J Archaeol Method Theory DOI 10.1007/s10816-012-9168-x
Reconsidering Raw Material Selection Skeletal Technologies and Design for Durability in Subarctic Alaska Amy V. Margaris
# Springer Science+Business Media New York 2013
Abstract Raw material selection is an essential facet of technological decision making. This analysis moves beyond more typical lithic studies, investigating links between raw material selection and practical tool function by integrating data from materials science and ethnohistory with an analysis of bone and antler tools. A case study of skeletal technologies from the Alaskan subarctic offers a fresh perspective on technological strategies, especially the selection of highly durable (fracture resistant) skeletal materials to create reliable tools for use in high-risk foraging contexts. Keywords Organization of technology . Raw material . Bone implements . Alaska
Introduction Archaeologists reverse engineer artifacts in order to uncover the problems these objects were designed to solve, and the associated skills and strategies employed by their unseen users. A rich literature explores such artifact design goals as reliability, expediency, flexibility, and portability, as well as the constraints a toolmaker faces when attempting to meet one or more of these goals (see discussions in Hayden et al. 1996; Nelson 1997; Schiffer and Skibo 1997). All technological strategies involve compromise, in part because each step in an artifact’s behavioral chain further reduces the universe of possible end products: the distance one is willing to travel to a good quarry can limit lithic nodule size; nodule size then limits flake dimensions. Yet, we can also view these linkages as empowering, in the sense that a toolmaker can use raw material selection—the first stage in any tool’s life history—as a means to profoundly influence how a tool will function in a given context. Prehistoric toolmakers clearly employed a wide range of materials, each with its own set of properties and technological potentials. But despite Nelson’s admonishment over A. V. Margaris (*) Department of Anthropology, Oberlin College, 305 King Building, 10 N. Professor Street, Oberlin, OH 44074, USA e-mail:
[email protected]
Margaris
20 years ago that archaeologists incorporate non-lithic materials into models of technological organization (1991:89), stone still attracts the lion’s share of theoretical attention among prehistorians. This study aims to widen the field of view by investigating facets of raw material selection and attendant artifact variability as they relate to a set of raw materials very distinct from stone: skeletal tissues. Antler, bones, and other “hard parts” of animal bodies form in biological rather than geological contexts, and each evolved in response to different selective pressures. Their resulting mechanical properties span a considerable range, making skeletal materials ideal subjects for examining the links between raw material choice and practical tool design. Stone is hard and brittle, and flintknappers worldwide typically choose varieties that are close-grained and break in a predictable fashion (Andrefsky 2005; Cotterell and Kamminga 1987; Whittaker 1994). These properties allow for the production of sharp edges and also give flaked stone tools a propensity for brittle fracture (Ellis 1997; Luedtke 1992; Waguespack et al. 2009). What, in contrast, are the intrinsic physical properties of diverse skeletal materials, and how do they delimit the potential uses of skeletal materials for technological purposes? Some important clues come from the fields of biology, engineering, and materials science, which have produced a rich literature on the properties of a diverse array of skeletal materials, from deer antler to bovid bone. Although archaeologists have long recognized antler’s particular fracture resistance (Albrecht 1977; Guthrie 1983; Knecht 1997; MacGregor 1985; MacGregor and Currey 1983; Pétillon et al. 2011; Pokines 1998; Stodiek 2000), few have investigated the conditions under which a broad range of skeletal materials are likely to be selected or rejected for technological purposes (but see Scheinsohn and Ferretti 1995). Such studies, guided by a firm understanding of the inherent mechanical properties and limitations of those materials, are essential for modeling the decisions peoples of the past made in organizing their time and labor around the production and use of tools created from diverse materials—rather than lithics or ceramics alone. In this study, I integrate three types of evidence, beginning with the results of controlled laboratory investigations which shed light on the mechanical properties (e.g., strength and fracture resistance) of various types of skeletal tissues. I then develop a case study of the design and use of skeletal technologies by protohistoric Alutiiq peoples of Alaska’s Kodiak Archipelago. Here, special conditions of preservation have yielded collections that contain a rich array of osseous artifacts, including objects crafted from caribou antler, whale bone, and bird bone. Ethnohistoric accounts from early Russian seafarers and missionaries to Kodiak document some of the tools’ uses, which help in inferring tool functions. By linking these three lines of evidence—materials science, archaeology, and ethnohistory—I illustrate a general-purpose framework for examining relationships between material properties, raw material choices, and the practical functioning of tools. In what economic and social contexts would a toolmaker attempt to closely match a material’s technical or formal qualities to a given desired use? What does a lack of correspondence reveal about how a toolmaker weighed other, perhaps competing design criteria? The results highlight a range of technological strategies that could be achieved with the use of osseous materials. In some cases, Alutiiq toolmakers sought to closely match a material to its intended context of use, especially in their use of antler, a remarkably fracture resistant material. Choosing to craft tools
Reconsidering Raw Material Selection
from fracture resistant materials, what I term design for durability, can be viewed as a risk-reduction strategy aimed at maximizing foraging outcomes when the cost of failure is high. Other choices highlight additional or competing constraints on tool investments. In the Kodiak region, this likely included the disparate costs associated with obtaining local versus distant raw materials—costs that, in the late prehistoric period, were also related to social status.
Mechanical Properties of Skeletal Materials In this section, I compare the properties of five types of osseous media that were widely used for tool construction in the past, including in the Kodiak region: (1) terrestrial mammalian limb bone, (2) antler, (3) ivory, (4) bird bone shafts, and (5) whale bone (summarized in Fig. 1). Current knowledge about these materials is mainly derived from controlled laboratory experiments. We can assume, however, that through observation, practice, and informationsharing, prehistoric toolmakers developed a similarly keen understanding of the limits of the materials available in their natural environment, and incorporated this knowledge into the designs of critical subsistence gear (Schiffer and Skibo 1997). In fact, Alutiiq patterns of raw material use offer insights for today’s scientists into the properties of certain biological materials. Material properties (unlike structural ones) are innate and dimensionless, meaning they do not vary with the geometry of an object. The material properties of skeletal media are quite diverse (Albrecht 1977; Currey 2002: Table 4.3: MacGregor and Currey 1983; Margaris 2009; McKittrick et al. 2010; Scheinsohn and Ferretti 1995; Zioupos et al. 1997) because hard animal tissues have evolved to serve an array of functions in life. The primary role of terrestrial limb bones is to provide structural support, so compact limb tissue tends to be stiff and strong. Antlers, in contrast, are used mostly for defense and protection. Moreover, they must grow extremely fast. Compact antler tissue is less highly mineralized than bone (Currey 1979), which makes it less stiff and strong but remarkably tough, or fracture resistant (Albrecht 1977; Currey 1979, 1988; MacGregor and Currey 1983; Margaris 2009). These properties constrain how the materials can be used.1 Ivory is often thought of as quite hard, but this is in fact true only of enamel, the highly mineralized coating that sheathes all or a portion of the tooth crown. The majority of a tooth (or tusk) is composed of another bony substance, dentin, which in humans has been shown to interface with the enamel layer to provide a special “soft zone” that absorbs chewing impacts (see review in Shahar and Weiner 2007). Like antler, dentin is less mineralized than compact bone, a relationship which translates almost universally across bony tissues into low stiffness and strength (MacGregor 1985:Table 2.1; Wegst and Ashby 2004), but high fracture resistance. Elephant and 1
Material stiffness (Young’s modulus) is defined as the stress/strain ratio as a material undergoes temporary (elastic) deformation, and ultimate strength is the maximum stress a material is capable of withstanding. Toughness is a measure of the amount of energy a material can absorb before breaking. See Currey (2002) for further explanation in the context of skeletal tissues.
Margaris Raw Material
Properties Stiffness
Strength
References Fracture Resistance (Toughness)
Bird bone (whole limb bone)
Biewener 1982 (painted quail, bobwhite); Scheinsohn and Ferretti 1995 (cormorant)
Compact limb bone (terrestrial mammal)
Currey 2002 (numerous specimens); MacGregor and Currey 1983 (bovid); Margaris 2009 (white-tailed deer, American elk); Scheinsohn and Ferretti 1995 (guanaco)
Porous whale bone
Currey 1988 (Atlantic whale); Kabel et al. 1999 (sperm whale); Scheinsohn and Ferretti 1995 (sperm whale)
Ivory (dentin)
MacGregor 1985 (elephant); Wegst and Ashby 2004
Antler (cortex)
Currey 2002 (red deer antler, reindeer antler); MacGregor and Currey 1983 (red deer antler); Margaris 2009 (reindeer antler)
Fig. 1 Skeletal tissues sort into broad functional categories. Their ensuing mechanical properties are shown here ranked from high (closed circle) to low (open circle). Ratings summarize experiments that compared multiple tissue types, but quantitative comparisons between studies are not always possible due to differences in testing conditions and procedures. References provide numerical data and details on the specific testing procedures used for each study
mammoth tusks are highly modified upper incisors, and their dentin is arranged in a series of nested cones around the pulp cavity (Krzyszkowska 1990). Walrus tusks, which are more germane to the Alaskan case study presented here, are upper canines and also differ somewhat from proboscidian ivory in that the pulp cavity houses a distinctive deposit of secondary dentin. This coarse-textured dentin (see Penniman 1952: Plate VIII) is more challenging than primary dentin to fracture in a controlled manner, but its mottled appearance, juxtaposed with the lighter surrounding dentin, offers a visual contrast that some toolmakers may have deliberately incorporated into their tools (e.g., LeMoine and Darwent 1998). Other types of skeletal media are best investigated as structures, such as the avian wing bones that prehistoric technicians frequently employed as natural tubes. Though thin-walled, bird bones are structurally strong and stiff as a result of their slender build (Biewener 1982:298), and system of internal struts which reduces buckling (e.g., Pennycuick 1967:Fig. 4). Their size, in concert with the bending and torsional flight motions to which avian proximal limb elements are adapted (Pennycuick 1967), make them especially useful for awls—instruments that are similarly engaged with a pressing or twisting motion. The fact that bird bones are both hollow and thin-walled also makes them convenient, portable containers for storing needles or pigments, and an acoustically ideal medium for flutes and whistles (Gál 2005). Finally, Alutiiq and other tool assemblages demonstrate the significant use of whale bone, a material that is in many ways poorly understood but whose distinctiveness reflects the cetacean’s secondary aquatic adaptation and limited need for buoyancy control (Wall 1983). Bone is metabolically expensive to produce, and because the cetacean skeletal system serves only limited roles in weight-bearing and buoyancy control (a problem whales solve through the ability to expel most air from their lungs during deep dives (Wall 1983)), some cetacean skeletal elements are lightweight and visibly porous, and lack a medullary cavity.
