I 85

The Cultural Construction ofChildhood in Scandinavia,3500 BC 1350 ADStig Welinder

A set of Swedish and N orwegian burial-grounds and churchyards from theNeolithic to the Early Middle Ages (3500 BC 1350 AD) is analysed asconcerns children's graves. Patterns of burial rituals corresponding tovarious concepts of childhood are constructed. Childhood is looked uponas a cultural construction independent of time and space. The basic growthprocess from infancy to adulthood is fundamental to the concept ofchildhood in all societies, but its transformation into burial ritual, materialculture, symbols, and ideology is varied.

Stig Welinder, Depctrtmet ofHumanities, Mid Sweden Utziversity,SE- 831 25stersund, Sweden.

All humans have been children. In manyplaces around the world today a third or moreof the living human population consists ofchildren. During prehistoric time that wasprobably the case everywhere.

The Swedish census from the 1750s is oneof the world's oldest, fairly accurate, popula-tion statistics (Hofsten 1986). During thattime 105 boys were bom for every 100 girls.Out of 1000 newborn children, 194 girls and214 boys died, while still in early infancy.Nevertheless, of the living population 33%was younger than 15 years. The basis of thiswas a high birth rate. The average number ofbirths per female was 4.8 children.

The picture may have been about the sameduring prehistoric time, although marriagepatterns are known to have changed during theMiddle Ages (Benedictow 1993),and little isknown before that. Remarkably little is knownabout all these prehistoric children.

THE HISTORICAL DEBATEDu ring the 1960s and 1970sPhilip Aries' viewthat parents did not show affection and love

for their children and did not take much care ofthem before the age of about seven years priorto the 16th and 17th centuries, provokeddebate. The reasons, according to Philip Aries,were the high number of children and the highinfant death rate (Aries 1960).Emotional tiesto children did not pay until the children hadpassed the first risky years of childhood.Actually, childhood did not exist in the modernsense. Already at the age of about seven thechildren behaved and took part in variouswork tasks like adults.

Philip Aries was accused of having genera-lised from too small a data base as concernstime and space and social class. French 17thcentury oil paintings of miniature personsdressed like adults do not present the contem-poraneous view of children. They conceptual-ise adults in becoming. The absence of anexplicit concept of childhood did not meanthat children were badly treated and regardedwithout affection (e.g. Pollock 1983).

In the pre-industrial Scandinavian society(Hagberg 1949; Wikman 1949) as well as inmost societies around the world (van Gennep

Current Swedish Archaeologn Voh 6, 1998

186 Stig Welinder

1960) individuals turned adult in a number ofsteps from birth and onwards. Now and thenthe steps were marked by rites of passage orfestivities. These steps formed the process ofbringing up the children, giving them know-ledge, work tasks and responsibility. Thisobservation will be of importance further onin the article.

The modern Western view of childhoodwas created towards the end of the 19thcentury, when the upper and middle classesdecided they could afford not to benefit fromtheir children's labour. A period of childhoodand adolescence defined by upbringing, edu-cation and culture became a norm, whichduring the 20th century spread to all classes(Bjurman 1981; Borgnakke 1981; Ohlander1993).People that grew up during the 1930sand 1940s tell about the importance of gettinga job and getting married. People that grew upduring the 1960s and 1970s tell about popmusic, fashion, films for adults only, and theimportance of getting a driver's license (Gillis1974; Olsson 1981; Andersson 1985; Hen-riksson 1991).

Modern research stresses the variation inthe views of children in various societiesthroughout history. Certainly also the liveslived by children were different within oneand the same society. In Sweden, for example,during the last decades death rates have beenhigher among boys, children in single-parentfami lies, children in urban areas, and childrenwith parents bom abroad (sterberg 1996).

THE ACHAEOLOGICAL DEBATEDuring the 1980s Scandinavian archaeologystressed the conspicuous lack of archaeologi-cal research on children and a conspicuouslack of data on prehistoric children. Typicalheadlines are: "Where have all the childrengone?" (Linderoth 1990), "The Stone Agechildren can we see them?" (Sandberg 1993),and "Did the Viking Berserk have any chil-dren?" (Linder 1995).

The basis of these views was, of course,data sets like the Early Medieval churchyardof Vsterhus (Tab. 1, population no. 26) with

51% children younger than 7 years and 57%younger than 14 years (Fig. I ).At most prehis-toric burial-grounds the percentages were con-siderably lower. Demographic research sug-gested there should be typically 3060%children (Welinder 1979; Sellevold 1989).They were simply not there.

Today the above statements do not hold

Fig. 1. The survival pattern of population no. 26Viisterhus (for a tnore sophistieated version, seeSi ven 1991 a, b).

true. Ahost of articles on children has emerged,most of them seminar papers and popularreading, which will not be referred to here (cf.Johnsen X Welinder 1995), all of them pub-lished in the Scandinavian languages. Thearticles by Grete Lillehammer, already classicin their field, opened the sluice-gate (1986,1989). She summarised almost all the con-cepts to be used in the following pages. Note-worthy is the distinction between the calendarage of a child, possibly to be defind from bonedata, i.e. the child's biological age, and thesocial or cultural age of the child, hopefullyreconstructable from burial rituals. The paral-lel to the sex-gender distiction is obvious, andno doubt the archaeological research on child-hood during the 1990s has had gender archae-ology as its prerequisite.

The deficit in children's graves may beapproached from at least three different an-gles:

(I) What happened between the points ofCurrent Stvedish Arhaeotogs:, Voh 6, 1998

The Cttltttral Construction of Cftildhood 187

time of death and burial? Were the children'sbodies disposed of in ways not recorded dur-ing routine burial-ground archaeology?

(2) What happened between the points oftime of burial and excavation? Did the smalland fragile children's skeletons, whether cre-mated or not, totally disappear due to biolog-ical and soil-chemical processes?

(3) What takes place during excavationand laboratory analysis? Are archaeologistsuninterested in children? Do archaeologistsexcavate and analyse data with blinkers on?

In addition to these questions, it is well worthreconsidering the question of whether child-ren's graves actually are too few. Our data setsuggests the opposite.

THE DATA SETOur data set consists of 26 different popula-tions of graves corresponding to one feature,one burial-ground, several burial-grounds, allthe burials in a province, or all burials inSweden from a defined period (Fig. 2, Tab. 1).We believe this to be a fair sample of burial-data from the Swedish Neolithic, Bronze Age,and Iron Age. A few Early Medieval popula-tions are included (population nos. 23, 26 and,dependent on definition, nos. 2425). Thelatter do not form a statistically, or in any otherrespect, representative sample.

From the Neolithic (population nos. I 2)few other children's graves are known andpublished in detail. From the Bronze Age(population no. 4) close to all chi ldren's gravesfrom Periods IIIVI are included in the popu-lations. At least that is the intention. Popula-tion no. 5 is a subset of population no. 4.

