Journal of Archaeological Science 157 (2023) 105824 Contents lists available at ScienceDirect Journal of Archaeological Science journal homepage: www.elsevier.com/locate/jas The role of salmon fishing in the adoption of pottery technology in subarctic Alaska Marjolein Admiraal a, b, *, Peter D. Jordan c, d, Helen M. Talbot a, Manon Bondetti a, Alejandro Serna a, e, Karine Taché f, Matthew von Tersch a, Jessica Hendy a, Krista McGrath a, g, Oliver E. Craig a, Alexandre Lucquin a a BioArCh Laboratory, University of York, Wentworth Way, Heslington, YO10 5NG, UK b Arctic Centre, Groningen Institute for Archaeology, University of Groningen, Aweg 30, 9718CW, Groningen, the Netherlands c Lund University, Department of Archaeology and Ancient History, Helgonavägen 3, LUX - Hus A, 223 62, Lund, Sweden d Global Station for Indigenous Studies and Cultural Diversity (GSI GI-CoRE), Hokkaido University, Northern Campus Research Building No.3 – 106, Kita 21, Nishi 11, Kita-ku, Sapporo, 001-0021, Japan e División Arqueología, Facultad de Ciencias Naturales y Museo, Universidad Nacional de La. Plata, 1900, La Plata, Argentina f Department of Historical Sciences, Université Laval, Pavillon Charles-De Koninck, 1030, Avenue des Sciences-Humaines Bureau 5309, Québec (Québec), G1V 0A6, Canada g Department of Prehistory and Institute of Environmental Science and Technology (ICTA), Universitat Autònoma de Barcelona, 08193, Bellaterra, Spain A R T I C L E I N F O A B S T R A C T Keywords: Ceramic technology makes an abrupt appearance in the New World Arctic at circa 2800 cal BP. While there is Lipid residue analysis general consensus that the ultimate source of these Alaskan pottery traditions lay in continental NE Asia, the Alaska motivations for the adoption of pottery in Alaska have remained unclear. Through organic residue analysis we Norton investigated the function of Norton pottery in Southwest Alaska, and the extent to which its function changed in Pottery later periods under the increasing northern influence of Thule culture in the region (from ca. 1000 cal BP). Our results show clear evidence of aquatic resource processing in all pottery vessels. Regional variability due to environmental and ecological differences are apparent in the pottery. The majority of Norton pottery was from inland riverine locations and the function of this early pottery was to process anadromous fish, with only limited evidence of other resources. After 1000 cal BP more sites appear on the coast, and while pottery technology changes dramatically at this time, this is not as clear in pottery function which remains aimed at local abundant aquatic resources. We hypothesize that pottery was adopted into Alaska as part of a riverine adaptation and suggest that targeted human exploitation of large riverine systems may have facilitated its expansion into Southwest Alaska. Furthermore, we suggest that this pattern might extend back into Siberia where Alaskan pottery originates. Author contributions tundral landscape is ill-suited to pyrotechnology, with limited supplies of fuel, and cold winters and damp summers limiting the manufacture M.A. and P.D.J. designed the research; M.A., H.M.T., M.V.T., M.B., A. and maintenance of ceramic vessels (Harry and Frink, 2009; Jordan and S., K.M.G. and A.L. did lipid residue analysis; M.A., J.H., O.E.C., and A.L. Gibbs, 2019a). Nonetheless, pottery was used by hunter-gatherer groups analyzed the data; M.A., P.D.J., O.E.C., and A.L. wrote the article. in Alaska for nearly 3 millennia (Anderson et al., 2017; Heizer, 1949). Before arriving in Alaska, pottery flourished in Northeast Siberia 1. Introduction amongst the Late Neolithic (~3000 cal BP) and Ust’ Belaia (3500 - 2500 cal BP) cultures of Northern Chukotka (Jordan et al., 2022; Kuzmin, The arrival of pottery in the New World Arctic around 2800 years ago 2014; Ponkratova, 2006). From a possible origin in the ceramic represents a remarkable new technological development (Fig. 1a). The Ymyiakhtakh culture (4200 - 2500 cal BP) of Yakutia (Ackerman, 1982; * Corresponding author. BioArCh Laboratory, University of York, Wentworth Way, Heslington, YO10 5NG, UK. E-mail address: marjoleinadmiraal@gmail.com (M. Admiraal). https://doi.org/10.1016/j.jas.2023.105824 Received 6 March 2023; Received in revised form 4 July 2023; Accepted 6 July 2023 Available online 20 July 2023 0305-4403/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 Fig. 1. Map of a) Alaska showing pottery sites of the Norton (ca. 2800-1000 cal BP, yellow), late prehistoric (including Thule and ancestral Yup’ik from ca. 1000 cal BP to contact, blue), and Koniag (ca. 500 cal BP to contact, orange) traditions, and post-3000 cal BP sites across the Bering Strait in Northeast Siberia (maroon) (a list of sites is available in Appendix Table B1. Translucent markers refer to sites that were not radiocarbon dated. The research area is marked by a rectangle corre­ sponding to the outline of map b) showing the Alaska Peninsula and Norton (yellow) and ancestral Yup’ik (blue) sites. Sites sampled for this study are marked with a number: (1) NAK-3, (2) DIL-161, (3) the Brooks River sites (n = 7), (4) Ugashik River 1 and 2, (5) NAK-10, (6) AK-3, (7) Kukak 1, (8) Mink Island, (9) UGA-29, (10) NAK-2, (11) Leader Creek, (12) UGA-28, see Table 1 and Appendix Table B2. Images reproduced from (Admiraal, 2020) and designed by Frits Steenhuisen. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) 2 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 Table 1 Site and sample information. For more detailed information see Appendix B. Fig. 1 # Culture/site Environment Calendar date (cal BP) Sampled sherds Reference Norton 1 Naknek River 3 Inland 2310–1848 8 Dumond (2011) 2 DIL-161 2130–2060 3 Bundy (2007) 1500–1300 3 3 Brooks River 10 undated 1 Dumond (1981) 4 Brooks River 14 1854 1 Dumond (2011) 5 Brooks River 7 1850 2 6 Brooks River 8 1230 1 7 Brooks River 5 1136 4 10 Ugashik River 1 1871 5 Henn (1978) 11 Ugashik River 2 undated 2 12 Naknek River 10 1685 1 Dumond (1981) 13 AK-3/MK-14 Pacific coast 1680 1 Clark (1977) 14 Kukak 1 1450 2 15 Mink Island 1620–1360 1 Hilton (2002) 960–755 1 16 Ugashik River 29 Bering Sea coast 896 1 Henn (1978) Ancestral Yup’ik 8 Brooks River 1 Inland 767–662 1 Dumond (2011) 9 Brooks River 20C 673 2 14 Kukak 1 Pacific coast 775 1 Clark (1977) 17 Naknek River 2 Bering Sea coast 667 1 Dumond (2011) 18 Leader Creek (Naknek River 8) Bering Sea coast 290–240 2 Dumond (2003) 295 3 19 Ugashik River 28 Bering Sea coast undated 1 Henn (1978) Dikov, 1979; Dumond and Bland, 1995), it entered the American role in coastal Norton sites, also south of the Seward Peninsula (Bock­ continent at the Bering Strait and spread rapidly along Alaska’s coastal stoce, 1979; Britton et al., 2013; Tremayne et al., 2018). At around 1000 margins with the Norton tradition (2800 - 1000 cal BP) (Anderson et al., cal BP influences originating from the Thule tradition further to the 2017; Dumond, 2016; Oswalt, 1955). Pottery remained restricted to the north reached across Alaska and brought change (Britton et al., 2013; broader coastal areas (Oswalt, 1955: pp. 40–41), and never truly spread Dumond, 2003; Farrell et al., 2014). In Southwest Alaska more sites further into the Alaskan hinterland (Fig. 1a). It first appeared after a appeared on the Bering Sea coast, and while anadromous fish and pan-Alaskan occupation hiatus that followed the disappearance of the caribou remained an important part of the diet, subsistence practices Arctic Small Tool tradition (ASTt) (Barton et al., 2018; Dumond, 2016; shifted more towards the marine environment. This change is best wit­ Tremayne and Brown, 2017). In the Northwest an early shift towards the nessed by the increased appearance of technologies for the open water exploitation of marine mammals occurred already among the ASTt hunt of large marine mammals (e.g. toggling harpoons), but also in more (Buonasera et al., 2015) but this only intensified with the local Choris abundant sites on the Bering Sea coast and in the transformation of and Norton stages (Britton et al., 2013; Tremayne et al., 2018). This was pottery technology. These late prehistoric ancestral Yup’ik groups possibly the result of the nearly year-round presence of sea-ice in the experienced varying levels of Thule influence throughout the region. north, which was absent for several months of the year further to the The pottery of these late prehistoric groups was tempered with crude south. These pronounced environmental differences have major effects mineral material, it was fragile, thick, and not as well-fired as that of the on species distribution (e.g. salmon runs were much more abundant in Norton tradition, which was often tempered with organic material Southwest Alaska) and helped shape subsistence strategies and likely (Figs. C1 and C2). These new traits likely originated in the northern also determined variation in pottery technology and function (Admiraal Arctic where local climate further