This is a general account of tool use by all Beverly cultures. As all had to adapt to the same environment and caribou herd, most had similar tools while some tool traits were culturally specific and are underlined in the tables. For brevity, cultures from earliest to latest are abbreviated: Northern Plano NP, Shield Archaic; SA; Pre-Dorset or Arctic Small Tool ASTt; with Early, Middle and Late Taletheilei as ET, MT and LT.
Adze Family Comparison
Adzes, axes, chisels, gouges and picks and their flakes fall under the adze family because most are ground shock-resistant basalt with similar traits. All are culturally affiliated and most are excavated. Adzes are long asymmetric unifacial tools with bifacially retouched hafts for wood or antler socket insertion. Unlike axes, their bits are perpendicular to the handle and sharpened by chipping and grinding the ventral surface. Prepared striking platforms on this surface may have a groove cut parallel to the bit for governing sharpening flake removal. This groove identifies a sharpening flake and permits grinding or use striae orientation. Its curvature matches that of the bit, while a sharpening flake from an adze side has a straight platform and edge. An axe resembles an adze except its bit is inline with its split attached handle. It has a bifacial and symmetrical bit, haft and midsection. Unless hand-hafted, chisels are smaller and have uni- or bifacial bits. Gouges are channel-ground for grooving wood. Picks are crude pointed earth diggers. Specific adze family data and analysis can be found in Gordon (1996).
There are 165 adze family tools, 137 in the tundra and 28 in the forest: 12 adzes (7%), 134 adze flakes (81%), 12 axes (7%), 2 chisels (1%), 3 gouges (2%) and 2 picks (1%). More numerous forest adzes reflect not only winter wood use but conservation when valuable tundra-specific shock-resistant basalt was unavailable. As the tundra has half as many adzes but 9 times as many flakes, extensive shaping and weight reduction of heavy adzes occurred before hunters moved south for winter. Adzes are in NP (1), ASTt (5), ET (5) & LT (1), but flakes in NP (27), SA (2), ASTt (3), ET (48), MT (41) & LT (12) imply adzes were used by all. Each in turn turned to wood resources in the winter forest to make tools used over the year. LT descendants or historic Chipewyan used metal axes, picks and chisels that were traded for fur.
Flat-bottomed and often ridged, most adzes have longitudinally and transversely striated bits, sides and bottom from their preparation and use. Six adzes are striated, 5 tundra and 1 forest, the latter with only normal longitudinal striae. Tundra adzes have 3 types of striae: transverse, longitudinal/transverse and diagonal/longitidinal; i.e., from normal head-on adze application through oblique to side application. Complex manufacturing and later use wear resulted in sharpening flakes with 8 types of striae, with 78 of 134 eroded, leaving 56 for examination. 20 (36%) have transverse striae, 9 (16%) diagonal, 10 (18%) longitudinal, 3 each (5% each) for transverse/diagonal, transverse/diagonal/longitidinal, and non-orientable (due to damaged striking platforms), and 7 (13%) longitudinal & transverse.
Like adzes, forest adze flakes have fewer striae types, with complex patterns missing, perhaps because adzes were only used there. Longitudinal striae resulting from normal head-on adze use is alike in both ranges: 9 of 50 in tundra flakes (18%) vs. 1 of 6 in forest (17%). Tundra and forest flake transverse striae have frequencies of 34 & 50%, diagonal as 16 & 17%, longitudinal as 18 & 17% and transverse/diagonal as 4 & 17%. On strial angle alone, functional differences in tundra and forest adzes are unproven. Function relates to worn area size and depth, with heavy wide pressure resulting in extensive striae. 135 flakes have traits traceable to their removal from specific adze locations, with ET with the most representative flake types and LT & SA the least. Position and frequency are lateral (71%), bit (25%), haft (2%) and dorsal ridge or keel (>1%). The ratio of lateral to bit flakes (ca. 3:1) reflects the elongated nature of adzes. Long narrow ET and MT haft thinning flakes match their adzes. Two ET ridge flakes are proximal to the bit and exclude the edge.
Flakes from known adze position number 121 & 14 in tundra & forest, with lateral and bit flake frequencies (68 & 79%, and 26 & 21%), respectively, plus the addition of haft and bit flakes (3 & 2%) in tundra sites. Turning to strial area, 31 (21%) are unstriated, while 37 (25%) and 78 (53%) have restricted or widespread striae. Tundra and forest flakes are 22 & 12% unstriated, 26 & 18 narrowly striated, and 51 & 71 widely striated. Greater forest adze use and curation is seen in more widespread striae. Of 146 adzes and flakes, 64 (44%) are unpolished, 6 (4%) lightly polished, 53 (37%) heavily polished, and 23 (16%) dorsally ridge polished. Polish varies from all types in ET to two in SA and ASTt. 129 tundra & 17 forest flakes are: unpolished 46 & 24%; lightly polished 4 & 6%; heavily polished 32 & 71%. Dorsal ridge polish is 18% and is unreported in forest sites. Higher frequency of unpolished tundra adzes and heavily polished forest adzes reiterate tundra basalt source and its heavy use in forest sites. 4 of 146 adzes and flakes have grooved platforms for precise sharpening. All are ET & MT and suggest tundra adze manufacture, with later sharpening flakes ungrooved. Platform angles are alike for all cultures (81-84 deg.), except for LT which is lower (mean 76 deg.). Bit platform angles vary widely (65-90 deg.), reflecting small specimen number or different use. Haft striking platform angles are <90 deg. throughout Taltheilei (68-70 deg.) and may indicate haft attachment via a socket.