Reconsidering Raw Material Selection
The compact-cancellous tissue boundary is gradual rather than abrupt, and the degree of porosity found in the cancellous tissue can vary considerably across a single skeletal element (de Buffrénil and Schoevaert 1998; Campbell-Malone 2007:135; Felts and Spurrell 1965). The properties of bony tissues are strongly affected by the number and distribution of a tissue’s pore spaces (Carter and Hayes 1977; Currey 1988; Gibson and Ashby 1997:Figs. 11.7 and 11.10). Several studies suggest that porous whale bone is no different, its stiffness falling toward the low end of a range of skeletal tissues (Currey 1988; Kabel et al. 1999; Scheinsohn and Ferretti 1995). For instance, the compact bone found in a fresh Atlantic whale rib is only slightly stiffer than compact reindeer antler (Currey 1988), one of the most flexible of all skeletal materials. Other key properties of porous materials, such as fracture resistance, have proven difficult to measure directly (Keaveny et al. 1997), although researchers are obtaining increasingly accurate estimates using such techniques as finite element analysis and optical metrology (Shahar and Weiner 2007). Later, however, I show how patterns of whale bone use evidenced in Alutiiq assemblages can also shed light on the properties of this interesting material. The results promise to expand our understanding of the uses of whale bone as a technological resource in and beyond the Kodiak region.
Alutiiq Environment and Subsistence in the Protohistoric Period Kodiak and its spray of neighboring islands stretch southward from the Kenai Peninsula within the gentle arc of the Gulf of Alaska (Fig. 2). A maritime climate insulates the island chain from dramatic seasonal temperature oscillations, but frequent storms challenge navigation by sea even today. Nonetheless, the sea provided an essential means of travel and subsistence for Alutiiq peoples who have occupied the Kodiak Archipelago for at least 7,000 years (Clark 1979; Fitzhugh 2003:40). Pre-contact Alutiiqs were foragers who preferred to settle along coastal shorelines and rivers to take advantage of the rich diversity of fish, shellfish, marine mammals, and birds that these environments offered. Ungulates are twentieth century introductions to the Archipelago (Clark 1975:204; Lantis 1952; Rausch 1969:217). The region harbors only a few endemic terrestrial mammals: the Kodiak brown bear and a handful of small-bodied taxa that include red fox, otter, and ground squirrel (Rausch 1969). Archaeologists divide Alutiiq prehistory into three Traditions: Ocean Bay (7500– 3500 BP), Kachemak (3500–800 BP) and Koniag (800–200 BP) (Clark 1974, 1979; de Laguna 1975; Jordan and Knecht 1988; Steffian and Saltonstall 2005). Two of the sites investigated here, Karluk One (KAR-001) and Settlement Point (AFG-015), date to the Koniag Tradition (Fig. 2; see Table 1 for site descriptions). Alutiiq settlement density on the Archipelago peaked in this period, and social stratification emerged (Fitzhugh 2003). The third site, the Afognak artel (AFG-016), was occupied in the early Russian period. Kodiak Island is notable as the locale where, in 1784, the Russian empire established its first permanent settlement in America. Alutiiq lifeways were forever altered through encounters with Russian fur traders eager for soft sea otter pelts and
Margaris
N
0
10 20 30 40 50 km
Alaska Peninsula
Afognak I.
it
of
tra
S
Afognak artel Settlement Point
ik el
Sh
Kodiak City
Karluk One Karluk River
Kodiak I. Gulf of Alaska Sitkalidak I.
Fig. 2 Location of the three study sites
charged with securing the Native labor required to procure them (Black 2004; Crowell 1997; Luehrmann 2008; Miller 2010). A number of Russian observers documented Alutiiq subsistence and tool traditions in the era of first sustained contact between Russian colonists and Koniag Alutiiq. These include the clergyman Ioasaf Bolotov, who lived on Kodiak between 1794 and 1799 (Bolotov in Black 1977), and the Russian Orthodox priest Gideon (Gideon 1989). During his 1804–1807 stay on Kodiak, Gideon was closely involved in the everyday lives of the Native population and left a sympathetic and detailed set of ethnographic observations. The Navy officer, Iurii Lisiansky, wrote an important account of his time spent on Kodiak between 1804 and 1805 (Lisiansky 1968). Finally, the journal of the young naval officer Gavriil Davydov, who wintered on Kodiak in 1802–1803, provides a wealth of ethnographic details (Davydov 1977). According to these sources, Alutiiq communities targeted seasonally abundant salmon, halibut, cod, and a variety of other fish species which were taken with hooks and multi-pronged leisters (Lisiansky 1968:206). Barbed harpoon and dart assemblies used in conjunction with throwing boards were essential for taking harbor seals, sea
Site type
Village
Village
Probably an artel (work station) occupied by conscripted Alutiiq laborers and Russian supervisors
Site name
Karluk One (KAR-001)
Settlement Point (AM 33; AFG-015)
Afognak artel (AM 34; AFG-016)
Table 1 Site description
Russian era circa A.D. 1786–1839
Koniag Tradition
Koniag Tradition
Occupation period
Four structures interpreted as Alutiiq work areas and/or residences with associated middens; moderate to good organic preservation; located east of Afognak River.
House and midden deposits; preservation generally poor except in carbonaceous shell-rich House 1/midden infill; located on beach berm near modern shoreline of Afognak Bay.
Waterlogged house floor and midden deposits; outstanding organic preservation; located on south shore of Karluk Lagoon before destruction by wave action.
Description of deposits
Woodhouse-Beyer (2001)
Partlow (2000); radiocarbon dates reported in Partlow (2000): 74, Table 4.01
Knecht (1995); radiocarbon dates reported in Knecht (1995):141, Table 4:1
References
Reconsidering Raw Material Selection
Margaris
otters, Steller sea lions, and dolphins from kayaks (baidarkas) (Lisiansky 1968:206). Alutiiqs also hunted seals at their haul-outs using spears or nets (Gideon 1989:56– 57). Alutiiq seamstresses fashioned sea lion and seal skins into kayak covers (Bolotov in Black 1977:84–5), and used gutskin to create waterproof kamleikas (Gideon 1989:62), jackets which were indispensable in the region’s frequently inclement weather. Littoral resources such as clams, mussels, and sea urchins provided valuable nourishment, especially in lean winter months (Gideon 1989:67), while sea birds taken with arrows (Davydov 1977:228), nets (Lisiansky 1968:205), or snares (Gideon 1989:65) were especially prized for their skins, which were made into parkas (Bolotov in Black 1977:85). Gideon (1989:65) tells us that one parka required the skins of thirty five puffins. Finally, although whaling is no longer undertaken in the region today, it was once the practice of specialists who were at once admired and feared (Lisiansky 1968:174, 209). Kodiak whalers worked alone, hunting from baidarkas using slate-tipped spears smeared with aconite, a poison made from monkshood root (Black 1987:9; Lisiansky 1968:202). The entire endeavor was marked by uncertainty, as whalers maintained no physical attachment to their prey. Instead, they inscribed their harpoon end-blades with personalized emblems in order to claim the kill, should a fortunate tide wash the whale to shore (Gideon 1989:68). Migrating humpback and fin whales were important sources of food and raw materials for Alutiiq communities. The meat and blubber of baleen whales were mixed with berries and consumed, often with fish, and the sinew was woven into bird nets (Bolotov in Black 1977:85). Alutiiqs also used whale bones as architectural elements (Black 1987) and for constructing portable tools. As we will see, whale bone was part of a diverse palette of skeletal tissues that Alutiiqs drew from when constructing their subsistence gear. In the next section, I describe the form and functioning of the five categories of tool components examined for this study. By determining the raw materials Alutiiqs chose for each type of tool, we gain insight into some of the practical design problems that various skeletal tissues are best suited to address, as well as local factors Alutiiqs weighed in making their technical decisions.