From the Iron Age the sample is somewhathaphazardly constructed. Some typical andsome spectacular burial-grounds from vari-ous parts of Sweden are included, in additionto one Norwegian burial-ground (populationno. 13).The intention is to show various kindsof burial rituals as concerns children's graves,and thus perhaps various views of childhood.Another intention is to contradict the abovetaphonomic points. There is no notable lack of

Fig. 2. The data set (cf. Tab. I).

children's graves, when all kinds ofburials areincluded in the analysis.

STARTING POINTSThe most obvious quality of the populations inthe data set is the inherent variation. Thedifferences in frequency and appearanceamong the various burial-grounds are note-worthy, even among those situated close toone another in time and space.

Nevertheless, the populations also havesome characteristics in common. Amongothers, objects unique to children's graves arerare. To our knowledge they are non-existent

Current Steedish Archateotogs; Voh 6, J998

188 Stig Welinder

before the Viking Age. During that periodsmall bronze jingle-bells were associated onlywith children and youths. They may have beentoys or been attached to the dress. The latter isknown for small children during the MiddleAges. Mirrors are also most often found inchildren's graves (Grslund 1973). Objectsespecially designed for small persons are rare,although a few small dress-ornaments andweapons have been recorded together withinfant skeletons at the Birka burial-ground(population no. 22). However, it is commonthat no or few grave-objects at all are associ-ated with the buried children (e.g. Tabs. 2, 6).

On the other hand, many children seem tohave been buried in special children's dress.When detectable from ornaments and dress-details in the graves, this may have been dressworn at ceremonies and festivities rather thaneveryday clothing, perhaps dress correspond-ing to traditional peasant costumes. Many ofthe children in the Early Iron Age graves onthe island of land (population nos. 8, 14) hadnotably more beads attached to their dress orused as ornaments than most of the adults.Elsewise, it was common that dress-orna-ments and dress-fittings were introduced intothe dress at an age of about 515 years (Tabs.58).

The overall impression is that the childrenhave not been treated in ways very differentfrom that of the adults. The children's gravesare there, and they are fairly numerous. Thechildren were not obviously neglected at in-terment, and we do not see any obvious reasonto conclude that the children were neglectedduring their too often too short life-spans. Thisview cannot be argued in detail, but we wantto quote the Indian poet Kalidasa (c. 380450AD), a contemporary to many of the childrenin our data set:

With thei r teeth half-shown in courselesslaughterand their efforts at talking so sweetly uncerlai n,when children ask to sit on his lapa man is blessed,even by the di rt of thei r bodies.

Cornelius Tacitus had little to write aboutchildren, although he too stressed that they"grew up naked and dirty" (Germania 20:1).

It is a paradox that childhood in prehi storyhas to be studied through populations of deadchildren. Thus, what we observe is not child-hood but burial rituals designed for children.The two are not synonymous, but we see noother reasonable method of study for themoment. Accordingly, the patterns we willreveal in the following pages are not patternsof various kinds of childhood, but patternsconstructed from various kinds of burialrituals. These patterns may correspond todifferent views of childhood. The minimumhope is that they correspond to various viewsof child-adult relations, which perhaps werenot relations within the existing householdsand communities, but relations conceptual-ised in the burial-grounds, that is abstract andidealized relations to be openly displayed.

THE SELECTION OF CHILDRENTO BE BURIEDOf the Neolithic populations 1520% arechildren's graves (Tab. 1).Of the Bronze Agepopulations, children's graves amount toabout 2030%.In the Iron Age burial-groundsof the kind commonly believed to have beenused by the inhabitants of one or a few farms,the variation is 1 62% children's graves. Ob-viously not all deceased children have beenarchaeologically recorded.

Especially the Iron Age burial-grounds inthe Stockholm area have few children'sgraves. Less than 10%, often about 56%,seems to be the common case (Tab. 3).Assum-ing a child mortality of about 3060%, ques-tions about missing children are justified.There are no data to explain this deficit in theStockholm area.

The corresponding burial-grounds onland contain about 1025% children'sgraves (Tab. 4), but in addition there arefeatures with lots of infant skeletons. Bjrby(population no. 14) is a burial-ground, orpossibly part of a much larger burial-ground,with 79% children's graves. At the Bredstra

Current Swedi sh Archaeotogy, Vol. 6, l998

The Cultural Construction of Childltood 189

34 burial-ground (population no. 8)children's graves compose c. 50%, but abouthalf of the children had been buried in one andthe same cairn. Another stone-setting, in theLnglt burial-ground (population no. 18),contained fourteen infant skeletons.

The same kind of mass grave for smallchildren is known from the island of Gotlandand from West Sweden (population nos. 3, 910). Here tens of infant skeletons have beenfound in Bronze Age cairns. At least some ofthese children, most of them very small, werenot buried during prehistoric time. Radiocar-bon dating suggests that the cairns were usedduring the Middle Ages as an alternative to theChristian churchyard (Lindqvist 1981).

A feature at the Vallhagar settlement site isalso worth mentioning. In the village's burial-ground there is only one child buried amongthe 140 adults. On the other hand, in a moundin the middle of the village there were twonewborn children (Tab. 11)buried in the earthabove a stonecist that contained an old man.

The examples demonstrate that thechildren were classified before interment fordifferent methods ofburial. Some were buriedlike the adults, some were buried in massgraves in features similar to those used foradults and occasionally together with a fewadults, and some were buried in ways still notdetected today. To find the norms used in thisclassification had better be a future task. Atpresent we will only implicitly hint at somepossibilities, when we discuss the followingmodels of childhood.

STEPWISE CHILDHOODWithin the Middle Neolithic Pitted Ware Cul-ture of the island of Gotland (population no. 1)the children's graves can be classified intofour groups according to age of the buriedchildren (Tab. 5). The groups form discretesteps in a sequence of stages from birth toadul thood.

The newborn children were not buried atall in the same burial-ground close to thesettlement site as the rest of the members ofthe settlement site group. From about the age

of 18 girls and boys were buried in equalnumbers in the same way as adults, althoughthe indigenous correspondence to adulthoodmay have been entered at the age of about 1213, at least for the boys.

The age-related steps in the burial-ritualpresumably correspond to steps in the grow-ing-up process of the children. They turnedadult in a series of steps. In life the childrenmay have been subdivided into age-groupswithin the community of the settlement site.They had no calendar age but belonged to astage along a defined series of more or lessage-related stages.

The same kind of stepwise childhood maybe inferred from the stepwise age-relatedburial-rituals reconstructed also for a numberof other populations in our data set from allprehistoric periods, from the Neolithic (Tab.12), the Late Bronze Age (Tab. 8), the EarlyIron Age (Tab. 6), and the Saami Late lron Ageand Early Middle Ages (Tab. 7).

