2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 119 Introduction An initial isotope diet study (Arneborg et al. 1999, Lynnerup 1998) indicated that the subsistence economy in Norse Greenland shifted from a predominantly terrestrial to a predominantly marine diet. This change is in accordance with the theories drawing upon the animal-bone record (McGovern 1985). This first isotope study was, however, limited. It included only a few samples, all of which were of human bone, and only the δ13C values were measured. The present study includes 123 samples from human and 213 from animal bones, evenly distributed between both domesticated and wild animals, and measurements of both δ13C and the δ15N (Nelson et al. 2012a, Nelson et al. 2012c, Nelson et al. 2012d [all this volume]). The samples are from 19 Norse sites (farms): 13 are from the Eastern Settlement (Fig. 1) and 6 are from the Western Settlement (Fig. 2). The sites are described in detail in Arneborg et al. (2012a [this volume]), and sampling strategies are described in detail in Arneborg et al. (2012a), Nelson et al. (2012a), Nelson et al. (2012b), Nelson et al. (2012c), Nelson et al. (2012d) (all this volume). Apart from the obvious question of whether the new study confirms the observed shift in the Norse diet, we also posed questions concerning how this shift occurred. Was it a gradual change or did it happen within the space of a few years? Did the shift reflect altered farming strategies? Were there differences between the two settlement areas, between farms, and between the sexes? What kind of marine food was consumed? The origin of the sampled humans was also considered. Were the bones of immigrants or of people who had grown up in Greenland? This paper is a summary of the archaeological results that can be deduced from the preceding technical papers discussing Norse Greenland diet on the basis of δ13C and the δ15N values of both human and animal samples. In Nelson et al. (2012a, 2012c [this volume]) the isotopic values of the terrestrial and wild animals of importance to the Norse dietary economy are presented and discussed, and the conclusion is that there is sufficient variation within each group to require detailed consideration in interpreting isotopic information on the humans. In Nelson et al. (2012b [this volume]), we present a small study on the Greenland Thule Culture to test the use of the isotopic method on Greenlandic material. All Thule Culture samples are from humans from before the recolonization of Greenland in 1721 and are thought to come from people that had a almost 100% marine diet (Gulløv 2012 [this volume]). Nelson et al. (2012c [this volume]) present the study on the Norse animal husbandry, and the carbon isotope values for the herbivores of dietary importance to the Norse were, in general, as expected for Human Diet and Subsistence Patterns in Norse Greenland AD c.980–AD c.1450: Archaeological Interpretations Jette Arneborg1,2,*, Niels Lynnerup3, and Jan Heinemeier4 Abstract - In this concluding paper of the JONA special volume on the Norse Greenland isotope study, we summarize the archaeological interpretations of the previous, technical papers. The study supports the conclusions and widens the results of our earlier limited study, i.e., that the diet of the Norse Greenlanders became more dependent on marine resources over time. The isotope data provide information at the level of the individual, and the study indicates that the Norse Greenlanders had an isotopically varied diet; there is no evidence that these differences were linked to sex or age. The shift in diet seems to have happened gradually, perhaps beginning during the initial settlement. The swiftness of the change, however, depended on where the immigrants settled; settlers in the southern part of the Eastern Settlement and in the Western Settlement may have adapted to the marine resources more rapidly than those in the central Eastern Settlement region. Social differences may partly explain the isotopically varied diet within Norse society; this explanation is, however, not without its reservations. Despite the changes in the dietary economy, and the increasing dependence on the marine resources, farming strategies remained unchanged. Climate change and unsustainable land-use practices have been proposed as two of the main reasons for the depopulation of the Norse Greenland settlements in the late 1400s, and it is obvious to draw attention to these factors when trying to explain the changes in the dietary economy. It is, however, more doubtful whether the environmental changes were, after all, the sole cause of the depopulation of the Norse Greenland settlement. The Norse Greenlanders apparently adapted well to their physical environment, and they could survive on the marine resources in as far as they were available. Special Volume 3:119–133 Greenland Isotope Project: Diet in Norse Greenland AD 1000–AD 1450 Journal of the North Atlantic 1Danish Middle Ages and Renaissance, Research and Exhibitions, The National Museum of Denmark Frederiksholms Kanal 12, DK-1220 Copenhagen, Denmark. 2Institute of Geography, School of GeoSciences, University of Edinburgh, Scotland, UK. 3Laboratory of Biological Anthropology, Section of Forensic Pathology, University of Copenhagen, Copenhagen, Denmark. 4AMS 14C Dating Centre, Department of Physics and Astronomy, Aarhus University, Ny Munkegade 120, DK-8000 Aarhus C, Denmark. *Corresponding author - jette.arneborg@natmus.dk. 2012 120 Journal of the North Atlantic Special Volume 3 animals living in a C3 environment, and the dataset provides a solid basis for isotopic dietary analyses of the Norse themselves. In Nelson et al. (2012d [this volume]) the measures of the stable carbon (δ13C) and nitrogen (δ15N) values of human bone collagen have been made for 80 individuals from an existing collection of Norse skeletal material. These data are interpreted with the aid of the data obtained for the wild fauna, for the Norse domestic animals, and for a number of Thule Culture individuals of about the same time period. Our study shows that the isotope data provide information at the level of the individual. The study indicates, furthermore, that the Norse had an isotopically varied diet, which must reflect dietary differences within Norse society. However, there is no evidence that these differences were linked to sex or age (Nelson et al. 2012d [this volume]), whereas chronological, geographic, and social differences are aspects that should be further pursued. Furthermore, there was, on the face of it, a greater relative consumption of marine protein in the Eastern Settlement compared to that of the Western Settlement (Nelson et al. 2012d [this volume]). This difference too should be examined in more detail. Only what were described as “robust” samples were included in the previous technical chapter on Norse dietary analysis (Nelson et al. 2012d [this volume]). This approach was taken because the validation of the method itself was of prime consideration. In this chapter, samples were designated “suspect” if they gave a low collagen yield or were suspected of containing possible remnants of preservatives (Nelson et al. 2012d [this volume]). However, as discussed in Nelson et al. (2012d [this volume]), identical isotopic results were obtained by two different laboratories with very different procedures and a focus on preservative removal. Therefore, in the following discussion all samples are included—with, of course, due account being taken of the problems raised in the preceding chapters. Furthermore, in Nelson et al. (2012d [this volume]) the Tjodhildes church material was omitted as possibly being from immigrants (i.e., from Icelanders) and therefore not suitable for the validation of the method of evaluating Greenland human dietary resources based on analysis of the local faunal isotope values. Again, in the following, the Tjodhilde samples are included in a belief that all samples should be studied in their own right and that archaeologically we cannot rule out beforehand the possibility that this small church served more than one generation of immigrant Norse Greenlanders. Figure 1. Map of the Norse Eastern Settlement with the sites included in the study. White is the inland ice, blue is the sea, and yellow is the land. The individual sites are described in detail in Arneborg et al. (2012a [this volume]). 2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 121 Chronology: Diet over Time All samples included in the study are presented in Arneborg et al. (2012a [this volume]). Forty-three human samples have been AMS dated at the AMS 14C Dating Centre at Aarhus University; 30 are from the Eastern Settlement and 13 are from the Western Settlement. In addition to these, AMS-dated human samples from unpublished archaeological excavations at churches Ø1, Ø23, Ø35 and Ø481 have been added to the dataset, as have the 1982 Anavik samples and human fragments found at the Vatnahverfi farm Ø167 (Arneborg et al. 2012a: tables 3–17 [this volume], Lynnerup et al. in Vebæk 1992:64ff.) (Table 1). An additional six samples from the Sandnes churchyard in the Western Settlement (samples # 184, 247, 249, 250, 254, 255) have been dated archaeologically (Arneborg et al. 2012a: table 13 [this volume]), as have one of the samples from Tjodhildes church (sample #165; Arneborg et al. 2012a: table 4 [this volume]) and seven samples (samples #220, 221, 222, 224, 226, 227, and 228) from Ø149, Narsarsuaq (Arneborg et al. 2012a: table 10 [this volume]). Figure 3 shows δ13C values for all the dated human samples sorted in time (cf. Arneborg et al. 2011a: tables 3–17 [this volume]), and the distribution demonstrates that the amount of marine protein in the Norse diet increased through time. With a few exceptions, the estimated mid-point value (δ13C= -16.3‰) between the endpoints δ13C= -19.2‰ and -13.4‰ for 100% terrestrial and 100% marine food protein, respectively (cf. Nelson et al. 2012d [this volume]), was crossed in the second half of the 13th century (Table 1). Because of the observed tight correlation between δ15N and δ13C values for the Norse humans (Nelson et al. 2012d [this volume]), we have not made similar plots of the time development in the δ15N values in Figure 3 and the following figures. Locals or Immigrants? The increasing marine element in the Norse diet is noticeable. There is, however, some levelling off from AD 1300 onwards (Fig. 3). We cannot, however, rule out the possibility that some of the samples, especially in the early period, may derive from immigrants, especially those from Ø29a Brattahlid– Tjodhildes Church in Qassiarsuk (Arneborg et al. 2012a [this volume]), which we have touched upon already. Forming the majority of the early Figure 2. Map of the Norse Western Settlement with the sites included in the study. White is the inland ice, blue is the sea, and yellow is the land. The individual sites are described in detail in Arneborg et al. (2012a [this volume]). 122 Journal of the North Atlantic Special Volume 3 samples, the Tjodhilde samples play an important role in the discussion about changing dietary habits. The Tjodhilde samples have a more terrestrial signature than the later samples in the Greenland dataset, and this finding, together with the early date for the church, has been one of the main arguments for categorizing the samples as possibly being from immigrants (Nelson et al. 2012d [this volume]). A further indication of at least some immigrants being present at Tjodhilde is the rather diverse oxygen isotope composition in the dental enamel of some of the humans buried there (Fricke et al. 1995), indicating different climatic early life conditions. The number and density of graves in the Tjodhilde churchyard was relatively high compared to the other early churches (but not when compared with the late medieval churchyards [Lynnerup 1998]). It appears from the archaeological excavations that, in some places, the burials were in at least two layers (confirmed by Svend Erik Albrethsen, Heritage Agency of Denmark [KUAS], Copenhagen, Denmark, who participated in the excavations; pers. comm. to J. Arneborg) with later burials disturbing older—and presumably forgotten— graves. Most of our samples are from intact skeletons, indicating that they were the last (or the first and only) to be buried on the specific spot where they were found. When compared with Icelandic samples (Tables 1 and 2, and Fig. 4) from the same period, the Tjodhilde samples are slightly more marine than those from contemporaneous Icelanders. In fact, only two Greenlandic samples in our dataset are as “terrestrial” as the bulk of the Icelandic values, and none of these are from Tjodhilde’s Church. Three scenarios for the interpretation of the isotope signal of each of the Tjodhilde Church individuals may be possible: 1) The individual was born and raised in Greenland, and belonged to an early generation of Norse Greenlanders that shifted to a slightly more marine diet than that of their Icelandic predecessors; 2) The individual was among the immigrants from Iceland Table 1. Summary of all AMS-dated human samples from Greenland in the present study. For complete information on the individual dates, please see Arneborg et al. (2012a [this volume]: tables 2–16). The quoted calibrated ages have been marinereservoir corrected based on δ13C values as described in Arneborg et al. (1999). For clarity, the calibrated ages are given as intercept or intercepts with the calibration curve (i.e., the most probable single calendar years) and the intercept intervals (in brackets) corresponding to ±1 sigma in the measured 14C age. Calibrated age intercept(s) and Project ID Lab ID δ13C (‰) VPDB δ15N 1 sigma range (in brackets) Tj#18 AAR-1275 -18.50 - 976 (894–996) Tj#28 AAR-1571 -18.00 - 985 (909–1017) Ø35#a AAR-7882 -19.87 7.68 1002 (983–1022) Tj#11 AAR-1267 -18.00 12.18 1020 (995–1043) Ø35#c AAR-7884 -17.68 12.24 1022 (1003–1033) Ø35#b AAR-7883 -16.74 14.16 1026 (1009–1042) Ø48#a AAR-7879 -20.00 7.99 1037 (1025–1155) V51#197 AAR-5257 -16.70 15.43 1038 (1021–1151) V51#240 AAR-5258 -16.40 15.28 1045 (1030–1116) Tj#12 