complicated the manufacture of and Knecht, 2019; Dumond, 2016; Frink and Harry, 2019). . pottery (Frink and Harry, 2008). Norton subsistence practices on the Southwest Alaskan mainland, While the archaeological record testifies to the importance of pottery especially on the Alaska Peninsula, encompassed a diversity of resources to the coastal peoples of Alaska, the reasons for investment in ceramic reflecting the different environmental zones of the area. The Alaska technology at this juncture in prehistory, and in such an extreme envi­ Peninsula is made up of an extensive coastal plain (the Bering Sea coast), ronment, remain poorly understood (Anderson et al., 2017; Jordan and a well-forested inland with abundant lakes and large rivers that drain Gibbs, 2019a). What drove the adoption of pottery technology in into the Bering Sea, and a substantial mountain range (the Aleutian Alaska? What was pottery used for? How did it evolve throughout Range) that creates a barrier between the Pacific coast and the rest of the Alaskan prehistory? The dispersal of pottery on the Alaska Peninsula peninsula. The substantial runs of anadromous fish that occur from encompassed different environmental and climate trajectories including spring to autumn in the many rivers south of the Bering Strait played a the distinct coastlines of the Subarctic Bering Sea and the more maritime significant role in Norton subsistence in Southwest Alaska (Dumond, climate of the Pacific coast. The large rivers and lakes of the Alaska 2016; Lutz, 1982). Anadromous fish in the region include Pacific salmon Peninsula support major salmon runs from spring to autumn. In contrast, (Oncorhynchus sp., i.e. Chinook: O. tshawytscha; Sockeye: O. nerka; Coho: the Pacific coast is isolated from the inland peninsula by the Aleutian O. kisutch; Pink: O. gorbuscha; and Chum salmon: O. keta), but also other Range and lacks large river systems (Fig. 1b). This, and the divergence species such as trout (e.g. Dolly Varden: Salvelinus malma) and whitefish between the two chronological ceramic traditions (before and after (e.g. Coregonae sp.). Caribou was also an important part of the Norton 1000 cal BP) makes the region an interesting case study to investigate diet, further supplemented by marine species, waterfowl and shellfish questions concerning pottery adoption drivers, functional patterns and depending on site location and the geographic distribution of changes throughout time and space. prey-species (Casperson, 2018; Coltrain, 2010; Dumond, 2016; Frink Recent efforts to answer similar questions have focused primarily on and Harry, 2019; Miszaniec et al., 2021). In Northern Alaska salmon late prehistoric pottery (e.g. Western Thule) (Anderson et al., 2017; runs were less abundant and marine mammals played a more important Farrell et al., 2014; Harry and Frink, 2009; Solazzo et al., 2008; Solazzo 3 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 and Erhardt, 2007). A common assumption is that the adoption of is reflected in our sample selection with 81% of Norton vessels coming coastal pottery was strongly connected to a maritime adaptation (Jordan from the interior Alaska Peninsula, while 66% of ancestral Yup’ik and Gibbs, 2019b). Indeed, the earliest pottery in Alaska is found around samples come from the coast. Norton sites on the interior river systems the Bering Strait where a maritime adaptation is more apparent than in have often been interpreted as fishing camps with semi-subterranean the southwest (Tremayne et al., 2018). Pottery may have been an houses and abundant net-sinkers present (Dumond, 1981, 2003, indispensable tool used for processing marine resources, particularly 2011). While ancestral Yup’ik sites are often found on the coast it is rendering marine mammal fats, an idea supported by ethnographic important to note that such sites are often located near river mouths that literature and Alaska Native oral histories (Anderson, 2019; Khvostof were also excellent strategic locations for the interception of anadro­ and Davydov, 1810; De Laguna, 2000; Heizer, 1949) and results from mous fish (Dumond, 2003). To investigate potential differences between other regions (Admiraal et al., 2019, Admiraal et al., 2020a,b; Farrell, the interior and exterior of the vessels, both in ceramic and carbonized 2013; Farrell et al., 2014). We aim to test this hypothesis through the surface residues we collected several samples per sherd: in total 46 characterization of organic residues left in the pottery vessels them­ ceramic samples and 56 carbonized surface residue samples, and one selves. We apply a combined approach of lipid residue analysis, com­ carbonized surface residue sample of a stone lamp from an earlier period pound specific isotope analysis and proteomics that can provide a as a reference (see Appendix Table B3). Additionally, lipids from faunal unique insight into past culinary practices and human diet (Evershed, remains (bone) of several archaeological sites in Southwest Alaska and 2008a; Heron and Craig, 2015; Solazzo et al., 2008). While lipid residue the Aleutian Islands were analyzed as reference materials. analysis and compound specific isotope analysis are routinely applied to pottery, the study of proteins preserved in pottery is still in its infancy 2.2. Lipid extraction and analysis (Hendy et al., 2018; Solazzo et al., 2008). As the preservation of faunal remains in Alaska Peninsula Norton and ancestral Yup’ik sites is limited, Lipids were extracted using acidified methanol and following this approach is especially valuable. Indeed, by applying lipid residue established protocols (Craig et al., 2013; Papakosta et al., 2015a; Col­ analysis and compound specific isotope analysis to pottery Anderson onese et al., 2015). In short, pottery samples were obtained by either et al. (2017) found that Thule pottery, as well as two Norton vessels at scraping off carbonized crusts with a sterile scalpel, or by drilling into the Cape Krusenstern site in Northwest Alaska, was predominantly used the surface of the sherd (3–5 mm depth) after removal of a 1 mm surface to process freshwater fish or anadromous fish. The site is located in an layer to avoid contamination. Bone samples were obtained by removing estuary setting on the Arctic Ocean. This result illustrates that a mechanically cleaned section of bone that was then crushed to a ho­ long-standing assumptions regarding the reasons for pottery adoption in mogenized powder (1 g) and solvent washed (3 × 2 mL dichlor­ Alaska, may be wrong. There is a significant gap in research regarding omethane/methanol 2:1 v/v wash) before extraction. Methanol was early Alaskan pottery function, and its role in the adoption of the added to the homogenized samples (4 mL–1 g of ceramic powder/bone, technology. Alternative drivers, deviating from a model of maritime 1 mL to 20 mg foodcrust) and the samples were sonicated for 15 min adaptations, need to be explored. Norton and the later ancestral Yup’ik after which sulphuric acid was added (800 μL to ceramic powder/bone, peoples of the Alaska Peninsula were known to exploit a diversity of 200 μL to foodcrust). Subsequently the acidified methanol mixture was resources over a variety of environmental settings including marine heated for 4 h at 70 ◦ C. After cooling and centrifugation, the supernatant mammals and fish on the Pacific and Bering Sea coasts, anadromous fish was transferred to a sterile vial and extracted using hexane (3 × 2 mL). on rivers, lakes and river mouths, and caribou in the interior (Bundy, To test for the presence of triacylglycerols and wax esters a selection of 2007; Casperson, 2018; Dumond, 2016). Here we investigate the func­ 10 samples were also subjected to solvent extraction (Evershed et al., tion of pottery with respect to these different subsistence foci. Our 1990). Briefly, a mixture of dichloromethane/methanol 2:1 v/v wash approach to explore spatial and temporal differences in pottery function was used to extract 1 g of ceramic powder (3 × 2 mL). All acid and on the Alaska Peninsula could shed new light on our understanding of solvent extracted samples were finally also derivatized using N,O-Bis prehistoric lifeways in this region and may reveal whether pottery was (trimethylsilyl)trifluoroacetamide (BSTFA). All samples were analyzed indeed used as part of a maritime adaptation. by gas chromatography-mass spectrometry (GC-MS) and GC-combustion-isotope ratio MS (GC-c-IRMS). Carbon and nitrogen 2. Materials and methods values of bulk foodcrust samples were obtained by elemental analysis-IRMS (EA-IRMS) according to existing protocols (Craig et al., 2.1. Sample selection 2013; Lucquin et al., 2016b; see Appendix A for further information on methodology). This represents the first large-scale organic residue study of Norton pottery and includes some of the earliest pottery from the North 2.3. Proteomics American Subarctic. Pottery of the Norton tradition is generally low in frequency, and the number of vessels per site