An interpretable stepwise childhood isdisplayed in the Early Medieval Vsterhuschurchyard (population no. 26). The steps arediscernible from the position of the graves inrelation to the church-building and the inher-ent gender segregation within the churchyard.In addition, the smallest children seldom gotan individual grave (Tab. 14). Before confir-mation, when still religiously ignorant and notsubject to full punishment by law, the childrengot no individual graves. They were related totheir female caretakers, in death as in life.After confirmation they were less dependenton their mothers, but still subordinate to theirfathers within the household. Accordinglythey were buried at the male side of the church.After becoming adults and assuming respon-sibility for themselves at the age of about 15,they were buried according to their gender(Tab. 15).

It is an open question whether or not thenorm stating that children could not be buriedinside the walls of the churchyard before bap-tism, was upheld or not. It has been suggestedthat the mass graves were the graves of theseunfortunate children. The seemingly high

Cttt rent Siuedish ttrchaeologv, Vnh 6, 1998

190 S&ig Welirler

number of children, all younger than 4 years,in the Fjlkinge burial-ground (population no,24) from the decades around 1000 AD havebeen suggested partly to be the unbaptisedchildren belonging to a community of a near-by, although unknown, Christian churchyard(Helgesson 1992b). The find of an infantskeleton beneath the floor of a dwelling-housein the Early Medieval town of Lund is remark-able in the context (Roslund 1990).

The process of becoming an adult withinthe Early Medieval European Catholic societyis merely a special case of the process ofundergoing a series of rites of passage on theway from childhood to adulthood (van Gennep1960).The steps may more or less correspondto the children's process of maturity as con-cerns motory, intellectual, and sexual ability.The latter quality stresses the process of grow-ing into a gender. This will be di scussed furtheron.

We presume that this kind of stepwiseprocess of growing up is what we see in theprehistoric burial-grounds referred to in ourabove examples.

CH ILDHOOD AS PART OF A LINEAR LIFEThe small number of children buried in theSaami Vivallen burial-ground (population no.23) around the tum of the last millennium gotnew kinds of dress-ornaments and dress-fittings at certain ages (Tab. 7). A stepwisechildhood may be inferred, where the first stepat about the age of 5 meant being recognisedas a person to be buried in the communalburial-ground.

The iron arrowheads recorded in twochildren's graves are reminiscent of thearrowheads, which were hung above the boys'cradles during historical time to make theboys skillful hunters. In the same way, thegirls got the wings, feet, and beak of a ptarmi-gan to make them cleanly, well-behaved andgood-natured (Drake 1979).

From the small number of graves at theVivallen burial-ground it may be suggestedthat new steps were entered in the late teensand at a mature age (Tab. 7).The stepwise life

continued after childhood along a linearcourse. Status and, possibly, gender variationwere embedded in the burial ritual, but thisalso seems to have been part of the maturingprocess. Life ran from birth, when smallchildren were not fully recognised as the samekinds of persons as elder children and adults,to mature age and explicitly recogniced pres-tige.

Life had a linear course from birth todeath. Childhood was the start of a linear life-span. This is the most common view amongWesterners of today.

CHILDHOOD AS PART OF A LIFE-CYCLEAmong non-Westerners it is a commonconception that time forms a cycle, both thetime of the World and the time of the humanlife, that is the human life is seen as a l ife-cyclewith its start and end in the same point (Vins-rygg 1988).The metaphor of cyclical time isthe annually recurring seasons or the ruralannual cycle.

The Late Bronze Age childhood was per-ceived as a series of stages according to astepwise childhood. In our reconstruction ofthe age-related burial ritual there are as manyas five steps (Tab. 8).At about the age of 15 thegrave-ritual was equal to that of the adults.This age may have defined the transition toadulthood. At about this age the burial ritualbecame gender-specific. This will be dis-cussed more thoroughly further on.

In the full Late Bronze Age population(no. 4) the number of grave-objects increasedtowards a maximum among the deceasedpersons in their late teens (Tab. 8). After thatwe have no full statistics available. The sub-population from Halland (no. 5) suggests thatthe number of objects in the male gravescontinued to increase along the men's life-span (Tab. 9), while the number of objects inthe female graves decreased notably. Asconcerns the male graves, weapons decreasedin number from their maximum occurrence inthe graves of men that died at an age of about1820 years.

The scanty data suggest that the female

C&.re& S&ve&iisi& Arcl&oeology, Vi&l. 6, l998

The Cwhrirol Cnns(union of Chilclhooil l91

burial ritual was linked to the female life-cycle mirroring pre-puberty, child-bearingage, and post-menopause age. Such a cyclicview of the female life in relation to femalefertility is nicely demonstrated in the femaleburial ritual of the Early Iron Age burial-ground (population no. 13) of Ula, wheresickles were buried together with women ofchild-bearing age, while young girls weregiven no grave-objects and old women wereburied in display-dress (Tab. 10). The last-mentioned fact, however, hints at stressingstatus at mature age towards the end of a l i nearlife. The male burial rituals of the two popula-tions suggest no similar explicit conceptuali-sation of a cyclic view of life, although theoccurrence of weapons in the Late BronzeAge graves suggests a climax in the middle ofthe men's life-span.

The fairly common view of life as a cyclemeans that childhood is the start of a run,which comes to an end at its startingpoint.Children are not newborn, they are reborn,having much or something in common withold people near the end of the cycle, or rather,they have something in common with thoserecently deceased, who are presumed to havebeen reborn as newborn children.

Westerners observe looks and idiosyncra-sies in small children in an attempt to recog-nise parents and grandparents. We know thatour findings are due to qualities transmittedbetween generations by DNA. Others knowthat lost generations are reborn in the newly-arrieved children. They may give the childrenthe names of dead ancestors, as do Swedes oftoday.

CHILDHOOD AS DISPLAY INTHE SOCIAL ARENAMany children were buried, but in variousways. Some children, in some societies themajority, were buried in ways hard to detecttoday. The variation is remarkable betweensocieties close in time and space as well aswithin one and the same society. One reasonfor this variation is that burial rituals wereformed to stage occasions of ceremonies and

display of status and power. This is valid alsofor the children's graves.

ECONOMIC DIFFERENTIATIONThe low percentage of children's graves in theStockholm area is notable. Especially theburial-grounds at Helg (population no. 21)are poor in children. There are just twochildren among 86 adults.

The character of the Helg settlement siteis disputed. It may have been one or a fewfarms, it may have been the living quarters ofpeople associated with a manor. In any case,the site is characterised by its wealth ofhandi-craft refuse, most conspicuously from metal-working, for example bronze-casting. Theidea is that in a community specialising inhandicraft, the children did not acquire theskill and status to be recogniced as persons tobe buried in the common burial-ground untilabout 15 years of age, while at least some didso at an earlier age in communities specialisingin farming.

