Megafauna forms a charismatic component of the North American world that has participated in Native American cultures through a range of utilitarian and symbolic avenues. Among them are extinct megafauna, which have largely been seen under the lens of short-lived, predatory interactions occurring at the very end of the Ice Age.
Proboscideans (mammoths, mastodons, and gomphotheres) provide a good illustration. Archaeological research on proboscidean–Native American interactions began with the 1929 discovery of mammoth bones in association with projectile points in Clovis, New Mexico (Howard Reference Howard1935). The North American archaeological record of proboscidean sites has since grown to encompass several dozen, many of them kill sites also dated within a short, roughly 500-year temporal interval at the end of the Late Pleistocene (Haynes Reference Haynes2022; Haynes and Huckell Reference Haynes and Huckell2007; Mackie et al. Reference Mackie, Surovell, O’Brien, Kelly, Spencer Pelton and Frison2020; Sanchez et al. Reference Sanchez, Holliday, Gaines, Arroyo-Cabrales, Martínez-Tagüeña, Kowler, Lange, Hodgins, Mentzer and Sanchez-Morales2014; Surovell and Waguespack Reference Surovell and Waguespack2008). The literature has accordingly centered on the importance of mammoth in the diet and economy of early Native North Americans (Chatters et al. Reference Chatters, Potter, Fiedel, Morrow, Jass and Wooller2024; Kilby et al. Reference Kilby, Surovell, Huckell, Ringstaff, Hamilton and Haynes2022; Waguespack and Surovell Reference Waguespack and Surovell2003), as well as on the relevance of human predation in understanding proboscidean extirpation (Broughton and Weitzel Reference Broughton and Weitzel2018; Haynes Reference Haynes2002; Martin Reference Martin, Martin and Wright1967; Meltzer Reference Meltzer2020; Surovell and Waguespack Reference Surovell and Waguespack2008; Surovell et al. Reference Surovell, Waguespack and Brantingham2005, Reference Surovell, Pelton, Anderson-Sprecher and Myers2016). The literature also discusses the functional interpretations of manufactured elongated proboscidean ivory and/or bone objects occasionally recovered at contemporaneous Late Pleistocene sites (Bradley Reference Bradley, Hahn, Menu, Taborin, Walter and Widemann1995; Hemmings Reference Hemmings2004; O’Brien et al. Reference O’Brien, Lyman, Buchanan, Collard and Langley2016; Sutton Reference Sutton2018). This uneven archaeological record has, however, limited our understanding of other facets of interactions between Native North Americans and proboscideans, including nonutilitarian uses of proboscidean products and interactions with proboscideans after their extirpation.
In this article, we explore the larger cultural aspects of interactions between proboscideans and Native Americans in Alaska—a northwestern region of North America (Figure 1)—during both the periods of coexistence and after proboscidean extirpation. Alaska (and adjacent Yukon, collectively known as eastern Beringia) was settled by people at the end of the Late Pleistocene in much the same process as the rest of the continent (Goebel and Buvit Reference Goebel and Buvit2011; Hoffecker et al. Reference Hoffecker, Powers and Goebel1993; Potter et al. Reference Potter, Doering, Lanoë, Nicholas Schmuck, Rasic, Holmes and Esdale2026). Given this timing, one may expect a similar record and understanding of interactions between Alaska Natives and the local proboscidean taxon (woolly mammoth, Mammuthus primigenius Footnote 1).
Map of Alaska and places and groups mentioned in text. The dashed line marks the boundary between Eskaleut- (Iñupiat/Inuvialuit, Yupiit, Unangax̂) and Na-Dene- (Dene, Tlingit, Eyak) speaking Alaska Native groups.
Figure 1 Long description
The map of Alaska displays various Native groups and geographic features. The main map includes regions such as Iñupiat, Inuvialuit, Gwich′in, Koyukon, Tanana, Yupiit and Unangax̂. Geographic features include the Bering Sea, Chukchi Sea and Beaufort Sea. Islands like St. Lawrence Island and Nunivak Island are marked. The inset map shows the Seward Peninsula with locations like Nome and St. Michael. A dashed line marks the boundary between Eskaleut- and Na-Dene-speaking groups. The map includes a scale bar indicating 500 kilometers.
Several factors, however, set Alaska apart from the rest of North America. Mammoth remains there are very numerous owing to favorable sedimentary and climatic preservation conditions and, possibly, reflecting higher living populations in the past (Guthrie Reference Guthrie2004, Reference Guthrie2006; Harington Reference Harington2003; Krasinski and Haynes Reference Krasinski and Haynes2010; Mann et al. Reference Mann, Groves, Kunz, Reanier and Gaglioti2013; Rowe et al. Reference Rowe, Bataille, Baleka, Combs, Crass, Fisher and Ghosh2024; Wooller et al. Reference Wooller, Bataille, Druckenmiller, Erickson, Groves, Haubenstock and Howe2021). Favorable conditions have led to comparatively well-preserved specimens, with occasional recovery of soft tissues. Last, as in nearby Wrangel Island, Chukotka (Vartanyan et al. Reference Vartanyan, Garutt and Sher1993), mammoths remained in parts of Alaska such as the Pribilof Islands well into the Holocene (Graham et al. Reference Graham, Belmecheri, Choy, Culleton, Davies, Froese and Heintzman2016; Guthrie Reference Guthrie2004; Veltre et al. Reference Veltre, Yesner, Crossen, Graham and Coltrain2008), some 6,000 years after the last recorded proboscidean dates in the rest of North America (Broughton and Weitzel Reference Broughton and Weitzel2018; Sanchez et al. Reference Sanchez, Holliday, Gaines, Arroyo-Cabrales, Martínez-Tagüeña, Kowler, Lange, Hodgins, Mentzer and Sanchez-Morales2014).
Here, we not only review existing but provide new archaeological and ethnographic data on the interactions between Alaska Natives and mammoths. We show how those interactions have encompassed nonpredation aspects and occurred long past mammoths’ extirpation. We further discuss how Alaska Native–mammoth relationships inform our larger knowledge of interactions between Native North Americans and extinct megafauna, in both the deep and more recent past.
Methods
This article synthetizes evidence contained within ethnohistorical documents, museum collections, and archaeological literature. In locating ethnohistorical evidence, we employed a snowball-like approach, given that ethnohistorical accounts tend to mention other sources known to the authors at the time. For material evidence, we conducted a semisystematic review of major archaeological/ethnological projects and publicly accessible collections at museums that are associated with such projects and/or are known to have acquired Alaska ethnographic material. Many specimens are listed in both publications and museum records; references in this article indicate the curation facility when known. We retained specimens labeled as “mammoth” (and linguistic variations; e.g., mammut, мамонт) or “mastodon” (common misnomer in old collections). Proboscidean ivory is generally recognizable because it differs from similar materials both macroscopically (dentine cones) and microscopically (Schreger lines; Locke Reference Locke2008), particularly to the trained eyes of archaeologists, artifact collectors, or museum curators. We verified the taxonomic identifications of objects curated at the University of Alaska Museum of the North—and of objects elsewhere—with associated pictures showing ivory macro/microscopic markers; we did not encounter any misidentifications. In this study, we also considered selected specimens marked as “fossil ivory” or “ivory” when associated pictures clearly showed them to be of proboscidean ivory. Our review is likely comprehensive for Pleistocene and Early Holocene sites. It is most likely not comprehensive for the Late Holocene, for which many mammoth specimens may remain hidden in collections/publications under the label “ivory,” as well as in nonpublicly available collections.