AAR-1268 -17.60 12.78 1065–1115 (1028–1171) Ø48#b AAR-7880 -17.30 12.97 1124–1153 (1042–1169) Tj#25 AAR-1568 -18.60 11.35 1165 (1046–1218) Tj#16 AAR-1272 -18.90 11.43 1169 (1061–1222) Tj#27 AAR-1570 -16.80 - 1172 (1063–1227) Tj#26 AAR-1569 -19.00 - 1175 (1061–1226) Tj#19 AAR-1276 -18.00 - 1192 (1122–1228) Ø35#d AAR-7881 -17.75 12.75 1213 (1167–1250) V7#175 AAR-5404 -17.80 14.28 1219 (1186–1261) Ø47#20 AAR-1437 -16.50 15.31 1233 (1170–1281) Ø47#22 AAR-1439 -18.70 14.01 1272 (1223–1290) Ø167 K-5889 -19.10 - 1275 (1265–1285) Ø111#205 AAR-6167 -16.20 16.94 1285 (1260–1295) Ø149#214 AAR-6147 -15.10 17.52 1290 (1280–1305) Ø149#216 AAR-6149 -14.20 17.54 1290 (1270–1305) V51#258 AAR-5261 -15.50 16.38 1294 (1285–1303) Ø111#208 AAR-6130 -15.50 16.42 1295 (1285–1305) Ø111#210 AAR-6131 -15.40 16.79 1295 (1285–1305) Ø47#21 AAR-1438 -17.60 - 1295 (1256–1392) V51#253 AAR-5259 -16.60 14.54 1296 (1287–1305) V51#1 AAR-1143 -15.10 15.33 1297 (1275–1317) Ø66#24 AAR-1442 -17.10 14.72 1297 (1279–1442) V51#256 AAR-5260 -16.60 14.58 1299 (1291–1309) V7#a K-4117 -16.60 - 1299 (1283–1323) Ø23#a AAR-8590 -16.20 16.14 1299 (1288–1314) V51#3 AAR-1145 -16.30 14.85 1301 (1282–1322) V51#5 AAR-1147 -16.50 14.89 1301 (1284–1320) Ø149#215 AAR-6148 -15.90 16.43 1305 (1290–1325) V51#6 AAR-1148 -15.80 15.40 1307 (1290–1328) Ø1#b AAR-8586 -16.50 15.00 1308 (1299–1324) Ø111#206 AAR-6128 -14.70 16.74 1320 (1300–1385) Ø111#207 AAR-6129 -15.40 16.35 1320 (1305–1390) Ø23#b AAR-8589 -15.60 16.07 1320 (1304–1388) Ø149#9 AAR-1265 -16.10 15.29 1322 (1301–1399) Ø149#213 AAR-6146 -15.30 16.63 1340–1390 (1320–1405) Ø149#8 AAR-1264 -14.70 17.39 1389 (1312–1414) V51#4 AAR-1146 -14.10 15.66 1390 (1323–1412) Ø66#23 AAR-1441 -15.80 - 1392 (1279–1317) V7#174 AAR-5403 -16.60 15.27 1394 (1323–1407) Ø149#10 AAR-1266 -16.20 16.11 1399 (1325–1418) V7#b K-4120 -15.00 - 1403 (1329–1427) Ø149#7 AAR-1263 -15.60 14.79 1404 (1329–1428) V51#2 AAR-1144 -15.10 15.18 1408 (1390 – 1428) Ø111#13 AAR-1269 -14.40 17.54 1418 (1329–1456) Ø1#a AAR-8585 -14.90 17.45 1426 (1412–1438) Ø111#15 AAR-1271 -16.60 15.57 1430 (1407–1447) Ø111#14 AAR-1270 -16.20 15.60 1437 (1413–1467) Ø23#c AAR-8591 -16.50 16.07 1448 (1436–1469) 2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 123 reflecting a rapid shift to a more marine diet after they settled in Greenland, but only partly reflected in their remains, due to slow bone collagen turnover which is estimated to be 5 to 20 years, depending on bone element, compactness, and age of the individual (for a full discussion see Jørkov et al. 2009 and references therein); or 3) The individual came from the West Fjords in Iceland, where a particularly marine diet was predominant (Sveinbjörnsdóttir et al. 2010). At least one of the samples in our dataset is definitely from an immigrant: one of the Greenlandic bishops who were buried at Gardar (Arneborg et al. 2012 [this volume]). As a high-status person, the bishop may have had a predominantly terrestrial diet. However, all the bishops in Norse Greenland were immigrants (Arneborg 1991), and therefore the isotope data most likely reflect the bishop’s diet during the time spent in his homeland of Norway (Fig. 3, Ø47#22). Apart from the bishop, another sample (Fig. 3, Ø167) deviates from the main trend. This sample is from the cranial remains of a 20–25-year-old male that were found in the passage of the centralized farm at the Vatnahverfifarm Ø167 (Arneborg et al. 2012a [this volume], Vebæk 1992:64). This is indeed an extraordinary place to find human remains and is difficult to explain. For all the other samples, with the exception of their date, our isotope study does not give convincing evidence on which to conclude whether they derive from immigrants or foreigners, and even given our reservations concerning the uncertainty of the provenance of some of the early samples, the apparent tendency for the dietary economy in Greenland to become more dependent on the marine resources over time is maintained. Marine Consumption: Eastern Settlement Compared to Western Settlement Without the chronological perspective, both the δ13C and the δ15N averages indicate that there was a greater relative consumption of marine protein in the Eastern Settlement compared to the Western Settlement (Nelson et al. 2012d [this volume]). When the chronological perspective is included, developments in the dietary economy of the two settlements are more or less identical (Fig. 5). The differences in δ13C are insignificant from c. A.D. 1300 onwards, and the apparently more marine diet in the Eastern Settlement can be explained in terms of the chronology. The majority of the late marine samples are all from the Eastern Settlement. Taking a closer look at our Eastern Settlement samples, there appear to be some geographical differences within the settlement (Fig. 6): a central region group, including the samples from inner-fjord Tunulliarfik and inner-fjord Igaliku fjord where Figure 3. δ13C values in all AMS-dated human samples. Time progression is left to right. Sample Tj#18 is AMS-dated AD 976 (894–996). Sample Ø111#14 is AMS-dated AD 1448 (1436–1469). See also Table 1. For a summary plot of all human δ15N versus δ13C values see Nelson et al. (2012d [this volume]). 124 Journal of the North Atlantic Special Volume 3 Table 2. Icelandic samples (from Sveinbjörnsdóttir et al 2010). The quoted calibrated ages have been marine-reservoir corrected based on δ13C values as described in Sveinbjörnsdóttir et al. 2010. Calibrated age 1 sigma range Site Id Sex Age 14C Age (BP) (68.2% probability) δ13C (‰) δ15N (‰) Lab ID Northern Iceland Brimnes vid Dalvik, Eyjafjardarsýsla, Nordurland DAV-A-9/6039 F 20 1150 ± 35 978–1027 -18.59 11.80 AAR-5860 Sílastadir í Glæsibæjarhreppi, Eyjafjadarsýsla SSG-A-1/1.10´47 II M Adult 1238 ± 35 810–950 -19.33 10.94 AAR-5863 Audbrekka, Hörgárdalur HFH-B-2, K83 F c. 20 949 ± 28 1216–1261 -18.15 13.05 AAR-5916 HFH-B-3, K84 F Adult 861 ± 36 1224–1269 -19.35 9.11 AAR-5917 Keldudalur KEH-A-05 F 50+ 1011 ± 37 1040–1160 -19.22 12.40 AAR-9234 KEH-A-08,K8 M 35–50 1065 ± 50 980–1150 -19.65 12.92 AAR-9235 KEH-A-20,K20 M 35–50 973 ± 39 1050–1210 -19.41 12.92 AAR-9236 KEH-A-07,K7 F? 50+ 1055 ± 50 980–1150 -19.86 11.21 AAR-9237 KEH-A-06,K6 F 20–35 1027 ± 49 1020–1160 -19.74 11.83 AAR-9238 KEH-A-11,K11 M 35–50 949 ± 37 1040–1160 -20.28 14.52 AAR-9239 KEH-A-07,K7 M? 50+ 1004 ± 29 1020–1160 -19.90 11.80 AAR-9240 KEH-B-08,K43 M 50+ 1054 ± 37 1020–1160 -19.18 11.44 AAR-9241 KEH-A-22,K22 M 50+ 1110 ± 42 970–1045 -19.24 12.70 AAR-9242 KEH-A-13,K13 M 50+ 1153 ± 50 895–995 -19.76 11.69 AAR-9243 KEH-A-28, K28 M 20–35 1265 ± 55 890–1010 -17.33 14.40 AAR-9244 KEH-B-16,K51 F 35–50 1135 ± 39 900–1020 -19.55 12.01 AAR-9245 KEH-B-10,K45 F 35–50 988 ± 42 1040–1170 -19.46 12.21 AAR-9246 KEH-B-07,K42 12–15 1156 ± 38 900–1020 -19.08 13.61 AAR-9247 KEH-B-15,K50 F? 1116 ± 44 900–1030 -19.58 10.90 AAR-9248 KEH-A-02,K2 F 20–35 1010 ± 39 1040–1170 -19.04 12.02 AAR-9249 KEH-A-15,K15 8–10 952 ± 38 1050–1220 -19.55 11.85 AAR-9250 KEH-A-29,K29 M? 15–20 979 ± 45 1040–1190 -19.41 11.48 AAR-9251 Kuml-01 F 1220 ± 30 780–940 -19.82 12.50 AAR-9252 Kuml-04 M 1150 ± 49 900–1030 -18.91 12.50 AAR-9253 Kuml-03 1148 ± 36 895–990 -19.86 12.00 AAR-9254 NW Peninsula Vatnsdalur, Patreksfjordur VDP-A-2 1330 ± 42 870–990 -16.76 13.80 AAR-5865 VDP-A-3 M 1320 ± 35 900–985 -16.37 15.50 AAR-5866 VDP-A-4 F 1263 ± 28 976–1022 -16.55 14.20 AAR-5867 VDP-A-5 F 1289 ± 37 890–975 -17.39 13.60 AAR-5868 VDP-A-6 M 1308 ± 30 810–940 -17.98 12.10 AAR-5869 VDP-A-7 M 1259 ± 26 895–985 -17.72 12.90 AAR-5871 Eastern Iceland Brú á Jökulsdal, Nordur-Múlasýsla BAJ-A-1, H149 F 30+ 1145 ± 34 890–1010 -19.63 9.40 AAR-5874 Straumur, Tunguhreppi, Nordur-Múlasýsla STT-A-2 1135 ± 35 960–1030 -19.21 9.38 AAR-5875 Vad í Skriddalur, Sudur Múlasýsla VAS-A-1, H148 M 30+ 1060 ± 35 1010–1150 -19.37 8.86 AAR-5877 Southern Iceland Skeljastadir í Thjórsárdal Grav 39d F 25+ 1075 ± 35 984–1032 -19.86 10.46 AAR-5880 Grav 41as M Adult 1094 ± 27 991–1021 -19.73 7.79 AAR-5883 Grav 16s F 907 ± 31 1155–1220. -20.10 8.38 AAR-5886 Grav 60s M 1058 ± 29 1010–1150. -19.35 7.44 AAR-5887 Grav 47 M Adult 922 ± 28 1155–1220 -19.78 6.89 AAR-5888 Grav 38 M Adult 1154 ± 66 890–1020 -19.51 6.95 AAR-5890 Grav 12s F Adult 1183 ± 40 895–990 -19.15 7.55 AAR-5891 Skálholt Bishop Páll M 40+ 918 ± 28 1165–1220 -19.54 9.98 AAR-5908 Akbraut Sudurland F ? Adult 905–1020 -19.97 AAR-5918 Skansinn, Vestmannaeyjar SVE-A-1 M 1210 ± 26 870–970 -19.68 8.00 AAR-5870 2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 125 towards an increasing dependence on marine resources through time is pronounced, whereas in the southern group the marine component is relatively steady throughout the sampled period; the samples most of our samples are from, and a southern group, including the samples from mid-fjord farm Ø149 and coastal farm Ø111 (Arneborg et al. 2012a [this volume]). In the central region group, the tendency Figure 4. δ13C values of Greenland samples compared with Icelandic samples (Sveinbjörnsdóttir et al. 2010). Tjodhilde’s Church samples are marked in green diamonds. Age increases from right to left. The earliest sample from Iceland is dated to AD 968 (894–991), see Table 2. The latest sample from Greenland is dated to AD 1448 (1436–1469), see Table 1. Figure 5. δ13C values of Greenland Eastern Settlement samples compared with Western Settlement samples. Age increases from right to left. 126 Journal of the North Atlantic Special Volume 3 are basically more marine than those from the central Eastern Settlement. Unfortunately, we only have samples for the southern group from the second half of the 13th century onwards. It is, however, worth noting that the southern samples are more marine compared to contemporary central region samples, indicating that from early times2 the economy of the southern Eastern Settlement farms perhaps depended more on marine resources than that of their kinsmen to the north. From the Western Settlement, the two Sandnes samples (Figs. 1 [V51#197], 3 [V51#240]) from the early period indicate that, in the early settlement period, the subsistence economy of the Western Settlement depended almost equally on animal husbandry and on hunting and fishing. Over time, the dietary economy changed in all three regions. Norse Greenlanders Compared to other Viking Age and Medieval Populations in Northern Europe Compared to other European populations in the North Atlantic and the Baltic (data from: Barrett and Richards 2004:261, Jay and Richards 2006, Kosiba et al. 2007:404, Linderholm et al. 2008:451, Müldner and Richards 2007:166, Privat and O’Connell 2002:785, Reitsema et al. 2010:1417, Sveinbjörndóttir et al. 2010), it is immediately apparent that none of the other populations have as marine a signal as the Norse Greenlanders; also, the span between lowest and highest value is wider for Greenland than for the others (with the reservation that only mean values were available for Birka and Ridanas, Sweden), indicating a more varied diet within the Norse Greenlandic society. It seems that the most comparable populations are Iceland, Ridanas on Gotland, Sweden, and Newark Bay on Orkney, UK (combining both Viking Age and Medieval Age for the latter), which is probably to be expected (Fig. 7). Especially in the first settlement period (ca. 980–1160), the Greenlanders are very much like the populations in both Iceland and the UK, though less marine than were the Viking Age population at Newark Bay. In period II (ca. 1160–1300), the dependence on the marine resources increased in Greenland at the same time as differences within the Greenlandic population increased (see also Fig. 3). In the last period from ca. 1300 onwards, the Greenlanders depended more on the marine resources than did any of the other above-mentioned populations. At the same time, the differences within the Norse Greenlandic population decreased drastically. Social Differences within Norse Greenland Society The dietary differences seen in our Greenlandic dataset may reflect either the location of the farm, i.e., coastal or inland, from where the sampled individual came or social differences within Norse society (Nelson et al. 2012d [this volume]); these underlying details are, however, almost impossible to reveal in our dataset when dealing with samples from the later parish churches. The small landnam-churches may be another matter because they only served an extended household group (family, slaves, and servants). In our dataset, the samples from Tjodhilde’s Church (Ø29a), Ø35, and Ø48 are from small landnam family churches. The anthropologist J. Balslev- Jørgensen (2001) and dentists V. Alexandersen and F. Prætorius (2003) argue that the Tjodhilde’s churchyard was divided up socially, with lower-ranking people buried on the north side of the church and people of high-status on the south side. This conclusion contrasts with that of Lynnerup (1998), who did Figure 6. δ13C values of Central Eastern Settlement samples compared with samples from the southern part of Eastern Settlement. Age increases from right to left. 2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 127 Later churches are presumed to have acted as public or parish churches, and the samples from these churches arise from a variety of farms of different social status and location in the landscape. Accordingly, we cannot discuss diet in relation to provenance and social position on the basis of our samples from these later churches. A few characteristic samples, however, give rise to considerations about origin. For instance, the 20–25-year-old female (V51#4) (Fig.3) who was buried in the Sandnes churchyard in the late 1300s could very well come from the low-status farm at nearby Niaquusat. More than 80% of her diet had come from the sea, and this proportion corresponds well with the amount of seal bones collected from the late-phase midden located in front of the farmhouse there (Fig. 8; McGovern 1985:115). Graves inside churches indicate high status, and at first, the three samples from the episcopal residence (Figs. 1 and 3, Ø47#20, Ø47#21, Ø47#22— all relatively terrestrial) from the north chapel were considered to come from high-status