The above is not valid for the farmingcommunity at Vallhagar (population no. 12).At its burial-ground there is only one childamong 140 adults (Tab. 11). At the nearbyburial-ground of Styrmansberg (populationno. 15), on the other hand, there are sevenchildren among 21 adults. Six of the childrenhad been put in the same grave as an adult.

The Styrmansberg site is a hill-fort. Someof the graves were inserted into the ruins of thestonewall of the hill-fort, others were moundssuperimposing a house-foundation and acattle-pen. Superficially the graves give a slov-enly impression. They do not resemble thegraves at the burial-grounds associated withthe Gotlandic farms, such as the Vallhagarfarm and burial-grounds. The landscapearound the hill-fort is not good arable. It is abrowsing and grazing area. This is stressed bythe regular occurrence of bones of cattle,sheep, lambs, pigs, and dogs in the graves.Animal bones are rare in the Vallhagar burial-grounds.

In our view the population at the Styr-mansberg burial-ground had lived and been

Cwvinv Sivedish A rchaeology, Vuh 6, l9W

192 Stig We(inder

buried as cattle-herders outside the farmingvillage. An economic system with severalgroups of people specialising in various sub-sistence tasks within a redistributi ve organisa-tion comes to mind (Odner 1972). Amongcattle-herders, children are useful at an earlyage in tending the animals.

In various kinds of societies with variouseconomic specialisations the children matureinto taking part in various tasks at a differentpace. This may be one reason for childrenbeing buried together with the adults in thecommon burial-ground in some societies, inothers not. The exceptions from this rule aremany, one being the Gotlandic Middle Neo-lithic burial-grounds (population no. I ), wherechildren were buried from about the age of 23 in the same place as the adults, while theywere not regularily buried together with toolsuntil about the age of 1213 (Tab. 5).

INBORN STATUSA classic debate concerns the idea that theexistence of prehistoric hierarchical societiescan be proved by the occurrence of children'sgraves with display grave-objects identical orsimilar to those in the graves of adults. Theidea is that status was inborn and, thus,ascribed to children at birth. It is a reasonableidea that some of the variation among chil-dren's burial rituals can be explained by statusvariation among the children.

The Bjurhovda burial-ground (populationno. 25) was used by the inhabitants of awealthy farm for a few decades after 1000AD.Both the deceased females and males form aranked society (Tab. 13).A small number ofwomen were cremated in dress with display-ornaments. Some of them wore the symbols ofmarried housewives. Another small numberof women wore dress with less striking orna-ments. Most women had no dress-ornamentsat all. The men were ranked as warriors. Asmall number of men were buried as mountedwarriors with a lance and a charger. Most menwere buried in plain dress with no dress-fittings (for the full argument, see Welinder1990).

There are children buried with the symbolsmarking the positions all along the scale ofrank. The small boys were not mounted war-riors, but one infant boy was buried togetherwith a spear and an axe that he was hardly ableto lift. In their early teens the boys got theirown charger and lance. One small girl wasburied with a full display-set of ornaments, infact one of the wealthiest in the burial-ground,while another small girl was buried togetherwith one of the high-ranking housewives.Another small child, the gender of whichcannot be suggested, was burried with nograve-objects at all.

At the upper hal fof the scale there are 42%children, at the lower half there are merely 5%(Tab. 13).It is close at hand to suggest that thetotal deficit in children's graves presented bythe overall percentage, 19% (Tab. I ), consistsof the children of the people of the farm thatwere not the mounted warriors and their wives.Some children were bom to acquire these orsimilar positions, some were not. The formerwere buried in the burial-ground of the farm,the latter were not.

The Vsterhus churchyard came into useat about the same time as the Bjurhovda burial-ground was deserted. Being a Christianchurchyard all children, at least the baptisedones, were buried inside the sacred area, butsome children were more equal than otherchildren. They were buried in cists with metalfittings, some of them in prestigious positionsclose to the church-building. Most of themwere boys, or at least buried at the southern,male side of the church.

The burial ritual of the Neolithic BattleAxe Culture was an exclusive tradition (Tab.12).From about the age of 34 some children,persumably both girls and boys, were buriedaccording to this ritual. Many of them wereburied together with adults, a few were buriedin graves of their own together with the samekinds of grave-objects as the adults. At the ageof about 12 years some children, or perhapsyoung adults, were asigned prestigeous ob-jects like copperornaments and battleaxes.The latter is valid for a boy.

Carrent Sieerlish Arehaeology, Voh 6, 1998

The Culturol Constrrtction of Childhood 193

The children of the Battle Axe Culturewere ranked in the same way as the adults.This is quite another view of children than thatpresented by the contemporaneous Gotlandicburial ritual discussed above.

In most societies adults are categorisedaccording to wealth, status, rank, and otherpositions. The above examples suggest thatthe same was valid also for children in someprehistoric societies. The simplest idea is thatthey were categorised in accordance with theirparents. How children were attributed toparents is, however, quite another problem,with the exception perhaps of the biologicalmother.

TO GROW INTO A GENDERThe osteological methods for sexing the skel-etons of children that have not passed pubertyare still uncertain. As concerns cremated bonesthere are, for all practical reasons, no methodsat all. Thus, the children's gender has to bedefined from the grave-objects in the chil-dren's graves without the, in fact, problematictest offered by sexing the skeletons. We have

seriated the children's graves in accordancewith Gebiihr 1992.

At the burial-grounds of the Viking Agetown ofBirka (population no. 22) 84 children'sgraves have been defined by the length of thegraves. This not very accurate method is theonly one available for the moment. It is worthtrying because of the wealth of objects inmany of the Birka graves and the presumablyrigid gender patterns in the Viking Age burialritual, for example as demonstrated by theBjurhovda burial-ground above.

A group of 18 girls' graves is easily recog-nisable by the sets of dress-ornaments andbeads. In addition there are keys, sickles,sewing-sets, and mirrors (Fig. 3). Anothergroup of 18 graves contains weights andweight-like objects, penannular brooches, andarrowheads, but never any of the above girls'objects. Presumably these graves are boys'graves. Together these two groups amount to43 % of the children's graves. The majority ofthe children's graves contains objects that arenot gender-indicative, for example knives,pots, slag, and buttons.

~~11 I-I I3~~

+ 0n+

Fig. 3. Seriation of the cltildren's graves at the Birka burial-grounds(population no. 22).

Current Su edish A rcluteotogy, Vol. 6, l 998

194 Stig Welinder

A group of boys seems to have been bomto become merchants, while this position wasonly achieved by adult women. This may be agender-characteristic of the early urban societyof Birka.

