I

DENTITY, GENDER, RELIGION AND ECONOMY: NEW ISOTOPE AND RADIOCARBON EVIDENCE FOR MARINE RESOURCE INTENSIFICATION IN EARLY HISTORIC

ORKNEY, SCOTLAND, UK James H. Barrett University of York, UK

Michael P. Richards Max Planck Institute for Evolutionary Anthropology, Germany, and University of Durham, UK

Abstract: Stable isotope measurements and radiocarbon dates on 54 burials from northern Scotland document trends in marine protein consumption from the late Iron Age to the end of the Middle Ages. They illuminate how local environmental and cultural contingencies interrelated with a panEuropean trend towards more intensive fishing around the end of the first millennium AD. Little use was made of marine foods in late Iron Age Orkney despite its maritime setting. Significant fish consumption appeared in the Viking Age (ninth to eleventh centuries AD), first in the case of some men buried with grave-goods of Scandinavian style but soon among both sexes in ‘Christian’ burials. There was then a peak in marine protein consumption from approximately the eleventh to the fourteenth centuries AD, particularly among men, after which the importance of fish-eating returned to Viking Age levels. The causes of these developments probably entailed a complex relationship between ethnicity, gender, Christian fasting practices, population growth, long-range fish trade and environmental change. Keywords: diet, ethnicity, fish trade, gender, Iron Age, Middle Ages, Orkney, Scotland, stable isotopes, Viking Age

I NTRODUCTION Recent zooarchaeological evidence has shown that there was a major increase in the importance of fishing in north-western Europe around or shortly after the end of the first millennium AD (Barrett et al. 1999, 2004a; Enghoff 1999, 2000; Makowiecki 2003:195; Perdikaris 1999; Van Neer and Ervynck 2003). This ‘fish event horizon’ involved a major increase in catches of cod, herring and related species, the ultimate causes of which are the subject of ongoing debate (Barrett et al. 2004a). Although a widespread phenomenon, the specific chronology of the fish event European Journal of Archaeology Vol. 7(3): 249–271 Copyright © 2004 SAGE Publications (www.sagepublications.com) and the European Association of Archaeologists (www.e-a-a.org) ISSN 1461–9571 DOI:10.1177/1461957104056502

250

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

horizon varied from region to region. Differences of this kind raise an important question: how did the widespread secular trend towards increased fishing interrelate with local environmental and cultural contingencies? Phrased another way, what were the causes of this phenomenon at a local scale? In some extremely maritime environments, such as coastal Arctic Norway, fish (cod in particular) was important throughout the first millennium AD, becoming more so in the eleventh and/or twelfth centuries as Norway became part of a pan-European fish trade (Christensen and Nielssen 1996; Perdikaris 1999). In contexts with greater agricultural potential, such as England, the transition was more dramatic. Within a few decades of AD 1000 the catch of herring probably increased fourfold and cod were landed on a meaningful scale for the first time (Barrett et al. 2004a). There was probably not even a word for cod in the Anglo-Saxon language (Sayers 2002). The Scandinavian colony of Orkney, an archipelago of approximately 40 islands off northern Scotland, provides an illuminating case study between these extremes. Orkney was part of Pictish Scotland during the middle centuries of the first millennium AD, before being colonized from Scandinavia around AD 850 (Barrett 2003; Ritchie 1985). The archipelago then functioned as the core of a semiindependent Norse chiefdom, known as the earldom of Orkney by the eleventh century, which at times controlled most of northern and western Scotland. Orkney was not formally incorporated into the Kingdom of Scotland until the late fifteenth century, and a Norse dialect was still spoken into the 1700s. The islands later served a pivotal role as a staging point in North Atlantic shipping and the same can probably be assumed of the Viking Age and Middle Ages (Thomson 2001:371–374). Based on a preliminary analysis of stable isotope and fish bone evidence, it has been argued that marine fishing (mainly for large cod family species) was intensified in Orkney in two stages: firstly as an introduction of Norwegian foodways during Viking Age colonization from Scandinavia (Barrett et al. 1999, 2001; Cerón-Carrasco 2002); and secondly as a response to the growing export market for dried fish in the eleventh and twelfth centuries (Barrett 1997; Barrett et al. 2000b:15–19). This article explores and tests this model by combining new and previously published radiocarbon dates and stable isotope data for human skeletons from Orcadian sites of Pictish (fourth to ninth centuries AD), Viking Age (ninth to eleventh centuries AD), and medieval (eleventh to sixteenth centuries AD) date. In particular, it combines dates published by Sellevold (1999), dates published by Barrett et al. (2000a, 2000b), isotope data published by Richards et al. (in press), and a series of new radiocarbon and stable isotope measurements. The two main assemblages are from Westness (Kaland 1996; Sellevold 1999) – a Pictish and Viking Age cemetery on the island of Rousay – and Newark Bay (Barrett et al. 2000a; Brothwell 1977; Brothwell et al. 1986; Lowe 2001; Richards et al. in press) – a Viking Age and medieval chapel and associated cemetery on Mainland, the largest of the islands (Fig. 1). However, data from other Viking Age and medieval burials excavated in Orkney, the adjacent mainland of northern Scotland, and the Western Isles of Scotland (also colonized by Scandinavians in the ninth and tenth centuries) have been included (see Table 1). This article asks four questions:

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

251

(i) When did Orcadians begin to consume significant amounts of marine protein? (ii) Was this a single or a two-stage process? (iii) Did it begin with Scandinavian colonization or only later in the Middle Ages? And, (iv) in so far as possible, what causal factors may have played the most important roles? These questions are critical in a wider context as they allow us to illuminate the interplay between what is now a well-known secular trend towards increased marine fishing in Europe on one hand, and local contingencies ranging from environment to ethnicity, religion, and gender on the other. In total, this article reports on 54

Figure 1. Location of Orkney and the main sites discussed.

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

252

Table 1. δ13C and δ15N values, C:N ratios, % marine carbon estimates, radiocarbon dates, mixed atmospheric/marine calibrations (using a range of ∆R values), grave-good dates and burial characteristics for Pictish to late medieval graves considered in this study. All analyses were on human bone collagen. Calibrations were performed using OxCal Version 3.9 (Bronk Ramsey 1995, 2001) and the datasets of Stuiver et al. (1998).

Site & burial Lab no.