people. The bishop (Ø47 #22), of course, had the required not find physical anthropological evidence for social differences. Although not at all statistically signifi- cant, but relevant for the general discussion on diet and social differences, a 25- to 30-year-old female in one of the north-side graves, displaying evident traces of having used her teeth as a tool, had subsisted on the same protein diet as one of the presumed highstatus young males in the common grave on the south side of the church. The example does not, however, really provide an answer to the question of dietary social differences. The woman may not have been buried on the north side of the churchyard because she was a servant but because she was a woman. Or being a servant, and having lived under the same roof as her master, her diet may have come from the same sources, although perhaps from different parts of the animal. Tjodhilde´s church at Brattahlid served a family group including servants/slaves and tenants, and the diversity seen in the samples may reflect this extended household, where some of its members, slaves or tenants, formed separate households with diets reflecting either their social status and/or the location of their home. Figure 7. Composite graph showing the Norse Greenland δ13C values (for the three settlement periods, conf. text) compared to values from Iceland (Sveinbjörndóttir et al. 2010); Giecz, Poland (Linderholm et al. 2008:451); Birka, Sweden (Reitsema et al. 2010:1417); Ridanas, Gotland (Kosiba et al. 2007:404); Fishergate, York (Müldner & Richard 2007 :166); Wetwang Slack, East Yorkshire (Jay and Richards 2006); Beringsfield, Oxfordshire (Privat and O’Connell 2002:785); Newark Bay, Orkney, Viking Age and Medieval Age (Barrett and Richards 2004:261). Vertical lines denote span between lowest and highest value; black crossbars denote mean values (when given in the publication). 128 Journal of the North Atlantic Special Volume 3 sity within the chronological groups can partly be explained by social and/or geographic differences. Among the domesticates, the zooarchaeological record shows that cattle were kept primarily for their secondary products of milk and cheese. This finding is in accordance with traditional Scandinavian dietary habits where milk was turned into long-life and storable dairy products. Sheep and goats were kept for both their secondary products (wool and milk) and their meat (McGovern 1985:102f.). Pigs, which in traditional Scandinavian societies were popular high-status source of meat, are represented in the animal-bone record only for the first centuries of settlement. Characteristically, most of the bones are from high-status farms such as Gardar (Ø47). Pigs seem to disappear around A.D. 1300 (McGovern et al. 2009:216). Apart from pigs, cattle were also prestigious, and there was a correlation between cattle, good pastures, and highstatus farms (Arneborg et al. 2012a [this volume]). In Greenland, on medium- and small-sized farms, goats and sheep replaced cattle over time (McGovern 1985). However, the very fact that even small sites with very limited pastures, for instance coastal Niaquusat (V48), had cattle shows the importance the Norse Greenlanders placed on dairy products. The zooarchaeological record also shows that seal was the prime meat supplier for the Norse Greenlandic households (McGovern 1985). Fish dignity to be buried in a chapel. The two other samples (a male aged 30–35, Ø47#20 and an 18/20–30- -year-old female, Ø47#21) are more problematic (discussed in Arneborg et al. 2012a [this volume]), and we cannot conclude with certainty that they represent high-status people. Consequently, neither can we say whether the two relatively terrestrial Gardar samples reflect chronology, geography, or social status. However, we do note that the differences in diet within the Norse Greenlandic society were particularly pronounced in the period from settlement to about A.D. 1300, and seemingly decreased thereafter (as apparent from the span in δ13C values; Fig. 7), although a possible effect of sampling bias should be taken into consideration (Fig. 6). Norse Diet: Subsistence and Economy Based on the human samples, our isotope study shows that the basic Norse dietary (subsistence) economy depended on a combination of animal husbandry and hunting, especially of marine mammals to a varying degree and also of reindeer. Through time, the dependence on hunting, especially of marine resources, increased. For the first generations of Greenlanders, the percentage of the marine protein was between 15 and 50%, increasing to between 50 and 80% in the last settlement period. (Nelson et al. 2012d [this volume]). As discussed above, the diver- Figure 8. The distribution of animal bones by species from three Western Settlement farms: 1) Niaquusat, a low-status farm on the coast settled from landnam to about 1350, 2) GUS, The Farm beneath the Sand, a middle-sized inland farm settled from landnam to about 1400, and 3) the high-status coastal farm Sandnes (Kilaarsarfik) settled from landnam to about 1400. 2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 129 bones are extremely rare, and our study does not add anything to the discussion concerning whether the Norse ate fish. Seal bones occur at all farms, although in relative larger proportions on the medium- sized and small farms. At the small Western Settlement coastal site of Niaquusat (V48), seal remains are dominant during the whole settlement period from c. A.D. 1000 to c. A.D. 1400, accounting for more than 80% of all bones in the midden deposits (McGovern 1985, Møhl 1982). Even here a tendency towards the increasing importance of seal can be noted (Fig. 8; McGovern 1985). At the medium- sized inland Western Settlement farm GUS, the initial phase (ca. A.D. 1000–1150) is characterized by 29% of the total bone assemblage deriving from marine animals. In the final phase (ca. A.D. 1300– 1400), this proportion increased to 44% (Enghoff 2003:90). The proportion of cattle appears to have been stable, whereas the proportion of seal increased at the expense of reindeer (Fig. 8). On the coastal high-status farm Sandnes (V51), the marine element seems to have remained more or less constant relative to the numbers of bones from domesticates and reindeer (Fig. 8; McGovern et al. 1996). In the Eastern Settlement, the results from the 2005–2006 excavations at high-status Brattahlid farm Ø29a reveal the same trends as seen in the Western Settlement. The number of sea mammals increases through time at the expense of domesticates (Fig. 9; McGovern and Pálsdóttir 2007). Also at the so-called landnam farm (Ø17a) in Narsaq (McGovern in Vebæk 1993), seal bones are abundant in the assemblage. As the name indicates, the excavated house is from the early Norse period; still, the faunal material reveals the same trend as that seen at other Norse farms. The proportion of marine mammals increases from the lower to the upper layers, and there is a similar increase in the number of sheep/goat bones compared with cattle (Fig. 9; McGovern et al. 1993:58 ff.). Reindeer protein (for a discussion on the identification and quantification of reindeer protein, see Nelson and Møhl 2003 and Nelson et al. 2012d [this volume]) is not traceable in the Eastern Settlement humans, and this finding is in accordance with the animal-bone record. Reindeer bones are not at Figure 9. The distribution of animal bones by species at two Eastern Settlement farms: 1) The so-called landnam farm Ø17a in Narsaq. This farm is located on the coast and the animal bones are from the first centuries of settlement. 