This article presents selected new data that consist of mammoth remains from recent excavations and/or analyses at the Swan Point and Mead sites in the Shaw Creek Flats of interior Alaska, which complement the published corpus at those and other sites. Descriptions of the sites and excavation methods can be found in the literature (Holmes Reference Holmes2001; Potter et al. Reference Potter, Holmes, Yesner, Graf, Ketron and Waters2014). Analytical methods follow those previously used for archaeofaunal assemblages from these sites (Lanoë and Holmes Reference Lanoë and Holmes2016; Potter et al. Reference Potter, Holmes, Yesner, Graf, Ketron and Waters2014). Taxonomic identification of one specimen from the Mead site was aided by collagen fingerprinting at the ZooMS laboratory (Michael Buckley, director). The new radiocarbon date in this article is calibrated with OxCal and IntCal20 (Bronk Ramsey Reference Bronk Ramsey1994; Reimer et al. Reference Reimer, Austin, Bard, Bayliss, Blackwell, Ramsey and Butzin2020).
Material Evidence of Mammoths
Before Extirpation
Areas of Alaska / eastern Beringia with potential to document interactions between people and live mammoths need to meet both criteria of (1) spatial and temporal overlap and (2) favorable preservation conditions of organic remains at archaeological sites. Currently, only the Tanana basin, in subarctic interior Alaska, meets those criteria, with other regions either having no evidence of early human presence (e.g., Pribilof Islands) or lacking adequate organic preservation at early archaeological sites (e.g., northern Alaska). Although mammoth blood residues have also been reported in association with several early eastern Beringian projectile points, these identifications are generally thought to be unreliable (Loy and Dixon Reference Loy and Dixon1998; Potter et al. Reference Potter, Reuther, Lowenstein and Scheuenstuhl2010).
Mammoth extirpation in the Tanana basin is dated no earlier than approximately 13.4–13.3 cal kyr BP (Guthrie Reference Guthrie2004, Reference Guthrie2006; Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021), some 500–900 years after the earliest known local archaeological sites (Goebel and Buvit Reference Goebel and Buvit2011; Potter et al. Reference Potter, Holmes, Yesner, Graf, Ketron and Waters2014, Reference Potter, Doering, Lanoë, Nicholas Schmuck, Rasic, Holmes and Esdale2026; Reuther et al. Reference Reuther, Holmes, Smith, Lanoë, Crass, Rowe and Wooller2023). Within that period, a dozen sites have been reported to preserve faunal remains (summarized in Goebel and Buvit Reference Goebel and Buvit2011; Potter et al. Reference Potter, Doering, Lanoë, Nicholas Schmuck, Rasic, Holmes and Esdale2026). Two of those, both in the Shaw Creek Flats, have yielded postcranial mammoth remains in association with human occupations (Table 1; Figure 2). Cultural Zone (CZ) 5 at the Mead site yielded two metapodials, including one directly dated to the time of occupation. Swan Point CZ4b yielded a minimum of eight neonate ribs recovered near a hearth (Lanoë and Holmes Reference Lanoë and Holmes2016; Rowe et al. Reference Rowe, Bataille, Baleka, Combs, Crass, Fisher and Ghosh2024). Specimen surfaces at both sites are heavily affected by root etching and weathering, and as a result, no clear butchery marks have been documented. One probable exception consists of a notch on one of the Swan Point ribs, whose morphology is consistent with dynamic (i.e., anthropogenic) percussion (Lanoë and Holmes Reference Lanoë and Holmes2016).
Selected mammoth postcranial remains from Alaska archaeological sites: (left) ribs of one-month neonate, Swan Point CZ4b; (right) juvenile or adult metapodial, Mead CZ5. UA = University of Alaska Museum; UAF = University of Alaska Fairbanks. (Color online)
Mammoth Remains from Late Pleistocene and Early Holocene Alaska / Eastern Beringia Archaeological Sites.
Table 1 Long description
The table lists mammoth remains from several Alaska archaeological site components, giving each component’s site age range, specimen type, minimum number of individuals, specimen dating range, and literature sources. Most dated specimens cluster in the late Pleistocene, commonly around 14 to 13 thousand years ago, including Broken Mammoth CZ4, Holzman C5a and C5b, Mead CZ5, and Swan Point CZ4b. Swan Point CZ4b has the largest variety and quantity, including a rib with a minimum of eight individuals, multiple tusk sections, two cheek teeth, and more than one thousand ivory fragments, with several specimen dates in the mid 14 to high 13 thousand year range. Mead CZ4/5 includes an ivory fragment dated much earlier, around 21 thousand years ago, standing out from the otherwise younger cluster. Some entries lack specimen dates or list the specimen age as unknown, so comparisons rely mainly on the dated items. Overall, the table indicates repeated mammoth material use or presence across multiple sites near the end of the Pleistocene, with one notably older dated fragment and several undated specimens that limit certainty about timing.
a UGAMS-27682, 11,570 ± 40 14C BP, δ13C = −21.92‰
b Older date (15.2–14.1 cal kyr BP) for this specimen (Potter et al. Reference Potter, Holmes, Yesner, Graf, Ketron and Waters2014) rejected due to contamination (Reuther et al. Reference Reuther, Holmes, Smith, Lanoë, Crass, Rowe and Wooller2023).
c Four younger dates (13.2–10.8 cal kyr BP) rejected due to contamination (Reuther et al. Reference Reuther, Holmes, Smith, Lanoë, Crass, Rowe and Wooller2023).
d Refined age from an originally reported date on the same specimen of 14.8–13.6 cal kyr BP (Holmes Reference Holmes, Goebel and Buvit2011).
Mammoth remains at pre-extirpation archaeological sites are more common in the form of ivory. Of the dozen sites with organic preservation, four—also in the Shaw Creek Flats—have yielded ivory specimens in various forms (Table 1).
Modified elongated objects (Figure 3a–b and d–g) have been documented at Swan Point CZ4b, Holzman Component 5a (C5a), and Broken Mammoth CZ4, with directly dated specimens alternatively contemporaneous with or older than the occupations (Table 1). Most specimens were manufactured on tangential tusk sections, except for the Swan Point object (Figure 3d) made of the nonsectioned, narrower, tusk of a juvenile. Some thinner objects (Figure 3a–b) are manufactured on ivory “splinters” that consist of one or a few dentine cones. All elongated specimens have a plano-convex or oval cross-section and are tapered, rounded, or pointed at one or both ends. Holzman and Broken Mammoth objects display striations and/or polish on preserved surfaces, with one (Figure 3a) presenting coarse striations or hatching on the bevels suggestive of hafting (Heppner Reference Heppner2017; Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021). At least one Broken Mammoth object (Figure 3e) also presents light transverse gouging suggestive of lashing (Heppner Reference Heppner2017).