is often unidentifiable due A selection of 11 ceramic samples was analyzed by proteomics in to the fragmented nature of the material which also limits the potential order to supplement lipid-based organic residue analysis and provide to reconstruct vessel form (Anderson et al., 2017). We aimed to explore further insight into processed foodstuffs. Protein extractions were per­ wider trends in Alaska Peninsula pottery use and extracted lipid residues formed on 100 mg of powdered ceramic material. Proteins were from a limited number of sherds per site, in addition to a small number extracted using Gel-Aided Sample Preparation, previously described in of late prehistoric pottery samples of ancestral Yup’ik sites to explore Hendy et al. (2018). Extracts were analyzed using LC-MS/MS at the regional temporal change in pottery function (see Appendix B Tables B2 Target Discovery Institute using the parameters previously described in and B3). A total of 37 Norton pottery sherds and 11 late pre­ Hendy et al. (2018). Samples were searched semi-tryptically using historic/ancestral Yup’ik sherds from 19 sites on the Alaska Peninsula Mascot (Matrix Science) against the Swiss-Prot database (d.o.a.: (Fig. 1b, Table 1 and Appendix Table B2) were sampled, covering a wide 01-05-2017) with the following propionamide (C) as a fixed modifica­ variety of landscapes ranging from the Bering Sea coast to the interior tion and deamidated (NQ), oxidation (M), propionamide (K), and pro­ river systems and the Pacific coast (Fig. 1). We targeted a variety of sites, pionamide (N-term) as variable modifications. Peptide mass tolerance regions and ecological zones to allow for comparison of organic residue was set at 10 pm m and fragment mass tolerance at 0.5 Da. Samples were data with spatial analysis of site distributions in Southwest Alaska to searched using a decoy strategy and an FDR correction of <1% of all further explore patterns in the occurrence of pottery in these different peptide spectral matches above the homology threshold was applied. environmental settings. Norton pottery is more often found in inland Proteins were considered confident identifications when matching the settings while ancestral Yup’ik pottery is more frequent at the coast. This criteria above, as well as being represented by at least two peptides. 4 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 Table 2 Lipid concentrations (expressed in mg g− 1) of ceramic and foodcrust (FC) samples on the Alaska Peninsula. Inland (ceramic) Inland (FC) Pacific coast (ceramic) Pacific coast (FC) Bering Sea coast (ceramic) Bering Sea coast (FC) range 0.028–4.3 0.7–10.4 0.4–4.4 0.5–16.6 1.2–11.1 1.1–37.8 mean 1.2 3.2 2.3 8.0 6.8 14.5 st dev 1.2 2.9 2.1 6.5 3.6 11.9 3. Results and interpretation nearly all samples, make up a lipid profile comparable to that of degraded aquatic products (Appendix Table B3). Aquatic biomarkers Lipid concentrations are generally high, ranging from 0.028 mg g− 1 were identified by GC-MS in nearly every sample and include ω-(o- to 37.8 mg g− 1 with a mean of 1.4 mg g− 1 for Norton ceramic samples, alkylphenyl) alkanoic acids (APAAs) with carbon length 16 to 22, and and 5.6 mg g− 1 for the late prehistoric ceramic samples. These values isoprenoid acids: TMTD (4,8,12- trimethyltridecanoic acid), pristanic greatly exceed the minimum amount required for interpretation (>5 μg acid (2,6,10,14- tetramethylpentadecanoic acid), and phytanic acid g− 1 for ceramic, and >100 μg g− 1 for charred deposits) and indicate (3,7,11,15-tetramethylhexadecanoic acid). APAAs are formed during excellent preservation (Craig et al., 2013; Evershed, 2008b). In one the prolonged heating of mono-, di- and tri-unsaturated fatty acids at a sherd (NAK8-13) extractable lipids accounted for almost 4% by total temperature of 200 ◦ C and higher, and therefore likely derive from mass, the highest recorded in an archaeological sample. Interestingly, cooking activities (Bondetti et al., 2021b; Evershed et al., 2008; Hansel variability in lipid concentration is observed between samples from et al., 2004). APAA C20/C18 ratios are above the threshold of 0.06 three defined regions: the inland (34 sherds, >15 km from the coast), (0.09–0.76 in our samples) as defined by Bondetti et al., 2021b, further Pacific Coast (6 sherds) and Bering Sea coast (8 sherds), with higher lipid supporting the presence of aquatic resources. Additionally, dihydroxy concentrations from the coastal regions than from the inland, and acids such as 11,12-dihydroxydocosanoic acid were identified in 17 out especially high lipid concentrations on the Bering Sea coast irrespective of 43 samples in the acidified methanol extracts following conversion to of site. Lipid concentrations from foodcrusts are generally higher than their TMS esters, confirming the aquatic origin of the vessel contents ceramic powder concentrations (see Table 2 and Appendix Table B3). (Hansel et al., 2011; Hansel and Evershed, 2009). Norton samples (n = 75) have considerably lower lipid concentrations The ratio of branched fatty acid (C15br and C17br) further supports than those of later ancestral Yup’ik samples (n = 27) (Mann Whitney U the aquatic nature of these samples (Appendix Table B3). Iso-branched = 263, z = 5.681 p = < 0.05), a trend likely also connected to fatty acids predominate in aquatic products (Demirci et al., 2021; Gar­ geographic site distribution. nier et al., 2018; Hauff and Vetter, 2010), while anteiso-branched fatty While protein analysis had been successful before on Arctic pottery acids are present in higher quantities in ruminants (Dudd et al., 1999; (Solazzo et al., 2008), our material did not yield any ancient proteins Hauff and Vetter, 2010), and even more so in beaver tissue (Demirci that could be confidently assigned to a dietary source (Appendix et al., 2021; Käkelä et al., 1996). The majority of our pottery samples Table B4). Identified proteins were from human keratin, most likely show high percentages of iso-branched fatty acids and are comparable to derived from contamination with skin and dust, and common laboratory aquatic species. However, seven of the Norton samples also show values reagents (e.g. trypsin). It is possible that the absence of proteins is due to more comparable to ruminants. While these samples also exhibit aquatic the high lipid content and low protein content of the original content biomarkers, it is interesting to note the possibility that several food­ which may have lowered the extraction efficiency. stuffs, including caribou, may have been mixed in these ceramic vessels, either at the same time or in separate cooking events. Isoprenoid acids are abundantly present in the samples and are 3.1. Molecular analysis degradation products of phytol, a constituent of chlorophyll which oc­ curs widely in plants and algae. In the freshwater and marine food web Several samples (n = 55) contain compounds that may have been phytanic acid is formed from phytol through the digestion of phyto­ formed during the firing of the pottery or during cooking on a coniferous plankton chlorophyll by invertebrates such as zooplankton (Avigan and wood-fueled fire (polycyclic aromatic hydrocarbons (PAH), benzene­ Blumer, 1968). Phytanic acid becomes integrated in the marine food polycarboxylic acids (BPCA) and abietic acid derivatives). Resinous web where it may accumulate and where catabolic processes may also compounds may also have been used to waterproof the pottery (Simo­ form other isoprenoid acids such as pristanic acid and the relatively neit et al., 2000; Oras et al., 2017). Non-specific plant biomarkers such stable TMTD (Ackman and Hooper, 1968). While in lesser abundances, as sterols (β-sitosterol) and terpenes (α-amyrin) were identified in four phytanic acid also occurs in ruminant animal tissues. In the rumen it is carbonized food crust samples. It is possible that the presence of plants is formed by bacterial oxidation and hydrogenation of phytol. We underestimated in our samples as the presence of lipid-rich aquatic oils discriminate between these sources by comparing the abundance of SRR may obscure the plant derived lipids that usually occur at low abun­ and RRR diastereomers of phytanic acid (SRR%) (Lucquin et al., 2016a). dance. Contamination by migration of lipids from the burial environ­ All SRR% pottery values plot within the aquatic range as compared to ment to the carbonized crusts must also be considered here (van Bergen modern references (Fig. 2a), although two outliers of Norton pottery et al., 1998). Seven samples have trace amounts of n-alkanes (C15–C29) (NAK3-1 and AK3-16) also are withing the range of ruminant values and and long-chain n-alkanols (C16–C30), identified in acidified methanol may represent the processing of caribou in these pots. extracts following conversion to their TMS esters (Appendix Table B3). That Alaskan coastal pottery was used to process aquatic species While this may hint at the processing of plants, abundances were too low comes as