The high frequencies of solid black and gray in tundra adze flakes suggest access to specific basalt dykes, as does the odd high frequency of gray-green in the forest. Both are site specific. 106 of 110 black flakes are from Grant Lake sites KkLn-2 & 4, while all 6 gray-green flakes are from IjOg-2 on Lake Athabasca. Sections are 64% keeled & 36% round-backed or planoconvex. 7 tundra adzes are 71% keeled & 29% round-backed, while 4 forest adzes are evenly split. Thus, tundra has more keeled rectangular adzes, possibly thicker for greater pressure needed in working antler rather than wood. A more thorough preparation of tundra basalt adzes ensured curation with many sharpenings, but ultimate discard is seen in scattergraphs of archaeological floors, with cultural phase-important traits and variables below:
Striae: 0=none; 1=transverse (perpendicular to bit edge or sharpening flake striking platform); 2=diagonal (midway between 1 & 3; 3=longitudinal (parallel to bit edge or flake striking platform); 4=1 & 2; 5=1, 2 & 3; 6=2 & 3; 7= unorientable due to missing striking platform; & 8=1 & 3.
Flake: 1=lateral; 2=bit; 3=haft; 4=complete; 5=ridge.
Area (of grinding or wear): 0=none; 1=restricted; 2=widespread.
Polish: 0=none; 1=light; 2=heavy; 3=facial ridges.
Grooved striking platform: 0=unknown due to platform destruction; 1=present; 2=absent.
Bitangle & striking platform = Face angle in degrees.
Burin Family Comparison
In the prolonged cold of 1500-600 B.C., the tundra and forest were occupied by Inuitlike Pre-Dorset bands of the Arctic Small Tool tradition (ASTt). It came from the Arctic coast on the heels of Shield Archaic Indian bands dispersing south to Saskatchewan. Both the latter and their predecessors, the Northern Plano Indians, made burins, but theirs and those of the Taltheilei Indians who followed the ASTt Inuitlike bands were very different from ASTt burins. Most Indian burins were made on lanceheads while ASTt burins were single purpose tools – more numerous and important – bifacial ones to slot resilient bone, antler and wood for insetting chert blades in harpoons, arrows and harpoons, and unifacial ones to plane tool bodies. There are 166 ASTt sites; 20 in tundra with 371 burin family tools, 247 with striae and 124 without. Of 45 forest sites only 7 have burin tools (10), 4 with striae and 6 without. Fewer forest tools relate to remote tundra chert sources. Of 121 tundra sites, 20 have 371 burin family tools, 3 of them, KjNb-3, 6 & 7, being stratified sites with most. Paradoxically, all are at a water-crossing where caribou and hunters were most transient on their southern migration. Compared to forest, tundra activities were more hectic (herd ambush, carcass retrievial, butchering, removing back sinew, wind-drying meat, skinning, scraping, smoking and tanning hides, and working antler and bone into tools with burins).
The burin family of slotters, burin planes, planes and primary and secondary spalls are defined and compared using functional analysis of their striae. The absence of shaped bone, antler and wood suggest organics were unpreserved in the acid soils. Most burins (and spall faces in new burins) have striae perpendicular to the longitudinal axis, but some unspalled blanks are striated. Slotters and planes often have striated tips, tapered bifacial haft, retouched and ground backs and mitten shape. Planes may be striated and bent in side view, the bent edge worn most. Thin slotter faces have bifacial parallel striae from slot wear, while thicker planes have unifacial striae on their flat surface from the high pressure needed to plane the sides of harpoon heads, bone points and art objects. Most slotters and planes have a retouched and ground edge opposite the spall face cutting edge. Extending from this edge into the haft was a channelled bone or antler backing for the pressure. Most burins have bifacially retouched and ground bases for hafting, while some large burins were hand-hafted. Spalls were also used as minute scrapers and gravers. Primary spalls are the first removed from a slotter or burin plane. They have triangular section, while later secondary spalls have rectangular section and striated uni- or bifacially according to their origin.