Five Types of Alutiiq Tools After inventorying each site’s osseous tool assemblage, I focused analysis on the five most common artifact types: (1) unbarbed arrow tips, (2) barbed harpoon points, (3) toggling fish harpoon tips, (4) pointed tools, and (5) woodworking wedges (Fig. 3). The sample, which comprised just over 300 objects, included complete artifacts as well as identifiable fragments and tool blanks. (Rare) refits were tallied as a single object. All of these artifacts functioned as parts of larger composite tools or toolkits. Fish Harpoon Tips Salmon and other anadromous fish were crucial resources taken in large and increasing numbers into the Developed Koniag Phase beginning around A.D. 1400 (Partlow 2000). Their rapid collection in summer and early fall was essential to ensure an
Reconsidering Raw Material Selection
Fig. 3 Five types of Alutiiq artifacts. Clockwise from top left: three fish harpoon components, tapered and blunt arrow tips, barbed sea mammal harpoon points, pointed tools shaped from bird bone, and other materials; woodworking wedges, splintered (left) and with possible lubricant cavity (right)
adequate winter’s supply, and weirs were probably used to aggregate the fish for easier capture (Birket-Smith 1953:42; Knecht 1995:199). Koniag fish harpoon tips consisted of two or more components, cupped together and bound at their bases to form an ingenious bi-convex point. This point was then set loosely over a slim foreshaft. A thrust of the spear would bring both tip and foreshaft all the way through the body of the fish: the tip then slipped off the foreshaft and toggled sideways, preventing the fish from wriggling off the spear. Because fish harpoons were used not just once but repeatedly in intense bouts of resource procurement, it was crucial that the technology be failsafe. They risked making frequent contact with shallow rocky creek beds, so impact resistance would
Margaris
have been a critical design feature. The design of the entire weapon head assembly in fact spoke to the goal of minimizing fracture: the arrangement of two to three components nested loosely together provided “give” upon impact and ensured that only the outer component felt the direct force of the strike. The collections yielded several variants of this basic design which I describe here, and for further clarification are illustrated in Fig. 4. The differences relate primarily to how the components articulated, but with the exception of spurred and stand-alone tips, all were expected to function similarly. Most had a straight, lipped base formed by the articulation of two point components, or “valves,” the smaller of which was countersunk within the larger. The larger of the two lashed pieces formed the working point that made direct impact with the fish and the rocky substrate below. The base formed a socket to accommodate a foreshaft. A few socketed valves featured a curved, basal spur. “Spurred components” were paired symmetrically so that both halves absorbed striking impacts. “Stand-alone” tips comprised the third variant. These were bilaterally barbed and socketed on one side but were otherwise symmetrically convex from front to back. The presence of socket cavities on basal or medial fragments clearly identified the fragments as portions of harpoon tip assemblies. Stand-alone socketed points lacked countersinks and were rounded on both sides, while the area where multiple components joined was flat. Finally, about 35 % of harpoon point components featured a different type of attachment mechanism that is “scarfed” rather than socketed. According to Knecht (1995:205–211), scarfed components were used in triplicate with the longer of two components cupping a third, smaller one within it. This complex design lent additional mobility to the assembly and reduced the risk of fracture upon impact. However, I was unable to identify any of the smallest (tertiary) valves among the collections. Their apparent absence from the Karluk assemblage could be explained if the use lives of these protected inner components were considerably longer than the rest of the assembly, so that they were
Fig. 4 Several variants of the fish harpoon design: a a typical, asymmetrical two-piece assembly with socket to accommodate a foreshaft; b symmetrical point with basal spurs and socket; c “stand-alone” tip with basal socket; d a three-piece assembly. Drawing of smallest valve after Amorosi (in Knecht 1995:207)
Reconsidering Raw Material Selection
simply discarded with less frequency. Alternatively, they may have been reworked onto other tools. The best raw material to complement the structural impact resistance built into fishing harpoon design was one that was effective at absorbing repeated, high-intensity stresses. But because penetration through the soft-bodied prey would have posed little challenge, stiffness and strength would have been a low priority. Barbed Harpoon Points Alutiiqs created complex, multicomponent harpoons used to take seals and sea lions (e.g., Lisiansky 1968:Plate III) through a two-step sequence of “strike and hold.” The point’s shouldered base slipped into a wide groove at the tip of a socket. The opposite end of the socket made a tongue-and-groove articulation with the wooden main shaft, which was equipped with a dart butt at its base to aid its launch with a throwing board. After a strike, the point detached from the socket but remained connected to the shaft via a sinew line tied through the hole in the harpoon’s base. The harpoon was not used to kill, but to incapacitate, so that with the help of floats, a hunter could track the creature’s movements in the water and deliver a fatal blow with a lance. All barbed harpoon points from the collections had pointed and barbed tips, with up to two additional unilateral barbs. (A single, bilaterally barbed harpoon from Karluk One with other unusual morphological characteristics was likely imported from outside the archipelago (Knecht 1995:226).) Barb edges could be straight or curved, but the morphology was always consistent within a single specimen. Other diagnostic features of barbed harpoon tips included a base whose single shoulder was always located on the harpoon’s barbed edge, and a line hole that was placed predictably in the crook of this shoulder where the base was widest, and at the farthest point from the unbarbed edge. Barbed sea mammal harpoon tips had to first possess enough strength and stiffness to strike effectively, penetrating the marine prey’s tough epidermis. Barbed harpoons were also designed to hold this connection to the prey despite its struggles (Julien 1982). In this second, “hold” phase of the sequence, the tip had to be resilient (tough) because if the harpoon broke, the quarry was lost. These two functional requirements posed a serious design challenge because material stiffness and strength (“strike”), on one hand, and toughness (“hold”) on the other, are generally at odds in skeletal tissues (Currey 2002:133–136; Margaris 2009). Thus, Alutiiq toolmakers may have been forced to compromise in maximizing a harpoon’s striking or holding abilities. Unbarbed Arrow Points The majority of arrows featured “stemmed” bases and unbarbed blades that were round or semi-circular in cross-section. Many had blunt tips (perhaps by design) and arrows with both sharp and blunt tips were included in the study. Alutiiqs used arrow tips with barbed blades in warfare (Gideon 1989:42) and possibly for taking bears (Birket-Smith in Knecht 1995:293), while unbarbed points were likely designed for
Margaris
taking small game and birds. Davydov presents this image of duck hunting with arrows over open water: When several baidarkas sight a flock of these birds [ducks] they paddle up to them, suddenly fire several arrows and shout for all they are worth. At this some of the ducks take fright, but before they can fly out of the water they have to dive. Thus they surely fall into the hands of the Americans, who give them no chance to draw breath, firing a constant barrage of arrows at them. In the end the ducks become so tired that they pass out, and are killed with paddles, or they escape ashore where they are caught by hand (Davydov 1977:228). Cormorants, anatids (ducks and geese), and alcids (murres and puffins) in fact comprise the majority of identifiable avian specimens recovered at Settlement Point (Partlow 2000:208). Because the pelts of spring cormorants, puffins, and ground squirrels were all used to construct parkas, it is likely that at least some Alutiiq arrows worked by stunning the animal (as the quotation above suggests) so as not to damage delicate skins. Knecht (1995:288) identified wooden self-tipped blunt arrows for birding among the Karluk One collection, and it is possible that unbarbed arrow points represented part of an alternative system for achieving the same end. An arrow point was subjected to considerable stress as it made contact with the prey or fell to the ground below. Strength and toughness were likely key desired properties of a successful and reusable arrow. High speeds would help diminish the need for stiffness, particularly if stunning the animal, rather than penetrating valuable pelts, was the primary goal. Woodworking Wedges Alutiiqs employed wedges in conjunction with hammerstones for splitting wood. Like their modern metal counterparts, most osseous wedges had a uniform width and tapered in thickness from a large butt to slender splitting edge. Their lengths varied considerably, as wedges were no doubt used for a range of tasks from heavy splitting of driftwood logs (Knecht 1995:439) to more delicate finishing work. According to local oral histories, fat was used to lubricate wedges in the past, and Knecht (1995:439) proposes that concavities found along the faces of some wedges may have been used to contain lubricants. Wedges, like fiber-working tools, remained in the control of the worker during use and underwent a fairly predictable form of loading. Ideally, the blow of a hammer created a pure compressive force that diffused across the width of the blade. In all but the stiffest instruments, however, a combination of compression and bending was the more realistic mode of loading. This means that wedges, and particularly those above a certain size, posed a mechanical quandary to woodworkers employing osseous materials. To support heavy impacts and successfully penetrate wood substrates, wedges needed to be both strong and fracture resistant. What is more, wedges also required stiffness so as not to buckle under a striking force, and to successfully relay the striking force directly into the wood. Stiffness and strength go hand in hand in bone and are inversely related to its degree of fracture resistance (Albrecht 1977; Currey 1969, 2002; Margaris 2009). All three
Reconsidering Raw Material Selection
properties would have been important to the successful functioning of wedges, and so wedges made of osseous materials represented a design compromise regardless of the specific type of bone or antler chosen. The best way to triangulate between these constraints was perhaps to create a wedge that was large, using a material that is good at resisting fracture. In the same way that a single twig is easier to snap than an entire bundle, “building bigger” imparts a degree of structural stiffness and strength that can help ameliorate the problems associated with using a material that is not very stiff or strong. Pointed Tools Pointed tools (after Owen 2005) comprised a rather loose category, defined here as pieces with at least one sharpened end created either through breakage or deliberate working. Many were long and slender, while others possessed a flat and wide gripping region and “nosed” end. They were admittedly a heterogeneous and informal group of objects and, as defined here, one that likely overlapped with several other artifact types. Awls and awl-like instruments run the risk of being conflated with any number of simply and similarly shaped instruments, from slender foreshafts (whose tips are narrowed but not pointed) to toggles. They might also have assumed different functions throughout their use lives. Nonetheless, some of these tools possessed diagnostic features that distinguished them from other artifact types, such as several made on land mammal bone that retained an articular end. A good number were also made of slender bird bones. Related technologies offer some clues to how pointed instruments might have been used, including sewing kamleikas (parkas), and kayak covers, and in creating the type of expertly woven objects recovered at the Karluk One site. All were hand-held instruments used to place direct pressure on the working material. In this crucial way, they differed from the arrows and barbed harpoon tips described here. Projectile technologies left the hand (and often the eye), but skin and fiberworking tools were used to maintain fine and continuous control over the stresses acting on the piece, which could be adjusted at will. Relatively heavy stresses might have been placed on sharp tips but in a gradual and constant manner; a skilled worker would have a visual and tactile sense of acceptable pressure. The result was a type of loading that sustained rather than sudden and adjustable rather than unmodulated. Tools used for piercing skin and hides needed first and foremost to be stiff, and able to withstand compressive, bending and torsional forces without buckling. Tools used in basketry and working of other fibers did not require very sharp tips. Fairly blunt but smooth working ends might have been preferred for separating or splitting adjacent fibers without snagging.