Pleistocene and Early Holocene Alaskan mammoth ivory technology: (a–b) Holzman C5a (from Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021); (c) Gerstle River C3 (from Potter Reference Potter2001); (d, i–k) Swan Point CZ4b (d and k from Lanoë and Holmes Reference Lanoë and Holmes2016); (e–g) Broken Mammoth CZ4; (h) Broken Mammoth CZ3; (l) Inmachuk River (from Gelvin-Reymiller et al. Reference Gelvin-Reymiller, Reuther, Potter and Bowers2006). UA = University of Alaska Museum; UAF = University of Alaska Fairbanks. (Color online)
Archaeological ivory specimens include large tusk sections from individuals of various ages at Swan Point CZ4b and a large adult tusk at Holzman C5b, with directly dated specimens at both sites contemporaneous with the occupations (Table 1). The adult tusk at Swan Point (Figure 3k) shows percussion marks suggesting its use as an anvil. It also served as an ivory blank, with an approximately 50 cm long negative delimited by deep lateral and transversal grooving of the outer tusk layer. The extracted splinter would have been of comparable thickness to and slightly wider (ca. 2.0–3.5 cm) than the splinter-based objects found at Holzman C5a (Figure 3a–b).
Additional ivory fragments of various sizes are associated with archaeological occupations at Swan Point CZ4b, Holzman C5a and C5b, and Mead CZ4/5; with directly dated specimens alternatively contemporaneous with or older than the occupations (Table 1). Swan Point CZ4b also yielded contemporaneous juvenile cheek teeth. At Swan Point CZ4b and Holzman C5a, most ivory fragments were documented in clusters in the vicinity of hearths (Lanoë and Holmes Reference Lanoë and Holmes2016; Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021). Most documented ivory fragments are too damaged to preserve technological traces, but some Swan Point specimens preserve tapered or stepped edges and/or striking platforms, characteristic of flakes produced by percussion or prying (Figure 3i–j).
After Extirpation
Mammoth ivory specimens continue to appear in Alaskan / eastern Beringian archaeological sites after mammoth extirpation (Table 1). Broken Mammoth CZ3 and Gerstle River C3 in the nearby Deltana Flats both yielded elongated specimens comparable in thickness and dimensions to the Holzman C5a specimens (Figure 3c, h), with the Broken Mammoth specimen also displaying beveling and striations. Mammoth osseous remains bearing similar technological traces have also been documented out of archaeological context and could represent manufacture either before or after local mammoth extirpation. An example from western Alaska is a grooved tusk (Figure 3l), directly dated to 41.0–39.6 cal kyr BP, which served as an ivory blank in a similar manner to the Swan Point tusk, with the negative of an approximately 70 cm long and 1.8–2.5 cm wide shaft delimited by lateral grooves and transversal chop marks (Gelvin-Reymiller et al. Reference Gelvin-Reymiller, Reuther, Potter and Bowers2006). Other examples arguably include mammoth bone flakes documented in locations of the Porcupine River basin, though definite attribution to technological action is less straightforward than for ivory (Bourgeon Reference Bourgeon2021; Cinq-Mars and Morlan Reference Cinq-Mars, Morlan, Bonnichsen and Turnmire1999).
Mammoth ivory objects are more common in Late Holocene archaeological contexts, particularly in western and northern Alaska (Table 2). Mammoth ivory forms a substantial component of archaeological assemblages in and near the Kobuk and Noatak valleys, near the Chukchi Sea coast, dated to roughly 1800–200 cal yr BP; with more isolated archaeological finds near the Bering and Beaufort Seas and in other areas of the Chukchi Sea. Documented objects are dominated by blunt arrowheads (Figure 4n), harpoon heads (Figure 4l), fishing lures (Figure 4j), combs (Figure 4w), net sinkers, tool handles, sled runners (Figure 4k), and shovel blades (Figure 4c). Rarer object types include arrow shaft straighteners, wrist guards, harpoon socket pieces, fishing hooks and spear prongs, spoons, pipe bowls, mouthpieces, wedges, cutting boards, fasteners, anthropomorphic and zoomorphic figurines (Figure 4u), labrets, and pendants (Figure 4u). Ivory blanks and fragments are found alongside finished objects at sites such as Kangiguksuk, Kotzebue, and Ekseavik, dated approximately 600–400 cal yr BP (Gelvin-Reymiller et al. Reference Gelvin-Reymiller, Reuther, Potter and Bowers2006; Giddings Reference Giddings1952; Hall Reference Hall1971; VanStone Reference VanStone1955). Collectively, they show extraction traces, including chopping with adzes, groove-and-splintering, and sawing with ground jade slabs—the latter being documented on mammoth molar fragments in addition to ivory (Figure 4v). Further manufacturing stages appear to have included whittling with stone knives, polishing with whetstones, and occasional engraving with stone knives and/or stone-/metal-tipped engravers (Hall Reference Hall1971).
Late Holocene and recent mammoth ivory objects: (a) ladle, collected in Utqiaġvik in 1888; (b) ladle, collected in Kotzebue in 1827; (c) shovel head, Ipiutak, approximately 1800–1100 cal yr BP (from Larsen and Rainey Reference Larsen and Rainey1948); (d) fat scraper, collected in Chukotka around 1909; (e) fat scraper, Point Hope, approximately 1800–1826; (f) socketed object with figures of a human and two bear heads, Bering Sea region, approximately 2300–1200 cal yr BP (drawn from photograph); (g) polar bear figurine, collected in Kotzebue in 1941; (h) caribou fawn-shaped arrow-shaft straightener, collected near Teller around 1880; (i) scraper, collected near Wainwright before 1923 (from Emmons Reference Emmons1923; no scale); (j) blackfish-shaped (Dallia) lure, Ekseavik, approximately 500 cal yr BP; (k) sled runner, Kotzebue, approximately 400 cal yr BP; (Continued).
Late Holocene Alaskan (and Nearby) Archaeological Sites with Mammoth Ivory Objects.
Table 2 Long description
The table lists archaeological locales in Alaska and nearby Yukon where mammoth ivory objects were found, organized by region, locale, site name, estimated age in calibrated years before present, and literature references. Dates range from about 2300 years before present at Gambell and Kukulik on Saint Lawrence Island to about 100 years before present at Cape Nome (Ayasayuk). Many entries cluster in the late first millennium to early second millennium before present, commonly around 1000 to 200 years before present, including Washout on Herschel Island, Tigara and Nukleet, and several sites in the Noatak and Kobuk valleys. The Beaufort Sea entry at Utqiaġvik (Birnirk) spans roughly 1500 to 400 years before present, while Point Hope includes an older Ipiutak range around 1800 to 1100 years before present and a younger Tigara range around 800 to 200 years before present. Interior river-valley sites in the Kobuk Valley include several single-point estimates around 700, 500, and 200 years before present, suggesting repeated late use rather than one brief episode. Some rows appear to contain formatting or transcription issues, including standalone age values without clear site labels, so exact counts and site-to-date matching may be incomplete.