We have applied the same seriation methodalso to the full population of Late Bronze Agechildren's graves (population no. 4) with itswell-defined age-related burial ritual indicat-ing a stepwise childhood (Tab. 8). However,the number of graves with a sufficient numberof different types of objects is very small. Allin all there are 22 graves and 11 types ofobjects, including four species of domesticanimals, to be seriated. Five of these gravesare osteologically sexed graves of juvenilepersons.

Two groups of mutually exclusive types ofobjects were formed by the seriation process.When adding graves with only one group-characteristic object, an interpretation depend-ent on gender and age may be suggested (Tab16). Some girls grew to become associatedwith the herds and the harvest at puberty,while some girls and most boys did not. Theidea is in agreement with the above discussionof the Bronze Age female childhood as part ofa cyclic life from birth to death to rebirth as ametaphor of the annual vegetational cycle.

Some boys of the Middle Neolithic BattleAxe Culture grew to become worthy of carry-ing a battle axe, others did not (Tab. 12).Within the contemporaneous Gotlandic hunt-ing-gathering tribes, some boys were disting-uished as seal-hunters when buried after theage of about 1213 (Tab. 5), others were not.It looks as if some young men were honouredat death as being more men than others. Therewas a real man's world to enter for the cunningand brave among the boys. Some failed toreach it.

What kind of boys were buried in dresswith bead-applications, elsewise found onlyon the dress of girls after puberty and adultwomen, never among the adult men (Anders-son et al. 1995)? In the old Scandinavianlanguages men that adopted the female sexualrole in homosexual relations, or were generally

weak and cowardly, were denoted "argr" or"argr madr". In Medieval Latin they were"effeminate". This may be the background fora few men that were buried at the female sideof the Vsterhus church.

Children did not simply grow into adultsas time went on. They had to learn how tobehave as an adult according to the traditionalnorms of their society, and they had to learn toconform to accepted female and male norms,that is they had to adapt to a gender.

CHILDREN'S GRAVES ANDCHILDREN ALIVEAll the children that figure in this article aredead and buried children. Today there simplyare no concepts, methods, and data to studyliving prehistoric children. There certainlywill be in the near future. The study of child-hood in prehistory is an increasing field ofinterest among young archaeologists. Certainlysome of them are conducting ethnoarchaeo-logical studies of li ving children just now, andmany of them are rethinking their own child-hood in an attempt to remember what it waslike to be a child, what they did, and what theydid that left tracks possible to distinguish fromthose of adults.

Hopefully all these archaeologists willcome to the same conclusion as we have asconcerns the dead and buried children. Child-hood is a cultural construct. There are patternsamong the amazing variation of burials andburial-grounds. Probably the patterns dis-cerned by us soon will be regarded as oversim-plified generalisations, but we think that theview that there are patterns will hold. Varioussocieties saw children in different ways, andwe see the different ways as different burialrituals.

We do not think that there are patterns ofevolutional change from the Neolithic to theEarly Middle Ages. There are likewise nosimple correlations between economy or socialstructure on the one hand, and the conceptuali-sation of childhood on the other. But there is abasic pattern ofchildren being dropped, totallyhelpless, by caring mothers and slowly

Current Swedish Arehaeology, Voh 6, /998

The Cultural Construction of Childhood 195

becoming more ski l led at taking care of them-selves and performing various tasks, finally tobecome those that take care of their ownnewborn children. This simple, perhaps self-evident, pattern is seen more or less in all ourpopulations from the Neolithic to the EarlyMiddle Ages. It was, however, turned intoconcepts, symbols, burial rituals, and materialculture in various ways in various societies,sometimes stressing the very process of grad-ually becoming an adult, sometimes stressingthe position of being bom the child of wealthyand prestigious parents, sometimes stressingsomething quite else within an unlimitedvariation ofhuman ways to culturally organisehuman interaction and childhood.

English revised by Laura Wrang.

A CKNOWLEDGEMENTSThe article has been compiled in collabora-tion with all in all about thirty students at thearchaeological departments at the universi-ties of Gteborg, Stockholm, Tromsttt, Ume,Uppsala, and stersund. Many of' them havechildren of' their own, although the numberincreased with only one during the courses inciuestion.

Current Sseedish Archaeology, Vol. 6, 199

196 Stig Welinder

TABLES

Tab. 1. The data ser.

Population Date Children Reference

1. MN Gotland2. The Battle Axe Culture

35002500 BC28002400 BC

c. 15 (24)c. 20 (13)

Andersson et al. 1995Andersson el al. 1995

3. Haralds Minne

4. LBA5. LBA Halland

1200-800 BC

1200500 BC1200500 BC

69 (20)

c. 30 (c.120)19 (10)

Lindman &Strmberg 1993Sallander 1995Lundborg 1972

6. LBA & EIA stergtland 1000 BC400 AD 1562 (c. 45)7. PRIA Vstergtland 400 BC100AD 1762 (301)8. EIA land 400 BC 600 AD 48 (35)

Bjrkhager 1995Lundin & Skoglund 1995Beskow Sjberg 1987,Hagberg et al. 1991

9. Suderbys10. Stora Vikers11.Fiskeby12. Vallhagar13. Ula14. Bjrby15. Styrmansberg16. Vrberg

17. Spnga

18. Lnglt

19. Hamre20. Lov21. Helg

22. Birka23. Vi vallen24. Fjlkinge25. Bjurhovda26. Vsterhus

1000I BC1000I BC1000 BC1000 AD700 BC700 AD100 BC100 AD1 200 ADI 200 ADI 600 AD

11000 AD

300500 AD

400700 AD5001000 AD7001000 AD

750950 AD900-1200 AD9501000 AD1000-1050AD1000-1350AD

?? (c. 30)?? (c. 20)15 (32)1 (I)20 (11)79 (77)25 (7)6 (4)

48 (14)

c. 70 (23)

18 (5)9 (6)3 (2)

17 (84)45 (5)62 (128)19 (8)57 (217)

Sjberg 1976Lindquist 1981Lundstrm 1965, 1970Stenberger 1955 a, bWelinder 1989Schultze et al. 1991Stenberger 1955 a, bFerenius 1971,Iregren 1972Biuw 1992,Sigvallius 1994Johansson-Lundh &Rasch 1991Welinder 1990Petr 1984Holmqvist el al. 1970,Sander 1990Grslund 1980Hallstrm 1944Helgesson 1992aWelinder 1990Gej vall 1960

Comment:Population nos. 2, 4 consist of all graves in Sweden, nos. I, 58 of all graves in a single province, nos. 1126 of the graves at I 4 burial-grounds each, and nos. 3, 910 of a single feature each. The dates refer tothe calibrated radiocarbon time- scale. The percentages of children's graves denote individuals youngerthan c. 15 years (infant, infans III). The intervals refer to minimum and maximum percentages amongseveral burial-grounds. The numbers in brackets denote the total number of children.