Alutiiq Strategies for Raw Material Selection A tool begins life as a raw material that is selected for use by a craftsperson. Patterns of raw material choice thus provide a starting place for identifying some key goals of the toolmaker as well as constraints on how their designs could be realized. Each of the five types of equipment described above was designed to withstand distinct,
Margaris
repeated stresses throughout its use life. By taking into account the intrinsic properties of the various materials Alutiiq toolmakers chose for their tools, we can begin to understand the ways Alutiiqs anticipated and prepared for these stresses, and to uncover other factors that influenced strategies for technological organization employed in the Alutiiq world and beyond. Each of the roughly 300 complete and fragmentary artifacts was identified as being made from one of five types of skeletal materials: (1) antler, (2) whale bone, (3) compact mammalian limb bone, (4) bird bone, or (5) ivory. Cetacean bone can be distinguished from antler by the former’s gradual, rather than abrupt, transition from cancellous to compact tissue. Artifacts were classed as large cetacean (i.e., whale) bone also based on size (e.g., those clearly constructed from whale rib segments) or, for smaller objects, the presence of large nutrient foramina, which were visible in the denser regions of the tissue (Krzyszkowska 1990). These tubes generally run parallel to the long axis of the element and, in cross-section, appear as dark pores (see Pétillon 2008:Figs. 2 and 3). If the material could not be confidently identified, an object was excluded from the analysis. Ethnohistoric information guided my arrangement of the five types of tool components along a continuum of inferred modes and rates of loading, from those required to resist sharp impacts (left) to those exposed to more slow and sustained stresses (right) (Fig. 5). Raw material frequencies were calculated for each artifact type, and a chi-square analysis carried out to test the null hypothesis that Alutiiqs’ selection of antler, whale bone, and bird bone for each of the five artifact types was random. Compact limb bone and ivory were excluded from the test due to small artifact counts (an interesting pattern that I will turn to later), and a few cells that were included in the test contained low counts (Table 2). The analysis nonetheless revealed that frequencies of antler, whale bone, and bird bone use differed significantly between the five artifact classes (X 2 [8, N = 293] = 282.6, p < 0.01). These results, paired with the tool functions shown in Fig. 5, reveal a robust pattern of raw material selection based on the anticipated need for fracture resistance, or durability.
100 %
106
73
44
32
49 ivory
80 %
bird bone compact limb bone
60 %
whale bone 40 %
antler
20 %
0%
sharp impacts
fish barbed arrows harpoons harpoons
wedges pointed tools low impacts or sustained stresses
Fig. 5 Alutiiq raw material preferences for each of five artifact types (adapted from Margaris 2009). Antler use declines, and whale bone use generally increases, as the need for fracture resistance diminishes
Reconsidering Raw Material Selection
Table 2 Counts of material types against artifact types
Antler
Whale bone
Compact limb bone
Bird bone
Ivory
Total
Fish harpoons
103
3
0
0
0
106
Barbed harpoons
52
15
6
0
0
73
Arrows
24
20
0
0
0
44
Wedges
2
30
0
0
0
32
Pointed tools
12
6
4
26
1
49
Total
193
74
10
26
1
304
Harpoons It is clear that Alutiiq toolmakers predominantly chose antler to create the highest impact tools: an overwhelming 97 % of fish harpoon points were made of antler, as were a majority of barbed harpoons used for sea mammal hunting. This trend is perhaps unsurprising from a purely materials perspective, as antler is far better than compact limb bone (or nearly any other documented skeletal material) at resisting fracture. What makes the pattern remarkable in the Alutiiq context is the fact that antler was not locally available to Alutiiq communities in the protohistoric era. The Russian priest Gideon recounts how Alutiiqs obtained caribou antler for making tools through trade networks that extended to the Alaska Peninsula. The northern and western inhabitants of Kad’iak [Kodiak] engaged in barter trade mostly with the Americans of Aliaksa [Alaska Peninsula]…. The former obtained from the Alaskans, in exchange for dentalium shell beads and amber, caribou antlers used for spear tips [foreshafts], caribou parkas, and also long caribou hair taken from the animals’ chest…(Gideon 1989:57). Karluk, the site yielding the largest assemblage studied here, is nearly 50 km from the Alaska Peninsula’s nearest point. According to Holmberg (1985:37–38), the dentalium shell Alutiiqs used in some of these transactions was itself secured through long-distance trade. Holmberg writes that Alutiiqs also made occasional forays to the Alaska Peninsula to hunt caribou directly (1985: 57); both activities further underscore the value protohistoric Alutiiqs placed on antler. In sum, to create tools whose functioning required maximum fracture resistance and durability, Alutiiqs prized caribou antler above all other materials—despite the inconvenience of acquiring it. This insight helps us to better understand Alutiiq strategies for dealing with the competing functional (“strike” versus “hold”) requirements of barbed harpoon points. Alutiiq harpoon makers might have preferred caribou compact limb bone for its penetrating power, versus caribou antler for its ability to resist fracture. The fact that limb bone proved to be the least popular raw material for harpoon manufacture suggests that Alutiiqs selected their materials not to maximize the tool’s striking and penetrating effectiveness, but rather to maximize its durability—its capacity, in other words, to remain intact following a successful strike and to be used again. This is not to deny that striking effectiveness was a crucial feature of harpoon performance. Indeed, the harpoon, a design found though out the world, has a sharp tip and barbs that are beautifully suited to first pierce and then hold. I instead argue that the harpoon’s necessarily complex morphology posed serious structural risks for which Alutiiqs sought to compensate by choosing tough fabrication materials.