Use of mammoth ivory was directly documented around the time of European contact mainly among the Iñupiat, and to a lesser extent, among the Yupiit; there was also an isolated instance of purported use of Yup’ik-Iñupiaq armor among the Tlingit in Sitka around 1804 (Table 3). The main objects mentioned to have been manufactured at the time of contact include ladles, scrapers and other tools handles, fat scrapers, labrets, and net sinkers; other rarer tools included spoons, sled runners, labrets, ornaments, and hammers. Sagdluaq, a Colville Iñupiaq, said in 1921–1924 to K. Rasmussen (Ostermann Reference Ostermann1952:152–153), “[Mammoth ivory] was fine ivory, still better than walrus tusk, and when we find a mammoth tusk we fashion weapons out of it, or we carve drinking horns [ladles] that are much larger and more beautiful than what can be made out of muskox horn.” Although ivory was the main mammoth product used, use of the fat of well-preserved mammoths was also reported among the Yupiit near Naknek on Bristol Bay, who used it to seal their kayaks (Dall Reference Dall1896).
Native Localities in Alaska and Adjacent Areas with Documented Use or Collection of Mammoth Ivory Objects.
Table 3 Long description
The table lists Alaska and nearby Arctic localities where mammoth ivory objects are reported, organized by region, Indigenous group, and locale, and noting whether ivory was documented in use or collected. Most entries fall along the Beaufort Sea, Chukchi Sea, and Bering Sea coasts, with fewer records in Southeast Alaska. The Beaufort Sea section includes Inuvialuit Atkinson Point and multiple Iñupiat locations, with Utqiaġvik showing both documented use and multiple collection references and linked nearby archaeological sites. In the Chukchi Sea region, Wainwright, Point Hope and Cape Lisburne, Kotzebue, and the Kobuk and Noatak Valleys include both documented use and collection, and Point Hope and Cape Lisburne are tied to major sites such as Ipiutak and Tigara. The Bering Sea region includes Wales and the Diomede Islands, Port Clarence and Teller, and several Yupiit areas in the Lower Yukon and Lower Kuskokwim, largely represented by collection records. Nearby Chukotka localities, including Naukan and Sireniki and Ungaziq, also show collection records, with Naukan linked to Uelen and Ekven archaeological sites. References are provided for each record, but the table compiles citations rather than counts or dates, so it indicates where documentation exists, not how common ivory objects were at each place.
Early European travelers/anthropologists collected numerous mammoth ivory objects among the Iñupiat and, to a lesser extent, among the Yupiit and the Inuvialuit (Table 3). Collected objects mirror those documented both in archaeological sites of the area and known to be in use at the time of contact. Common objects accordingly include ladles (Figure 4a–b), scrapers and other tools handles (Figure 4i, s), fat scrapers (Figure 4d–e), labrets (Figure 4y), net sinkers (Figure 4q), arrowheads, harpoon heads (Figure 4m), and fishing lures; less common object types include adze heads, wrist guards, bolas, harpoon counterweights (Figure 4x) and socket pieces (Figure 4f), arrow-shaft straighteners (Figure 4h), spoons, bowls, combs, anthropomorphic and zoomorphic figurines (Figure 4o), armor slats and shin guards, mouthpieces, needles and needle cases, buckles and fasteners (Figure 4t), pendants, snow goggles (Figure 4r), snow knives, and unmodified ivory/cheekteeth specimens used as amulets (Figure 4p). These poorly provenienced objects likely reflect manufacture and/or use contemporaneous with their collection, as well as collection from nearby archaeological sites dating to the last approximately 2,000 years (Hollowell Reference Hollowell2004). Some specimens can indeed be assigned to specific archaeological cultures on stylistic grounds, such as harpoon heads and socketed objects of the Old Bering Sea culture dated to approximately 2300–1200 cal yr BP (Figure 4f, m) and harpoon counterweights of the Thule culture around 1000–300 cal yr BP (Figure 4x).
Some Alaska Native groups at the time of contact—for instance, the Iñupiat at Elephant Point near Kotzebue in the 1820s-1830s—knew of and used natural accumulations of mammoth remains as ivory mines (Belcher Reference Belcher1861; Beechey Reference Beechey1832; Gough Reference Gough1973; von Chamisso Reference von Chamisso1835; von Kotzebue Reference von Kotzebue1821). Alaska Natives in general appear to have been well aware of accumulations of mammoth (presumably alongside other megafauna) remains (Gilmore Reference Gilmore1908; Jacobsen Reference Jacobsen1884; Maddren Reference Maddren1905), some of which are named accordingly, including nen’yex deddha’ (“underground driftwood,” explained by the Tanana Dene speaker to refer to bones) near Minto (Kari Reference Kari2024) and taatl’enhledlo denh (“bones lying underwater,” Koyukon Dene) near Tanana (Jetté and Jones Reference Jetté and Jones2000). Some groups from mammoth ivory-rich regions, such as the Iñupiat of the Kobuk/Noatak areas, similarly used raw ivory and/or ivory objects as staple trading items (Murdoch Reference Murdoch1892; Rainey Reference Rainey1947).
Use of mammoth ivory and cheekteeth for crafting has continued since contact among Alaska Natives, particularly the Iñupiat and the Yupiit, with the added use of metal and power tools (Chan Reference Chan2013; Jolles and Dyachkova Reference Jolles, Dyachkova, Fitzhugh, Hollowell and Crowell2009; Ray Reference Ray1961). Commonly manufactured objects include ornaments, beads, figurines, and tool handles (Alaska State Museums; Anchorage Museum; Burke Museum; Larsen and Dickey Reference Larsen and Dickey1982; National Museum of National History; Ray Reference Ray1961, Reference Ray1977; e.g., Figure 4g), with occasional tusk scrimshawing (e.g., works by Angokwazhuk at the Royal Ontario Museum; Ray Reference Ray1977). Collection and sale of archaeological mammoth ivory objects has also continued (Hollowell Reference Hollowell2004).
One Yup’ik and Iñupiaq term (also used by other Inuit groups) for both “mammoth” and “mammoth ivory”—keligvak (and alternate spellings, Table 4)—may reflect the use of mammoth ivory as raw material. Keligvak combines kelig and -vak (“one associated with”). Jacobson (Reference Jacobson2012) relates the first morpheme to kelig (“to scrape”). It could refer to the frequent use by the Iñupiat of mammoth ivory for fat scrapers and (to a lesser extent) skin scraper handles (Lincoln et al. Reference Lincoln, Goodwin, Goodwin, Ongtowasruk, Senungetuk and Weyiouanna2010; Murdoch Reference Murdoch1892; Nelson Reference Nelson1900); also see Figure 4:d–e, i, s). If this interpretation is correct, then the term keligvak came to designate both ivory and mammoths by successive metonymies, from tool to raw material, and from raw material to animal.
Summary of Mammoth-Related Oral History and Names in Alaska and Adjacent Areas.
Table 4 Long description
The table lists Indigenous groups and localities in Alaska and nearby regions, marking which kinds of beings mammoths are described as (sky, aquatic, underground, terrestrial) and whether stories mention hunting, danger, or magical power, alongside recorded names and sources. Dene (Kaska) portray the mammoth as terrestrial and link it with hunting and danger, with several name spellings recorded. Koyukon accounts describe both aquatic and terrestrial aspects and uniquely note magical power. Tutchone entries emphasize underground and terrestrial traits and include hunting. Iñupiat and Inuvialuit entries largely focus on name variants across regions; where traits are marked, Kotzebue Sound includes a sky association and danger in one record, while several Kotzebue and Nunamiut records emphasize aquatic and or terrestrial traits. Nunamiut accounts repeatedly mark terrestrial traits, sometimes paired with sky or aquatic traits, and one record notes danger. Yupiit entries include underground-only and aquatic-focused descriptions; Kuskokwim records combine sky and aquatic traits, with hunting noted in one. Blank cells indicate traits not reported in that specific source, so differences may reflect documentation choices rather than true absence in the tradition.