Cttrrenr Swedish Archtteology, Voh 6, 1998

The Cutturat Construction of Childhood 197

Tab. 2. The percentage of children and adults according to number of grave-obj ects(AOT according to Hedeager 1978) at the burial-g round ofFiskeby (population no. 11).

AOT Early Iron Age Late Iron AgeChildren Adults Children Adults

25

691013

7921

403920

7525 15

4229

Tab. 3. The percentage of children s graves at a selection of1ron Age burial groundsin the vicinity of Stockholm (population nos. 1617, 2021).

Children (%) Total number of individuals16. Vrberg17. Spnga 156

157a157b158, 160, 162

20. Lov, Viking AgeVendel Period

21. Helg

56684

1652

8588901282195688

Comment:Population no. 17 Spnga is subdivided into individual burial-grounds, and population no. 20Lov into periods.

Tab. 4. The percentage of children at the largest Early Iron Age burial-grounds on land (population nos. 8, 14, 18).Population

Common burial-grounds

Population no. 8Bredstra 3-4Bredstra 40Grdslsa 74Algutsrum 13

Special populations

14. Bjrby18. Lnglt

Children (%)

23 (c. 50)102419

79c. 70

Total no. of individuals

35 (c.60)317616

97c. 35

CommenttThe numbers in brackets for Bredstra 3-4 include one special feature with parts of c. 25skeletons of children.

Current Ssvedish Arehaeotogy, Vol. 6, 1998

198 Srig Wetinder

Tab. 5. The stepwise childhood during the Middle Neolithic on Gotland(population no. I).Start of age interval Characteristics of graves (all of them inhumation graves)

23

56

Recognisable burials are close to non-existent

No or very few grave-objectsNo or few dress-ornaments

Tusks of boares are introducedAxes, hunting and fishing gear, and other tools are introducedA new dress with numerous beads is introducedFew female graves

1819 (until c. 20) Equal number of female and male gravesComment: For details, see Andersson et al. 1995:Tab. 1.

Tab. 6. The stepwise childhood during the Late Bronze Age and Early Iron Age in sterg tland (population no. 6).

Age interval (years) Characteristics of graves (all of them cremation graves)

Infant (01)

Infans I (07)Infans II (514)

Juvenilis (1024)

The number of graves is smallMiniature cinerary urnsNo grave-objectsNo or few grave-objectsMetal dress-objects are introduced(e.g. needles and belt fittings)No or few grave-objects. Metal objects are rare

Comment: For details, see Bjrkhager 1995:4852.

Tab. 7. The stepwise childhood and adult life-span in the Saami population at Vivallen(population no. 23).

Age

2030

&35

Characteristics of graves (all of them inhumation graves)

No recognisable graves

Iron arrowheadsA few metal dress-objectsNew dress-ornaments are introduced, or a new kind of dress(e.g. beads, penannular brooches, finger-rings)A divison into graves with no objects and graves with the sameobjects as above (possibly females and males in both groups)All graves contain grave-objects. Some graves are richlyfurnished with exotic display-objects (possibly only males)

Comment: The total number of graves is small.

Current Swedish Archaeotogy, Vol. 6, 1998

The Cultural Construction of Childhood 199

Tab. 8. The stepwise childhood during the Late Bronze Age(population nos. 4, 910).Start of age interval Characteristics of graves (Mostly cremation graves)

45

79

Few individual gravesDouble, triple and mass gravesNo grave-objectsNo or few grave-objectsA few grave-objectsTools and dress-ornaments as often as the adults, but a smallernumber of various types

c. 14 (until c. 2024) As above. Razors and tweezers as often as the adultsWeapons more often than the adults

Comment: For details, see Sallander 1995:40&3.

Tab. 9. The life-cycle during the Late Bronze Age in Halland (population no. 5).Age interval Number of objects per grave

Females Males In det.

Infans III (014)Juvenilis (1420)Adultus (2040)Maturus (4060)Senilis (&60)

01.20.80.5

1.01.51.8

0.2

Comment: The table is based on a small number of graves.

Tab. 10. The female life-cycle (number of graves) at the Early lron Ageburial-ground of Ula (population no. 13).Object types Infans I-II

(014)Juvenilis(1420)

Adultus Maturus Senilis(2040) (4060) (&60)

Dress-ornamentsSicklesNo objects

2 21(9)

Comment: From Welinder 1989:Tables 56.

Current Swedhsh Archaeology, Vot. 6, 1998

200 Stig Wetinder

Tab. 11. The chi ldren in relation to adults at the Early Iron Age Vallhagarburial-grounds and at the Stynnansberg hill-fort (population nos. 12, JS).

Types of graves Styrmansberg VallhagarBurial-grounds Settlement site

Graves with a single child (1)Graves with only adults (1)Double graves with one or twochildren and one adult (2)

15

6 (6)

140

1 (2)Comment:(1) = number of individuals, (2) = number of graves, number of children in brackets.

Tab. 12. The stepwise childhood in the Middle Neolithic Battle Axe Culture(population no. 2).

Start of age interval Females Males In det.

34

Recognisable graves are very few

Generally the same grave-objects as adultsDouble graves together with an adult are common

c. 12 Copper objectsare introduced

Battle axes andcopper objectsare introduced

Comment:Most children's graves contain no or few objects. The same is valid for most adult graves.

Current Swedish Archaeology, Vol. 6, 1998

The Culzurol Conszrucrion of Childhood Tol

Tab. 13. Status display at the Bj urhovda burial-grouzzd (population zzo. 25).

Age Females

Characteristics Number ofof graves individuals

Males

Characteristics Number ofof graves individuals

Total numberof children

015 A full set of 1breastornamentsand a pendantwith amongothers a needlecase

The equipmentof a mountedwarrior

015 A full set ofbreastornaments

A spear andan axe

015 An occasional 2dress-om ament

Only belt-fittings 1

No gender det.

015 No or few objects, no gender-related objectsc. 20

Comnmen:The gender heading is based on osteological analyses as concerns the adult individuals.For the children it is based either on the grave-object or the presence in a gendered adult grave. Individual sthat cannot be unanimously labeled as concerns age are excluded. For details, see Welinder 1991.

Tab. 14. The stepwise childhood in the Early Medieval Vsterhus churchyard(population no. 26).