Margaris
Engineers have learned that holes and notches on an object’s surface are natural stress concentrators that can lead to fracture (Lipson and Juvinall 1963). Koniag Tradition style harpoons, which featured a basal line hole and unilaterally arranged barbs, would seem to represent a high-risk mechanical design. The area around line holes was severely strained as a hooked animal was tugged in the water, and harpoon barbs are nothing more than a series of notches in reverse! Realizing the precariousness of these design elements, Alutiiq craftsmen consistently placed the harpoon’s line hole at the widest portion of the base, where it was strongest and best able to support the stress. Overall, they used raw materials and the structural design of harpoons in very clever and complementary ways in order to achieve balance between two conflicting but equally crucial sets of performance characteristics. Finally, about 20 % of barbed harpoons were fabricated not from antler but cetacean bone. One explanation for the use of whale bone could be its ready availability in the Kodiak region as a result of planned hunts and beach scavenging. Other factors, however, emerge below in discussions of the three remaining tool types. Pointed Tools A total of 49 pointed tools and tool fragments were recovered from the three study locales. Most (88 %) of the identified specimens were complete, but given their morphological simplicity, it is possible that some fragments were overlooked. Both fish harpoon and barbed harpoon points were designed to withstand sudden stresses, and impacts greater than those encountered by hand-held fiber and skin-working tools. Yet, these pointed tools were made from an array of osseous materials with diverse properties. Stiff and strong bird bone comprised 53 % of the total identifiable sample: many preserved one epiphysis and a sharp tip produced by spiral fracture. Interestingly, only a small percentage was constructed from the intrinsically stiff limb bone of terrestrial animals. Other media included whale bone, ivory (used for only a single specimen in this region where walrus were absent and ivory was a trade good), and tough and flexible antler, which accounted for nearly a quarter (24 %) of the sample. The diversity of materials observed here could reflect the difficulty of classifying artifacts with similar, simple forms (see Griffitts 2006; LeMoine 1994; Owen 2005), as well as actual functional diversity within the group. Indeed, the waterlogged Karluk One site yielded a rich array of organic materials and products including grass, feathers, hair, and basketry (Knecht 1995) that were likely manipulated using a variety of tools represented in this assemblage. That the pointed tools in this study were not a functionally homogenous group is supported by the fact that their lengths also ranged widely, from 3.3–16.4 cm. Some of this variation might also reflect wear and resharpening. Finally, one interpretation of the diversity of materials seen for pointed tools is that there were simply few constraints on their design, as they were probably easy to construct, maintain, and perhaps replace. Based on their formal simplicity, one might infer that awls and other pointed tools were used for tasks that were not especially timesensitive or in other ways risky, with a corresponding low threshold for acceptable
Reconsidering Raw Material Selection
designs. However, in regions of the world where climatic extremes frequently threaten human survival, personal apparel and other types of stitched equipment are critical forms of subsistence gear (e.g., Binford 1977) whose improper functioning can be disastrous. Alutiiq Elder Nick Ignatin, for example, recalls how in the past, Alutiiq men brought a sewing kit (kakiwik), on board their kayaks in case the boat’s skin cover should need repairs at sea (in Crowell et al. 2001:151). It is clear that a fuller functional analysis of this diverse set of instruments is called for in order to understand who likely used them, and for what purposes. Wedges Next, the middle of the chart (Fig. 5) indicates the appreciable use of whale bone. Large cetacean bones were used to create all but two of the 32 wedges in the sample, many of which are rib segments whose lateral margins were left unworked. Few were definitively broken, but most showed signs of battering on the butt and blade ends. Pounding caused butt ends to crumple and created longitudinal splits emanating from the blade end which, in a few cases, had completely split the wedge in two. Whale bone is among the more massive materials available to traditional toolmakers (see Betts 2007; McCartney 1979), especially in terms of width, the greatest dimensional limitation for wedges. Antler offers the same benefit of a solid rather than tubular macrostructure but cannot offer the same girth. The two antler wedges were among the most slender of the group, while similarly sized wedges were also crafted from whale bone (Fig. 6). (Although I examined only osseous tools for this study, my informal survey of the full Karluk collection showed that Alutiiqs did use some wood, as well as whale bone, for wedges.) Size was clearly one factor in Alutiiq’s preferred use of whale bone for wedge construction, and in this case study, we can add ready access to the list. But a third reason for its selection on Kodiak and beyond may relate to the bone’s particular spongy architecture and its attending mechanical behavior. All of the smallest wedges in the Kodiak collections contained some cancellous tissue; whale rib wedges were almost completely cancellous. Materials scientists categorize cancellous tissue as a type of “cellular solid,” a set of very porous materials whose low relative density makes them lightweight, buoyant, and elastic. Other examples of cellular solids include honeycombs, polystyrene packing foam, and wood. These materials’ light
Fig. 6 Whale bone wedges were produced along a size continuum that overlaps the dimensions of the two antler specimens
Margaris
weight and superior shock absorbance (Gibson and Ashby 1997:2–11) parallel those of the cancellous tissue found in long bone epiphyses and the spongy “filling” in flat bones of the human skeleton. Like other cellular solids, cancellous tissue that is lightly compressed deforms elastically, meaning it returns to its original shape when the stress is released. With increased compression, individual trabeculae or pore (cell) walls begin to buckle or fracture, which causes open cells to collapse. The damage incurred by crushing is irreparable. Critically, however, up to the point when the opposing walls of cells compress enough to make direct contact, the tissue is becoming denser and hence, stronger (Gibson and Ashby 1997:439) rather than weaker. Wedges must be both stiff and strong in order to support heavy loads and transfer them ably through wood, and many of the Kodiak wedges indeed showed damage from the repeated application of heavy compressive loads. Clearly, accrued strength through use would have greatly benefited the performance of bone wedges. Thus, I argue that Alutiiqs predominantly chose whale bone for wedges for a number of reasons: the material’s ready availability and large size (which confers a degree of stiffness and strength at the structural level), to which we can now add potential for increased strength with use-related damage. None of these factors, however, fully explains why Alutiiqs frequently selected whale bone to create the final artifact category investigated here: arrows. Arrows Bird bone is the only material of this study too small for Alutiiq arrow manufacture, yet at both sites from which they were recovered (Karluk One and Settlement Point), arrows were crafted in roughly equal numbers from whale bone and antler only. We have already seen how Alutiiqs preferred to work with antler when creating tools that needed to resist fracture. Thus, we can infer that durability (rather than stiffness, conferred through the use of compact limb bone) was one important criterion in arrow design, just as it was for fish and sea mammal harpoons. Why then were roughly half of all arrows in the sample crafted not from antler, but from whale bone? Earlier, I cited evidence that cetacean bone is nearly as flexible as antler, a feature that in bony tissues is often correlated with high fracture resistance. Yet, these two materials differed greatly in their accessibility. Together, whale bone’s properties and its patterns of use on Kodiak raise the intriguing possibility that Alutiiqs viewed it as a sort of “poor man’s” antler: a resource that was locally available, with a degree of fracture resistance that fell short of antler but was nonetheless superior to that of land mammal limb bone. On one hand, perhaps arrows made of whale bone were widely regarded as more “disposable” than their antler counterparts—easy to make, lose, and replace. On the other hand, not all Alutiiq toolmakers may have had the opportunity to choose between high-quality but scarce antler, and whale bone, the poorer-quality but more plentiful local option. This suggestion is not intended to downplay the importance of whales and whalers in Alutiiq society. Indeed, whaling was a seasonal activity carried out by ritual specialists, so there were potential restrictions on material access. But it is worth considering the possibility that in the ranked society that developed on Kodiak in the Koniag Tradition, one in which
Reconsidering Raw Material Selection
only rich men commandeered slave labor, owned sturdy boats for long-distance travel, and other means of accruing social and political power (Fitzhugh 2003), trade items like antler obtained from the distant mainland were especially valued as tokens of prestige, in addition to serving more utilitarian functions. Further research is needed to address the intriguing question of whether antler’s distribution was socially limited in a way that whale bone was not.
Discussion Skeletal Materials, Risk, and Design for Durability In this study, I integrate artifact analysis, ethnohistory, and materials data in order to explore how raw material selection can be used to help carry out a number of tool design strategies. In some cases, bone or antler tools are revealed as solutions to technological quandaries in which toolmakers sought to balance a number of potentially conflicting goals, from maximizing a tool’s use life to minimizing raw material acquisition costs. The study also draws attention to the ways raw material selection can be used to strategically affect foraging outcomes via a tool’s practical functioning (Fig. 7), evidenced most robustly here in Alutiiqs’ use of tough materials (antler and whale bone) to create tool components used in high-risk foraging contexts (Bamforth and Bleed 1997; Torrence 1989). Alutiiqs used salmon harpoons, for example, to capture a sizable resource package whose availability was predictable, but also ephemeral. Failure to obtain adequate quantities of anadromous fish also came at a high cost because Alutiiqs relied on salmon stores as an important late winter/early spring food source. Alutiiqs drew on antler’s natural resilience to create harpoon tips capable of withstanding repeated impacts with the shallow, rocky creek beds in which salmon were captured and
sharp impacts
flexible flexible & & very probably tough tough
antler
ivory (dentin)
sharp impacts, pounding/ crushing
TechnoFunction
flexible & tough, increased strength through compression
porous whale bone
low impacts or sustained stresses
Mechanical Property
Raw Material
strong & stiff
terr. limb bone
bird bone tubes
Fig. 7 Relationships between raw materials, their properties, and the practical functioning of osseous tool components. Moving upwards on the chart: some raw materials possess mechanical properties which make them ideally suited for maximizing one desired tool function, such as resisting impacts. Conversely, a top– down reading models expected raw material choices based on technological strategies that emphasize effective tool functioning over other, potentially competing design criteria
Margaris
further amplified the tool’s reliable functioning by employing a multi-component, impact-resistant tip assembly. Similarly, Alutiiqs used antler and whale bone to make barbed harpoons whose capacity to resist fracture was essential during the dangerous struggle to capture valuable sea mammals. These examples illustrate how selection of highly fracture resistant materials can be a powerful means for creating tools which are highly durable. Design for durability, in turn, represents one pathway toward a larger goal of optimizing tool reliability (sensu Bleed 1986). These results corroborate and expand on conclusions from studies that have compared the use-related performance characteristics of stone and organic weapons tips. (See Pétillon et al. 2011 and Pokines 1998 for reviews of the experimental literature.) Both experimental and ethnographic evidence (Ellis 1997) suggest that flint and obsidian tips tend to have shorter use-lives than their organic counterparts. They can hold sharper tips and edges, and are more likely to fracture in a wound, causing greater damage to the prey. Stone points are brittle, however, often fracturing on impact. Alutiiq whaling lance tips, ground thin from fragile slabs of slate, may exemplify this type of design. In contrast, antler projectile points have good penetrating power and often sustain minimal damage during impacts with target carcasses. These projectile studies do not address the performance characteristics of tools intended for use outside of terrestrial hunting contexts, however, while Alutiiq fish and sea mammal harpoons offer a wider perspective on the role of raw material selection in carrying out tool design strategies. Two related points are worth mentioning here. First, bone and antler are best at resisting fracture when used wet rather than dry (Currey 1979; MacGregor 1985; Yamada 1970), a feature which would amplify the durability of tools used in aqueous contexts. Second, the role of raw material selection in technological strategizing is far from straightforward. Alutiiq tools underscore how design choices regarding a tool (or tool component) material and a tool’s overall structure can play off each other in a number of ways. We have already seen how durability was built into the structural designs, as well as material choice, of Alutiiq harpoon tips. As Knecht has noted (1995:205), multi-component fish harpoons employed a telescoping design that would “give” upon impact. Alutiiq barbed harpoons were similarly designed to prevent fracture, through the placement of the line hole along the edge of the base, where the harpoon was widest (and strongest)—a featured shared by some harpoons of Magdalenian design (e.g., Julien 1982). A tool’s material and structural properties need not echo each other, however, and design at one scale can be used to compensate for a perceived but unavoidable limitation at another. For example, as a trade-off for lowered procurement or manufacturing costs, a toolmaker may select a raw material whose properties are considered less than ideal for a given use. The overall size and shape of the tool can then be designed to counteract or compensate for sub-par raw material properties. Such was almost certainly the case for Alutiiq woodworking wedges, which required toughness, strength and stiffness—three properties that do not tend to co-occur in skeletal tissues. By selecting large segments of whale bone for their wedges, Alutiiq toolmakers could create tools that were stiff and strong by nature of their sheer size, even if the raw material itself is not particularly so.