Mammoth Stories
Mammoths form a substantial component of the oral history of several Native groups in Alaska and neighboring areas (Table 4). Mammoths are most often narrated as underground beings (Yupiit, Iñupiat, Koyukon Dene), and sometimes as sky (Iñupiat; Figure 5b) or aquatic (Yupiit, Iñupiat; Figure 5a, d) beings. Figurative depictions reflected this cosmological understanding (e.g., Figure 5a) until access to European knowledge of live proboscideans’ morphology (Figure 5b, c) sometime after around 1885 (Townsend Reference Townsend1887). In addition, some Dene groups—such as the Lower Tanana, Kaska, and the Tutchone—mention terrestrial beings, which, though not always specifically identified, are generally understood to refer to and/or include mammoths. Other Dene groups—such as the Gwich’in, Dena’ina, or Ahtna—although aware of mammoths, do not appear to have elicited names or stories associated with them (James Kari, personal communication 2025).
Native mammoth representations in Alaska and adjacent areas: (a) aquatic mammoth (equgaarpak), bottom of qantaq or wooden bowl, collected on Nelson Island (Yup’ik) in 1879 (described in Nelson Reference Nelson1900); (b) sky people and sky dogs hunting a sky mammoth (keligvak) while “early man” looks on, Simon Paneak (Nunamiut Iñupiaq), 1968 (from Campbell Reference Campbell1998; no scale); (c) mammoth, walrus tusk, collected in Utqiaġvik (Iñupiaq) around 1912 (from Chan Reference Chan2013; no scale); (d) aquatic/underground mammoth (keligvak), walrus ivory drill bow, collected on Nunivak Island (Yup’ik) in 1905 (redrawn from Gordon Reference Gordon and Winston1917; no scale); (e) string figure of an aquatic mammoth (keligvak), Mackenzie Delta (Inuvialuk), 1913–1918 (from Jenness Reference Jenness1923; no scale). NMAI = National Museum of the American Indian, Smithsonian Institution; NMNH = Smithsonian Institution, National Museum of Natural History, Department of Anthropology. (Color online)
Regardless of their nature (sky, terrestrial, underground, or aquatic being), in stories, mammoths tend to be portrayed as game animals. Mammoth hunters were sometimes members of the same community as the storyteller, as told by Sagdluaq, a Colville Iñupiaq, to K. Rasmussen in 1921–1924:
[A man] caught sight of something which was tearing up the earth’s surface. . . . The man gazed and gazed at the lumps of earth moving until the tusks of a mammoth emerged. . . . So he drew his bow and shot an arrow down into the ground a little behind and under the place where the tusks were sticking up. . . . And there, lying right on top of the ground, was a big, big animal, a mammoth, which had fallen forward on its knees and had died of its wound [Ostermann Reference Ostermann1952:153].
Another example comes from W. Charlie, a Northern Tutchone:
The father cut a large pole and sharpened the end. He then hid away in the bushes. The big animal came up the trail from the lake to the house where the man’s son was waiting on top of the ice house. The son clubbed the animal from the top of roof over and over again. The big animal couldn’t climb up to get that man because the sides of the house were slippery from the ice. All the time he was doing this, his father jumped out from the bushes and speared the big animal in the soft place behind his front legs and killed him [Tagé Cho Hudän Interpretive Centre 2025].
Alternatively, hunters were otherworldly beings, such as sky people, as illustrated by Simon Paneak, a Nunamiut Iñupiaq (Figure 5b), and narrated by Pegliruk, a Kobuk Iñupiaq:
[Ataogoraachuak] sees some great big thing going along barely touching the ground. . . . That big animal is going along just above the ground. And now this man sees three men following that something, and they have spears, and they are walking without touching the ground like the big animal. . . . After a while they come to that thing with ivories that has been killed down by a lake. This is a great, big thing with ivories growing out of its head [Giddings Reference Giddings1961:71–72].
A Dene example comes from the Kaska, who tell the story of cultural hero Yamodeyā (Traveler) using mammoth sinew to craft arrows in his quest to overcome other otherworldly beings (Kaska Tribal Council 1997). Terms for mammoths among the Koyukon Dene (Table 4) also reflect their status as the underground prey of underworld spirit people (Jetté Reference Jetté1911). The words combine the morphemes nen (“earth”), yeh (“down”), k’e (animal subject), and one or several additional morpheme(s) between tlen’ (“meat”) and tl’en (“bones”; Jetté and Jones Reference Jetté and Jones2000). The overall meaning therefore encompasses “underground game animals” or their bones. Some mammoth-bone concentrations are named accordingly, including naaghedeneeł ne kk’aatl’one (“cut bank of the underworld people”) near Tanana (Jetté and Jones Reference Jetté and Jones2000).
Mammoths can be associated with danger (Table 4). Several Dene groups have stories of mammoths chasing, killing, and sometimes eating people, as told by W. Charlie, a Northern TutchoneFootnote 2:
The man’s wife and her young twin brothers saw something coming away down on the ice. Thinking that it was their brother-in-law, they ran down the trail to meet him. Their sister, the man’s wife, walked behind. As the shape got closer they realized that it wasn’t a man but a very large animal. It was too late for the two boys. The animal killed and ate them. Their sister saw what happened and hid under the snow until the animal passed by [Tagé Cho Hudän Interpretive Centre 2025].
Mammoths can also be associated with magical power that may be taken advantage of, whether by ivory crafters, as told by Sagdluaq, a Colville Iñupiaq, to K. Rasmussen in 1921–1924—“Everything made of mammoth tusk has the protective force of an amulet” (Ostermann Reference Ostermann1952)—or by shamans following encounters with mammoth beings or fossils, as told by Pegliruk, a Kobuk Iñupiaq:
Now, when Ataogoraachuak gets home, he takes some jade, and he finds out he can cut it and drill it with his little finger. He is a good angatkok [shaman] after that. Those people that hunt the big thing with ivories live up in the air about halfway above us now. They are angatkoks. When it is bad weather around here, those people up there are hunting around for something. When it is fine weather, they are staying at home. Only angatkoks can see them [Giddings Reference Giddings1961:72–73].
One additional Yup’ik term for mammoths, equgaarpak (Table 4), appears to be descriptive. It combines equgaa and -rpak (“large”). Jacobson (Reference Jacobson2012) relates equgaa to equk (“to carry on one’s shoulder”) and equg (“thing carried on one’s shoulder”), suggesting that equgaarpak are either large animals that carry something near their head or animals that carry something large near their head; the alternative quugaaq omits the size factor. Quugiinraq, from quugaaq and -linraq (“product of”), accordingly refers to mammoth ivory (Jacobson Reference Jacobson2012). Consequently, equgaarpak probably describes mammoths as tusk-carrying beings, as also suggested by their depiction with prominent tusks (Figure 5a). The Kaska Dene negūtīh mirrors equgaarpak given that it combines ne (iterative), gū (areal), and -tīh (“carrying one large object”; Kaska Tribal Council 1997).