δ13C (radiometric dating)

δ13C (AMS dating)

δ13C (ultrafiltered)

δ15N (ultrafiltered)

–21.17

–21.10 –21.00

C:N

Estimated % marine carbon

Date bp

10.83 11.00

3.35 3.30

0 0

1410+60 1600+70

–21.00 –21.00 –20.93 –20.84 –20.76 –20.70

10.41 11.26 11.47 11.00 10.37 11.80

3.51 3.46 3.28 3.26 3.52 3.40

0 0 1 2 3 3

1530+60 1480+40 1410+30 1360+70 1330+120 1190+50

Westness 32 1963b

T-7470 TO-7195

–21.6

25 28a 21 10 26 5

T-6529 T-7471 T-7468 T-6527 T-6817 T-6532

–19 –19.2 –17.1 –20.7 –17.3 –21.1

5

TO-7532

–20.70

11.80

3.40

3

1300+30

5 combined

–20.70

11.80

3.40

3

1271+26

36

T-6532 & TO-7532 T-6530

–21.2

–20.63

11.74

3.63

4

920+50

24 20 Naust Naust Naust combined 12

T-7466 –19.5 T-6528 –20.1 AA-50703 AA-53127 AA-53127 & AA-50703 T-6815 –18.7

–20.55 –20.46 –18.42 –18.42 –18.42

10.88 11.01 14.38 14.38 14.38

3.24 3.26 3.33 3.33 3.33

5 6 29 29 29

1400+40 1400+60 255+35 305+40 277+26

–18.20

14.80

3.40

31

1240+70

11

T-6814

–18.10

13.70

3.30

32

1270+60

–17.80 –17.80

–19.1

Newark Bay 71/5 TO-6933

–19.74

–20.31

10.27

3.33

8

1170+50

69/36

–19.32

–20.25

8.61

3.27

8

1090+40

10.10 10.50 10.92

3.30 3.30 3.36

9 11 12

1190+40

TO-7182

68/16 69/4a 68/16A

TO-7174

–20.47

–20.20 –20.00 –19.95

71/9 69/34

TO-7181

–21.76

–19.80 –19.65

10.50 10.88

3.30 3.35

13 15

930+40

68/20

AA-54931

–19.70

–19.63

10.41

3.26

15

1055+40

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

253

Mixed atmospheric/marine calibration, 95.4% range (∆R = –79+17)

Mixed atmospheric/marine calibration, 95.4% Grave-good range (∆R = 91+30) dating

Age

Sex

gravegoods References

460–780 250–620

430–780 250–640

adult adult

female female

no no?

400–660 430–660 540–690 540–880 400–1000 680–990

380–660 430–670 530–770 540–890 400–1000 680–1000

c.850–950

child adult adult adult adult adult

? female male female male female

no no no no no yes

650–860

650–890

c.850–950

adult

female

yes

660–880

660–890

c.850–950

adult

female

yes

1010–1250

1000–1270

c.850–950

adult

female

yes

560–770 530–800 1650–1950+ 1520–1950+ 1640–1950+

540–780 530–870 1670–1950+ 1649–1950+ 1670–1950+

adult adult adult adult adult

female male male male male

no no no no no

710–1030

710–1160

c.850–950

adult

male

yes

690–990

710–1040

c.850–950

adult

male

yes

720–1020

710–1030

infant

?

no

860–1050

810–1160

adult

female

no

720–1000

720–1020

adult adult adult

female female male

no no no

1020–1250

1030–1280

adult adult

female female

no no

890–1160

890–1220

adult

female

no

this study, Sellevold 1999 this study, Sellevold 1999; Barrett et al. 2000b this study, Sellevold 1999 this study, Sellevold 1999 this study, Sellevold 1999 this study, Sellevold 1999 this study, Sellevold 1999 this study, Graham-Campbell & Batey 1998; Sellevold 1999 this study, Graham-Campbell & Batey 1998; Sellevold 1999

this study, Sellevold 1999, James Graham-Campbell pers comm. this study, Sellevold 1999 this study, Sellevold 1999 this study, Sellevold 1999 this study, Sellevold 1999

this study, Graham-Campbell & Batey 1998; Sellevold 1999 this study, Graham-Campbell & Batey 1998; Sellevold 1999 Barrett et al. 2000a, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Richards et al. forthcoming Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

254 Table 1. continued

Site & burial Lab no. 69/85

TO-7186

69/11

TO-7180

δ13C (radiometric dating)

δ13C (AMS dating)

C:N

Estimated % marine carbon

Date bp

10.80

3.29

15

1000+40

–19.60

9.85

3.32

16

1380+30

10.60 11.50 10.10 11.50 11.03

3.30 3.30 3.30 3.30 3.27

16 17 17 18 18

1130+50

δ13C (ultrafiltered)

δ15N (ultrafiltered)

–19.63 –20.69

69/84 69/104 69/2 69/NO 69/104B

TO-7189

–17.42

–19.60 –19.50 –19.50 –19.40 –19.38

69/33 70/1

TO-7192

–19.11

–19.30 –19.23

10.80 12.04

3.30 3.29

19 20

1010+60

70/5

TO-6939

–19.13

12.17

3.30

21

520+30

71/3

TO-7193

–19.32

–19.07

10.37

3.26

21

1200+40

69/105

TO-7190

–20.29

–18.66

11.63

3.29

26

990+50

70/30 69/99

TO-7187

–18.47

–18.60 –18.51

12.70 12.04

3.30 3.24

27 28

1060+40

1968/2

TO-6940

–19.70

–18.44

12.98

3.27

28

550+40

70/6

TO-6937

–18.80

–18.40

14.39

3.41

29

700+40

69/37

TO-6944

–18.13

14.70

3.28

32

910+50

68/14

GU-10955

–18.40

–18.03

13.35

3.34

33

630+50

69/107

TO-7191

–18.62

–17.87

12.05

3.31

35

1340+60

69/8

TO-7178

–19.41

–17.84

13.70

3.28

35

830+30

68/5

AA-54930

–17.70

–17.82

13.37

3.29

35

590+40

69/67a

TO-7183

–17.74

13.70

3.30

36

990+40

71/8

TO-6934

–17.81

–17.72

13.79

3.28

36

1160+60

71/8

AA-54938

–18.20

–17.72

13.79

3.28

36

1020+45

71/8 combined 71/7 70/37

TO-6934 & AA-54938

–17.72

13.79

3.28

36

1070+36

AA-54934

–17.40

–17.70 –17.52

12.90 13.72

3.30 3.31

37 39

1130+40

69/69

TO-7184

–20.84

–17.50

14.38

3.25

39

910+40

69/83

TO-7185

–17.47

13.77

3.30

39

1060+40

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

255

Mixed atmospheric/marine calibration, 95.4% range (∆R = –79+17)

Mixed atmospheric/marine calibration, 95.4% Grave-good range (∆R = 91+30) dating

Age

Sex

gravegoods References

990–1210

990–1240

adult

female

no

640–780

640–870

adult

female

no

770–1050

810–1170

adult adult adult adult adult

female female male female female

no no no no ?

970–1250

990–1270

adult adult

? female

no no

1390–1620

1410–1640

adolescent male

no

770–1010

770–1040

adult

female

no

1010–1250

1020–1280

adult

male

no

970–1190

990–1240

adult adult

? female

no no

1320–1630

1410–1640

adult

female

no

1290–1430

1290–1470

adult

female

no

1040–1300

1110–1400

infant

?

no

1300–1490

1380–1630

adult

male

no

660–960

680–1000

adult

female

no

1210–1390

1250–1420

adult

?

no

1320–1530

1400–1640

adult

male

no

1030–1260

1040–1300

adult

male

no

810–1160

880–1220

adult

male

no

1010–1240

1030–1290

adult

male

no

990–1190

1020–1260

adult

male

no

890–1160

970–1230

adult adult

female male

no no

1070–1310

1160–1410

adult

female

no

1000–1210

1020–1270

adult

male

no

this study, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Richards et al. forthcoming Richards et al. forthcoming Richards et al. forthcoming Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming this study, Richards et al. forthcoming this study, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming this study, Richards et al. forthcoming this study, Richards et al. forthcoming this study, Richards et al. forthcoming

Richards et al. forthcoming this study, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

256 Table 1. continued

Site & burial Lab no.

