 1989) has contributed to defining the islanders as diligent people who lived a modest life in harsh and isolated circumstances close to nature. The archipelago came to represent intact nature, an old-fashioned way of living, and freetime experiences, and the natural esthetics of the archipelago was placed on a par with the fascination of Lappish nature (Huuhtanen 1983).
The image of the archipelago has supported the concept of the modesty of the archipelagian culture and its submission to the harsh circumstances even in the learned interpretations concerning the past. This accounts, at least partly, for the idea of the archipelago as a many-thousand-year-old wild and desolate region. Kenneth Gustavsson has paid attention of the description of the medieval monastery of Kökar in the archipelago of Åland by Reinhold Hausen, the State Antiquarian (1850–1942): it was
(…) a poor convent on a few out-of-the-way and almost forgotten islands where the main task of the brothers was to pacify the wild islanders” (Gustavsson 1994: 494).
Contrary to Hausen’s image, the historical records of Åland indicate that the inhabitants of Åland and the archipelago of Åboland maintained active trade contacts with Stockholm as early as the 16th century; the fishermen took the main surplus of their catch to be sold there (Friberg 1983; see also Kerkkonen 1978 and Orrman 1991 a: 262–264). Nor can the peasants or fishermen of later days be described as isolated or wild (Villstrand 1993). The socio-economic recession of the archipelago, the emigration of the population, and the change from a central production area into the present-day economic and social periphery began only in the early 20th century, largely as a consequence of the economic crisis of coastal fishing. The adverse development has continued until recently (e.g. Vainio 1981; Eklund 1994; Andersson 1998).
The empirical props of the periphery hypothesis – the image of the archipelago as a remote and desolate wilderness – were the comparison of frequencies of artefacts in Åboland with those of the mainland coast of South-Western Finland on one hand and the absence of Iron-Age burials in the archipelago on the other hand. The severe natural conditions in the archipelago, and the insecurity caused by the political contradiction between the east and the west were regarded as responsible for these phenomena; the political contradiction led often to dangerous or unstable circumstances, particularly in the Viking Age. Another alternative was provided by Kivikoski’s notion according to which the absence of finds was only seeming: in the Viking Age and in the period of Crusades, the archipelago was inhabited by a Christian Swedish population which, due to its burial customs without artefacts, was archaeologically invisible (see chapter 2).
The comparison between find frequencies involved certain questionable traits since the field archaeological research situations in the regions to be compared were not comparable. In the south-western part of the Finnish mainland, more than 600 Iron Age sites have been excavated during the past 100 years. In particular, the parishes of Laitila, Kalanti, and the region of Turku have been well represented when the sites for excavation have been chosen. The focal point has been directed into investigations of graves and cemeteries, particularly into Iron Age cremations, inhumations and burial mounds. This obviously reflects the interests of many scholars of the first half of the 20th century, particularly those of Helmer Salmo, Alfred Hackman, A.M. Tallgren, and Ella Kivikoski (see Vuorinen 2000 b). Of course, the construction of houses and roads has directed where the rescue excavations were located, too.
In the early 20th century and long after the Second World War, Åboland was disregarded in the field archaeological activity. Despite certain commentary statements (Cleve 1941), no field archaeological investigations were launched which might have brought any further elucidation on the periphery hypothesis. The results of the comparisons between the find frequencies in the archipelago and on the mainland were thus as arbitrary as the choice of the regions to be compared.
What then was the reason for Åboland being disregarded in field archaeology? Salo is probably quite right in accentuating the sparseness of antiquarian and research resources and the cumbersome accessibility of archaeological remains in the archipelago (Salo 1990 c). Another factor must have been the unimportance of the archipelago for the construction of Finnish national prehistory as well as the socioeconomic peripheralization of the archipelago in the 20th century: the archipelago was seen as a region of prehistorically little significance which had, besides, been under the sea water for several millenniums. The rather late launch of construction projects might be a third factor. Extensive road construction projects during which archaeological remains are often unearthed, started in the archipelago as late as 1945 with the first building phase of the Skärgårdsvägen (”the Archipelagian Road”); previously the traffic communications were provided by steamships and ferryboats (Vikström 1994; Westerlund 2000). Thousands of summer cottages have been built in the archipelago since the 1960s (e.g. Andersson 1999) but they are mostly located close to the shores, and therefore these building projects have not affected archaeological remains to the same extent as on the mainland.