2) Brattahlid (Qassiarsuk), a high-status coastal farm. 130 Journal of the North Atlantic Special Volume 3 all frequent in the Eastern Settlement assemblages (Fig. 9; McGovern 1985). In the 2005–2006 midden excavations at Brattahlid, “Caribou bones are present in low but consistent frequency throughout the phases ...” (McGovern and Pálsdóttir 2007). The Norwegian priest Ívar Bárðarson mentions that reindeer hunting at Renø —Reindeer Island— in the Eastern Settlement was restricted and could only take place with the permission of the bishop. We may assume that a small reindeer population prompted these restrictions and that permission was given only to the few. In the Western Settlement, reindeer bones appear in the middens in larger quantities. At the high-status farm Sandnes, reindeer played a substantial role in the diet, and the same is the case at the medium-sized GUS farm (Fig. 8). Most of the bones were from the meatbearing part of the animal, and the importance of reindeer decreased slightly through time (McGovern et al. 1996:111f.). At the medium-sized inland farm GUS, between 13 and 28% of all mammal bones were of reindeer, showing a decrease through time (Fig. 8; Enghoff 2003:41). Even at the small farm at Niaquusat, reindeer is present, although to a much lesser extent (Fig. 8; McGovern 1985:115f.). As may have been the case in the Eastern Settlement, McGovern et al. (1996:112) suggest that reindeer meat was mainly restricted to the elite. In general, the Western Settlement individuals who were analyzed had obtained less than 25% of their protein from reindeer (Nelson et al. 2012d [this volume]), and it is not possible, on this basis, to draw any conclusions concerning social differences. On the whole, Norse subsistence strategies were faithful to those which the initial settlers brought with them from their homelands in Scandinavia and in Iceland. Traditionally, the Norse subsistence economy depended on a combination of dairy products and meat, both from their livestock, supplemented with hunting of both sea mammals and terrestrial animals (e.g., Kaland and Martens 2000, McGovern 2000, McGovern et al. 2009). In Greenland, the central Eastern Settlement temperate region—at least in the initial settlement period— offered the best conditions for the traditional Norse pastoral economy, whereas the Low Arctic conditions in the southern part of the Eastern Settlement and in the Western Settlement very probably gave rise to a more marine-based economy. The human samples from the early period (most of them are from the central Eastern Settlement region) may reflect the fact that the Greenland immigrants in the initial settlement period tried to establish what McGovern (2000:331) calls “the homeland’s ideal farmyard” with relatively large numbers of cattle, pigs, sheep, and goats and a less significant dependence on the marine resources. Constantly, they had to adjust to their new environment and to natural and man-made changes. Farming Strategies One of the goals of the study was to examine whether the shift in the Norse diet reflects altered farming strategies. Norse animal husbandry strategies were based on grass. The isotope signals in the livestock do not reveal any significant differences between the farming strategies of Eastern and the Western Settlements, and there are no signs of changes through time (Nelson et al. 2012c [this volume]). The carbon isotope values show that “hungerfeeding” (i.e., feeding with, for instance, fish and/or seal refuse), known from Iceland in modern times (e.g., Bruun 1928:273, Hooker 1813:348), was not the case at any time in Norse Greenland (Nelson et al. 2012c [this volume]). Compared to the other samples, cattle and goats from the small Western Settlement farm V48 at Niaquusat have relative high δ15N values, and manuring of the pasture and/or the hayfield appears to be an obvious explanation for this (discussed in Nelson et al. 2012c [this volume]). This may not necessarily have been deliberate manuring. The Niaquusat site is surrounded by mountains and is too small to have had regular pastures. Winter fodder must have been collected in the immediate vicinity of the farm buildings, and cattle and goats must have grazed the midden where household and stable waste accumulated over the years. In contrast, the Niaquusat sheep do not show the same raised δ15N values, perhaps because they grazed mountain pastures away from the farmhouses, reflecting the fact that cattle and goats were kept primarily for their milk and were kept close to the houses for daily milking, whereas sheep were kept primarily for the wool and meat and grazed on more distant pastures. Bones of pigs have been found at only a few sites and they are restricted to the early settlement period (McGovern 1985:86). Their isotope data indicate that the pigs were fed on a marine diet, most probably fish and seal offal (Nelson el al. 2012c [this volume]). In Iceland, pig remains are mostly limited to the first centuries following the landnam (Ólafsson et al. 2006:397ff.), whereas in the Faeroes there seems to have been substantial pig keeping well into the 13th century (McGovern et al. 2004). In Iceland and the Faroes, pigs were fed on terrestrial foodstuffs (T. McGovern, Hunter College, CUNY, New York, NY, USA and S. Arge, Føroya Fornminnissavn, Tórshavn, Faroe Islands, personal comm.), and reduction in available woods or marshland is one of the explanations offered for the decrease in pig keeping, especially in Iceland (McGovern et al. 2009:216) The same explanation cannot apply to Norse Green2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 131 fodder was grown in fertilized infields. The manure was composed of cattle dung, whereas dung from sheep was used as fuel. Besides the infields, many farms also obtained hay from meadows and bogs that were never fertilized. The best hay came from the manured infields and was given to the cattle (see also: Adalsteinsson 1991:291). Evidently, manuring practices in Norse Greenland still need to be looked into, and until then, the interpretation of the raised δ15N values from Niaquusat must be viewed with some reservation. Norse Greenland Dietary Economy: Concluding Remarks The results of this isotope study support those of the first limited study: the diet of the Norse Greenlanders became more dependent on marine resources over time. The changes seem to have happened gradually, beginning during the initial settlement. The swiftness depended, however, on where the immigrants settled. Settlers in