δ13C (radiometric dating)

δ13C (AMS dating)

δ13C (ultrafiltered)

δ15N (ultrafiltered)

–15.60

–17.47

C:N

Estimated % marine carbon

Date bp

13.77

3.30

39

1070+45

–17.47

13.77

3.30

39

1064+30

69/83

AA-54932

69/83 combined 99/6

TO-7185 & AA-54932 OxA-10409

–16.79

–17.4

13.9

3.4

40

1015+37

70/39

AA-54935

–16.90

–17.31

14.41

3.31

41

965+40

70/35 70/2

TO-6935

–16.95

–17.30 –17.29

13.90 13.92

3.30 3.30

41 41

1060+50

71/2

AA-54937

–15.00

–17.09

14.57

3.33

43

1075+35

CC4

TO-6941

–16.68

–16.99

14.11

3.30

45

920+40

68/12

TO-7173

–17.56

–16.90

14.35

3.29

46

930+40

70/15

AA-54933

–17.80

–16.88

13.52

3.32

46

1060+35

SK001

OxA-10407

–16.81

–16.6

14.1

3.4

49

1070+36

69/104A

TO-7188

–16.28

15.58

3.33

52

1030+40

71/1

AA-54936

–15.80

–15.41

15.59

3.24

62

1075+35

SUERC2894

–19.70

–19.60

11.50

3.30

16

1350+35

SUERC2895

–19.70

–19.60

11.50

3.30

16

1265+35

–19.60

11.50

3.30

16

1308+25

–17.80

15.10

3.30

36

895+35

Other Balnakeil

Balnakeil

Balnakeil combined

SUERC-2894 & SUERC-2895

Bu of Cairston 009 SUERC-1201 Cnip C

GU-3485

–18.3

–18.66

11.56*

3.29

26

1150+50

Cnip B

GU-3489

–20.8

–18.35

11.77*

3.28

29

1150+50

Scar 134 Scar 134 Scar 134 Scar 134 combined

AA-12596 AA-54929 SUERC-1176 AA-54929, SUERC-1176 & AA-12596

10.6 10.6 10.6 10.6

3.5 3.5 3.5 3.5

8 8 8 8

1040+60 910+35 1125+40 1009+24

*Not ultrafiltered

–21.10 –18.30 –20.90

–20.3 –20.3 –20.3 –20.3

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

257

Mixed atmospheric/marine calibration, 95.4% range (∆R = –79+17)

Mixed atmospheric/marine calibration, 95.4% Grave-good range (∆R = 91+30) dating

Age

Sex

gravegoods References

990–1210

1020–1270

adult

male

no

1000–1190

1020–1260

adult

male

no

1020–1240

1040–1290

adolescent male

no

1040–1290

1050–1390

adult

no

990–1220

1020–1280

adult ? adolescent female

no no

990–1200

1020–1270

adult

no

1120–1310

1190–1410

adolescent male

no

1120–1310

1190–1410

adult

male

no

1020–1220

1040–1290

adult

female

no

1020–1220

1040–1290

adult

female

no

1030–1260

1050–1330

adult

female

?

1030–1250

1060–1330

adult

male

no

640–830

650–890

c.850–950

child

male

yes

this study, Graham-Campbell & Batey 1998, Powell et al. 1991

680–950

680–980

c.850–950

child

male

yes

this study, Graham-Campbell & Batey 1998, Powell et al. 1991

660–870

660–940

c.850–950

child

male

yes

1150–1310

1180–1400

780–1060

860–1190

adult

male

no

810–1120

880–1210

child

?

no

880–1190 1020–1250 780–1030 980–1160

880–1220 1020–1270 770–1050 980–1210

adult adult adult adult

male male male male

yes yes yes yes

male

female

no

late Viking Age late Viking Age late Viking Age late Viking Age

this study, Richards et al. forthcoming

this study, Brothwell pers comm. this study, Richards et al. forthcoming Richards et al. forthcoming this study, Richards et al. forthcoming this study, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming Barrett et al. 2000a, Richards et al. forthcoming this study, Richards et al. forthcoming this study, Brothwell pers comm. this study, Richards et al. forthcoming this study, Richards et al. forthcoming

this study, Melissa Melikian pers comm. Neighbour pers comm. in Barrett et al. 2000b Neighbour pers comm. in Barrett et al. 2000b Owen & Dalland 1999 this study this study

258

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

burials for which both dates and rigorous stable isotope measurements (δ13C and δ15N) are available. An additional 11 adult burials from Newark Bay which have stable isotope data but lack radiocarbon dates have been included to illustrate gender patterns first noted by Richards et al. (in press, see later).

M ETHODS Bone collagen carbon and nitrogen isotope values are indicators of the sources of dietary protein consumed by humans over their lifetimes (Katzenburg 2000; Mays 2000; Sealy 2001). Collagen is the best preserved, and most abundant, bone protein and is routinely extracted for radiocarbon dating and isotopic analysis. There exist a number of criteria that can be used to assess the quality and preservation of the extracted collagen (e.g. C:N ratio, collagen yield, %C and %N in collagen extract: Ambrose 1990; DeNiro 1985) so in most cases we can be reasonably sure that the original in vivo δ13C and δ15N values of the human or animal are being measured. Mammal bone collagen δ13C and δ15N values are long-term averages (e.g. 5–20 years) of dietary protein δ13C and δ15N values (Ambrose and Norr 1993). In the context of this article, human δ13C values are indicators of the amount of marine protein consumed by humans over the few years before they died, with values of δ13C = –21 ± 1‰ indicative of approximately 100% terrestrial protein, and –12 ± 1‰ indicative of approximately 100% marine protein (see later in this article). Nitrogen isotopes generally reflect trophic position, with consumer bone collagen values enriched by 2–4‰ over the protein they consume (Schoeninger and DeNiro 1984). In marine ecosystems there are many more trophic levels than in terrestrial systems, and therefore top-level consumers such as marine mammals have much higher δ15N values (e.g. 15–20‰) compared to top-level terrestrial carnivores from temperate environments (e.g. 9–10‰). In a number of cases, earlier δ13C values measured in association with radiometric (‘conventional’) or AMS radiocarbon dating also exist (see Table 1). Until recently, many ‘conventional’ radiocarbon laboratories measured δ13C values of bone collagen as part of the dating process, as they are necessary for the correct interpretation of the 14C date. As the end goal of this procedure was not dietary reconstruction, these δ13C measurements are not always as accurate and precise as those made specifically for dietary analysis. Therefore, the first group of δ13C measurements cannot be used for dietary reconstruction. It is clear that there are non-trivial differences between the new values published here and those produced in association with earlier AMS dates. The mean difference between the 36 matching assays is 0.8 ± 0.8‰. In this instance the discrepancies are likely to derive from two sources. Firstly most of the values from AMS dating do not have associated C:N ratios with which to assess the original collagen preservation. Many may therefore be affected by diagenesis. Samples with irregular C:N values (e.g. outside the range 2.9–3.6) have been excluded from the new dataset. Secondly the new samples were processed using ultrafilters, following procedures outlined in Brown et al. (1988) which result in better quality collagen extracts, and which were not routinely applied when many of the AMS samples were measured. In