no surprise. However, the aquatic spectrum encompasses a to properly assess this. A selection of samples (n = 10) was subjected to huge variety of species and accompanying human subsistence strategies. solvent extraction and silylation to further explore the presence of plant Indeed, it takes a completely different strategy to fish for salmon than it biomarkers that may be lost by acidified-methanol extraction. Eight does to hunt marine mammals on the open water. The ratio of isoprenoid samples out of ten contained mono- and di-acylglycerols but no further acids TMTD and phytanic acid may be of further interpretational value diagnostic compounds were detected. and could offer a potential means to discriminate between (groups of) Most ceramic vessels presented strong molecular evidence for the marine species (Ackman and Hansen, 1967; Ackman and Hooper, 1968; processing of aquatic resources. The presence and high relative abun­ Cox et al., 1972; Ratnayake et al., 1989). We observed differences be­ dance of monounsaturated (C16:1 to C26:1) and polyunsaturated (C18:2, tween modern anadromous fish and other aquatic species including seal C20:2, and C22:2) fatty acids, in addition to more common medium and and shellfish, where anadromous fish has low quantities of TMTD and long-chain saturated (C9 to C32), and dicarboxylic (C7–C15) fatty acids in 5 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 Fig. 2. A) Percentage of SRR diastereomer in total phytanic acid in Norton (yellow) and ancestral Yup’ik (blue) pottery, compared with modern ruminant and aquatic resources; and b) ratios of 4,8,12-TMTD and phytanic acid (logarithmic scale) in Norton and ancestral Yup’ik pottery and modern references (see Appendix Table B3 and Table B5). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) high phytanic acid quantities (t (2.19), p = 0.03) compared to other compound specific isotopes further corroborate that the majority of aquatic species, especially seal (see Ackman and Hooper, 1968). Our samples are aquatic in origin (Appendix Table B6). Modern reference pottery data exhibits low TMTD/phytanic acid ratios similar to anad­ values of known origin in three different aquatic groups (marine species, romous fish (Fig. 2b). To fully explore the potential of the TMTD/phy­ anadromous fish and freshwater fish) allow for further differentiation tanic acid ratio further research is necessary. The influence of metabolic within the aquatic spectrum after correction to account for recent processes and the position of the species in the marine food web, as well changes in the isotopic content of atmospheric CO2. Marine organisms as post-depositional and degradation processes are currently poorly are generally more enriched in 13C than other aquatic species such as understood. anadromous fish and freshwater fish, although variation exists depending on feeding habits. Seventy percent of Norton pottery (28/40) 3.2. Isotopic analysis values plot within the range of anadromous fish species (Fig. 3). Some outliers have fatty acids less enriched in 13C that may indicate the 3.2.1. Compound specific isotopes of individual fatty acids processing of freshwater fish, and two early Norton samples (i.e. To further distinguish between potential sources of the residues we DIL161-1004; NAK3-1) plot towards ruminant values. Seven Norton measured the carbon isotope values (δ13C) of individual fatty acids C16:0 samples (of which four are from the Pacific coast) show a stronger and C18:0 using GC-combustion-isotope ratio MS (GC-C-IRMS). These marine signature with more enriched values. Compared to Norton (n = Fig. 3. Carbon isotope measurements of individual fatty acids C16:0 and C18:0 for the three distinct regions: the Alaska Peninsula inland, the Bering Sea coast, and the Pacific Coast (Appendix Table B6). Norton samples are in yellow ancestral Yup’ik in Blue. Shapes refer to the presence (filled circles) and absence (open circles) of aquatic biomarkers. The data are compared to reference values of modern tissue and bone from the Northern Hemisphere plotted in 66,8% confidence ellipses, the dotted diagonal line represents the ruminant offset boundary of Δ13C = − 1 (Appendix Table B7; Bondetti et al., 2021a; Choy et al., 2016; Courel et al., 2020; Craig et al., 2011; Cramp et al., 2014; Horiuchi et al., 2015; Lucquin et al., 2016b; Pääkkönen et al., 2020; Taché and Craig, 2015). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) 6 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 36), pottery of the ancestral Yup’ik (n = 9) does not show statistically mainly from riverine sites, show no correlation between δ13C and δ15N significant differences in isotopic values of fatty acid C16:0 (Mann- (Pearson r = 0.30, df = 24, p = 0.143), indicating a more varied source Whitney U = 123, z = 1.092, p = 0.274) but do show slightly more of carbon, that could reflect freshwater or anadromous fish but may also enriched values (late prehistoric mean C16:0 = − 23.9; Norton mean be the result of mixtures with plants and terrestrial animals (Fig. 4). C16:0 = − 24.3). Samples from the late prehistoric period are mainly from the coast, except for two samples from the Brooks River (BR20) that are 3.2.3. Elemental composition of charred surface deposits indeed less enriched than the other late prehistoric samples, likely The atomic C:N ratio of foodcrust samples and the Δ13C offset be­ reflecting the processing of anadromous fish (Appendix Table B3). This tween compound specific and bulk data can crudely inform on the data is further corroborated by carbon and nitrogen stable isotope values composition of the charred deposits (Robson et al., 2022). High C:N of human remains from the interior Brooks River area (BR5) (Coltrain, ratios indicate a lipid rich sample (Admiraal et al., 2019), or a sample 2010). Coltrain’s results indicate that late prehistoric ancestral Yup’ik rich in carbohydrates from plants (Bondetti et al., 2020). A low C:N ratio people in this area subsided on a diet of terrestrial animals and anad­ shows a greater proportion of nitrogenous compounds such as proteins romous fish such as salmon, with the possible addition of marine animal (Gibbs et al., 2017). Most of our foodcrust samples have intermediate products from the coast. atomic C:N ratio values (n = 31/37 = 6.24–14.10), reflecting lipid-rich, but not pure lipid products (Admiraal et al., 2019; Heron et al., 2013). 3.2.2. Carbon and nitrogen analysis of foodcrusts This is supported by the positive correlation (Pearson R = 0.625, df = The bulk carbon (δ13C), nitrogen (δ15N), %C and %N values of 34, p = < 0.05) between the C:N ratio and the lipid concentrations of the charred deposits from the surface of 38 pottery vessels were determined lipid extract of the same sample. The offset of compound specific δ13C by EA-IRMS (Appendix Table B6). This approach is complicated both by values (average δ13C16:0 and δ13C18:0, measured on the lipid extract) and the charring process and the different quantities of proteins, lipids and the foodcrust bulk 13C value, is of interest because it can show whether carbohydrates in the vessel contents prior to charring (Heron and Craig, the two techniques are essentially measuring the same component (i.e. 2015), but has been shown to be effective at distinguishing different lipids: exhibiting a minimal offset), or if the bulk analysis is picking up sources and substances (e.g. oil vs protein-rich foodstuffs; Gibbs et al., on other components such as proteins and carbohydrates, resulting in a 2017). In Fig. 4, the δ15N values obtained from the foodcrusts are larger offset due to the isotopic difference between fats, proteins and compared with archaeological bone collagen values from Alaska and carbohydrates (Admiraal et al., 2020b; Post et al., 2007; Robson et al., Canada, adjusted to consider the tissue to collagen offset (ca.