Discussion
Utilitarian and Symbolic Functions of Mammoths in Alaska
Alaskans / eastern Beringians probably ate mammoths before their extirpation. Recovery at sites in the Shaw Creek Flats of postcranial mammoth remains with little to no other economic use strongly suggests food consumption, regardless of the absence of butchery traces from lack of adequate bone surface preservation. This case is strongest for the ribs at Swan Point CZ4b, which are consistent with eating and discarding, hearthside, of the highly nutritious half-rack (minimally) of a neonate mammoth. Such a high-value anatomical package is also more likely to have been acquired by predation rather than by scavenging, although either may have occurred, as argued for the Eurasian Paleolithic (Konidaris et al. Reference Konidaris, Barkai, Tourloukis and Harvati2021). In contrast, we have not encountered reliable evidence of mammoth eating after extirpation—for instance, from scavenging of well-preserved fossil carcasses.
Technological use of mammoth products (particularly ivory), unlike dietary use, is documented for the entire timeline of human presence in Alaska. Two regions/periods—the Shaw Creek Flats about 14–13 kyr BP and the Kobuk-Noatak region of the last 2,000 years—provide evidence of the entire manufacturing process, from material sources to end products, including production implements and by-products.
Tusk acquisition in the Kobuk-Noatak occurred in mammoth-rich locales (e.g., Elephant Point) or from opportunistic encounters (e.g., eroding Kobuk and Noatak riverbanks [Giddings Reference Giddings1961; Hall Reference Hall1971]). Pleistocene people of the Shaw Creek Flats similarly appear to have targeted fossil ivory in at least some cases, as shown by the >5 kyr offset between some ivory objects and archaeological occupations (Table 1). Well-preserved fossil ivory would have been more efficient to manufacture than fresh ivory (Christensen Reference Christensen1999; Steguweit Reference Steguweit2015). Objects that lack a radiocarbon offset with the occupation may, in contrast, reflect acquisition of (sub)fossil ivory or fresh ivory from recently dead mammoths (killed by people or not), including fresh ivory stored/rotten for later use as suggested by Wygal et alia (Reference Wygal, Krasinski, Holmes, Crass and Smith2021) for the tusk recovered at Holzman C5b. Tusks still in alveoli may then have been extracted (at collection site or at workshop) by chopping near the pulp cavity (Pitulko, Pavlova, et al. Reference Pitulko, Pavlova and Nikolskiy2015)—as probably shown by the basal bevel of the Swan Point CZ4b juvenile tusk (Figure 3d)—or by twisting. The presence of cheekteeth at workshop sites such as Swan Point CZ4b and Kangiguksuk indicates that, in some cases, people may have collected cheekteeth, heads, skulls, and/or crania alongside tusks.
Initial manufacture took place at workshop sites, at Holzman C5a and Swan Point CZ4b, in well-delimited activity areas near hearths. In the Shaw Creek Flats, probable preforms consisted of ivory sections of varying thickness (Figure 3d-e, f, h) produced by grooving, percussion, or prying—techniques well known in ivory and antler work of the Eurasian Upper Paleolithic (Christensen Reference Christensen1999; Hahn et al. Reference Hahn, Menu, Taborin, Walter and Widemann1995; Heckel and Wolf Reference Heckel and Wolf2014; Khlopachev and Girya Reference Khlopachev and Girya2010; Pitulko, Pavlova, et al. Reference Pitulko, Pavlova and Nikolskiy2015; Semenov Reference Semenov1964). Manufacturing implements indubitably included burins for grooving (Hirasawa and Holmes Reference Hirasawa and Holmes2025; Holmes Reference Holmes, Goebel and Buvit2011)—as elsewhere across the Paleolithic world (Semenov Reference Semenov1964)—as well as tools such as hammers, anvils (Figure 3k), and wedges, some or all of which consisted of ivory by-products themselves (Pitulko, Pavlova, et al. Reference Pitulko, Pavlova and Nikolskiy2015; Semenov Reference Semenov1964). Preform manufacture during the Late Holocene saw the addition of new techniques, including chopping and sawing, made possible by novel ground-stone implements (adzes and saws). Preforms were then relocated to habitation sites for final manufacture, as has been argued of Broken Mammoth CZ4 objects (Heppner Reference Heppner2017); traded, as in the Kobuk-Noatak area; or possibly completed at the same workshop site.
Final stages of manufacture during the Pleistocene consisted of whittling or abrasion, as documented from striations and polish on the Broken Mammoth and Holzman objects (Heppner Reference Heppner2017; Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021), and commonly employed in Paleolithic ivory work (Semenov Reference Semenov1964). Those techniques continued to be used during the Late Holocene, with finer work permitted by novel implements such as iron-tipped engravers. Desired end products during the Pleistocene appear to have consisted of elongated, thicker (Figure 3g) or thinner (Figure 3a–b) “rods,” both of which are generally associated with weapon parts (points or foreshafts; Heppner Reference Heppner2017; Pitulko, Pavlova, et al. Reference Pitulko, Pavlova and Nikolskiy2015; Sutton Reference Sutton2018; Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021). During the Late Holocene, preferred end products similarly included weapon parts (see Sagdluaq 1921–1924 above [Ostermann Reference Ostermann1952]; also Figure 4l–n, x) and also large, three-dimensional objects such as ladles, fat scrapers, and shovel blades (see Sagdluaq 1921–1924 above [Ostermann Reference Ostermann1952]; also Figure 4a–i).
Symbolic functions of mammoth resources are known only in Late Holocene Alaska, during which manufactured ivory objects have communicated information about the relationships between the object makers, object owners, and mammoth animals and/or their ivory. This symbolism reflects the cosmological status of mammoths, shared across many Alaska Natives, as supernatural beings that are intrinsically powerful, dangerous, or associated with other powerful supernatural beings. This cosmological importance may derive from continued interactions with mammoth fossils, deep-time memories of live mammoth and other extinct megafauna, or a combination of both. Among the Iñupiat, people who interact with mammoth beings have been imbued with magical power (e.g., shamans, as in Pegliruk above)—as have objects manufactured of mammoth ivory (see Sagdluaq 1921–1924 above [Ostermann Reference Ostermann1952]) and/or unworked cheekteeth fragments (Figure 4p), as part of the larger Yup’ik/Iñupiaq tradition in which depictions of animals and body parts (amulets), modified or not, come to embody attributes of those animals (Hill Reference Hill2011a; Murdoch Reference Murdoch1892; Rasmussen Reference Rasmussen1929). Crafting likely participated in the makers’ social status (Inomata Reference Inomata2001) and augmented the symbolic value of ivory objects by conveying notions of labor and rarity, contributing to making objects such as ivory ladles highly valued trade items and personal possessions (Murdoch Reference Murdoch1892).