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

259

sum, although indicative of general patterns, the error ranges that probably apply to the δ13C values associated with earlier AMS dates make them poor guides to the diet of single individuals and thus of limited value for marine reservoir correction of radiocarbon dates (see later). They are not used further in the present article. Given that the individuals under consideration consumed marine foods, it is necessary to calibrate the radiocarbon dates by combining atmospheric and marine calibration curves. This process, marine reservoir correction, is necessary because there is an offset of approximately 400 years between radiocarbon dates of contemporary terrestrial and marine carbon – when the latter is derived from the surface ocean depths accessible to pre-industrial societies – (Stuiver et al. 1998). To perform mixed calibration, δ13C values are first used to estimate a per cent marine carbon (and associated error) for each sample. This is done by estimating what δ13C values should be associated with 100 per cent marine and 100 per cent terrestrial diets in the region of interest and then assuming a linear interpolation between them (Ambrose et al. 1997; Arneborg et al. 1999; Barrett et al. 2000a; Chisholm et al. 1982; Richards and Hedges 1999; Richards and Mellars 1998). A local ∆R value must then be chosen to quantify the difference between the global terrestrial-marine offset and local conditions (Ascough et al. 2004; Harkness 1983; Reimer et al. 2002). Per cent marine carbon estimates and ∆R values can then be used with the mixed calibration option of radiocarbon calibration software (e.g. Bayliss et al. 2004:569). Here this has been done using OxCal Version 3.9 (Bronk Ramsey 1995, 2001) and the marine and terrestrial calibration curves of Stuiver et al. (1998). Although superficially routine, marine reservoir correction suffers from uncertainty regarding the marine and terrestrial δ13C endpoints and how much ∆R values vary through space and time. For the purposes of this study, the marine and terrestrial δ13C endpoints are nominally set at –12‰ and –21‰ respectively. The former value matches the mean of eight seal specimens from Newark Bay (–12.0; ± 0.5‰: Richards et al. in press) and is consistent with values for cod family fish from Orkney once diet-collagen offsets are taken into consideration (Barrett et al. 2000a:543). The terrestrial end-point (–21‰) is, after rounding, the value for an inland population from Norway (Johansen et al. 1986:756) and the lowest exhibited in the Orcadian dataset (Table 1). It is also broadly consistent with cattle and caprine samples from Newark Bay that produced mean δ13C values of –21.8 ± 0.2‰ and –21.8 ± 0.8‰ respectively (Richards et al. in press). Humans on a 100 per cent terrestrial diet would share these sheep and cattle values (i.e. c.–22‰) if entirely vegetarian, but should have slightly more positive δ13C figures assuming that they ate a mixed diet which included animal protein, due to the so-called trophic level shift in δ13C values of approximately 1‰ (e.g. Schoeninger and DeNiro 1984). It is difficult to quantify the error associated with these estimates, but ± 1‰ may be appropriate based on the standard deviations associated with the Newark Bay fauna (see earlier) and seven modern cod family fish samples from Orkney (± 0.9‰ at one standard deviation: Barrett et al. 2000a:543). This produces an error of approximately ± 11 per cent around the resulting estimates of per cent marine carbon. Both can be incorporated into the radiocarbon calibration procedure (Bronk Ramsey 1995, 2001).

260

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

Variability in local ∆R values is an equally difficult problem. Based on the University of Washington’s Marine Reservoir Correction Database (Reimer and Reimer 2004), values calculated from known age modern material collected in northern Scotland range from 91 ± 30 to –13 ± 30. However, new work comparing marine and terrestrial samples from secure archaeological contexts by Reimer et al. (2002) suggests that a value of –33 ± 93 may be more appropriate and Ascough et al. (2004) argue that the relevant value could be as low as –79 ± 17 at c.2000 radiocarbon years BP. These lower values imply that the impact of the marine reservoir effect may be less than previously thought. Here we use both –79 ± 17 and 91 ± 30 to explore the degree to which uncertainty regarding the value of ∆R is likely to influence the interpretation of results. This wide range of ∆R will also help account for the possibility that some individuals may be migrants from other regions, such as Scandinavia.

W ESTNESS Westness is a cemetery of 32 burials on the island of Rousay in Orkney (GrahamCampbell and Batey 1998; Kaland 1996; Sellevold 1999). Most burials are unaccompanied inhumations oriented east–west, which are assumed to be preViking Age (Pictish). However, a subset of eight burials was oriented southeast–north-west and contained Scandinavian grave-good assemblages of mid ninth- to mid-tenth-century date (Graham-Campbell and Batey 1998; James Graham-Campbell pers. comm.). These accompanied graves could be dated so finely by typology due to the inclusion of numerous objects and their status as sealed single events. Nine of the ‘Pictish’ and four of the ‘Scandinavian’ burials have both radiocarbon dates and new stable isotope assays and can thus be calibrated. The grave-good and radiocarbon dates do not overlap for one of the ‘Scandinavian’ burials (grave 36, see Table 1). This discrepancy could indicate that grave 36 was interred later in the Viking Age than the grave-goods suggest, but it is more likely a product of the statistical nature of radiocarbon dating. One additional burial (near the naust, or boathouse, away from the cemetery) was a post-medieval or even twentieth-century addition and is not considered further. All data are provided in Table 1 and Figure 2 plots the relationships between calibrated date (two sigma range) or approximate grave-good date (where available) and δ13C value assuming ∆R values of –79 ± 17 and 91 ± 30 respectively. As expected, the unaccompanied burials oriented east–west all have radiocarbon dates most consistent with the pre-Viking Age, Pictish, period. Strikingly, they all have very terrestrial δ13C values (from –21.1‰ to –20.5‰) and δ15N values consistent with relatively modest levels of marine protein consumption. There are no differences between males and females. Despite their maritime environment, on an island of approximately 10 km in length, these individuals clearly chose to base their subsistence strategy predominantly on arable and pastoral farming. The sea was not even used for fodder, as a marine signal would be passed up the food chain via livestock (Barrett et al. 2000a). Given an error range of approximately ± 11 per cent, this does not mean they ate no marine protein (cf. Hedges 2004). It does

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

261

– – – – – – – – – – – – – –

– – – – – – – – – – – – – –

Figure 2. δ13C values and 2σ radiocarbon (or grave-good where available) date ranges for Pictish to Viking Age burials from Westness, Orkney. Each bar represents one skeleton. Solid bars are females, open bars are males and the hatched bar is a child of indeterminate sex. (a) Calibrations using a ∆R of –79 ± 17. (b) Calibrations using a ∆R of 91 ± 30. See Table 1 for further details.