+ 2‰; 2022). The Δ13C offsets (Δ13C FAmean- δ13Cbulk) show marginally greater Fernandes et al., 2014). Thirty-three of 38 (87%) foodcrust samples had range (− 8.23 to 2.26) in Norton pottery, than in the ancestral Yup’ik δ15N values between 10‰ and 15‰, within the lower range expected for pottery (− 4.47 to 1.72), indicating that the late prehistoric ancestral aquatic organisms and closer to the range of anadromous fish and Yup’ik pottery may have been used to process slightly more lipid-rich freshwater fish than to marine organisms (Knudson and Frink, 2011), substances. There is no statistically significant difference between C:N but also in the range of wild non-ruminants such as bear, fox and small ratios of Norton (n = 26) and ancestral Yup’ik pottery (n = 11) (Man­ game. The δ13C values of the charred deposits are harder to compare n-Whitney U = 136, z = 0.216, p = 0.828) nor between coastal (n = 11) with bone collagen measurement as their value depends on their lipid and inland (n = 26) samples (Mann Whitney U = 113, z = 0.980, p = content, which is relatively depleted in 13C. Nevertheless, there is a 0.327). However, when comparing the pottery data to Aleutian stone positive correlation between δ13C and δ15N (Pearson r = 0.92, df = 8, p bowls, previously interpreted to have been used for oil rendering pur­ ≤ 0.005) for the late prehistoric ancestral Yup’ik samples (mainly poses (Admiraal et al., 2019), the lower C:N ratios (Mann Whitney U = coastal), as may be expected for mixtures of marine resources at 306, z = 3.288, p = < 0.05, stone bowls n = 31, pottery n = 37) and different trophic levels (fish, marine mammals). The Norton foodcrusts, smaller Δ13C offsets in the pottery samples (n = 32) versus the stone Fig. 4. A) carbon and nitrogen values of bulk carbonized residues of Norton (yellow) and ancestral Yup’ik (blue, Appendix Table B6), against δ15N values of modern references on the left (Admiraal et al., 2019; Britton et al., 2013; Byers et al., 2011; Choy et al., 2016; Coltrain et al., 2016, 2004; Knudson and Frink, 2011; Marsh et al., 2017; Misarti et al., 2009; Pauly et al., 1998; West and France, 2015, see Appendix Table B8). δ15N values of collagen were corrected to allow comparison to carbonized surface residues by assuming that the δ15 N in the surface residues are derived from protein in animal tissue and the Δ15Ntissue–collagen = ~+2‰ (Fernandes et al., 2014); b) Δ13Coffset (δ13C FAmean- δ13Cbulk) of Norton (yellow) and ancestral Yup’ik (blue) inland sites (circles); sites on the Pacific coast (triangles); and sites on the Bering Sea coast (squares), Aleutian stone bowl values are shown for reference (downward triangles, Admiraal et al., 2019). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) 7 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 bowls (n = 19) (Mann Whitney U = 95, z = 4.222, p = < 0.05) may (Mann Whitney U = 0, z = 12.636, p = < 0.05; PC = 12, BSC = 17, I = indicate that the pottery of the Alaska Peninsula was not a specialized 73). This supports the notion that more lipid-rich species were processed tool used for the rendering of oil, but had a more diverse use in general. in the pottery there (e.g. seal). This might reflect the processing of different species in these areas. Interestingly, while marine resources 4. Discussion were processed on both the Pacific and Bering Sea coasts, the prey-species between these marine regions may have differed as well. 4.1. The function of Norton and ancestral Yup’Ik pottery This argument is supported by faunal assemblages (where preserved) (Dumond, 2003, 2016), human isotope data from the Pacific coast and Organic residue analysis of Norton and late prehistoric ancestral the Brooks River (Coltrain, 2010) and other inferences on subsistence Yup’ik pottery from the Alaska Peninsula provided unambiguous evi­ focus in these areas (e.g. the presence of harpoons, net sinkers, etc.) dence for the processing of aquatic resources in all the ceramic vessels (Dumond, 2003; Schaaf, 2008). The Pacific coast is isolated from the rest tested here (n = 49). Our main finding is that the vast majority of the of the Alaska Peninsula by the Aleutian Range and lacks large salmon Norton pottery we tested from the interior Alaska Peninsula was used to rivers. It is possible that, while contemporary with other Norton sites, process salmonids or other anadromous fish while coastal resources the early ceramic presence on the Pacific coast may actually be cultur­ were present in the pottery from coastal sites. We found limited evidence ally distinct from the Norton on the Bering Sea coast and in the Alaska of the addition of other resources, such as caribou and plants. However, Peninsula interior (Clark, 1977: p. 82). we caution that this may be the result of masking of less lipid-rich re­ sources (e.g. from plants) by lipid-rich aquatic oils. Interestingly, some 4.3. The drivers of pottery adoption in Southwest Alaska of the earliest pots in the region, from inland site locations on the Naknek River (NAK-3) and Alagnak River (DIL-161), have yielded clear Our results indicate that the tested Norton pottery from Southwest marine values beyond the range of anadromous fish (Appendix Alaska was used for a variety of cooking purposes: pottery from the Table B6). These sites are located 20 and 38 km from the coast respec­ interior was mainly used to process anadromous fish, with the possible tively, indicating the transport of coastal resources inland over distances addition of terrestrial game like caribou, and plant resources (which are of more than a day’s walk (~15 km) as early as 2120 cal BP (Bundy, likely underrepresented in our samples due to masking by the lipid-rich 2007; Dumond, 2011). This suggests that marine resources were still of aquatic oils), and even some marine species both at coastal and inland importance to these early settlers of the Alaska Peninsula. locations. The use of the pottery for cooking, as opposed to cold storage, The ancestral Yup’ik pottery tested in this study (n = 11) mainly is evidenced by the presence of heating biomarkers such as APAAs, originates from Bering Sea coastal sites and yielded abundant marine BPCA’s and PAH’s (see Appendix Table B3). Pottery was likely a lipids. Some late prehistoric pottery from the Alaska Peninsula inland (i. household item and used on a daily basis to cook meals. However, e. Brooks River sites, n = 2) was used to cook anadromous fish, a result pottery function may not have been limited to household use and could further corroborated by human isotope data (Coltrain, 2010). Based on also have played an important role in the seasonal processing of vast our limited sample of ancestral Yup’ik pottery we cannot confidently amounts of salmon throughout the ceramic period. Highly productive suggest a major change in pottery function from 1000 cal BP. High lipid salmon runs in Southwest Alaska occur from spring to autumn, allowing concentrations, a few high C:N ratios and smaller Δ13C offsets between for the mass harvesting of fish during this time. These large surplus lipid and total carbon in foodcrusts may indicate the processing of more catches would have demanded a rapid turnover to preserve the fish for lipid-rich species, likely marine mammals. While this interpretation is in winter (Tushingham and Bettinger, 2013). The drying and smoking of line with ethnohistoric information about pottery use in Alaska salmon is a well-known practice in Alaska and around the world (Fie­ (Anderson, 2019; Heizer, 1949), we stress that further research is nup-Riordan, 2007; Henry et al., 2018). But fish could also be preserved necessary to confirm such patterns. Indeed, pottery data from the Cape through cooking. Pottery would have provided a controlled way to Krusenstern site (900–300 cal BP) indicates that anadromous and process these resources quickly and efficiently for long-term storage freshwater fish remained important among the late prehistoric in­ (Jordan and Gibbs, 2019b). Subsequently, these pots could have been habitants of this site (Anderson et al., 2017). Perhaps the connection of used to rehydrate and cook dried fish in the winter season. Our data pottery technology to the processing of riverine resources lingered shows that while Norton pottery seems to have been predominantly used longer than expected, even within the strongly maritime-adapted to process aquatic resources, it did not seem to have a very specialist tradition of the Western Thule. function (e.g. for rendering oil). Instead it was likely used as a cooking vessel to prepare dishes in which fish was the main ingredient. 4.2. Geographically dependent pottery use 4.4. Implications for the origins of alaskan pottery traditions The direct evidence for pottery function we have generated through organic residue analysis is in line with suggested subsistence strategies As the function of early pottery in Alaska is becoming clearer, further of the Norton and ancestral Yup’ik traditions in Southwest Alaska questions now arise in relation to ancestral pottery traditions of (Dumond, 2016; Shaw, 1998), and supports the notion that differences Northeast Siberia (Yakutia, Kamchatka and Chukotka), thought to be the in subsistence are specific to the geographically dependent distribution ultimate source of the Alaskan ceramic traditions. While technological of prey-species (Miszaniec et al., 2021, p. 19; Moss, 2012). Indeed, the comparisons between pottery from Alaska and Siberia have been drawn strongest differences in pottery use on the Alaska Peninsula are observed before (Ackerman, 1982; Oswalt, 1953), reconstructing pottery function on a geographic scale between three defined regions (Pacific coast (PC), can shed new light on this question. The abundant presence of Norton Bering Sea coast (BSC), and inland (I) locations). The differences be­ sites on interior riverine locations and the predominant use of the pot­ tween the three regions are apparent in compound specific isotopes tery at those sites to process anadromous resources is interesting. We (Mann Whitney U = 0, z = 8.378, p = < 0.05, PC = 6, BSC = 6, I = 33) have