The second prop of the periphery hypothesis was the empirical generalization which had ensued as a consequence of the excavation projects on the coast: the cairns were an explicitly Bronze-Age form of burial rituals, which vanished gradually in Finland Proper to be substituted by other forms of graves or cemeteries in the Iron Age. The fact that the cairns are (so far) the only Metal-Age type of burial form found in the archipelago, would, according to this conception, demonstrate that no burials took place in the archipelago during the Iron Age. While several cairns were obviously from the Iron Age (see e.g. Cleve 1948), the investigators resorted willingly to regarding them as indications of alien people, as sporadic sailors’ graves which had been left there by Vikings. This was a forced conceptual construction presented for lack of anything better because nothing suggests that the Iron-Age cairns should be alien or sporadic in this milieu; on the contrary, their morphology and choice of site indicate a continuity of the cairn tradition and a knowledge of local circumstances. The threat induced by Vikings in Åboland – irrespective of its imminence – does not account for the desolation of the archipelago, unlike the coast of the mainland, more than 1000 years before the Vikings, and its remaining desolate. On the other hand, there is no evidence of any Christian mortuary ritual without artefacts, if the excavated cairns with no artefact finds cannot be regarded as such.
In the light of the present-day evidence, it is not easy to find any ecological reasons for why the natural conditions in the archipelago should have been so harsh that – as suggested by Tallgren – Stone-Age and Bronze-Age fishermen, hunters or peasants could not have made their living in the archipelago (Tallgren 1931 a; Tallgren 1931 b). One of the main advantages of the archipelago is the geodiversity of the mosaic of land and sea. Olavi Granö and Markku Roto described in the 1980s the variation of the number of islands and the length of the shoreline in squares of 25 km (5 x 5 km) on the coast of Finland (Granö et al. 1986). According to their report the number of islands was more than 100, and the shoreline was longer than 75 km for every 25 square kms in many parts of the southwestern archipelago. The total length of the shoreline in the archipelago of Åboland is almost a third of the total length of the whole shoreline in Finland (Granö & Roto 1991). Due to land uplift the islands are no more the same as in the Bronze Age and the Iron Age, but a study of earlier land uplift phases (figures 61 and 62) demonstrates that the mosaic of land and sea in the archipelago was principally as kaleidoscopic in the prehistoric period as it is today. The great topographic diversity of the archipelago infers a number of islands of varying shapes and sizes, inlets, capes, straits, and stretches of open sea with varying animal and plant habitats within a relatively small area. The most specific traits of the archipelago include small vegetational patterns, conspicuous contrasts, and numerous ecotones:
The distance from the open shoreline to the dark interior of the woods is short, yet the wood is overshadowed in turn by a bare rock outcrop supporting windblown downy birch” (Lindgren 2000: 14).
Easily crossed waterways which, at least partly, were sheltered by the archipelago connected the land and the sea with each other. Consequently,the topographic diversity of nature must have provided good possibilities for fishing, hunting, seal hunting, fowling, and foraging. The Neolithic settlement of the archipelago of Åland was obviously based on maritime resources and their seasonal availability. The most essential natural resource that the Stone-Age population of the archipelago lacked, in comparison with the mainland residents, was the rapids abounding with salmon (Núñez 1986; Núñez 1991; Núñez 1994; Núñez 1996; Väkeväinen 1982).
The many small dales filled with fine-grained sediments and the meadows developing on gently sloping shores provided also potential soil for small-scale agriculture and cattle-breeding for the inhabitants of the archipelago. The clay in the archipelago belongs to the light Litorina clay-type in southwestern Finland, and tilling this soil was possible also with light implements (Orrman 1991 a: 200–201; Orrman 1991 b). The investigations conducted by Birgitta Roeck Hansen and Aino Nissinaho in Laitila, in South-West Finland demonstrate the small size of Iron-Age cultivations (Roeck Hansen & Nissinaho 1995). It is therefore difficult to vindicate the notion presented by Nikander suggesting that the tillable land areas on the island of Kimitoön should have been insufficient to Iron-Age peasants (Nikander 1942: 30–31).
The archaeological material discussed above contains two aspects which are not easily compatible with the periphery hypothesis: on one hand the Iron-Age cemeteries and individual graves in the archipelago, and on the other hand the conservatism and the continuity which are characteristic of the morphology of the cairns, as well as of the location and the spatial references of the grave sites.