Reconsidering Raw Material Selection
A Fresh Look at Whale Bone Alutiiq technological expertise also offers several insights into the practical uses of whale bone. While of no significance to contemporary engineers, the ribs and vertebrae of whales were once valued materials for the construction of permanent architecture and portable tools among numerous cultures, including those occupying arctic and subarctic regions of North America (Black 1987; Clark 1974, 1979; Crowell et al. 2001; Heizer 1956; Jordan and Knecht 1988; McCartney 1979; Savelle 1997; Steffian 1992;) as well as coastal California (Harrington 1928:134; Rogers 1929:209; Wake 1997, 1999), and among groups from Pleistocene France (Pétillon 2008) to Iron Age Scotland (Hallén 1994). Whale bone possesses several properties that would prove desirable for toolmakers in certain contexts. First, it offers a good deal of material from which to construct sizable tools (Betts 2007). Its abundance also makes it possible to compensate easily for errors in manufacture (poor-quality objects can simply be discarded), and to experiment without fear of exhausting raw material supplies. Second, along the spectrum of skeletal materials, whale bone is pliant rather than stiff (Currey 1988; Kabel et al. 1999; Scheinsohn and Ferretti 1995), which is useful for tools intended to flex with use. Next, there is good evidence that whale bone is a very fracture resistant material—perhaps nearly as tough as antler. Although there are methodological problems in directly testing the toughness of porous bone (Keaveny et al. 1997), Scheinsohn and Ferretti’s (1995) proposal that whale bone is more fracture resistant than land mammal (in their case, camelid) bone is supported and augmented by Alutiiq patterns of raw material selection. Alutiiq toolmakers’ frequent overlapping uses of whale bone and antler suggest that cetacean bone is not only tougher than land mammal bone, but its toughness is a close second to ultratough antler. Finally, to these properties, we can add that as a type of cellular solid, porous whale bone likely possesses the interesting property of gaining strength as it is compressed. Although cetacean bone is not an inherently strong material, its size allows toolmakers to “build big,” thereby adding a degree of strength at the structural scale that only increases as the tool is stressed in compression, as would be expected when used to perform pounding or crushing activities. These general statements about cetacean bone are supported by some newly recognized patterns of raw material selection evidenced at the French Magdalenian site of Isturitz. At this inland locale—in reverse of the Kodiak example—reindeer antler was locally plentiful and cetacean bone a distant resource. Nonetheless, resident Magdalenian toolmakers made remarkably similar raw material choices, selecting antler and whale bone seemingly interchangeably to produce their wedges and projectile points (Pétillon 2008). Evaluating Tool Investment Costs It is evident that skeletal materials can be used to serve a number of technological roles, but at what costs? We know that in the Alutiiq context, design for durability was set against a backdrop of high material acquisition and labor costs. Given the geographic, and likely social, obstacles most Alutiiqs faced in obtaining caribou antler from the mainland, it is not surprising that many salmon harpoon tips were resharpened until little or no use-life remained in the tips. The formal complexity of
Margaris
both fish and sea mammal harpoons also hints at significant production costs. Fishing harpoon tips in particular had a highly specialized function, and considerable time and skill were required to produce their complex, carefully fitted tips in anticipation of their later use. Alutiiq fish harpoons, in other words, epitomized Bleed’s (1986) concept of a highly reliable tool. Barbed harpoon tips had fewer parts and were less specialized— ethnohistoric sources report their use to capture a range of sea mammal species—but were nonetheless expensive to produce in terms of the time and labor required to extract a blank, carefully shape the tool’s complex contours, and drill a line hole. Ethnographic and experimental together data support the notion that transforming skeletal media into formal tools can be a very a labor-intensive activity. Seen another way, biological materials like bone and wood are incredibly versatile, lending themselves to complex modification through shaping by soaking and steaming, and a range of blank reduction techniques that can involve chopping, grooving and wedging, splitting, abrading, scraping, and whittling, to name a few. However, modeling tool investment strategies demands a better understanding not only of the absolute time and labor investments associated with any particular tool but also of how they compare with the costs of investing in alternative technologies. For instance, Ugan et al. (2003) offer a set of heuristic models for comparing degree of investment in tools within the same general category (i.e., different shapes of fish hooks), which suggest that “tech investment” should increase when doing so significantly reduces time spent obtaining and processes resources. I suggest that researchers can build on their approach by investigating technological variation and change at both (1) larger and (2) smaller scales, e.g., (1) modeling how technological shifts take place not only within but also between tool categories (e.g., a shift from fish spears to fish hooks) as Bettinger et al. (2006) advocate, and (2) quantitatively assessing the time and labor inputs associated with creating a given tool form from a range of competing materials. Late Dorset communities of the Canadian High Arctic crafted harpoons from bone, antler, and ivory (LeMoine 2005), for example, Magdalenians from both antler and bone (Julien 1982). Although the end products may appear very similar, the underlying skills, knowledge, manufacturing tools, and labor schedules required to produce, repair, and maintain them likely were not, as a result of both innate differences in the materials themselves, and those differences which result from local ecological and cultural factors. It is easy to produce a slender tubular bone awl through spiral fracture, but to create a similar pointed tool from bone, antler, or ivory requires a more complex and specialized debitage sequence (e.g., Gelvin-Reymiller and Reuther 2010). Timed studies which compare, for instance, rates of groove and splinter blank production on antler versus terrestrial mammal bone will help build estimates of tool costs and their variability across materials. With these data in hand, we could begin to read strategic tool investment decisions not only from the materials represented in archaeological assemblages but also in lower-than-expected rates of raw material use. In the Alutiiq study, land mammal limb bone was among the stronger types of available skeletal materials, yet was seldom used in the five tool types examined. The pattern cannot be explained by acquisition costs alone nor by a mismatch between the size and geometry of mammalian long bones and those of the intended tool. Although Kodiak’s local environment supported only small-bodied mammalian taxa (with the exception of brown bear), through mainland trade or hunting Alutiiqs presumably could have obtained
Reconsidering Raw Material Selection
long bones from the same caribou which provided valuable antler. And the fact that terrestrial mammalian bone was selected to create (a small number of) harpoons and pointed tools demonstrates that its formal properties were not a total impediment to its use. Along the spectrum of skeletal tissues, limb bone is relatively brittle and stiff (Currey 2002; Margaris 2009), and when designing tools for use in high-risk foraging contexts, Alutiiqs clearly found land mammal bone less desirable than tougher materials. It is interesting then that in the lower risk contexts in which most pointed skin and fiber working tools were presumably employed, Alutiiqs turned most often to bird bone. In addition to being stiff, bird bone was also locally available, and could be quickly and easily shaped to suit. Similarly, timed tool manufacturing data might help explain the near absence of ivory tools in the Kodiak collections, as well as evaluate more generally when, and for whom, ivory working was considered “worth the trouble” (sensu Ugan et al. 2003). Walrus ivory and caribou antler are similar in their mechanical properties and their status as imports in the Kodiak Archipelago, yet they saw very different patterns of use. Unlike antler, ivory representation at the study sites is limited primarily to small, ornamental objects such as labrets which could be easily imported as finished objects. Could this disparity be explained by differences in manufacture techniques, skills, or scheduling constraints associated with working the two raw materials? What might a better understanding of the costs of ivory tool investment also reveal about the social and, perhaps, symbolic values Alutiiqs places on these and other non-local raw materials (e.g., Steffian and Saltonstall 2005)? What is more, this type of cost/benefit analysis can be extended to any number of similarly under-explored materials. Wood species, for example, also differ greatly in such technical properties as hardness and density (Gibson and Ashby 1997), and although wood itself rarely preserves in archaeological contexts, lithic woodworking tools offer a durable record of wood’s long history of use. In sum, the fact that toolmakers in the past employed a wide range of tool media suggests that technological solutions to the problems of making a living also took more forms than we have thus far recognized. The bottom–up analysis I employ here, which begins with a firm understanding of the inherent properties of the materials themselves, illustrates a framework for investigating relationships between raw material selection and artifact variability which is both nuanced and broadly applicable.