analyzed some of the earliest pottery in Alaska, albeit not the oldest with more depleted values at inland sites, likely reflecting the processing which occurs in Northwest Alaska (Anderson et al., 2017; Dumond, of anadromous fish versus marine species on the coast. TMTD/phytanic 2016; Tremayne et al., 2018) and remains untested. The main use of acid ratios (Mann Whitney U = 37, z = 8.762, p = < 0.05; PC = 7, BSC = these pots to process riverine fish raises the question whether pottery 8, I = 37) are higher on the coasts than in the inland. This is in line with entered the New World Arctic as part of a riverine adaptation (Admiraal, our analysis of modern reference data which shows lower TMTD/phy­ 2020; Jordan et al., 2022). Pottery sites in Eastern Siberia are often tanic acid ratios in anadromous fish. In turn, lipid concentrations are located along large river systems (e.g. the Lena, Kolyma and Anadyr highest on the Bering Sea coast as opposed to the Pacific coast and inland rivers, see Fig. 1a). Pottery in this area may have been used in the same 8 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 way as in Alaska. Especially the environments of Southwest Alaska with Groningen. The project was co-supervised by co-authors PDJ and OEC. its major river drainage systems (e.g. the Yukon-Kuskokwim, Ayukalik, This work was supported by the University of Groningen, Faculty of Arts, and Naknek rivers) could have reflected origin regions in interior the Arts and Humanities Research Council (AH/L0069×/1), and the Northeast Siberia. Organic residue analysis of Northeast Siberian pottery European Research Council (ERC-2015-AdG No 695539). We are (Admiraal et al. forthcoming) and further work on early Alaskan pottery grateful to Pamela Endzweig and the Museum of Natural and Cultural assemblages will help resolve this question. History in Eugene, Kathryn Myers and the Katmai National Park Service, as well as Aaron Crowell and the Smithsonian Arctic Studies Center in 5. Conclusions Anchorage for facilitating the sampling of the ceramic vessels. We thank the Zeehondencentrum Pieterburen as well as Ester Oras for supplying Here we examined the function of pottery in Southwest Alaska to reference materials, and Kristine Korzow Richter for identification of better understand the adoption of ceramic technology in this sub-Arctic faunal remains through ZooMS. Furthermore, we thank Frits Steen­ environment. Through lipid residue and stable isotope analysis we huisen for creating the maps used in this research, and extend our presented unequivocal evidence that Southwest Alaskan pottery was gratitude to Shinya Shoda and Laura Llorente Rodriguez for their used to process aquatic resources. Norton sites are predominantly pre­ valuable contribution to MA’s lab training. sent at inland locations along major rivers and the pottery at these sites was predominantly used to process anadromous fish, although there is Supplementary data some variation with contributions of caribou and potentially plant products. Interestingly, we also found evidence for the presence of Supplementary data to this article can be found online at https://doi. marine resources at some of the earliest inland sites of the Alaska org/10.1016/j.jas.2023.105824. Peninsula (NAK-3 and DIL-161), as well as on the Pacific coast. We observe that pottery function may have included daily cooking prac­ References tices, but we infer that pottery could also have been a valuable tool to process fish during seasonal mass harvests, although likely not for the Ackerman, R.E., 1982. The neolithic-bronze age cultures of Asia and the Norton phase of alaskan prehistory. Arctic Anthropol. 19, 11–38. rendering of oil. Although pottery technology drastically changes under Ackman, R.G., Hansen, R.P., 1967. The occurrence of diastereomers of phytanic and the influence of the Western Thule from Northern Alaska at around pristanic acids and their determination by gas-liquid chromatography. Lipids 2, 1000 cal BP, this transition is not as clear in pottery function. Instead, 357–362. Ackman, R.G., Hooper, S.N., 1968. Examination of isoprenoid fatty acids as differences in pottery function seem to be linked to shifting site locations distinguishing characteristics of specific marine oils with particular reference to with environmental differences that deviate from the generally occupied whale oils. Comp. Biochem. Physiol. 24, 549–565. range of environments by Norton and ancestral Yup’ik groups (i.e. Admiraal, M., 2020. The adoption of pottery into the New World: exploring pottery function and dispersal. In: Southwest Alaska through Organic Residue Analysis. Ph. riverine vs. coastal subsistence foci; Dumond, 2000; Shaw, 1998). D.). University of Groningen. Indeed, this geographically-dependent function of pottery is also Admiraal, M., Knecht, R., 2019. Understanding the function of container technologies in observed during the Norton period with results pointing to marine prehistoric Southwest Alaska. In: Ceramics in Circumpolar Prehistory: Technology, Lifeways and Cuisine. Cambridge University Press, pp. 104–127. resource processing on the Pacific coast. Admiraal, M., Lucquin, A., Drieu, L., Casale, S., Jordan, P.D., Craig, O.E., 2020a. We have also explored the potential implications of our results for Leftovers: the presence of manufacture-derived aquatic lipids in Alaskan pottery. the adoption of pottery in this part of the world. While pottery tech­ Archaeometry 62, 346–361. nology first entered Alaska among maritime adapted groups (i.e. Choris/ Admiraal, M., Lucquin, A., von Tersch, M., Craig, O.E., Jordan, P.D., 2020b. The adoption of pottery on Kodiak Island: insights from organic residue analysis. Quat. Norton in Northwest Alaska) organic residue analysis studies on early Int. 554, 128–142. Alaskan pottery so far indicates its main function was aimed at anad­ Admiraal, M., Lucquin, A., von Tersch, M., Jordan, P.D., Craig, O.E., 2019. Investigating romous and freshwater fish rather than marine resources. While sam­ the function of prehistoric stone bowls and griddle stones in the Aleutian Islands by lipid residue analysis. Quat. Res. 91, 1003–1015. pling strategies may play a role in this conclusion, this is also supported Anderson, S.L., Tushingham, S., Buonasera, T.Y., 2017. Aquatic adaptations and the by organic residue analysis at the estuary Cape Krusenstern site in adoption of Arctic pottery technology: results of residue analysis. Am. Antiq. 82, Northwest Alaska (Anderson et al., 2017). We hypothesize that anad­ 452–479. Anderson, S.L., 2019. Ethnographic and archaeological perspectives on the use life of romous and freshwater fish played an important role in the expansion of Northwest Alaskan pottery. In: Ceramics in Circumpolar Prehistory: Technology, pottery in Alaska. It is possible that inland routes along large rivers could Lifeways and Cuisine. Cambridge University Press, pp. 128–151. have been feasible for the spread of this technology. This idea has im­ Avigan, J., Blumer, M., 1968. On the origin of pristane in marine organisms. J. Lipid Res. 9, 350–352. plications for the dispersal of pottery in Siberia, where early pottery sites Barton, L., Shirar, S., Jordan, J.W., 2018. Holocene human occupation of the central are often found along large rivers and only appear on the coast with the Alaska Peninsula. Radiocarbon 60, 367–382. rise of maritime adaptations in the area (Jordan et al., 2022). The Bockstoce, J., 1979. The Archaeology of Cape Nome, Alaska. UPenn Museum of Archaeology. function of these Siberian pottery vessels remains unclear and further Bondetti, M., González Carretero, L., Dolbunova, E., McGrath, K., Presslee, S., research is now underway to clarify this important issue (Admiraal et al. Lucquin, A., Tsybriy, V., Mazurkevich, A., Tsybriy, A., Jordan, P., Heron, C., forthcoming). Additionally, more research into the variability and Meadows, J., Craig, O.E., 2021a. Neolithic farmers or Neolithic foragers? Organic changing function of pottery in the different Alaskan environments is residue analysis of early pottery from Rakushechny Yar on the Lower Don (Russia). Archaeol. Anthropol. Sci. 13, 141. needed to better understand spatiotemporal patterns of adoption and Bondetti, M., Scott, E., Courel, B., Lucquin, A., Shoda, S., Lundy, J., Labra-Odde, C., transformation, and to close gaps in knowledge. Drieu, L., Craig, O.E., 2021b. Investigating the formation and diagnostic value of ω-(o-alkylphenyl)alkanoic acids in ancient pottery. Archaeometry 63, 594–608. Bondetti, M., Scott, S., Lucquin, A., Meadows, J., Lozovskaya, O., Dolbunova, E., Declaration of competing interest Jordan, P., Craig, O.E., 2020. Fruits, fish and the introduction of pottery in the Eastern European plain: lipid residue analysis of ceramic vessels from Zamostje 2. The authors declare that they have no known competing financial Quat. Int. 541, 104–114. Britton, K., Knecht, R., Nehlich, O., Hillerdal, C., Davis, R.S., Richards, M.P., 2013. interests or personal relationships that could have appeared to influence Maritime adaptations and dietary variation in prehistoric Western Alaska: stable the work reported in this paper. isotope analysis of permafrost-preserved human hair. Am. J. Phys. Anthropol. 