The graves on the islands of Stora Ängeskär and Långfuruholm in Dragsfjärd (060, 063) were, for a long time the only reported cairns in Åboland which had been assuredly constructed after the Bronze Age, the former in the Viking Age, and the latter in the Late Iron Age (or the Middle Ages?). Subsequently, a number of cairns have been discovered in their vicinity in the archipelago of Dragsfjärd; their small size and low-lying location resemble those of the graves at Stora Ängeskär and Långfuruholm, with which they seem to be of roughly the same age. The cemetery at Furunabb, Houtskär, consisting of 12 cairns, was dated back to the middle of the Iron Age in the investigations of the National Board of Antiquities. The graves at Sundbergen, Nagu must have been constructed in the Late Iron Age although their location on a high rock with a wide viewshed is very similar to the description presented in textbooks traditionally to the exalted sites of Bronze-Age ”stoves of beasts”. The distribution of the shore zone datings of grave sites (figure 57) indicates that 58 burial sites in Åboland are located so low that they were below sea level or in the shore zone during the Bronze Age. Consequently, 15 per cent of the graves in Åboland must logically date back to the Iron Age, and the burial sites at Sundbergen indicate that although some graves were constructed in a terrain which had emerged out of the sea in the Bronze Age, they are to be dated back to the Iron Age. The distribution of the shore zone datings did not reveal any such interruption or discontinuity which might be expected if the archipelago had remained for 15 centuries a desolate wilderness where only fishermen and hunters from the mainland coast moved very rarely during the hunting and fishing season.
Iron Age cairns are particularly abundant in the outer archipelago zone that still constituted a skerry zone for one or two millennia ago. It seems that all cairns in the so-called area of combined activities of the Southwestern Archipelago National Park date back to that period. The total area of the area of combined activities is 3000 km, the land covering 150 km of the total area. The mean density of burial cairns is thus 0.25 cairns per square kilometer (for even higher figures see Tuovinen 1985: 34–45). Judging from the statistics given by Juha-Matti Vuorinen the mean density of burial cairns in the northern part of Finland Proper – one of the best known areas in Finland due to many field archaeological projects during the 20th century – is 0.19/km. Because Vuorinen’s statistics include both Bronze Age and Iron Age cairns, the difference of densities must have been considerably greater than the difference implied by the statistics above. The greater density of Iron Age cairns in the outer archipelago zone is furthermore emphasized by the fact that the land uplift has increased considerably the total land area of the archipelago since the prehistoric period while the process has affected the mainland coast only in regions where there are seashores. The major original difference has been partly evened by the differential increase of land area. The gradient of burial cairns from the mainland coast to the outer archipelago is thus not only a stochastic time gradient, but it is probably also a gradient of density of cairns.
When tracing the characteristics of the graves in Group P and in Group R we paid attention to the morphological and spatial differences between the two groups but, simultaneously, it became evident that apart from the differences, there was an obvious continuity between them. The very appearance of the grave, a simple, externally structureless heap of stones was similar in the two groups. Although cairns have been categorized by archaeologists, and although their variation has been frequently described, the external substance of the cairn is, nevertheless, monotonously uniform, and accentuates, in its own way, the reiteration of the mortuary ritual. On an average, the graves in Group P are larger than those in Group R but it is not quite clear whether they were larger even at the time of their construction, or whether they have grown in size gradually due to later interments and subsequent addition of stones. Being older than the graves in Group R, they have had time to grow in size, too.
There are differences between the graves belonging to Group P and to Group R concerning the differences of height in the surroundings of the grave sites; the grave sites in Group P are located in a terrain with great differences of height. Great differences of elevation may naturally be considered to be results of a cultural choice but, at the same time, they are linked with land uplift. When the surface of the peneplane emerged out of the sea after the Ice Age, the highest tops of rocks turned into the first islands, and the dales and plains filled with fine-grained sediments were the last to dry up. The oldest areas of dry land are thus characterized by a sharp relief while the relief of the youngest areas is gently sloping. If graves of various ages are found in a landscape of land uplift, the oldest tend to be located in the areas with a sharp relief, and the youngest in those with a gently sloping relief. Consequently, the most significant differences characterizing the graves in Group P and in Group R have a context both to the prehistorical mortuary ritual and to the passing of time. The differences and the similarities are thus intertwined.