the southern part of the Eastern Settlement, and in the Western Settlement, may have adapted to the marine resources more rapidly than those in the central Eastern Settlement region. In the first centuries of settlement, the dietary differences within the society were pronounced, perhaps indicating that access to terrestrial food was restricted and only an option for the well-off part of the population. Despite the changes in the dietary economy and the increasing dependence on the marine resources, we see no attempts to prevent or discourage this development. Farming strategies remained unchanged. Qualitative changes such as “hunger-feeding” the domesticated animals were never the case. Instead, the Norse farmers presumably reduced the number of domesticates and replaced the resource-demanding cattle with the less-demanding sheep and goats. Climate change and unsustainable land-use practices have been proposed as the main reasons for the depopulation of the Norse Greenland settlements in the late 1400s, and it is obvious to draw attention to these factors when trying to explain the changes in the dietary economy. It is more doubtful, after all, whether the changes in the Norse dietary economy directly caused the depopulation of the Norse Greenland settlement. The Norse Greenlanders apparently adapted well to their physical environment, and they could survive on the marine resources as long as these lasted. Culturally and socio-economically there may, however, have been barriers that were difficult to cross. The old saying “you are what you eat” may not hold the whole truth after all. Humans are more than this, and in order to explain the depopulation of the Norse Greenland settlements, ideology, percepland where pigs were fed on marine foodstuffs and most probably were penned. Another explanation for the declining number of pigs in Greenland could be the workload connected with pig keeping. As described above, the δ15N values for cattle and goats from Niaquusat differ from those of the other cattle and goats sampled. If we accept that the raised nitrogen values, especially for the Niaquusat samples, are due to deliberately/non-deliberately manured pastures and hayfields, we also have to conclude that, in general, the results of the isotope study show the Norse did not manure their pastures, since no other samples from domesticates show the same elevated values. This conclusion supports the ideas of Simpson et al. (2002:440) who, on the basis of geoarchaeological studies, have not found any traces of Norse manuring practices in the Qassiarsuk – Brattahlid region. Simpson and Adderley (2007:44) point out that intensive fertilization of arable land requires large numbers of domestic livestock in order to generate the manure, as well as labor to compost it and transport it to the fields. The Norse Greenlanders may have been short of both manpower and stalled domestic livestock to produce the required manure. Instead, they had to rely on natural soil fertility and productivity In contrast to the geoarchaeological study, Commisso and Nelson (2006, 2007, 2010) argue that raised nitrogen values in modern plants growing on the Norse sites indicate deliberate Norse fertilization of the infields in the Middle Ages. Manuring also seems to have been practised at the bishop’s see Gardar (Buckland et al. 2009). Also at GUS, in the Western Settlement, there are indications of fertilized fields (Schweger 1998:16), and a new study of ancient DNA in soil samples indicates that waste from byres and stables was spread over the adjacent fields (Hebsgård et al. 2009). At GUS, the anthropogenic layer thins out with distance from the farm building, and the question is whether we are talking about disposal of waste or deliberate fertilizing as part of a land management strategy—or perhaps both. In the North Atlantic, fertilizing as part of a land management strategy is bound up with the infield-outfield system and with fenced infields. The practice is seen in Western Norway by the Early Iron Age and should be perceived in a mixed farming context, including agriculture and animal husbandry. Inside the fence, in the infield, the land was cultivated for cereals. The outfield was pasture (Øye 2005:362). The infield-outfield system, including the fenced infield, was an integral part of the cultural package the immigrants brought with them to the North Atlantic islands from Norway. When Daniel Bruun (1928:264ff) travelled around Iceland in the late 1800s, he described how grass for winter 132 Journal of the North Atlantic Special Volume 3 Hooker, W.J. 1813. Journal of a tour in Iceland in the summer of 1809. Vol. I. Harvard University, London, UK. 503 pp. Jay, M., and M.P. Richards. 2006. Diet in the Iron Age cemetery population at Wetwang Slack, East Yorkshire, UK: Carbon and nitrogen stable isotope evidence. Journal of Archaeological Science. 33(5):653–662. Jørkov, M.L.S., J. Heinemeier, and N. Lynnerup. 2009. The Petrous Bone: A new sampling site for identifying early dietary patterns in stable isotopic studies. American Journal of Physical Anthropology 138:199–209. Kalland, S.H.H., and I. Martens. 2000. Farming and daily life. Pp. 42–54, In W.W. Fitzhugh and E.I. Ward (Eds.). Vikings: The North Atlantic Saga. Smithsonian Institution Press, Washington, DC, USA. 432 pp. Kosiba, S.B., R.H. Tykot, and D. Carlsson. 2007. Stable isotopes as indicators of change in the food procurement and food preference of Viking Age and Early Christian populations on Gotland (Sweden). Journal of Anthropological Archaeology 26:394–411 Linderholm, A., C. Hedenstierna Jonson, O. Svensk, and K. Lidén. 2008. Diet and status in Birka: Stable isotopes and grave goods compared. Antiquity 82:446–461. Lynnerup, N. 1998. The Greenland Norse: A Biologicalanthropological study. Meddelelser om Grønland, Man and Society 24. Copenhagen, Denmark. 149 pp. McGovern, T.H. 1985. Contributions to the paleoeconomy of Norse Greenland. Acta Archaeologica 54-1983:73– 122. McGovern, T.H. 2000. The demise of Norse Greenland. Pp. 327–339, In W.W. Fitzhugh and E.I. Ward (Eds.) Vikings: The North Atlantic Saga. Smithsonian Institution, Washington, DC, USA. 432 pp. McGovern, T.H., T. Amorosi, S. Perdikaris, and J. Woollett. 1996. Zooarchaeology of Sandnes V51: Economic change at a chieftain’s farm in West Greenland. Arctic Anthropology 33(2):94–122. McGovern, T.H., and A. Pálsdóttir. 2007. Bone remains. Pp. 22–36, In R. Edvardsson (Ed.) Archaeological Excavations at Qassiarsuk 2005–2006. 109 pp. Available online at http://www.nabohome.org/publications/fieldreports/ BrattahlidE29NFieldReport07.pdf. Accessed 10 February 2012. McGovern, T.H., C. Amundsen, S. Perdikaris, R. Harrison, and Y. Krivogorskaya. 2004. An Interim report of a Viking-age and Medieval archaeofauna from Undir Junkarinsfløtti, Sandoy, Faroe Islands. Norsec Zooarchaeology Laboratories Report No 13. CUNY Doctoral Program in Anthropology, Brooklyn College Zooarchaeology Laboratory, Hunter College Bioarchaeology Laboratory, New York, NY, USA. McGovern, T.H., S. Perdikaris, I. Mainland, P. Ascough, V. Ewens, Á. Einarsson, J. Sidell, G. Hambrecht, and R. Harrison. 