262

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

imply, however, that it is unlikely to have been a significant element of their dietary routine. In contrast, the Viking Age burials show distinctions between the males and females. The two females, both with grave-goods (grave 5 and grave 36), have relatively terrestrial δ13C values (–20.7‰ and –20.6‰ respectively). In contrast, the two males (grave 11, buried in a fishing boat, and grave 12), both with gravegoods, have clearly marine signals (–18.1‰ and –18.2‰ respectively). These two males had high δ15N values (13.7‰ and 14.8‰), indicative of a diet with a fair amount of marine protein. Similarly, a male burial and an unsexed child of approximately six years of age from the Viking Age cemetery of Cnip, in the Western Isles of Scotland, produced δ13C values of –18.7‰ and –18.4‰ respectively (Dunwell et al. 1995; Tim Neighbour pers. comm. in Barrett et al. 2000b). A male (based on grave-goods) child burial of Viking Age date excavated at Balnakeil in the north of mainland Scotland (Powell et al. 1991) had a δ13C value of –19.6‰ which also indicates some consumption of marine protein, but not to the same extent as the individuals from Cnip (Table 1). These differences between Pictish/Viking Age and female/male burials do not apply universally. A Viking Age male also buried in a boat at Scar in the island of Sanday, Orkney (Owen and Dalland 1999), produced the terrestrial δ13C value of –20.3‰ (Table 1). Nevertheless, the data suggest that some Viking Age males buried in explicitly Scandinavian style had far more maritime diets and lifestyles than their Pictish predecessors.

N EWARK B AY Newark Bay is a multi-period chapel and settlement site on Mainland, the largest island in Orkney. It produced over 250 burials during salvage excavations in 1968–1972 and (on a much smaller scale) 1999–2000 (Barrett et al. 2000a; Brothwell 1977; Brothwell et al. 1986; Lowe 2001). In its earlier phases, the chapel may represent one of the earliest Viking Age churches in Orkney. It produced two midtenth-century Anglo-Saxon coins (Brothwell 1977; Morris and Emery 1986:340), and approximately five of the 36 calibrated radiocarbon dates for the site are most consistent with a Viking Age origin (Table 1). Its foundation thus pre-dated or shortly followed the traditional conversion of the Orkney earldom in AD 995 (Barrett 2002). The church and its associated cemetery then continued in use throughout the Middle Ages, providing an unparalleled view of dietary trends from the point at which Westness left off to as late as the sixteenth or seventeenth century. With the exception of one burial (69/104, 69/104A or 69/104B) that included a jet-like bracelet and one (70/28) that may have been associated with an antler comb, none included grave-goods (Don Brothwell pers. comm. 2004). The burials from Newark Bay can be divided into three overlapping groups based on their date and degree of marine protein consumption (Fig. 3). Firstly, the small number of Viking Age examples yielded moderately marine δ13C values of –20.3‰ to –17.9‰. These are all more positive (and thus more marine) than the nine ‘Pictish’ burials from Westness and one case (69/107) has a more marine

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

263

– – – – – – – – – – – – – –

– – – – – – – – – – – – – –

Figure 3. δ13C values and 2σ radiocarbon date ranges for Viking Age to late medieval burials from Newark Bay, Orkney. Each bar represents one skeleton. Solid bars are females, open bars are males and hatched bars are indeterminate. (a) Calibrations using a ∆R of –79 ± 17. (b) Calibrations using a ∆R of 91 ± 30. See Table 1 for further details.

264

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

signal than the Viking Age men from that site. Secondly, a large number of values from a period between the eleventh and fourteenth centuries produce a strong marine ‘spike’. Many exceed –18‰ (c.30% marine protein) and one reaches –15.4‰ (c.60% marine carbon). Thirdly, the small number of late medieval burials return to moderately marine values, between –19.1‰ and –17.8‰. The Newark data as a group demonstrate a strong correlation between δ13C and δ15N (r = –0.98). This observation suggests that the marine foods consumed were predatory species near the top of the marine food chain, such as the large cod family fishes known from zooarchaeology in Orkney (Barrett et al. 1999; Richards and Hedges 1999). The Newark Bay data also demonstrate illuminating gender-based patterns. Firstly, several of the Viking Age burials with marine isotope signatures are of women. This observation could mean either that the practice of fish eating had spread from men to women by the time burial began at Newark Bay (the mid- to late tenth century) or that earlier Viking Age examples of women with marine isotope values have simply not yet been found due to the small sample from Westness. Secondly, although not evident from the Viking Age burials alone, when all the data from the various time periods are combined, men did eat more marine protein, on average, than women at Newark Bay. Not all female δ13C values are predominately terrestrial as was found at Westness, but the larger sample size (including the undated samples) is amenable to statistical testing and there is a significant difference between men and women (Mann–Whitney U = 116.5, p = 0.046).

D ISCUSSION In part these results confirm earlier hypotheses, but they also require new explanations. It does seem likely, as previously suggested, that fishing was not an important economic or social activity in Pictish Orkney. Similarly, it would appear that a more maritime oriented subsistence strategy was introduced to Orkney in the Viking Age, probably by AD 850–950 based on the burials dated by gravegoods (or a little later if these represent first generation immigrants). Conversely, it is new information that this maritime strategy (and perhaps its related world view) can be associated with particular individuals. This implies the introduction of occupational or social differences in Viking Age Orkney, and possibly of a novel division of labour by gender, that had profound impacts on diets. Why, for example, did the occupant of the boat burial at Westness (grave 11) have a far more marine diet than the man interred in a similar boat at Scar? Both were high-status graves associated with weapons. Should we assume, for example, that only one lived a life at sea as a Viking (pirate?) in the strict sense of the word despite equity of status at death? Alternatively, can it be inferred that one was predominately a fisherman by occupation? These questions cannot be fully answered, but are indicative of a very different society from one where diets were similar between individuals. For Newark Bay the questions become yet more complex. Despite generous error margins for per cent marine carbon and the use of alternative ∆R estimates,