The similarity of the graves in Group P and in Group R is demonstrated by the fact that, on an average, there are no great differences in the ratio of the elevation of grave sites and the surrounding topography although local variation in one direction or the other is perceptible. In the Iron Age, the graves were constructed at elevated sites in the same way as in the Bronze Age. There are no conspicuous differences, either, in the location of the cairns in reference to the other cairns and cairn cemeteries in the vicinity. In the Iron Age, new graves were still constructed preferably beside old ones, as before, or at traditional sites resembling those of old graves, as demonstrated by the graves on the island of Nötö in Nagu and at Hyppeis in Houtskär. Frequently, for instance at Lillandet in Nagu and on the island of Kimitoön, the graves of the two groups are located within sight from each other, even closer. The memory of Bronze-Age graves, their significance for the constructors, seems to have been preserved until the Iron Age. Other traits, which the graves of the two groups have in common, are: the slope of the terrain at the grave site and the dissection of the landscape surrounding the grave site.
In the Iron Age, maritime sites were selected as grave sites, as it had been done in the Bronze Age. The viewsheds from these were, however, even more maritime than in the Bronze Age, contrary to what might be thought on account of land uplift. This is explained by the fact that in course of the gradual emergence of the archipelago out of the sea, cairns were constructed step by step further and further out in the archipelago with the rise of new islands and their growing size. The spatial references of the grave sites were systematic: the same regions were included in the viewsheds from several grave sites, or the viewsheds contained similar elements of landscape. The spatial references showed in certain cases great precision: the grave sites in the southwestern part of Houtskär, for instance, were selected so that the same water areas, two stretches of open sea, called the Hästö fjärden and the Svinö fjärden, were visible from several individual grave sites. According to my opinion, the systematism of spatial references derives from local knowledge of nature and from the traditional system of significances concerning natural places. Thus, the cairns and the grave sites in Åboland do not imply an interruption of the tradition but indicate continuity from the Bronze Age to the Iron Age and beyond.
Other investigations conducted in Åboland are in accord with this. The finds associated with the dwelling site and the trading place at Kyrksundet, Dragsfjärd, date mainly back to the 10th and the 11th centuries, and according to Torsten Edgren the site has at that time been inhabited all year round (Edgren 1995 a; Edgren 1995 b; Edgren 1996; Edgren 1997 b; Edgren 1999 b: 7–18). Other indicators of settlement are the dwelling sites on the island of Kimitoön from the Early Iron Age (Asplund 1997 a), and the hillfort on the island of Borgholm in Iniö, which has been possibly used in the Late Iron Age (Tuovinen et al. 1992).
The vegetation period in the archipelago begins early in spring when the snow melts and the earth warms up. Severe frosts are virtually non-existent during the vegetation period. The frostless period is the longest in the country: it lasts about six months (Solantie 1987; Solantie 1990). Even if the numerical values of the climatic parameters during the early phases of agriculture – during the subboreal and the subatlantic chronozones (Mangerud et al. 1974) – had not been the same as today, the more favourable climate in the archipelago in comparison with that on the mainland, was probably a significant factor in prehistorical agriculture which had to accommodate to extreme northern circumstances (Zvelebil & Rowley-Conwy 1986). The pollen analyses conducted by Irmeli Vuorela revealed in Åboland Cerealia occurrences which belong to the oldest in the country (Vuorela 1991 b; Vuorela & Hicks 1995; Vuorela 1999). Magnus Fries discovered indications of Bronze-Age agriculture on Åland islands as early as the early 1960s (Fries 1961) but, at that time, they were apparently without any cause considered questionable (Irmeli Vuorela, pers.comm.).
The earliest indication of primitive agriculture in the present-day archipelago has been discovered in a sample obtained from the bog of Lalaxkärret in Nagu. The radiocarbon date of the absolute Cerealia limit of the sample is 3600 90 uncal BP, calibrated (Stuiver & Reimer 1993) 2110 – (1960) – 1880 BC. In archaeological chronology the date refers to the Neolithic culture of Kiukais[2]; it is one of the oldest traces of agriculture in Finland. The pollen curve shows also possible indications of cultivation in about 2600 BP. Due to peat harvesting there is a gap of 2000 years in the history of the surface layer of the swamp, and therefore the pollen material could not be investigated continuously until the present day (Vuorela 1991 a; Vuorela 1991 b).