The Place of Mammoths in Alaskan Cultures
Mammoths have long been part of a larger network of interactions between Alaska Natives and animals/resources, with each resource offering both economic and symbolic uses. Choosing to actively harvest and use mammoth products may have depended on availability and functional characteristics relative to those of other resources.
Availability of mammoth resources has varied in Alaska over the last 14 cal kyr BP. Live mammoth populations probably peaked in unglaciated areas (excluding much of southern Alaska) around 15–14 cal kyr BP, then declined and became extirpated over much of the state by about 13 cal kyr BP (Guthrie Reference Guthrie2006; Mann et al. Reference Mann, Groves, Kunz, Reanier and Gaglioti2013). Colder-climate and alkaline aeolian sediments in unglaciated areas would also have favored the preservation of older, fossil tusks until about 12 cal kyr BP, as have the permafrost of northern and western Alaska during the Holocene. Both core areas/periods of mammoth use therefore offered high availability in mammoth resources. The Shaw Creek Flats and adjacent middle Tanana valley appear to have been favorable mammoth habitats around 15–13 cal kyr BP (Guthrie Reference Guthrie2006; Rowe et al. Reference Rowe, Bataille, Baleka, Combs, Crass, Fisher and Ghosh2024) and would have provided both live animals and fresh carcasses, as well as probable natural accumulations of fresh ivory (so-called cemeteries). In contrast, frequent encounters of well-preserved mammoth tusks in the Kobuk-Noatak region during the Late Holocene may reflect local abundance during the Pleistocene (Guthrie Reference Guthrie2004), favorable preservation conditions in a region covered by permafrost, high rates of permafrost erosion in riverbanks and coastlines, or higher human population density.
Active use of mammoth resources in both core areas/periods reflects both their availability and functional characteristics relative to other resources. In the Pleistocene Shaw Creek Flats, food from a fresh mammoth carcass would have come in a package larger than for any other available animals at the time and would have presumably been associated with a powerful symbolic meaning given the size and dangerousness of the animal—bearing in mind that predation would have likely been both more costly (danger) and rewarding (prestige, amounts of food) than scavenging. Eastern Beringians nonetheless ate other resources alongside mammoths, including horse and waterfowl (Lanoë and Holmes Reference Lanoë and Holmes2016), both of which may have provided higher returns than mammoth on a seasonal and/or local basis. In parallel with food, mammoth ivory offered a dense, fairly elastic, three-dimensional material with a natural elongated shape that was also symbolically charged (e.g., magical in the Late Holocene Kobuk-Noatak). Although mammoth ivory has been used to manufacture a wide range of objects, these combined mechanic and symbolic characteristics have made it a material of choice for hunters’ weapons as well as large, curated, domestic objects, which tend to combine utilitarian and symbolic purposes rather than fulfill only one function. Other materials have similarly been preferred to mammoth ivory depending on their availability, symbolic value (generally strong for materials derived from animal sources [Hill Reference Hill2011a]), and mechanical properties—for instance, softer and more common materials (wood, bone) for expedient objects (Gelvin-Reymiller and Reuther Reference Gelvin-Reymiller and Reuther2010; Margaris Reference Margaris2014); harder and more resistant materials (ground stone) for strikable objects, such as adzes (Neffe Reference Neffe2023); more elastic and resilient materials (antler) for projectile points (Guthrie Reference Guthrie, Clutton-Brock and Grigson1983; Lee and Goebel Reference Lee and Goebel2016; Lynch Reference Lynch2023; Margaris Reference Margaris2014); and naturally preshaped materials (e.g., walrus ivory) for specific objects such as drill bows (Chan Reference Chan2013).
The lower intensity of use of mammoth resources in other areas/periods of Alaska also reflects their relative availability and functionality. Eastern Beringians of the Shaw Creek Flats may have largely stopped eating mammoth by around 13.8 cal kyr BP, preferring gregarious ungulates, waterfowl, and fish (Halffman et al. Reference Halffman, Potter, McKinney, Finney, Rodrigues, Yang and Kemp2015; Lanoë et al. Reference Lanoë, Reuther, Holloway, Holmes and Kielhofer2018; Potter Reference Potter2018; Potter et al. Reference Potter, Holmes, Yesner, Graf, Ketron and Waters2014, Reference Potter, Halffman, McKinney, Reuther, Finney, Lanoë and Andrés López2023; Wygal et al. Reference Wygal, Krasinski, Holmes, Crass and Smith2021), which, because of their high abundance, must have offered better returns than mammoths. Ivory (presumably [sub]fossil), in contrast, remained in active use locally until approximately 13 cal kyr BP. For most of Alaska Native history, however, other materials have been preferred over mammoth ivory. This preference surely reflects—at least partly—the lower availability of mammoth ivory relative to materials encountered as part of Holocene lifeways, including elk (Potter Reference Potter2007; Potter et al. Reference Potter, Irish, Reuther, Gelvin-Reymiller and Holliday2011, Reference Potter, Holmes, Yesner, Graf, Ketron and Waters2014), caribou (Esdale Reference Esdale2008; Giddings Reference Giddings1952, Reference Giddings1962; Holmes Reference Holmes2008), walrus (Fitzhugh et al. Reference Fitzhugh, Hollowell and Crowell2009; Hill Reference Hill, Braje and Rick2011b; Mason and Rasic Reference Mason and Rasic2019), and whale (Mason Reference Mason1998; McCartney Reference McCartney2003) economies, or relative to locally abundant materials such as fossil walrus ivory on the Bering Sea islands (Chan Reference Chan2013; Geist and Rainey Reference Geist and Rainey1936). Mammoth resources, particularly ivory, have continued to be used in an episodic manner from chance encounters of well-preserved fossils or trade from core areas such as the Kobuk-Noatak area during the Late Holocene. Ivory in marginal areas has been employed for similar but more restricted purposes than in core areas, often embodying strong symbolic meaning (e.g., traded ivory ladles) but lacking, for instance, consistent attribution of magical power.
Use of mammoths in Alaska is comparable to use in nearby arctic/subarctic Eurasia. Some regions of northern Sakha provided abundant mammoth resources—including both live animals and concentrations of carcasses—to Upper Paleolithic peoples, who created visual representations and used mammoths for food and particularly ivory for weapon parts and ornaments (Pitulko et al. Reference Pitulko, Pavlova and Nikolskiy2017, Reference Pitulko, Pavlova, Dorozhkina, Nikolskiy and Basilyan2024; Pitulko, Pavlova, et al. Reference Pitulko, Pavlova and Nikolskiy2015). Larger quantities of archaeological mammoth material compared to Alaska likely reflect the longer temporal overlap with live mammoths rather than fundamental differences in the intensity of interaction. As in Alaska, relationships between mammoths and Native Northern Eurasians after mammoth extirpation have been most strongly expressed in areas that favor mammoth ivory preservation, such as northern Sakha and the nearby New Siberian Islands (Larmer Reference Larmer2013; Nikolskiy et al. Reference Nikolskiy, Sulerzhitsky and Pitulko2011; Nordenskiöld Reference Nordenskiöld1880). Ivory objects found at Holocene sites such as Zhokhov, dated approximately 8.7 cal kyr BP (Pitulko, Ivanova, et al. Reference Pitulko, Ivanova, Kasparov and Pavlova2015), or the Cape Baranov sites, dated around 1500–400 cal yr BP (Okladnikov and Beregovaya Reference Okladnikov and Beregovaya2008), resemble those found at contemporaneous Alaskan sites. Ivory objects among recent Native Eurasians also mirror Alaska in combining utilitarian and symbolic functions (Bronshtein and Shirokov Reference Bronshtein and Shirokov2014), the latter as part of a vibrant tradition of magical and shamanic implements informed by cosmological narratives of mammoths as powerful supernatural, underground, or aquatic beings (Arzyutov Reference Arzyutov2019; Bogoras Reference Bogoras1909; Ivanov Reference Ivanov1949; Janhunen Reference Janhunen, Osada and Endo2011; Leroi-Gourhan Reference Leroi-Gourhan1935; Lot-Falck Reference Lot-Falck1963; Lukin Reference Lukin2021; Prokof’eva Reference Prokof’eva1949; Senkevich-Gudkova Reference Senkevich-Gudkova1949; Serikov and Serikova Reference Serikov and Serikova2005). Native Eurasian terms for mammoths similarly reflect this understanding, including the original “mammoth” that purportedly derives from an Uralic mānoŋt (“earth tusk/horn”; Janhunen Reference Janhunen, Osada and Endo2011).