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

265

the strongest marine spike probably post-dates AD 1000. Thus it is likely to be a second stage of maritime intensification. It is precisely contemporary with the explosion of marine fish consumption in England, but Orkney lacks the burgeoning urban markets that probably account for this more southerly development (Barrett et al. 2004a). How then can the fish event horizon be explained in this context? Several possibilities are ideologically based. The Westness data imply that fish consumption was not a low-status practice. The two males with marine isotope signatures both had richly accompanied graves. If originally practised by a migrant Scandinavian elite, the eleventh-century surge in fish consumption could thus reflect status emulation and ideological realignment by an indigenous population. In the eleventh century, Norse earls were successfully centralizing a previously heterarchical region (Barrett et al. 2000b:4–9; Thomson 2001:69–87). Conversely, it may be relevant that Newark Bay is one of the earliest known Viking Age church sites in Orkney. Christian fasting practice, which allowed the consumption of fish during Lent and other fast days, was experiencing a renaissance elsewhere in Europe in the tenth and eleventh centuries (e.g. Kornexl 1998; Woolgar 2000). It is conceivable that the earliest converts in Orkney were exceptionally rigorous in their adherence to this doctrine. A further possibility, raised previously on the basis of preliminary evidence, is that the Orcadian fish event horizon was associated with the production of stockfish for export to expanding markets – including England (Barrett 1997). Semi-specialized fish-processing middens did appear in northern Scotland for the first time in the eleventh to the twelfth centuries (Barrett et al. 2000b). It is conceivable that an expansion in fishing to serve export markets led to the consumption of more fish in Orkney, particularly by the fishermen themselves. The fish trade argument would be stronger, however, if there was no increase in marine protein consumption as fish processing expanded, if the increasing catch was only for export (Barrett et al. 2000b:19). As it stands, the growth of fish trade remains a possible rather than definitive explanation. Finally, the potential impact of environmental factors in their widest sense must be considered: the carrying capacity of the terrestrial economy; population growth; the productivity of the sea; and storminess or other limits on sea-going. The end of the first millennium was probably the height of the so-called medieval warm period in north-western Europe (Barber et al. 2003; Dahl-Jensen et al. 1998). It was correspondingly a time of growing agricultural production and population – processes for which there is local evidence in northern Scotland (Dyer 2002:26; Simpson 1997; Thomson 2001:53). The fish event horizon in Orkney may thus mark a period of maritime resource intensification associated with land-based economic and demographic growth. Its end could then be explained by the well-known population and economic crashes of the fourteenth and fifteenth centuries, either in Orkney itself or in distant markets (Thomson 1984). The impact of climate change on fish productivity must be considered, but meaningful conclusions cannot yet be drawn. Cod and herring are arcto-boreal species, thus temperature affects them differently in different parts of their ranges.

266

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

In the North Sea, warmer temperatures depress production, while in northern waters such as the Norwegian and Barents Seas, warm weather increases productivity (Barrett et al. 2004b and references therein). Orkney occupies an intermediate zone where relationships between these variables are difficult to predict. It might be argued that climatic amelioration improved the opportunities for seafaring at this time, but the medieval warm period was a process of several centuries’ duration rather than an event. On present evidence radically improved conditions in the eleventh and twelfth centuries cannot be assumed. In sum, the new stable isotope and radiocarbon dating evidence presented here has helped clarify the chronology of maritime resource intensification in northern Scotland. It began in the Viking Age, with some men and (a little later?) women eating high proportions of marine protein. It then continued, reaching its most extreme level in the fish event horizon of the eleventh to the fourteenth centuries. In the late Middle Ages marine food consumption probably returned to Viking Age levels. These results help clarify our understanding of causation. Identity (as ethnicity and perhaps gender) was probably important in the Viking Age, followed by demographic, economic and possibly religious pressures in the eleventh to the fourteenth centuries. In this later period gender may also have continued to play a role in defining appropriate dietary practices. Subsequently, the importance of fish consumption may have dwindled in the face of population and economic crashes – locally or in distant urban markets.

A CKNOWLEDGEMENTS This study was funded (as part of the Viking Age Transitions Project) by the Social Sciences and Humanities Research Council of Canada and the Innovation and Research Priming Fund of the University of York. The samples and age and sex data for Newark Bay were provided by Theya Molleson of the Natural History Museum, London. Don Brothwell of the University of York provided additional samples and information regarding this site. Samples and site information regarding Westness and Balnakeil were provided by Alison Sheridan of the National Museums of Scotland. The Scar samples were provided by Anne Brundle of the Orkney Museum. The Bu of Cairston sample was provided by Melissa Melikian of AOC Archaeology. James Graham-Campbell kindly provided information regarding the grave-goods from Westness. We would like to thank Ben Fuller, Mandy Jay and Gundula Müldner for useful discussions about the data, as well as assistance in sample preparation.

R EFERENCES AMBROSE, S.H., 1990. Preparation and characterization of bone and tooth collagen for isotopic analysis. Journal of Archaeological Science 17:431–451. AMBROSE, S.H. and L. NORR, 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In J.B. Lambert and G. Grupe (eds), Prehistoric Human Bone: Archaeology at the Molecular Level: 1–37. Berlin: Springer-Verlag.

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

267

AMBROSE, S.H., B.M. BUTLER, D.B. HANSON, R.L. HUNTER-ANDERSON AND H.W. KRUEGER, 1997. Stable isotopic analysis of human diet in the Marianas archipelago, Western Pacific. American Journal of Physical Anthropology 104:343–362. ARNEBORG, J., J. HEINEMEIER, N. LYNNERUP, H.L. NIELSEN, N. RUD and A.E. SVEINBJORNSDÓTTIR, 1999. Change of diet of the Greenland Vikings determined from stable carbon isotope analysis and 14C dating of their bones. Radiocarbon 141:157–168. ASCOUGH, P. L., G.T. COOK, A.J. DUGMORE, J. BARBER, E. HIGNEY and E.M. SCOTT, 2004. Holocence variations in the Scottish marine radiocarbon reservoir effect. Radiocarbon 46:1–10. BARBER, K.E., F.M. CHAMBERS and D. MADDY, 2003. Holocene palaeoclimates from peat stratigraphy: macrofossil proxy climate records from three oceanic raised bogs in England and Ireland. Quaternary Science Reviews 22:521–539. BARRETT, J.H., 1997. Fish trade in Norse Orkney and Caithness: a zooarchaeological approach. Antiquity 71:616–638. BARRETT, J.H., 2002. Christian and pagan practice during the conversion of Viking Age Orkney and Shetland. In M. Carver (ed.), The Cross Goes North: 207–226. Woodbridge: Boydell Press. BARRETT, J.H., 2003. Culture contact in Viking Age Scotland. In J.H. Barrett (ed.), Contact, Continuity and Collapse: the Norse Colonization of the North Atlantic: 73–111. Turnhout: Brepols, Studies in the Early Middle Ages. BARRETT, J.H., R.A. NICHOLSON and R. CERÓN-CARRASCO, 1999. Archaeoichthyological evidence for long-term socioeconomic trends in northern Scotland: 3500 BC to AD 1500. Journal of Archaeological Science 26:353–388. BARRETT, J.H., R.P. BEUKENS and D.R. BROTHWELL, 2000a. Radiocarbon dating and marine reservoir correction of Viking Age Christian burials from Orkney. Antiquity 74:537–543. BARRETT, J., R. BEUKENS, I. SIMPSON, P. ASHMORE, S. POAPS and J. HUNTLEY, 2000b. What was the Viking Age and when did it happen? A view from Orkney. Norwegian Archaeological Review 33:1–39. BARRETT, J.H., R.P. BEUKENS and R.A. NICHOLSON, 2001. Diet and ethnicity during the Viking colonisation of northern Scotland: evidence from fish bones and stable carbon isotopes. Antiquity 75:145–154. BARRETT, J.H., A.M. LOCKER and C.M. ROBERTS, 2004a. ‘Dark Age Economics’ revisited: the English fish bone evidence AD 600–1600. Antiquity 78:618–636. BARRETT, J.H., A.M. LOCKER and C.M. ROBERTS, 2004b. The origins of intensive marine fishing in medieval Europe: the English evidence. Proceedings of the Royal Society B 271:2417–2421. BAYLISS, A., E. SHEPHERD POPESCU, N. BEAVAN-ATHFIELD, C. BRONK RAMSEY, G.T. COOK and A. LOCKER, 2004. The potential significance of dietary offsets for the interpretation of radiocarbon dates: an archaeologically significant example from medieval Norwich. Journal of Archaeological Science 31:563–575. BRONK RAMSEY, C., 1995. Radiocarbon calibration and analysis of stratigraphy: the OxCal program. Radiocarbon 37:425–430. BRONK RAMSEY, C., 2001. Development of the radiocarbon program OxCal. Radiocarbon 43:355–363. BROTHWELL, D., 1977. On a mycoform stone structure in Orkney, and its relevance to possible further interpretations of so-called souterrains. Bulletin of the Institute of Archaeology 14:179–190. BROTHWELL, D., D. TILLS and V. MUIR, 1986. Biological characteristics. In R.J. Berry and H.N. Firth (eds), The People of Orkney: 54–88. Kirkwall: The Orkney Press. BROWN, T.A., D.E. NELSON and J.R. SOUTHON, 1988. Improved collagen extraction by modified Longin method. Radiocarbon 30:171–177.