Vuorela has discovered almost equally old evidence of Stone-Age cultivation in the swamp of Isokärret in Kimito: the date of the absolute Cerealia limit is 3360 100 uncal BP (cal 1770 – (1660) – 1540 BC). Agriculture stabilized in about 2400 BP (empiric Cerealia limit), and to judge from its indications, it was probably slash-and-burn-cultivation. The cultivation of rye started in the Viking Age in about 900 AD, which is shown by the sharp increase in the Cerealia pollen frequencies (rational Cerealia limit). At the same time, permanent field cultivation was introduced, and the sample bears also evidence of a simultaneous change in the flora as a result of pasturage (Asplund & Vuorela 1989; Vuorela 1990; Vuorela 1991 a).
At Mossdalen, Kimito, the earliest indications of woodland clearings in the pollen samples from the Bronze Age, ca 3000 BP (Vuorela 1991 b), suggest slash-and-burn-cultivation. Also the absolute Cerealia limit (2530 110 uncal BP) dates back to the Bronze Age, approximately 780 BC (Asplund & Vuorela 1989).
The curves of pollen samples collected from the swamp of Mossen in Korpo reveal indications of woodland clearing more than 2000 years ago. The age of the absolute Cerealia limit is 1990 90 BP, calibrated 110 BC – (4) – 100 AD. The Iron-Age cultivation in the surroundings of the swamp has apparently not been continuous, even if the pollen material suggests two more intense periods of cultivation during the Iron Age. In about 600 BP, i.e. about 1400 AD, effective agricultural activity was started in the surroundings of the bog, the deciduous trees were cut down, and the terrain was cleared for cultivation (Vuorela 1991 a; Vuorela 1991 b).
Thus, the pollen stratigraphy demonstrates that the sporadic phase of agriculture which is represented by slash-and-burn-cultivation from the Kiukais culture to the Pre-Roman Iron Age, is followed by increased area of cultivation and more intensive human impact on the flora in the Early Iron Age. Eventually the cultivation of rye gains ground in the Late Iron Age and the Middle Ages. The final transition to field cultivation takes place at the same time. The complete picture is, however, still rather inaccurate due to the small number of investigations. The inaccuracy is caused, among other things, by the sparse pollen production of the oldest crops, barley and wheat (Vuorela 1991 b: 130–131).
The mother tongue of the majority of the present-day population of Åboland is Swedish. The population descends from Central Sweden, and they carried out the colonization of the archipelago which according to most researchers took place in the 12th century. The place name material of the region consists of many ingredients which represent contacts between Swedish and Finnish. The recent investigations concerning place names in Åboland include an extensive mapping work and interpretations of place names, they determine the various name strata and their reciprocal age relationships. This allows a comparison of onomastic and archaeological results with each other although the temporally remote linguistic and material cultural ingredients are not entirely compatible.
The oldest stratum of the name material in Åboland is represented by the names of Finnish origin which have then been borrowed into Swedish, and simultaneously adopted a phonetic structure to suit the Swedish language. Approximately 1000 place names of the total of 40000 in the region are originally Finnish. Ritva Liisa Pitkänen divides the names into three age strata (Pitkänen 1990):
Half of the loan-names originate from the time of colonization and the preceding centuries. These names include e.g. Kvivlax < *Kuivalahti, ’dry cove’.
A number of names originate from the beginning of the first millennium (e.g. Lemlax < *Lemmenlaksi ’Lempi’s cove’ (called after a person named Lempi).
The oldest names were given during the last millennium BC (e.g. Rosklax < *Ruskonlaksi).
The Finnish names are often associated with islands, coves or sea inlets, capes, or other natural places. When they were borrowed into Swedish, they often became names of settlement, for instance Hyppeis (< *Hypöinen) and Mossala (< *Musta-salo). The Finnish name material includes, apart from names of natural places, also names associated with fishing, cattle-raising, agriculture, villages, houses, rear buildings, smithies, mills, harbours, borders, means of communication, persons, as well as pyhä- and hiisi-names (Pitkänen 1985). During the Iron Age, hiisi-names originally denoted places marked off within inhabited areas. They were associated with positive religious meanings, but after the coming of Christianity they changed to designate beasts (Anttonen 1996: 116–123). Part of the names may descend from the Estonian language. The borrowing of names has, according to Pitkänen, required a permanent and organized co-operation, common interests, and agreements between populations with different mother tongues. An additional prerequisite must have been a bilinguality of a certain degree. Hostile contacts could not have resulted in Finnish names being borrowed into Swedish, the same applies to contacts between settlers from Sweden and Finnish seasonal fishermen (Pitkänen 1985 a: 352–353; Pitkänen 1985 b; Pitkänen 1992). On the other hand, the use of the borrowed names must have been continuous, otherwise they would not have been preserved to the present-day investigators of names (Pitkänen 1990; Zilliacus 1997). The Finnish material of loan-names is thus not in accordance with the notion that the archipelago of Åboland would have been only a usufructuary of travelling fishermen and slash-and-burn cultivators of the Iron Age.