2009. The archaeofauna. Pp. 168–252, In G. Lucas (Ed.). Hofstaðir. Institute of Archaeology Monograph Series–1. Reykjavík, Iceland. 440 pp. Müldner, G., and M.P. Richards. 2007. Stable isotope evidence for 1500 Years of human diet at the City of York, UK. American Journal of Physical Anthropology 133:682–697. Møhl, J. 1982. Ressourceudnyttelse fra norrøne og eskimoiske affaldslag belyst gennem knoglematerialet. Grønland 8–9:286–295. tion, interactions between economies (subsistence and exchange economy), and cultural contacts (Norse-Europe and Norse-Inuit) also have to be taken into consideration. Literature Cited Adalsteinsson, S. 1991. Importance of sheep in early Icelandic agriculture. Acta Archaeologica 61-1990:285– 291. Alexandersen, V., and F. Prætorius, 2003. Hvad tænder kan fortælle om de første nordboere. Aktuel Arkæologi 3:12–15. Arneborg, J., 1991. The Roman Church in Norse Greenland. Acta Archaeologica 61-1990:142–150. Arneborg, J., J. Heinemeier, N. Lynnerup, H.L. Nielsen, N. Rud, and Á.E. Sveinbjörndóttir. 1999. Change of diet of the Greenland Vikings determined from stable carbon isotope analysis and 14C dating of their bones. Radiocarbon 41(5):157–168. Arneborg, J., N. Lynnerup, J. Heinemeier, J. Møhl, N. Rud, and Á.E. Sveinbjörnsdóttir. 2012a [this volume]. Norse Greenland dietary economy ca. AD 980–ca. AD 145: Introduction. Journal of the North Atlantic Special Volume 3:1–39. Balslev Jørgensen, J. 2001. Nordbogravene ved Brattahlid. Grønland:81–99. Barret, J.H., and M.P. Richards. 2005. Identity, gender, religion and economy: New isotope and radiocarbon evidence for marine resource intensification in Early Historic Orkney, Scotland, UK. European Journal of Archaeology 7(3):249–271 Bruun, D. 1928. Fortidsminder og Nutidshjem paa Island. Gyldendalske Boghandel, København, Denmark. 416 pp. Buckland, P.C., K.J. Edwards, E. Panagiotakopulu, and J.E. Schofield 2009. Palaeocological and historical evidence for manuring and irrigation at Garðar (Igaliku), Norse Eastern Settlement, Greenland. The Holocene 19(1):105–116. Commisso, R.G., and D.E. Nelson 2006. Modern plant δ15 N values reflect ancient human activity. Journal of Archaeological Science 33:1167–1176. Commisso, R.G., and D.E. Nelson 2007. Patterns of plant δ15 N values on a Greenland Norse farm. Journal of Archaeological Science 34:440–450. Enghoff, I.B. 2003. Hunting, fishing, and animal husbandry at The Farm Beneath The Sand, Western Settlement. Meddelelser om Grønland, Man and Society 28. Copenhagen, Denmark. 104 pp. Fricke, H.C., J.P. O’Neil, and N. Lynnerup. 1995. Oxygen isotope composition of medieval human tooth enamel from the medieval Greenland: Linking climate and society. Geology 23:869–72. Gulløv, H.C. 2012. Archaelogical commentary on the isotopic sudy of the Greenland Thule culture. Journal of the North Atlantic Special Volume 3:65–76. Hebsgaard, M.B., J. Arneborg, M. Thomas, P. Gilbert, P. Heyn, M.E. Allentoft, M. Bunce, K. Munch, R. Nielsen, C. Schweger, and E. Willerslev 2009. “The Farm Beneath the Sand” An archaeological case study on ancient “dirt” DNA. Antiquity 83:430–444. 2012 J. Arneborg, N. Lynnerup, and J. Heinemeier 133 Vebæk, C.L. 1992. Vatnahverfi: An inland district of the Eastern Settlement in Greenland. Meddelelser om Grønland, Man and Society Vol.17. Copenhagen, Denmark. 132 pp. Vebæk, C.L. 1993. Narsaq: A Norse landnáma farm. Meddelelser om Grønland, Man and Society Vol. 18. Copenhagen, Denmark. 85 pp. Øye, I. 2005. Farming and farming systems in Norse societies of the North Atlantic. Pp. 359–370, In A. Mortensen and S.V. Arge (Eds.). Viking and Norse in the North Atlantic. Select Papers from the Proceedings of the Fourteenth Viking Congress, Tórshavn, 19–30 July 2001. Annales Societatis Scientiarum Færoensis Supplementum XLIV. Tórshavn, Faroe Islands. 445 pp. Endnotes 1The archaeological excavations took place in 2002 and 2003 under the direction of Jette Arneborg. 2According to one radiocarbon date from lower culture layers at Ø149, the settlement here dates to around A.D. 1000 (K-5573: A.D. 1025 (1010-1150)) (Vebæk 1991:73). Nelson, D.E., and J. Møhl 2003. Radiocarbon dating Caribou antler and bone. Are they different? Arctic 56:262–265. Nelson, E., J. Møhl, J. Heinemeier, and J. Arneborg. 2012a [this volume]. Stable carbon and nitrogen isotopic measurements of the wild animals hunted by the Norse and the Neo-Eskimo people of Greenland. Journal of the North Atlantic Special Volume 3:40–50. Nelson, E., N. Lynnerup, and J. Arneborg. 2012b [this volume]. A first dietary study of the Greenlandic Thule Culture. Journal of the North Atlantic Special Volume 3:51–64. Nelson, E., J. Heinemeier, J. Møhl, and J. Arneborg. 2012c [this volume]. Isotopic analysis of the domestic animals of Norse Greenland. Journal of the North Atlantic Special Volume 3:77–92. Nelson, E., J. Heinemeier, N. Lynnerup, J. Arneborg, and Á.E. Sveinbjörnsdóttir. 2012d [this volume]. An isotopic analysis of the diet of the Greenland Norse. Journal of the North Atlantic Special Volume 3:93–118. Ólafsson, G., T.H. McGovern, and K.P.Smith. 2006. Outlaws of Surtshellir Cave: The underground economy of Viking Age Iceland. Pp. 395–405, In J. Arneborg and B. Grønnow (Eds.). Dynamics of Northern Societies. Publications from the National Museum–Studies in Archaeology and History. Vol. 10. Copenhagen, Denmark. 415 pp. Privat, K.L., and T.C. O’Connell. 2002. Stable isotope analysis of human and faunal remains from the Anglo- Saxon cemetery at Berinsfield, Oxfordshire: Dietary and social implications. Journal of Archaeological Science 29:779–790. Reitsema, L.J., D.E. Crews, and M. Polcyn. 2010. Preliminary evidence for medieval Polish diet from carbon and nitrogen stable isotopes. Journal of Archaeological Science 37:1413–1423. Schweger, C. 1998. Geoarchaeology of the GUS site: A preliminary framework. Pp. 14–18, In J. Arneborg and H.C. Gulløv (Eds.). Man, Culture, and Environment in Ancient Greenland. The Danish National Museum and Danish Polar Center, Copenhagen, Denmark. 212 pp. Simpson, I.A., W.P. Adderley, G. Guðmundsson, M. Hallsdóttir, M.Á. Sigurgeirsson, and M. Snæsdóttir. 2002. Soil limitations to agrarian land production in premodern Iceland. Human Ecology 30(4):423–443. Simpson, I., and W.P. Adderley. 2007. Geoarchaeological investigation at Qassiarsuk (Brattahlið) Greenland. In R. Edvardsson (Ed.) Archaeological Excavations at Qassiarsuk 2005–2006. Available online at http:// www.nabohome.org/publications/fieldreports/BrattahlidE29NFieldReport07. pdf. Accessed 10 February 2012 Sveinbjörnsdóttir Á.E., J. Heinemeier, J. Arneborg, N. Lynnerup, G. Olafsson, and G. Zoëga. 2010. Dietary reconstruction and reservoir correction of 14C dates on bones from pagan and early Christian graves in Iceland. Radiocarbon 52(2–3):682–696. Vebæk, C.L. 1991. The Church topography of the Eastern Settlement and the excavation of the Benedictine convent at Narsarsuaq in the Uunartoq Fjord. Meddelelser om Grønland, Man and Society Vol. 14. Copenhagen, Denmark. 81 pp.