268

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

CERÓN-CARRASCO, R., 2002. Of fish and men, de iasg agus dhaoine: aspects of the utilization of marine resources as recovered from selected Hebridean archaeological sites. Unpublished PhD thesis: University of Edinburgh. CHISHOLM, B.S., D.E. NELSON and H.P. SCHWARCZ, 1982. Stable-carbon isotope ratios as a measure of marine versus terrestrial protein in ancient diets. Science 216:1131–1132. CHRISTENSEN, P. and A.R. NIELSSEN, 1996. Norwegian Fisheries 1100–1970: main developments. In P. Holm, D.J. Starkey and J. Thor (eds), The North Atlantic Fisheries, 1100–1976: National Perspectives on a Common Resource: 145–176. Esbjerg: North Atlantic Fisheries History Association. DAHL-JENSEN, D., K. MOSEGAARD, N. GUNDESTRUP, G.D. CLOW, S.J. JOHNSEN, A.W. HANSEN and N. BALLING, 1998. Past temperatures directly from the Greenland ice sheet. Science 282:268–271. DENIRO, M.J., 1985. Post-mortem preservation and alteration of in vivo bone collagen isotope ratios in relation to paleodietary reconstruction. Nature 317:806–809. DUNWELL, A.J., T.G. COWIE, M.F. BRUCE, T. NEIGHBOUR and A.R. REES, 1995. A Viking Age cemetery at Cnip, Uig, Isle of Lewis. Proceedings of the Society of Antiquaries of Scotland 125:719–752. DYER, C., 2002. Making a Living in the Middle Ages. London: Yale University Press. ENGHOFF, I.B., 1999. Fishing in the Baltic region from the 5th century BC to the 16th century AD: evidence from fish bones. Archaeofauna 8:41–85. ENGHOFF, I.B., 2000. Fishing in the southern North Sea region from the 1st to the 16th century AD: evidence from fish bones. Archaeofauna 9:59–132. GRAHAM-CAMPBELL, J. and C.E. BATEY, 1998. Vikings in Scotland: an Archaeological Survey. Edinburgh: Edinburgh University Press. HARKNESS, D.D., 1983. The extent of natural 14C deficiency in the coastal environment of the United Kingdom. PACT 8:351–364. HEDGES, R.E.M., 2004. Isotopes and red herrings: comments on Milner et al. and Liden et al. Antiquity 78:34–37. JOHANSEN, O.S., S. GULLIKSEN and R. NYDAL, 1986. δ13C and diet: analysis of Norwegian human skeletons. Radiocarbon 28:754–761. KALAND, S.H.H., 1996. En vikingtidsgård og -gravplass på Orknøyene. In J.F. Krøger and H. Naley (eds), Nordsjøen: Handel, Religion og Politikk: 63–68. Karmøy Kommune: Vikingfestivalen. KATZENBERG, M.A., 2000. Stable isotope analysis: a tool for studying past diet, demography and life history. In M.A. Katzenberg and S.R. Saunders (eds), The Biological Anthropology of the Human Skeleton: 305–327. New York: John Wiley and Sons. KORNEXL, L., 1998. Benedictine Reform. In P.E. Szarmach, M.T. Tavormina and J.T. Rosenthal (eds), Medieval England: an Encyclopedia: 119–120. London: Garland Publishing. LOWE, C., 2001. Newark. Discovery and Excavation in Scotland (new series) 1:66. MAKOWIECKI, D., 2003. Historia Ryb I Rybolówstwa W Holocenie Na Nizu Polskim W Swietle Badan Archeoichtiologicznych. Poznan: Instytut Archeologii i Etnologii Polskiej Akademii Nauk. MAYS, S., 2000. New directions in the analysis of stable isotopes in excavated bones and teeth. In M. Cox and S. Mays (eds), Human Osteology: in Archaeology and Forensic Science: 425–438. London: Greenwich Medical Media. MORRIS, C.D. and N. EMERY, 1986. The chapel and enclosure on the Brough of Deerness, Orkney: survey and excavations, 1975–1977. Proceedings of the Society of Antiquaries of Scotland 116:301–374. OWEN, O. and M. DALLAND, 1999. Scar: a Viking Boat Burial on Sanday, Orkney. Phantassie: Tuckwell Press.