Acknowledgments I thank first and foremost the Alutiiq Museum and Archaeological Repository for providing access to collections and their continued support over many years. Jennie D. Shaw and Amy Steffian drafted the base map used here, and Amy, along with Steve Kuhn, Michael Schiffer, Jason Haugen, and three anonymous reviewers, offered shrewd and constructive comments on the manuscript. A special thanks to Mollie Callahan and Allison Davis for their technical assistance, and to William Fitzhugh, Stephen Loring, Link Olson, Charles Potter, and Kate Wynne for important cetacean conversation. Funding for this research was provided by the NSF (OPP-0424901), the University of Arizona Social and Behavioral Sciences Research Institute, and the Early American Industries Association.
References Albrecht, G. (1977). Testing of materials as used for bone points in the Upper Paleolithic. In H. Camps-Fabrer (Ed.), Méthodologie appliquée à l’industrie de l’os préhistorique (pp. 124–199). Paris: CNRS.
Margaris Andrefsky, W. (2005). Lithics: macroscopic approaches to analysis. Cambridge: Cambridge University Press. Bamforth, D. B., & Bleed, P. (1997). Technology, flaked stone technology, and risk. In C. M. Barton & G. A. Clark (Eds.), Rediscovering Darwin: evolutionary theory and archaeological explanation (pp. 109– 139). Archaeological Papers of the American Anthropological Association, No. 7. Washington: American Anthropological Association. Bettiner, R. L., Winderhalder, B., & McElreath, R. (2006). A simple model of technological intensification. Journal of Archaeological Science, 33, 538–545. Betts, M. W. (2007). The Mackenzie Inuit whale bone industry: raw material, tool manufacture, scheduling and trade. Arctic, 60, 129–146. Biewener, A. A. (1982). Bone strength in small mammals and bipedal birds: do safety factors change with body size? Journal of Experimental Biology, 98, 289–301. Binford, L. R. (1977). Forty-seven trips. A case study in the character of archaeological formation processes. In R. S. V. Wright (Ed.), Stone tools as cultural markers: change, evolution and complexity (pp. 24–36). Canberra: Australian Institute of Aboriginal Studies. Birket-Smith, K. (1953). The Chugash Eskimo. Nationalmuseets Skrifter, Etnografisk Raekke VI. Copenhagen: National Museum of Denmark. Black, L. T. (1977). The Konyag (the inhabitants of the island of Kodiak) by Ioasaf [Bolotov], 1794–1799, and by Gideon, 1804–1807. Arctic Anthropology, 14, 79–108. Black, L. T. (1987). Whaling in the Aleutians. Études/ Inuit Studies, 11, 7–50. Black, L. T. (2004). Russians in Alaska 1732–1867. Fairbanks: University of Alaska Press. Bleed, P. (1986). The optimal design of hunting weapons: maintainability or reliability. American Antiquity, 51, 737–747. Campbell-Malone, R. (2007). Biomechanics of North Atlantic Right Whale bone: mandibular fracture as a fatal endpoint for blunt vessel-whale collision modeling. Ph.D. thesis, Cambridge: Massachusetts Institute of Technology/Woods Hole Oceanographic Institution. Carter, D. R., & Hayes, W. C. (1977). The compressive behaviour of bone as a two-phase porous material. Journal of Bone and Joint Surgery, 59A, 954–962. Clark, D. W. (1974). Contributions to the later prehistory of Kodiak Island, Alaska. Archaeological Survey of Canada Paper No. 20, Mercury Series. Ottawa: National Museums of Canada. Clark, D. W. (1975). Technological continuity and change within a persistent maritime adaptation: Kodiak Island, Alaska. In W. Fitzhugh (Ed.), Prehistoric maritime adaptations of the circumpolar zone (pp. 203–227). The Hague: Mouton. Clark, D. W. (1979). Ocean Bay: An early North Pacific maritime culture. Archaeological Survey of Canada Paper No. 86, Mercury Series. Ottawa: National Museums of Canada. Cotterell, B., & Kamminga, J. (1987). The formation of flakes. American Antiquity, 52, 675–708. Crowell, A. L. (1997). Archaeology and the capitalist world system: a study from Russian America. New York: Plenum Press. Crowell, A. L., Steffian, A. F., & Pullar, G. L. (Eds.). (2001). Looking both ways: heritage and identity of the Alutiiq people. Fairbanks: University of Alaska Press. Currey, J. D. (1969). Mechanical consequences of varying the mineral content of bone. Journal of Biomechanics, 2, 1–11. Currey, J. D. (1979). Mechanical properties of bone tissues with greatly differing functions. Journal of Biomechanics, 12, 313–319. Currey, J. D. (1988). The effect of porosity and mineral content on the Young’s modulus of elasticity of compact bone. Journal of Biomechanics, 21, 131–139. Currey, J. D. (2002). Bones: structure and mechanics. Princeton: Princeton University Press. Davydov, G. I. (1977). Two voyages to Russian America, 1802–1807, R. Pierce. (Ed.), C. Bearne (Trans.). Kingston: The Limestone Press de Buffrénil, V., & Schoevaert, D. (1998). On how the periosteal bone of the delphinid humerus becomes cancellous: ontogeny of a histological specialization. Journal of Morphology, 198, 149–164. de Laguna, F. (1975). The archaeology of Cook Inlet, Alaska. Anchorage: Alaska Historical Society. Ellis, C. J. (1997). Factors influencing the use of stone projectile tips: An ethnographic perspective. In H. Knecht (Ed.), Projectile Technology (pp. 37–74). New York: Plenum Press. Felts, W. J. L., & Spurrell, F. A. (1965). Structural orientation and density in cetacean humeri. The American Journal of Anatomy, 116, 171–203. Fitzhugh, B. (2003). The evolution of complex hunter-gatherers. New York: Kluwer Academic. Gál, E. (2005). New data on bird bone artefacts from Hungary and Romania. In H. Luik, A. M. Choyke, C. E. Batey, & L. Lõugas (Eds.), From hooves to horns, from mollusc to mammoth (pp. 325–338). Oxford: Oxbow Books.