Age intervals Characteristics of graves

0I Nearly all at the female side Mass graves with 3-23 children atthe female sideA few individual graves, mostly atthe male side

1 7

714

&14

Predominantly at the female side

Predominantly at the male side

Segregated according to sex

Individual graves

Currenz Srvedish Archaeology, Vot. 6, /998

202 Stig Welinder

Tab. /5. The ideal Catholic childhood and the Vsterhus churchyard (population no. 26).Start of Social status Social responsibilitiesage interval

Interpretation in relation tothe Vsterhus churchyard

0 Newborn

Baptism

None

Religiously ignorantCannot distinguish thecommunion from anordinary mealDependent on its mother

Buried in the female part ofthe churchyardMass graves

68 Confirmation Subordinate to its fatherNot subject to full punish-ment according to law

Individual gravesPredominantly in the malepart of the churchyard

1215 Adulthood Allowed to make agreementsSubject to full punish-ment according to lawTo be married

Buried totally in agreementwith the adults

Comment: A simplified schedule, partly based on Nilsson 1989, 1994.

Tab. /6. Late Bronze Age (population no. 4) gender-groups of types of obj ects.Grave-obj ects Female Male In det.

)10 yrs. )10 yrs. &5 yrs. 5-10 yrs. )10 yrs.Sickles, cattle, sheep/goat, pigsKnives, armlets, rings, lambs

Comment: The table is based on osteological sex-determinations. Figures denote number of individuals.

REFERENCESAndersson, B. 1985. Ungdomsrens historia. In:

Lundmark, L.-E. & Stridsman, Kj. (Ed). Senkommer en annan tid. En antologi om barnsoch ungdomars livsformer. Socialstyrelsen,Stockholm. Pp. 2145.

Andersson, G. , Welinder, S.& Westeson, . 1995.Barndomens grnser under mellanneolitikum.In: Johnsen, B.& Welinder, S.(Eds). Arkeologiom barn. Societas Archaeologica Upsaliensis,Uppsala. Pp. 2941.

Aries, Ph. 1982 (1960). Barndomens historia.Gidlunds. Stockholm.

Benedictow, O.J. 1993.The medi eval demog raphi csystem of the Nordic countries. Middelalder-forlaget. Oslo.

Beskow Sjberg, M. (Ed). 1987. lands jrn-ldersgravflt. Volym I. Riksantikvariembe-tet, Statens historiska museer. Stockholm.

Bjurman, E.L. 1981.Vad ska du bli min lille vn?om barnuppfostran frr i tiden. In: Bjurman,EL. (Ed). Barnhistoria. Om barns villkor frr.Esselte Studium, Nacka. Pp. I23.

Biuw, A. 1992. Norra Spnga. Bebyggelse ochsamhlle underjtirntildern. Stockholmsmono-grafier 76. Stockholm.

Bjrkhager, V. 1995.Gravskicket under barnaren.En studie av stgtska gravar frn vergngenmellan bronslder och jrnlder. In: Johnsen,B. & Welinder, S. (Eds). Arkeologi om barn.Societas Archaeologica Upsaliensis, Uppsala.

Current Svvedish Archaeology, Vol. 6, /998

The Cultural Construction of Childhood 203

Pp. 43-56.Borgnakke, K. 1981. Inte riktigt barn, inte riktigt

vuxna. In: Clausen, C. (Ed). Barnet blir barn.En antologi om barndomens historia. FrlagetBarrikaden, Stockholm. Pp. 127166.

Drake, S. 1979(1918).Vsterbottenlapparna underfrra hlften av 1800-talet. Tv frlggare. Ume.

Ferenius, J. 1971. Vrby och V rberg. En studie ijIimldemsbebyggelsehistoria. Studies in North-European Archaeology. Series B. Theses andpapers published in offset 1. Stockholm.

Formisto, T. 1993.An osteologi cal analysi s ojhumanand ani nu&l bones from Levi&nluhta. Vammala.

Gebiihr, M. 1992.Alter und Geschlecht. Aussage-mglichkeiten anhand des archologischen undanthropologischen Befundes. In: Stjernquist, B.(Ed). Prehi stori cg raves as a source ofi nfor&na-tion. Symposium at Kastlsa, land, May 2123, 1992. Kungl. Vitterhets Historie och An-tikvitets Akademien, Stockholm. Pp. 7382.

Gej vall, N. -G. 1960. Westerhus. Medi eval popula-tion and church in the light ofskeletal remains.Kungl. Vitterhets Historie och AntikvitetsAkademien. Stockholm.

van Gennep, A. 1960 (1908).The vites ofpassage.University of Chicago Press, Chicago.

Gillis, J. 1974. Youth and history. Academic Press,New York.

Grslund, A.-S. 1973.Barn i Birka. Tor 15.Uppsala.Pp. 161-179.

1980.BirkaIV. Theburial customs. Astudyofthegraves on Bjrk. Kungl. Vitterhets Historieoch Antikvitets Akademien, Stockholm.

Hagberg, L. 1949. Seder och tro vid mrkestillfl-lena i barnets liv. Nordisk kultur 20. Pp. 1440.

Hagberg, U.E., Stjernquist, B.& Rasch, M. (Eds).1991. lands jarndldersgravflt. Voly&n II.Riksantikvariembetet, Statens historiska mu-seer, Stockholm.

Hallstrm, G. 1944.Gravfltet p Vivallen i Funs-dalen. Fornv rdaren 8. stersund. Pp. 306344.

Hedeager, L. 1978.Processes towards state forma-tion in Early Iron Age Denmark. In: Kris-tiansen, K. (Ed). New di rectionsin Scandinavi-an archaeology. Nationalmuseet, Copenhagen.Pp. 217-223.

Helgesson, B. 1992a. En mrklig frdelning ettsenvikingatida gravflt i Fjlkinge, Skne.META 92:1-2.Lund. Pp. 4043.

1992b. 79 barngravar i Skne. Alla dessa sm,som aldrig ftt chansen att vxa upp. .. Populrarkeologi 1992:4.Pp. 1214.

Henriksson, B. 1991.Att skapa sig en ungdom: omdet sv ra samspelet mellan frihet och kontroll.Slutrapport frn projektet "Ung p 90-talet" iVlisters. Clacio. Stockholm.

Hofsten, E. 1986. Svensk befolkni ngshistoria. Ni-

g ra grunddrag i utvecklingen fr n I750. Rabn& Sjgren, Stockholm.

Holmqvist, W. , Lamm, K. & Lundstrm, A. (Eds).1970. Excavations at Helg 111. Report for19601964. Kungl. Vitterhets Historie ochAntikvitets Akademien, Stockholm.

Iregren, E. 1972.Vrb och Vrberg II. Studie avkremerat &nnni sko- och dj urbensmaterial frn

jrnldern. Theses and Papers in North-Euro-pean Archaeology 1. Stockholm.

Johansson-Lund, K. &. Rasch, M. 1991.Lngltssocken. In: Hagberg, U.E., Stjernquist, B. &Rasch, M. (Eds). 1991.lands jrnlders-g ravf lt. Volym II. Riksantikvariembetet, Stat-ens historiska museer, Stockholm. Pp. 263338.