Kurt Zilliacus derives the parish name Houtskär from the Finnish name of an island *Hauta-salo in which the word hauta refers to old graves, maybe prehistoric cairns with significance to those who introduced the place name; the word salo refers to a large (wooded) island. According to him, the name reflects the respect shown toward the graves by the bygone generations (Zilliacus 1997). This interpretation opens the opportunity to try to understand the relationship between those who had given the name to the place and the graves, many of which could be very old when the name was given. It means that the graves still involved living significances and interpretations when the parish name Houtskär was introduced. Which graves, then, among those which have been preserved until our day, could have been objects of such respect? In the first place probably those which had been constructed at elevated and prominent sites with extensive viewsheds. Such graves are characterized by an elevated site but also by their relatively high position in reference to their adjacent surroundings. Kummelbergen at Björkö, Kummelbergen at Kittuis, Trollberg and Västerskogen at Medelby, Ekholm and Svinö Svälteskär at Hyppeis, and Resberg at Nåtö belong to these graves. The both registered grave sites called Kummelbergen in Houtskär belong thus to this group. Even this aspect, for its part, suggests that ancient graves were noted when giving names to various places.
The Swedish place names do not include name types deriving from the Primitive Germanic language spoken in Scandinavia in the prehistoric period. Therefore the old theory of onomastics, according to which a population speaking a Primitive Germanic language would have lived on the coast of Finland before the Christian era, has been abandoned (Naert 1995: 15–18; Häkkinen 1996: 169–170; Zilliacus 1997). According to the prevailing opinion, the Swedish place names date back to the 12th century, at the earliest.
Also the Swedish material includes place names of various ages. The oldest village names are of the type -by, -böle, and -boda in Kimito, and -by in Korpo (Zilliacus 1994: 13–24; 41–46). In his recently published research work, Bertel Fortelius has, however, come to the conclusion that the village names Aspö, Björkö, Kälö, and Brånskär (Brunskär) in the outer archipelago of Korpo, which originally were natural names, are likely to date back to the Iron Age (Fortelius 1999: 437–455).
The notion that Åboland had been an uninhabited desolate wilderness for 1500 years compelled its advocates to assume specific circumstances and events, which have not been verified empirically. The periphery hypothesis has relied on implicit conceptions, which may have been formulated on somebody’s writing desk, of the isolated and harsh natural conditions in the archipelago. This has resulted in indiscriminate interpretations, and the notion retained its vitality because no field investigations which would have thrown light on the hypothesis were started for decades. The problem has remained: was there a gap in the settlement or is there a gap in the field archaeological studies and evidence of the settlement? The morphological conservativity of the cairns, the relative locations of the graves, the local knowledge revealed by the spatial references of the grave sites, and the datings of the graves indicate, however, a continuity from the Bronze Age to the Iron Age and beyond. The results of other field archaeological investigations, of pollen analyses, and of the onomastic studies are in accordance with this. Of course, these arguments are not adequate to create a consistent and complete image of the Iron-Age settlement in Åboland, its character and sources of livelihood, but they do give sufficient evidence to justify the abandonment of the hypothesis concerning the peripheral nature of the archipelago.
| [1] | I will not discuss here rare phenomena which are characterized by the fact that their very discovery and detection require a fairly extensive number of samples, and thus a particularly strenuous archaeological work, especially when the phenomenon is concentrated in regional clusters (Nance 1983: 312–318), which is not unusual. These phenomena are represented by rarities which may occur, say, once in a century. |
| [2] | At the Late Neolithic dwelling site of Kotirinne (Niuskala), Turku, a cereal grain was found in the 1980s (Vuorela & Lempiäinen 1988). Today the site is located on the mainland (thus outside the research area of the present study), but during the Kiukais culture it was surrounded by an inner archipelago landscape. |