151, 448–461. Bundy, B., 2007. A Norton tradition village site on the Alagnak River, southwest Alaska. Acknowledgements Alaska J. Anthropol. 5, 1–22. Buonasera, T.Y., Tremayne, A.H., Darwent, C.M., Eerkens, J.W., Mason, O.K., 2015. Lipid The authors dedicate this work to the late Dr. Don Dumond, whose biomarkers and compound specific δ13C analysis indicate early development of a dual-economic system for the Arctic Small Tool tradition in northern Alaska. invaluable support, guidance and advice provided the backbone of this J. Archaeol. Sci. 61, 129–138. study. This research is part of MA’s PhD project at the University of 9 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 Byers, D.A., Yesner, D.R., Broughton, J.M., Coltrain, J.B., 2011. Stable isotope chemistry, Frink, L., Harry, K.G., 2019. An exploration of Arctic ceramic and soapstone cookware population histories and Late Prehistoric subsistence change in the Aleutian Islands. technologies and food preparation systems. In: Ceramics in Circumpolar Prehistory: J. Archaeol. Sci. 38, 183–196. Technology, Lifeways and Cuisine. Cambridge University Press, pp. 152–167. Casperson, M., 2018. Walrus, Seal, and Seabird Faunal Remains from Summit Island in Garnier, N., Bernal-Casasola, D., Driard, C., Pinto, I.V., 2018. Looking for ancient fish Bristol Bay, Alaska: the Subsistence Practices of Norton Peoples in an Island products through invisible biomolecular residues in the Roman production vats from Environment (2740–980 Cal BP). the Atlantic Coast. J. Marit. Archaeol. 13, 285–328. Choy, K., Potter, B.A., McKinney, H.J., Reuther, J.D., Wang, S.W., Wooller, M.J., 2016. Gibbs, K., Isaksson, S., Craig, O.E., Lucquin, A., Grishchenko, V.A., Farrell, T.F.G., Chemical profiling of ancient hearths reveals recurrent salmon use in Ice Age Thompson, A., Kato, H., Vasilevski, A.A., Jordan, P.D., 2017. Exploring the Beringia. Proc. Natl. Acad. Sci. U. S. A 113, 9757–9762. emergence of an “Aquatic” Neolithic in the Russian Far East: organic residue analysis Clark, G.H., 1977. Archaeology of the Alaska Peninsula: the Coast of Shelikof Strait 1963- of early hunter-gatherer pottery from Sakhalin Island. Antiquity 91, 1484–1500. 1965. Hansel, F.A., Bull, I.D., Evershed, R.P., 2011. Gas chromatographic mass spectrometric Colonese, A.C., Farrell, T., Lucquin, A., Firth, D., Charlton, S., Robson, H.K., detection of dihydroxy fatty acids preserved in the “bound” phase of organic residues Alexander, M., Craig, O.E., 2015. Archaeological bone lipids as palaeodietary of archaeological pottery vessels. Rapid Commun. Mass Spectrom. 25, 1893–1898. markers. Rapid Commun. Mass Spectrom. 29, 611–618. Hansel, F.A., Copley, M.S., Madureira, L.A.S., Evershed, R.P., 2004. Thermally produced Coltrain, J.B., 2010. Alaska Peninsula Stable isotope and radioisotope chemistry: a study ω-(o-alkylphenyl)alkanoic acids provide evidence for the processing of marine in temporal and adaptive diversity. Hum. Biol. 82, 613–627. products in archaeological pottery vessels. Tetrahedron Lett. 45, 2999–3002. Coltrain, J.B., Hayes, M.G., O’Rourke, D.H., 2004. Sealing, whaling and caribou: the Hansel, F.A., Evershed, R.P., 2009. Formation of dihydroxy acids from Z- skeletal isotope chemistry of Eastern Arctic foragers. J. Archaeol. Sci. 31, 39–57. monounsaturated alkenoic acids and their use as biomarkers for the processing of Coltrain, J.B., Tackney, J., O’Rourke, D.H., 2016. Thule whaling at Point Barrow, Alaska: marine commodities in archaeological pottery vessels. Tetrahedron Lett. 50, the Nuvuk cemetery stable isotope and radiocarbon record. J. Archaeol. Sci.: Report 5562–5564. 9, 681–694. Harry, K., Frink, L., 2009. The Arctic cooking pot: why was it adopted? Am. Anthropol. Courel, B., Robson, H.K., Lucquin, A., Dolbunova, E., Oras, E., Adamczak, K., 111, 330–343. Andersen, S.H., Astrup, P.M., Charniauski, M., Czekaj-Zastawny, A., Ezepenko, I., Hauff, S., Vetter, W., 2010. Quantification of branched chain fatty acids in polar and Hartz, S., Kabaciński, J., Kotula, A., Kukawka, S., Loze, I., Mazurkevich, A., neutral lipids of cheese and fish samples. J. Agric. Food Chem. 58, 707–712. Piezonka, H., Piličiauskas, G., Sørensen, S.A., Talbot, H.M., Tkachou, A., Heizer, R.F., 1949. Pottery from the southern eskimo region. Proc. Am. Phil. Soc. 93, Tkachova, M., Wawrusiewicz, A., Meadows, J., Heron, C.P., Craig, O.E., 2020. 48–56. Organic residue analysis shows sub-regional patterns in the use of pottery by Hendy, J., Colonese, A.C., Franz, I., Fernandes, R., Fischer, R., Orton, D., Lucquin, A., Northern European hunter-gatherers. R. Soc. Open Sci. 7, 192016. Spindler, L., Anvari, J., Stroud, E., Biehl, P.F., Speller, C., Boivin, N., Mackie, M., Cox, R.E., Maxwell, J.R., Ackman, R.G., Hooper, S.N., 1972. Stereochemical studies of Jersie-Christensen, R.R., Olsen, J.V., Collins, M.J., Craig, O.E., Rosenstock, E., 2018. acyclic isoprenoid compounds. iii. the stereochemistry of naturally occurring Ancient proteins from ceramic vessels at Çatalhöyük West reveal the hidden cuisine (marine) 2,6,10,14-Tetramethylpentadecane. Can. J. Biochem. 50, 1238–1241. of early farmers. Nat. Commun. 9, 4064. Craig, O.E., Saul, H., Lucquin, A., Nishida, Y., Tache, K., Clarke, L., Thompson, A., Henn, W., 1978. Archaeology on the Alaska Peninsula. The Ugashik Drainage. University Altoft, D.T., Uchiyama, J., Ajimoto, M., Gibbs, K., Isaksson, S., Heron, C.P., of Oregon Anthropological Papers, Eugene, Oregon. Jordan, P., 2013. Earliest evidence for the use of pottery. Nature 496, 351–354. Henry, A., Zavadskaya, E., Alix, C., Kurovskaya, E., Beyries, S., 2018. Ethnoarchaeology Craig, O.E., Steele, V.J., Fischer, A., Hartz, S., Andersen, S.H., Donohoe, P., Glykou, A., of fuel use in northern forests: towards a better characterization of prehistoric fire- Saul, H., Jones, D.M., Koch, E., Heron, C.P., 2011. Ancient lipids reveal continuity in related activities. Ethnoarchaeol. Ethnoarchaeol. 10 (2), 99–120. culinary practices across the transition to agriculture in Northern Europe. Proc. Natl. Heron, C., Andersen, S., Fischer, A., Glykou, A., Hartz, S., Saul, H., Steele, V., Craig, O., Acad. Sci. USA 108, 17910–17915. 2013. Illuminating the late mesolithic: residue analysis of “blubber” lamps from Cramp, L.J.E., Jones, J., Sheridan, A., Smyth, J., Whelton, H., Mulville, J., Sharples, N., northern europe. Antiquity 87, 178–188. Evershed, R.P., 2014. Immediate replacement of fishing with dairying by the earliest Heron, C., Craig, O.E., 2015. Aquatic resources in foodcrusts: identification and farmers of the Northeast Atlantic archipelagos. Proc. Biol. Sci. 281, 20132372. implication. Radiocarbon 57. de Laguna, F., 2000. Travels Among the Dena: Exploring Alaska’s Yukon Valley. Hilton, M.R., 2002. Evaluating site formation processes at a higher resolution: An University of Washington Press, Seattle. archaeological case study in Alaska using micromorphology and experimental Demirci, Ö., Lucquin, A., Çakırlar, C., Craig, O.E., Raemaekers, D.C.M., 2021. Lipid techniques (PhD). University of California, Los Angeles. residue analysis on Swifterbant pottery (c. 5000–3800 cal BC) in the Lower Rhine- Horiuchi, A., Miyata, Y., Kamijo, N., Cramp, L., Evershed, R.P., 2015. A dietary study of Meuse area (The Netherlands) and its implications for human-animal interactions in the Kamegaoka culture population during the Final Jomon period, Japan, using relation to the Neolithisation process. J. Archaeol. Sci.: Report 36, 102812. stable isotope and lipid analyses of ceramic residues. Radiocarbon 57, 721–736. Dikov, N.N., 1979. Ancient Cultures of the Northeastern Asia. Nauka. Jordan, P.D., Ponkratova, I., Diakonov, V.M., Solovyova, E., Yamahara, T., Kato, H., Dudd, S.N., Evershed, R.P., Gibson, A.M., 1999. Evidence for varying patterns of Admiraal, M., 2022. Tracking the adoption of early pottery traditions into maritime exploitation of animal products in different prehistoric pottery traditions based on Northeast Asia: emerging insights and new questions. In: Fitzhugh, J.C.I.P. (Ed.), lipids preserved in surface and absorbed residues. J. Archaeol. Sci. 26, 1473–1482. Maritime Prehistory of Northeast Asia. Springer, New York. Dumond, D.E., 2016. Norton hunters and fisherfolk. In: Friesen, M., Mason, O. (Eds.), Jordan, P., Gibbs, K. (Eds.), 2019a. Ceramics in Circumpolar Prehistory: Technology, The Oxford Handbook of the Prehistoric Arctic. Oxford University Press, p. 395, 371. Lifeways and Cuisine. Cambridge University Press. Dumond, D.E., 1981. Archaeology on the Alaska Peninsula: The Naknek Region, 1960- Jordan, P., Gibbs, K., 2019b. Cold winters, hot soups and frozen clay. In: Ceramics in 1975. University of Oregon Anthropological Papers. Eugene. Circumpolar Prehistory: Technology, Lifeways and Cuisine. Cambridge University Dumond, D.E., 2000. The Norton tradition. Arctic Anthropol. 