Johnsen, B. & Welinder, S. (Eds). Arkeologi ombarn. Societas Archaeologica Upsaliensis,Uppsala.

Lillehammer, G. 1986.Barna i Nordens forhistorie.Dr&ft metodegrunnlaget og kildenes ba:rekraft.KAN 2. Bergen. Pp. 321.

1989. A child is bom. The child*s world in anarchaeological perspective. Norwegian Ar-chaeological Review 22:2. Oslo. Pp. 89105.

Lindman, G. & Strmberg, B. 1993.Gravform ochgravskick tv arkeologiska underskningarisdra lvsborg. Riksantikvariembetet ochstatens historiska museer. Rapport UV 1993:1.Stockholm,

Linder, L. 1995. Hade den vikingatida brsrkenngra barn? In: Johnsen, B. & Welinder, S.(Eds). Arkeologi om barn. Societas Archaeo-logica Upsaliensis, Uppsala. Pp. 6977.

Linderoth, Th. 1990. Vart har alla barnen tagitvgen? Populr arkeologi 8:3. Pp. 1416.

Lindquist, M. 1981.Mylingar offer, utsatta barneller frhistoriska barnbegravningar? Gotl ndsktarkiv 53. Visby. Pp. 92.

Lundborg, L. 1972.Underskningar av brons lders-hgar och bronsldersgravari sdra Halland.Hallands museum 2. Halmstad.

Lundin, I. & Skoglund, M. 1995.Gravfltens min-sta. Om barngravar under frromersk jrnlder.In: Johnsen, B.&Welinder, S. (Eds). Arkeologiom barn. Societas Archaeologica Upsaliensis,Uppsala. Pp. 5768.

Lundstrm, P. 1965.Gravfiilten vid Fiskebyi Norr-kping. II. Fornlmningar och fynd. Kungl.Vi tterhets Historie och Antikvitets Akademien,Stockholm.

1970. Gravfilten vid Fiskeby i Norrkping. I.Studier i ett totalunderskt komplex. Kungl.Vit terhets Historie och Antikvitets Akademien,Stockholm.

Nilsson, B. 1989.De sepultris. Gravrtten i corpusjuris canonici och i medeltida nordisk lag-stiftning. Almqvist & Wiksell, Stockholm.

Current Swedish Archaeology, Vol. 6, 1998

204 Stig Welinder

1994. Kvinnor, mn och barn p medeltidabegravningsplatser. Projektet Sveriges krist-nande. Publikationer 3. Uppsala.

Odner, K. 1972. Ethno-historic and ecological set-tings for economic and social models of an IronAge society: Valldalen, Norway. In: Clarke,D.L. (Ed). Modelsin Archaeology. Methuen,London. Pp. 623-651.

Ohlander, A.-S. 1993. Det borttrngda barnet.Opuscula Historica Upsaliensis 10. Uppsala.

Olsson, L. 1981.Frn barnarbete till ungdomsarbets-lshet. In: Bjurman, E.L. (Ed). Barnhistoria.Om barn villkor frr. Esselte Studium, Nacka.Pp. 95100.

Petr, B. 1984. Arkeologiska underskningar pLov. 4. Bebyg g els earkeologisk analys. StudiesinNorth-EuropeanArchaeology 10.Stockholm.

Pollock, L.A. 1983. Forgotten children. Parent-child relations from 1500 to 1900. CambridgeUniversity Press, Cambridge.

Roslund, M. 1990.Nittiotalets medeltidsarkeologioch det dda barnet. Fornvnnen 85. Stockholm.Pp. 283292.

Sallander, K. 1995.Brons ldems barn. Unpublishedseminar paper. Arkeologiska institutionen,Uppsala universitet.

Sandberg, M. 1994. Stenldersbarn kan vi sedem? Fjlnir 13:1.Uppsala. Pp. 514.

Sander, B. 1990. Helg gravflt 116 kronologi,korologi. Unpublished seminar paper. Arkeolo-giska institutionen, Stockholms universitet.

Schultze, H. , Beskow Sjberg, M. , Broberg, B. Ik.Holgersson, K. 1985. Ett barngravflt frnromerskj m lder. Fornl mni ng 89, Bj rby, Kast-lsa socken, land. Arkeologisk underskning1972, 1976/76. RA Rapport UV 1984:35.Stockholm.

Sellevold, B.J. 1989.Kvinners dfldelighet i forbin-delse med svangerskap og fdsel i forhistorisktid og middelalder, belyst ut fra studier avskjelettmaterialer. Skriftserie frn Centrum frkvinnoforskning vid Stockholms universitet 1.

Stockholm. Pp. 7996.Sigvallius, B.1994.Funeral pyres. Iron Age crema-

tions in North Splnga. Theses and Papers inOsteology l. Stockholm.

Siven, C.H. 1991a.On estimating mortalities fromosteological data. Internati onal Journal ofAn-thropology 6:2. Firenze. Pp. 97110.

1991b. On reconstructing the (once) livingpopulation from osteological age data. Interna-ti ona I Journal ofAnth ropology 6:2.Fi ren ze. Pp.111118.

Sjberg, A. 1976.Bronsldersrsen vid Suderbys iVsterhejde. Gotllindskt Arkiv. Visby. Pp. 136137.

Stenberger, M. (Ed). 1955a. Vallhagar a Migra-tion Period settlement on Gotland/Sweden.Part I. Ejnar Munksgaards Forlag, Copenhagen.

1955b. Vallhagar a Migration Period settle-ment on Gotland/Sweden. Part 11. EjnarMunksgaards Forlag, Copenhagen.

Vinsrygg, S 1988. Archaeology as if peoplemattered. A discussion of humanistic archaeo-logy. Norwegian Archaeological Review 21:I.Oslo. Pp. 1 20.

Welinder, S. 1979.Prehistoric demography. CWKGleerup, Lund.

1989.An experiment with the analysis of sex andgender of cremated bones. Tor 22. Uppsala. Pp.29-41.

1990. Mcinniskor i Vsterstrakten fr 1000 rsedan. Vsters kommun, Vsters.

1991. Den frsta generationen stadsbor iVsters. In: Wik, B. (Ed), Sentrum periferi.Sen tra og sen trums dan nelser gj ennom fe&rhi s-torisk og hi storisk tid. Gunneria 64. Trondheim.Pp. 387393.

Wikman, K.R.V. 1949. Livets hgtider. Nordiskkultur 20. Albert Bonniers frlag, Stockholm.

sterberg, V. 1996.Social structure and chi Idren 'slife chances. An analysisofchi ld mortality in Swe-den. Swedish Institute for Social Research 26.Stockholm.

Current Swedi sh Archaeology, Vol. 6, 1998