Extinct Megafauna among Alaska Natives and North Americans
Mammoths’ status as extinct animals have not set them fundamentally apart from extant animals/resources among Late Holocene Alaska Natives. For instance, Iñupiat appear to have related to mammoths and walruses (aiviq/asveq) in similar ways. As for mammoth, acquisition of walrus resources has not depended on the hunting of live animals, given that much of the ivory (tugkaraq/tuugaaq Footnote 3) has been harvested from fossil/subfossil deposits (Chan Reference Chan2013; Geist and Rainey Reference Geist and Rainey1936; Ray Reference Ray1961). Both mammoth and walrus ivory, whether fossil or fresh, have seen very similar technological uses due the shared biomechanical properties of each material (Chan Reference Chan2013; LeMoine and Darwent Reference LeMoine and Darwent1998; Margaris Reference Margaris2014). Walrus ivory, like mammoth ivory, has been favored to manufacture objects of high symbolic value, such as weapon parts, highly curated domestic implements, figurines, and amulets—some of them endowed with hunting and/or shamanic magic, reflecting similar narratives of mammoth and walrus as powerful cosmological beings (Chan Reference Chan2013; Fitzhugh et al. Reference Fitzhugh, Hollowell and Crowell2009; Hill Reference Hill2011a, Reference Hill, Braje and Rick2011b, Reference Hill2017; LeMoine et al. Reference LeMoine, Darwent, Darwent, Helmer, Lange, Whitridge and Hill2024).
The status of mammoths in Late Holocene Alaska provides a lens through which one may approach human–proboscidean relationships in other postextirpation regions/periods of North America. Earlier Alaska Natives of the Early Holocene interior Alaska may have related to mammoths in a very similar way, given that they also used well-preserved ivory for weaponry (Figure 3c) that undoubtedly carried high symbolic meaning, reflecting both the hunting power and the rarity of the material. In contrast, relationships with extinct proboscideans in temperate and tropical regions appear to have forsaken economic uses, presumably due to poor preservation and loss of biomechanical function of proboscidean ivory in these climates. Documented proboscidean objects instead have served exclusively symbolic purposes, for instance in Otoe (Great Plains) magical bundles (Harrington and Skye Reference Harrington and Skye1913). This more limited relationship mirrors a generally fuzzier presence in oral histories in which proboscideans tend to be grouped among other large beasts likely encompassing multiple megafauna taxa (Cohen Reference Cohen2002; Echo-Hawk Reference Echo-Hawk2000; Edmonston Reference Edmonston1953; Eiseley Reference Eiseley1945, Reference Eiseley1946; Lankford Reference Lankford1980; Mayor Reference Mayor2005; Strong Reference Strong1934), rather than be specified as in arctic Alaska and northern Eurasia.
Human–proboscidean relationships at times of coexistence resembled those after extirpation. Mammoth uses among eastern Beringians of the Shaw Creek Flats were similar to those of walruses among Iñupiat/Yupiit, in that they included both food and ivory resources, each of which may have been acquired in integrated or entirely separate processes. Such interactions were likely at play elsewhere in temperate and tropical North America during the Pleistocene, with active use of varied proboscidean resources depending on local availability and functionality (utilitarian and/or symbolic) relative to other resources. Ivory may generally have been more important than suspected, as suggested by the archaeological record in Alaska but also of the Eurasian Paleolithic (see above), which suggest that extinct proboscideans were economically and symbolically important sources of raw material. Proboscideans there undoubtedly carried powerful symbolism—potentially magical—reflective of cosmological narratives that in the Pleistocene were further informed by visual experiences of live animals (Barkai Reference Barkai2021; Hussain and Floss Reference Hussain and Floss2015).
The record of Alaska Natives–mammoth interactions contributes to our understanding of the cultural relevance—beyond short-lived predation—of extinct megafauna and, more broadly, larger animals. Like Alaska mammoths, extinct megafauna in times of coexistence with ancient peoples may have been preferentially used for symbolically charged uses, including utilitarian items over or in addition to food uses. Such symbolic nonfood focus of megafauna is illustrated by the use of whale bone for projectiles among nonwhaling Paleolithic societies, despite an abundance of mechanically better-suited materials (McGrath et al. Reference McGrath, Van Der Sluis, Lefebvre, Charpentier, Rodrigues, Álvarez-Fernández and Baleux2025), or by the disproportionate use of great auk beaks for burial clothing in the Archaic Maritimes (Tuck Reference Tuck1971). As in Alaska, many megafauna taxa have also likely retained their cultural salience in Native North America long past their extinction. Traces of those undoubtedly may be found in symbolically charged—and, occasionally, utilitarian—fossils, and alongside a rich corpus of oral history (Echo-Hawk Reference Echo-Hawk2000; Mayor Reference Mayor2005).
Acknowledgments
Excavations at the Mead and Swan Point sites benefited from support by Barbara Crass and from the efforts of numerous crew members. Angela Linn (University of Alaska Museum of the North) and Karina Ceron and Katherine Skaggs (American Museum of Natural History) aided in accessing museum records. An earlier version of this article was presented at the Elephant Archaeology symposium organized by Rowan Flad at the 90th Annual Meeting of the Society for American Archaeology in Denver, Colorado. This article was greatly improved from conversations with and comments from Erica Hill (University of Alaska Southeast), James Kari (University of Alaska Fairbanks), Jessica MacLellan (Wake Forest University), Ashley Sharpe (Smithsonian Tropical Research Institute), and three anonymous reviewers.
Funding Statement
Various stages of this research were funded by the U.S. National Science Foundation (awards OPP-1223119 to Potter, OPP-2329997 to Potter and Reuther, and OPP-2329998 to Lanoë) and the University of Alaska Museum of the North. Additional support was provided by the Otto William Geist Memorial Fund at the University of Alaska and by Nancy Eliason to the University of Alaska Museum of the North.
Data Availability Statement
All original data are presented in the article.
Competing Interests
The authors declare none.