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

269

PERDIKARIS, S., 1999. From chiefly provisioning to commercial fishery: long-term economic change in Arctic Norway. World Archaeology 30:388–402. POWELL, J., J. POWELL, D. LOW and R. GOURLAY, 1991. Balnakeil Bay Viking burial. Discovery and Excavation in Scotland 1991:46. REIMER, P. and R. REIMER, 2004. Marine Reservoir Correction Database. http://radiocarbon.pa.qub.ac.uk/marine/ (accessed on 8 April 2004). REIMER, P.J., F.G. MCCORMAC, J. MOORE, F. MCCORMICK, and E.V. MURRAY, 2002. Marine radiocarbon reservoir corrections for the mid- to late Holocene in the eastern subpolar North Atlantic. The Holocene 12(2):129–135. RICHARDS, M.P. and R.E.M. HEDGES, 1999. Stable isotope evidence for similarities in the types of marine foods used by late Mesolithic humans at sites along the Atlantic coast of Europe. Journal of Archaeological Science 26:717–722. RICHARDS, M.P. and P.A. MELLARS, 1998. Stable isotopes and the seasonality of the Oronsay middens. Antiquity 72:178–184. RICHARDS, M.P., T.I. MOLLESON and B.T. FULLER, in press. Stable isotope palaeodietary studies of humans and fauna from the multi-period site of Newark Bay, Orkney. Journal of Archaeological Science. RITCHIE, A., 1985. Orkney in the Pictish kingdom. In C. Renfrew (ed.), The Prehistory of Orkney: 183–209. Edinburgh: Edinburgh University Press. SAYERS, W., 2002. Some fishy etymologies: Eng. Cod, Norse porskr, Du. Kabeljauw, Sp. Bacalao. NOWELE: North-Western European Language Evolution 41:17–30. SCHOENINGER, M.J. and M.J. DENIRO, 1984. Nitrogen and carbon isotopic composition of bone collagen from marine and terrestrial animals. Geochimica et Cosmochimica Acta 48:625–639. SEALY, J., 2001. Body tissue chemistry and palaeodiet. In D.R. Brothwell and A.M. Pollard (eds), Handbook of Archaeological Sciences: 269–279. Chichester: John Wiley and Sons. SELLEVOLD, B.J., 1999. Picts and Vikings at Westness: Anthropological Investigations of the Skeletal Material from the Cemetery at Westness, Rousay, Orkney Islands. Oslo: NIKU (Scientific Report 010:1–62). SIMPSON, I.A., 1997. Relict properties of anthropogenic deep top soils as indicators of infield management in Marwick, West Mainland, Orkney. Journal of Archaeological Science 24:365–380. STUIVER, M., P.J. REIMER and T.F. BRAZIUNAS, 1998. High-precision radiocarbon age calibration for terrestrial and marine samples. Radiocarbon 40:1127–1151. THOMSON, W.P.L., 1984. Fifteenth-century depression in Orkney: the evidence of Lord Henry Sinclair’s Rentals. In B.E. Crawford (ed.), Essays in Shetland History:125–142. Lerwick: The Shetland Times Ltd. THOMSON, W.P.L., 2001. The New History of Orkney. Edinburgh: Mercat Press. VAN NEER, W. and A. ERVYNCK, 2003. The late mediaeval heyday of the Flemish marine fishery: a fish-eye view. In M. Pieters, F. Verhaeghe, G. Gevaert, J. Mees and J. Seys (eds), Colloquium: Fishery, Trade and Piracy – Fishermen and Fishermen’s Settlements in and around the North Sea Area in the Middle Ages and Later: 40–43. Oostende: Instituut voor het Archeologisch Patrimonium (Rapport 13). WOOLGAR, C.M., 2000. ‘Take this penance now, and afterwards the fare will improve’: seafood and late medieval diet. In D.J. Starkey, C. Reid and N. Ashcroft (eds), England’s Sea Fisheries: the Commercial Sea Fisheries of England and Wales since 1300: 36–44. London: Chatham.

270

EUROPEAN JOURNAL OF ARCHAEOLOGY 7(3)

B IOGRAPHICAL N OTES James H. Barrett is a medieval archaeologist and zooarchaeologist with particular research interests in early historic Scotland, the emergence of intensive sea fishing in Europe and the political economy of ‘peripheries’. His study of maritime societies has also involved him in research on dietary reconstruction using stable isotope analysis and the marine reservoir correction of radiocarbon dates. Address: Department of Archaeology, The King’s Manor, University of York, York, YO1 7EP, UK. [email: [email protected]] Michael P. Richards is a biomolecular archaeologist specializing in the method, theory and application of stable isotope analysis. His research ranges from the Palaeolithic to the Middle Ages and from the Pacific to the Mediterranean. His previous work on Scottish material includes studies of the Mesolithic–Neolithic transition. Address: Department of Human Evolution, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, D-04103 Leipzig, Germany, and Department of Archaeology, University of Durham, South Road, Durham, DH1 3LE, UK. [email: [email protected]]

A BSTRACTS Identité, sexe, religion et économie: nouvelles preuves isotopiques et radiocarbones de l’intensification de l’exploitation des ressources marines en Orkney médiévale, Ecosse, GrandeBretagne James H. Barrett et Michael P. Richards Les analyses par isotopes stables et les données radiocarbones de 54 inhumations de l’Écosse du Nord indiquent l’évolution de la consommation de protéines marines de la fin de l’Âge du Fer jusqu’à la fin du Moyen Âge, et montrent comment une tendance paneuropéenne envers une pêche plus intensive vers la fin du premier millénaire ap. J-C est étroitement liée aux influences environnementales et culturelles locales. Malgré sa situation maritime, on n’avait guère recours à une alimentation marine à Orkney pendant l’âge du fer récent. La consommation de poisson devenait significative pendant l’Âge des Vikings (9e–11e siècle ap. J-C), comme on peut le constater d’abord dans les sépultures de plusieurs hommes enterrés avec du mobilier funéraire de style scandinave, mais bientôt également dans des tombes ‘chrétiennes’ des deux sexes. La consommation de protéines marines atteignit son plus haut niveau, notamment parmi les hommes, du 11e au 14e siècle environ, tandis que par la suite l’importance de la consommation de poisson retomba au niveau atteint pendant l’Âge des Vikings. Ces développements ont probablement été causés par des relations complexes entre ethnicité, sexe, pratique chrétienne du jeûne, croissance de la population, commerce de poissons à longue portée et changements environnementaux. Mots clés: alimentation, Âge du Fer, Âge des Vikings, Moyen Âge, isotopes stables, Orkney, Ecosse, commerce du poisson, ethnicité, sexe

BARRETT & RICHARDS: IDENTITY, GENDER, RELIGION AND ECONOMY

271

Identität, Gender, Religion und Ökonomie: Neue Isotopen- und Radikarbondaten als Belege für die Zunahme der Nutzung mariner Ressourcen auf den frühgeschichtlichen Orkney-Inseln, Schottland, UK James H. Barrett und Michael P. Richards Messungen stabiler Isotope und Radiokarbondaten von 54 Bestattungen aus dem nördlichen Schottland dokumentieren Trends in der Konsumption marinen Proteins von der späten Eisenzeit bis zum Ende des Mittelalters. Sie werfen ein Licht darauf, wie der lokale Beitrag von Unwelt und Kultur mit den pan-europäischen Trends zu einer Intensivierung des Fischfangs am Ende des ersten Jahrtausends AD zusammenhing. In der späten Eisenzeit wurden marine Nahrungsmittel auf den Orkney-Inseln – trotz ihrer maritimen Lage – nur in geringen Umfang genutzt. Eine signifikante Verwendung von Fisch lässt sich erst für die Wikingerzeit (9.–11. Jahrhundert AD) nachweisen: anfangs im Fall einiger Männer, die mit Grabbeigaben in skandinavischem Stil bestattet wurden, doch bald auch bei „christlichen” Gräbern beiderlei Geschlechter. Vom 11. bis zum 14. Jahrhundert lässt sich dann ein Höhepunkt in der Nutzung von marinem Protein, besonders bei männlichen Individuen, festhalten. Danach geht die Bedeutung des Verzehrs von Fisch auf das Niveau der Wikingerzeit zurück. Die Gründe dieser Entwicklungen liegen möglicherweise in der komplexen Beziehung zwischen kultureller Zugehörigkeit, Gender, christlichen Fastenpraktiken, Bevölkerungswachstum, dem Fischhandel über lange Strecken und in Veränderungen der Umwelt. Schlüsselbegriffe: Nahrung, Eisenzeit, Wikingerzeit, Mittelalter, stabile Isotope, Orkney-Inseln, Schottland, Fischhandel, kulturelle Zugehörigkeit, Gender