37, 1–22. Press, pp. 1–16. Dumond, D.E., 2003. Archaeology on the Alaska Peninsula: the Leader Creek Site and its Käkelä, R., Hyvärinen, H., Vainiotalo, P., 1996. Unusual fatty acids in the depot fat of the Context. University of Oregon Anthropological Papers, Eugene. Canadian beaver (Castor canadensis). Comp. Biochem. Physiol. B Biochem. Mol. Dumond, D.E., 2011. Archaeology on the Alaska Peninsula: the Northern Section. fifty Biol. 113, 625–629. years onward, University of Oregon Anthropological Papers, Eugene. Khvostof, N.A., Davydov, G.I., 1810. Dvukratnoie Putieshestvi V Ameriku, p. 2. St. Dumond, D.E., Bland, R.L., 1995. Holocene prehistory of the northernmost north Pacific. Petersburg. J. World PreHistory 9, 401–451. Knudson, K.J., Frink, L., 2011. Nitrogen isotope analysis in the Arctic: identifying fish Evershed, R.P., 2008a. Organic residue analysis in archaeology: the archaeological processing and marine resource use through ethnoarchaeological soil analysis on biomarker revolution. Archaeometry 50, 895–924. Nelson Island, Alaska. Alaska J. Anthropol. 9, 17–54. Evershed, R.P., 2008b. Experimental approaches to the interpretation of absorbed Kuzmin, Y.V., 2014. The neolithization of Siberia and the Russian far east: major organic residues in archaeological ceramics. World Archaeol. 40, 26–47. spatiotemporal trends (the 2013 state-of-the-art). Radiocarbon 56, 717–722. Evershed, R.P., Copley, M.S., Dickson, L., Hansel, F.A., 2008. Experimental evidence for Lucquin, A., Colonese, A.C., Farrell, T.F.G., Craig, O.E., 2016a. Utilising phytanic acid the processing of marine animal products and other commodities containing diastereomers for the characterisation of archaeological lipid residues in pottery polyunsaturated fatty acids in pottery vessels. Archaeometry 50, 101–113. samples. Tetrahedron Lett. 57, 703–707. Evershed, R.P., Heron, C., Goad, J., 1990. Analysis of organic residues of archaeological Lucquin, A., Gibbs, K., Uchiyama, J., Saul, H., Ajimoto, M., Eley, Y., Radini, A., Heron, C. origin by high-temperature gas chromatography and gas chromatography mass P., Shoda, S., Nishida, Y., Lundy, J., Jordan, P., Isaksson, S., Craig, O.E., 2016b. spectrometry. Analyst 115, 1339–1342. Ancient lipids document continuity in the use of early hunter-gatherer pottery Farrell, T., 2013. A Lipid Residue Analysis of Pottery Sherds from the Yup’Ik Eskimo Site through 9,000 years of Japanese prehistory. Proc. Natl. Acad. Sci. U. S. A 113, of Nunalleq (Alaska). 3991–3996. Farrell, T.F.G., Jordan, P., Taché, K., Lucquin, A., Gibbs, K., Jorge, A., Britton, K., Lutz, B.J., 1982. Population pressure and climate as dynamics within the arctic small tool Craig, O.E., Knecht, R., 2014. Specialized processing of aquatic resources in tradition of Alaska. Arctic Anthropol. 19 (2), 143–149. prehistoric Alaskan Pottery? A lipid-residue analysis of ceramic sherds from the Marsh, J.M., Mueter, F.J., Iken, K., Danielson, S., 2017. Ontogenetic, spatial and Thule-Period site of Nunalleq, Alaska. Arctic Anthropol. 51, 86–100. temporal variation in trophic level and diet of Chukchi Sea fishes. Deep Sea Res. Part Fernandes, R., Meadows, J., Dreves, A., Nadeau, M.-J., Grootes, P., 2014. A preliminary 2 Top. Stud. Oceanogr. 135, 78–94. study on the influence of cooking on the C and N isotopic composition of multiple Misarti, N., Finney, B., Maschner, H., Wooller, M.J., 2009. Changes in northeast Pacific organic fractions of fish (mackerel and haddock). J. Archaeol. Sci. 50, 153–159. marine ecosystems over the last 4500 years: evidence from stable isotope analysis of Fienup-Riordan, A., 2007. Yuunnaqpiallerput. In: Fienup-Riordan, A. (Ed.), The Way We bone collagen from archeological middens. Holocene 19, 1139–1151. Genuinely Live, Master Work of Yup’Ik Science and Survival. University of Miszaniec, J.I., Darwent, J., Darwent, C.M., 2021. Small game, estuaries, and nets: new Washington Press. perspectives on Norton culture coastal adaptations from a shell midden in Norton Frink, L., Harry, K.G., 2008. The beauty of “ugly” Eskimo cooking pots. Am. Antiq. 73, Sound, Alaska. J. I. Coast Archaeol. 16, 317–341. 103–120. 10 M. Admiraal et al. Journal of Archaeological Science 157 (2023) 105824 Moss, M.L., 2012. Understanding variability in Northwest coast faunal assemblages: Meadows, J., Hartz, S., Craig, O.E., Heron, C.P., 2022. Light production by ceramic beyond economic intensification and cultural complexity. J. I. Coast Archaeol. 7, using hunter-gatherer-fishers of the Circum-Baltic. Proceedings of the Prehistoric 1–22. Society 88, 25–52. Oras, E., Lucquin, A., Lõugas, L., Tõrv, M., Kriiska, A., Craig, O.E., 2017. The adoption of Schaaf, J.M., 2008. Mink Island, Katmai national Park and preserve, pacific coast of the pottery by north-east European hunter-gatherers: evidence from lipid residue Alaska peninsula. Alaska Park Sci. 7, 35–39. analysis. J. Archaeol. Sci. 78, 112–119. Shaw, R.D., 1998. An archaeology of the central yupik: a regional overview for the Oswalt, W., 1953. Northeast asian and alaskan pottery relationships. SW. J. Anthropol. 9, yukon-kuskokwim delta, northern bristol bay, and nunivak Island. Arctic Anthropol. 395–407. 35, 234–246. Oswalt, W., 1955. Alaskan pottery: a classification and historical reconstruction. Am. Simoneit, B.R.T., Rogge, W.F., Lang, Q., Jaffé, R., 2000. Molecular characterization of Antiq. 21, 32–43. smoke from campfire burning of pine wood (Pinus elliottii). Chemosphere Global Pääkkönen, M., Evershed, R.P., Asplund, H., 2020. Compound-specific stable carbon Change Sci. 2, 107–122. isotope values of fatty acids in modern aquatic and terrestrial animals from the Baltic Solazzo, C., Erhardt, D., 2007. Analysis of lipid residues in archaeological artifacts: sea Sea and Finland as an aid to interpretations of the origins of organic residues mammal oil and cooking practices in the Arctic. In: Barnard, H., Eerkens, J.W. (Eds.), preserved in archaeological pottery. Journal of Nordic Archaeological Science Theory and Practice of Archaeological Residue Analysis. British Archaeological (JONAS) 19, 3–19. Reports, Oxford, pp. 161–178. Papakosta, V., Smittenberg, R.H., Gibbs, K., Jordan, P., Isaksson, S., 2015. Extraction and Solazzo, C., Fitzhugh, W.W., Rolando, C., Tokarski, C., 2008. Identification of protein derivatization of absorbed lipid residues from very small and very old samples of remains in archaeological potsherds by proteomics. Anal. Chem. 80, 4590–4597. ceramic potsherds for molecular analysis by gas chromatography–mass spectrometry Taché, K., Craig, O.E., 2015. Cooperative harvesting of aquatic resources and the (GC–MS) and single compound stable carbon isotope analysis by gas beginning of pottery production in north-eastern North America. Antiquity 89, chromatography–combustion–isotope ratio mass spectrometry (GC–C–IRMS). 177–190. Microchem. J. 123, 196–200. Tremayne, A., Darwent, C.M., Darwent, J., Eldridge, K., Rasic, J.T., 2018. Iyatayet Pauly, D., Trites, A.W., Capuli, E., Christensen, V., 1998. Diet composition and trophic revisited: a report on renewed investigations of a stratified Middle-to-Late Holocene levels of marine mammals. ICES J. Mar. Sci. 55, 467–481. coastal campsite in Norton Sound, Alaska. Arctic Anthropol. 55, 1–23. Ponkratova, I., 2006. Pottery industries in the North of the Russian far east. In: Tremayne, A.H., Brown, W.A., 2017. Mid to Late Holocene population trends, culture Dumond, D.E., Bland, R.L., Bland, R.L. (Eds.), Tran.), Archaeology in Northeast Asia: change and marine resource intensification in Western Alaska. Arctic 70, 365. on the Pathway to Bering Strait. University of Oregon Anthropological Papers. Tushingham, S., Bettinger, R.L., 2013. Why foragers choose acorns before salmon: University of Oregon, Oregon, pp. 129–159. storage, mobility, and risk in aboriginal California. J. Anthropol. Archaeol. 32 (4), Post, D.M., Layman, C.A., Arrington, D.A., Takimoto, G., Quattrochi, J., Montaña, C.G., 527–537. 2007. Getting to the fat of the matter: models, methods and assumptions for dealing van Bergen, P.F., Nott, C.J., Bull, I.D., Poulton, P.R., Evershed, R.P., 1998. Organic with lipids in stable isotope analyses. Oecologia 152, 179–189. geochemical studies of soils from the Rothamsted Classical Experiments—IV. Ratnayake, W.M.N., Olsson, B., Ackman, R.G., 1989. Novel branched-chain fatty acids in Preliminary results from a study of the effect of soil pH on organic matter decay. Org. certain fish oils. Lipids 24, 630–637. Geochem. 29, 1779–1795. Robson, H.K., Lucquin, A., Admiraal, M., Dolbunova, E., Adamczak, K., Czekaj- West, C.F., France, C.A., 2015. Human and Canid dietary relationships: comparative Zastawny, A., Fitzhugh, W.W., Gumiński, W., Kabaciński, J., Kotula, A., Kukawka, S., stable isotope analysis from the Kodiak Archipelago, Alaska. J. Ethnobiol. 35, Oras, E., Piezonka, H., Piličiauskas, G., Sørensen, S.A., Thielen, L., Wetzel, G., 519–535. 11