Reconsidering Raw Material Selection Gelvin-Reymiller, C., & Reuther, J. (2010). Birds, needles, and iron: Late Holocene prehistoric Alaskan grooving techniques. Alaska Journal of Anthropology, 8(1), 1–22. Gibson, L. J., & Ashby, M. F. (1997). Cellular solids: structure and properties (2nd ed.). Cambridge: Cambridge University Press. Gideon, H. (1989). The round the world voyage of Hieromonk Gideon 1803–1809. R. A. Pierce (Ed.), L. T. Black (Trans.). Kingston, Ontario: The Limestone Press. Griffitts, J. L. (2006). Bone tools and technological choice: Change and stability on the Northern Plains. Ph.D. thesis, Tucson: University of Arizona. Guthrie, R. D. (1983). Osseous projectile points: Biological considerations affecting raw material selection and design among Paleolithic and Paleoindian peoples. In J. Clutton-Brock & C. Grigson (Eds.), Animals and archaeology: hunters and their prey. BAR International Series 163 (pp. 273–294). Oxford: Archaeopress. Hallén, Y. (1994). The use of bone and antler at Foshigarry and Bac Mhic Connain, two Iron Age sites on North Uist, Western Isles. Proceedings of the Society of Antiquaries of Scotland, 124, 189–231. Harrington, J. P. (1928). Exploration of the Burton Mound at Santa Barbara, California. Forty-fourth annual report of the Bureau of American Ethnology, 1926–27 (pp. 23–168). United States Government Printing Office: Washington, DC. Hayden, B., Franco, N., & Spafford, J. (1996). Evaluating lithic strategies and design criteria. In G. H. Odell (Ed.), Stone tools: theoretical insights into human prehistory (pp. 9–45). New York: Plenum Press. Heizer, R. F. (1956). Archaeology of the Uyak Site, Kodiak Island, Alaska. Anthropological Records (17th ed., Vol. 1). Berkeley: University of California Press. Holmberg, H. J. (1985). Holmberg’s ethnographic sketches. The Rasmuson Library Historical Translation Series, Vol. I. M. W. Falk, & M. W. (Eds.), F. Jaensch (Trans.). Fairbanks: University of Alaska Press. Jordan, R. H., & Knecht, R. A. (1988). Archaeological research on western Kodiak Island, Alaska: The development of Koniag culture. In R. D. Shaw, R. K. Harritt, & D. E. Dumond (Eds.), The late prehistoric development of Alaska’s native people (pp. 225–306). Anchorage: Aurora IV, Alaska Anthropological Association Monograph Series. Julien, M. (1982). Les Harpons Magdaléniens. 17th Supplément á Gallia Préhistoire. Paris: CNRS. Kabel, J., van Rietbergen, B., Dalstra, M., Odgaard, A., & Huiskes, R. (1999). The role of an effective isotropic tissue modulus in the elastic properties of cancellous bone. Journal of Biomechanics, 32, 673–680. Keaveny, T. M., Panilla, T. P., Crawford, R. P., Kopperdahl, D. L., & Lou, A. (1997). Systematic and random error in compression testing of trabecular bone. Journal of Orthopedic Research, 15, 101–110. Knecht, H. (1997). Projectile points of bone, antler, and stone: Experimental explorations of manufacture and use. In H. Knecht (Ed.), Projectile technology (pp. 191–212). New York: Plenum Press. Knecht, R. A. (1995). The late prehistory of the Alutiiq people: Culture change on the Kodiak Archipelago from 1200–1750 A.D. Ph.D. thesis, Bryn Mawr: Bryn Mawr College. Krzyszkowska, O. (1990). Ivory and related materials: an illustrated guide. Bulletin Supplement, University of London Institute of Classical Studies No. 59. London: Institute of Classical Studies. Lantis, M. (1952). Eskimo herdsmen: Introduction of reindeer herding to the natives of Alaska. In E. H. Spicer (Ed.), Human problems in technological change (pp. 127–148). New York: John Wiley and Sons. LeMoine, G. (1994). Use wear on bone and antler tools from the Mackenzie Delta, Northwest Territories. American Antiquity, 59, 316–334. LeMoine, G. (2005). Understanding Dorset from a different perspective: Worked antler, bone, and ivory. In P. D. Sutherland (Ed.), Contributions to the study of Dorset Palaeo-Eskimos (pp. 133–144). Ottawa: Canadian Museum of Civilization. LeMoine, G., & Darwent, G. M. (1998). The walrus and the carpenter: late Dorset ivory working in the High Arctic. Journal of Archaeological Science, 25, 73–83. Lipson, C., & Juvinall, R. C. (1963). Handbook of stress and strength: design and material applications. New York: The Macmillan Company. Lisiansky, U. (1968). A voyage round the world in the years 1803, 4, 5, & 6: Performed by order of His Imperial Majesty Alexander the First, Emperor of Russia, in the ship Neva. London: J. Booth. Luedtke, B. E. (1992). An archaeologist’s guide to chert and flint. Archaeological research tools 7. Los Angeles: Institute of Archaeology, University of California. Luehrmann, S. (2008). Alutiiq villages under Russian and U.S. rule. Fairbanks: University of Alaska Press. MacGregor, A. G. (1985). Bone, antler, ivory and horn—the technology of skeletal materials since the Roman period. London: Croom Helm. MacGregor, A. G., & Currey, J. D. (1983). Mechanical properties as conditioning factors in the bone and antler industry of the 3rd to the 13th century A.D. Journal of Archaeological Science, 10, 71–77.
Margaris Margaris, A. V. (2009). The mechanical properties of marine and terrestrial skeletal materials: implications for the organization of forager technologies. Ethnoarchaeology, 1, 163–183. McCartney, A. P. (Ed.). (1979). Archaeological whale bone: A northern resource. First report of the Thule conservation project. University of Arkansas Anthropological Papers No. 1. Ottawa: National Museums of Canada and Department of Indian and Northern Affairs. McKittrick, J., Chen, P.-Y., Tombolato, L., Novitskaya, E. E., Trim, M. W., Hirata, G., et al. (2010). Energy absorbent natural materials and bioinspired design strategies: a review. Materials Science and Engineering: C, 30, 331–342. Miller, G. A. (2010). Kodiak kreol: communities of empire in early Russian America. Ithaca: Cornell University Press. Nelson, M. C. (1991). The study of technological organization. In M. B. Schiffer (Ed.), Archaeological method and theory (Vol. 3, pp. 57–100). Tucson: University of Arizona Press. Nelson, M. C. (1997). Projectile points: form, function, and design. In H. Knecht (Ed.), Projectile Technology (pp. 371–384). New York: Plenum Press. Owen, L. R. (2005). Distorting the past: Gender and the division of labor in the European Upper Paleolithic. Tubingen: Kerns Verlag. Partlow, M. A. (2000). Salmon intensification and changing household organization in the Kodiak Archipelago. PhD thesis, Madison: University of Wisconsin. Pennycuick, C. J. (1967). The strength of the pigeon’s wing bones in relation to their function. Journal of Experimental Biology, 46, 219–233. Pétillon, J.-M. (2008). First evidence of a whale-bone industry in the western European Upper Paleolithic: Magdalenian artifacts from Isturitz (Pyrénées-Atlantiques, France). Journal of Human Evolution, 54, 720–726. Pétillon, J.-M., Bignon, O., Bodu, P., Cattelain, P., Debout, G., Langlais, M., et al. (2011). Hard core and cutting edge: experimental manufacture and use of Magdalenian composite projectile tips. Journal of Archaeological Science, 38, 1266–1283. Pokines, J. T. (1998). Experimental replication and use of Cantabrian lower Magdalenian antler projectile points. Journal of Archaeological Science, 25, 875–886. Rausch, R. L. (1969). Origin of the terrestrial mammalian fauna of the Kodiak Archipelago. In T. N. V. Karlstrom & G. E. Ball (Eds.), The Kodiak Island refugium: Its geology, flora, fauna, and history (pp. 216–234). Calgary: The Boreal Institute, University of Alberta. Rogers, D. B. (1929). Prehistoric man of the Santa Barbara coast, California. Santa Barbara: Santa Barbara Museum of Natural History. Savelle, J. (1997). The role of architectural utility in the formation of archaeological whale bone assemblages. Journal of Archaeological Science, 24, 869–885. Schiffer, M. B., & Skibo, J. M. (1997). The explanation of artifact variability. American Antiquity, 62, 27– 50. Shahar, R., & Weiner, S. (2007). Insights into whole bone and tooth function using optical metrology. Journal of Materials Science, 42, 8919–8933. Scheinsohn, V., & Ferretti, J. L. (1995). The mechanical properties of bone materials in relation to the design and function of prehistoric tools from Tierra Del Fuego, Argentina. Journal of Archaeological Science, 22, 711–717. Steffian, A. F. (1992). Fifty years after Hrdlicka: Further excavation of the Uyak site, Kodiak Island, Alaska. In R. H. Jordan, F. de Laguna, & A. F. Steffian (Eds.), Contributions to the anthropology of southcentral and southwestern Alaska (pp. 141–164). Fairbanks: Anthropological Papers of the University of Alaska Vol. 24. Steffian, A. F., & Saltonstall, P. G. (2005). Tools but not toolkits: traces of the Arctic Small Tool Tradition in the Kodiak Archipelago. Alaska Journal of Anthropology, 3(2), 17–49. Stodiek, U. (2000). Preliminary results of an experimental investigation of Magdalenian antler points. In C. Bellier, P. Cattelain, & M. Otte (Eds.), La chasse dans la préhistoire/ Hunting in prehistory, SRBAP (Anthropologie et Préhistoire 111) (pp. 78–90). Bruxelles: Société Royale Belge d'Anthropologie et de Préhistoire. Torrence, R. (1989). Retooling: Towards a behavioral theory of stone tools. In R. Torrence (Ed.), Time, energy, and stone tools (pp. 57–66). Cambridge: Cambridge University Press. Ugan, A., Bright, J., & Rogers, A. (2003). When is technology worth the trouble? Journal of Archaeological Science, 30, 1315–1329. Waguespack, N. M., Surovell, T. A., Denoyer, A., Dallow, A., Savage, A., Hyneman, J., et al. (2009). Making a point: wood- versus stone-tipped projectiles. Antiquity, 83, 786–800.
Reconsidering Raw Material Selection Wake, T. A. (1997). Bone artifacts and tool production in the native Alaskan neighborhood. In K. G. Lightfoot (Ed.), The archaeology and ethnohistory of Fort Ross, California (pp. 248–278). Berkeley: Berkeley Archaeological Research Facility. Wake, T. A. (1999). Exploitation of tradition: Bone tool production and use at Colony Ross, California. In M.-A. Dobres & C. R. Hoffman (Eds.), The social dynamics of technology: practice, politics and world views (pp. 186–208). Washington, DC: Smithsonian Institution Press. Wall, W. P. (1983). The correlation between high limb-bone density and aquatic habitats in recent mammals. Journal of Paleontology, 57(2), 197–207. Wegst, U. G. K., & Ashby, M. F. (2004). The mechanical efficiency of natural materials. Philosophical Magazine, 84(21), 2167–2186. Whittaker, J. C. (1994). Flintknapping: making and understanding stone tools. Austin: University of Texas Press. Woodhouse-Beyer, K. E. (2001). Gender relations and socio-economic change in Russian America: An archaeological study of the Kodiak Archipelago, Alaska, 1741–1867. Ph.D. thesis, Providence: Brown University. Yamada, H. (1970). Strength of biological materials. New York: Williams and Wilkens. Zioupos, P., Currey, J. D., Casinos, A., & de Buffrénil, V. (1997). Mechanical properties of the rostrum of the whale Mesoplodon densirostris, a remarkably dense bony tissue. Journal of Zoology, London, 241, 725–737.