Striae: Production vs. Use Wear:
As most burin family tools have striae perpendicular to their axes, it is easy to assume intentional shaping by grinding as in Dorset burinlike tools, rather than use wear. But the Dorset ground most tools and were mediocre knappers compared to their ASTt ancestors who ground few tools and were superb knappers. Further, all striated tools are in sandy sites, while tools from rocky or muddy sites are unstriated. Contaminating sand grit likely striated tools by being caught between the burin and object worked because: (1) striated or ground non-burin tools are rare; (2) wear and grinding lines on the face opposite the striae are superficial, with tool thinning through flake removal; (3) striae cross the face when only the spall facet edge needs sharpening; (4) this edge is easier to resharpen by spalling than grinding; (5) in non-burin family tools like bifacial scrapers, edges are rounded, with deep striae across rather than along the edge. These striae are impossible and undesirable to grind, their origin likely due to worked damp skin becoming sand-contaminated in typical windy blowout sites; (6) bone and antler would polish rather than striate if clean; and (7) planes are striated over their concave face, a phenomenon difficult to reproduce with whetstones, but reproducible by planing the sand-contaminated convex sides of bone and antler tools.
Distribution of Striated Burin Family Tools:
Striated and unstriated tools are in 4 forest sites each, with KeNi-4 having both. They are in 20 tundra sites, with 9 sites having both types. 128 unstriated tools include 36 slotters (34 tundra – 3 stratified & 33 surface; 2 forest), 26 burin planes (24 tundra – 20 strat.; 2 surf.; 2 forest – both surf.), 25 planes (24 tundra – 6 strat.; 18 surf.; 1 forest – surf.), 14 primary spalls (all tundra – 6 strat.; 8 surf.) and 27 secondary spalls (26 tundra – 14 strat.; 12 surf.; 1 forest – surf.). The remaining 251 striated tools consist of 39 slotters in KjNb-6 & 7 in the tundra, and one from a forest site. There are 43 tundra and 2 forest striated burin planes. 16 stratified planes are from KjNb-6 & 7. There are 27 tundra and 2 forest surface planes. Striated unspalled planes include 18 tundra and 1 forest, all 8 stratified ones from KjNb-7. Primary striated spalls number 24, all tundra, 18 stratified ones from KjNb-6 & 7. 124 secondary striated spalls are mainly KjNb-3, 6 & 7.
Spalling and its Effects nn Burin Size, Use and Strial Angle:
For metric and strial measurements of burin family tools, two longitudinal axes are used as reference points: (1) main burin or plane axis running from haft to tip roughly parallel to and midway between the haft edges; and (2) axis outside the tip in heavily spalled burins which have become tapered. There, an imaginary line parallel to the longitudinal axis and behind the spallface is used for orientation. In parrot-beaked burins, the haft midpoint axis may be well within the tip but way off center, with angles >90 deg. as they “hang over” their spallscars and body. Dihedral burins maintain their 90 deg. strial angle as spall removal is usually even on both sides. Since spalls with 90 deg. striae come from parrot-beaked, dihedral and angle burins, it might be possible to determine where on the orginal burin the spall originated if we have enough of its tip to recreate it. This is the tip angle and measures the taper at the spall tip at its striking platform.
Burin strial angle is that angle between the longitudinal axis and its striae. Strial angles in burins may change by: (1) their use as chisels and gravers; (2) increased spalling because spalls are not parallel-sided, but thin from burin tip to hinge fracture. Unlike burin axes (1) or (2), the spall longitudinal axis runs midway along its length irrespective of how many spalls are removed. Change in spall strial angle may be seen as follows: a burin resembles a mitten with fingers and thumb together as imaginary spalls, the thumb being the primary spall. Striae are perpendicular lines drawn from thumb tip across the middle finger joints, so spall strial angles perpendicular to the hand will also be perpendicular to each spall. But spalls unlike fingers taper, so if fingers are spread, simulating spall removals, their “strial angle” will increase. New burins and early spalls often have 90 deg. angles, but more spalling (sharpening) may increase angles to 130 deg. as the burin is tilted so that the spall face remains perpendicular to the object worked. If we have spalls with angles filling the 90-130 deg. sequence, it is possible to re-assemble burins. If the burin is tilted back so the spallface retains the same angle of attack to the medium worked, angles in sequential spalls may be identical. To compare spalls and burins, a common measurement is needed – the angle between striae and spall facet. As spalls often curve, their longitudinal axes must often be approximated. To obtain strial angle, proximal or tip and distal or hinge ends must be known, as must the ventral face with its bulb of percussion and the distal face with its worn facet edges.
Non-Strial Analysis of Burin Family Tools:
Other traits than striae elicit burin function. Burin family tools may or may not have carination, a spallface, backing, ground haft or base, spallscars on sides or faces or both, back wear or retouch, prescar preparation and angle the spallface makes with the tip. Slotter primary and secondary spalls have similar length because slotters and burin planes have similar length, but more than half of their spalls are snapped. Primary spalls are shorter than secondary because they are the first removed. They are also thicker and narrower because they retain the original burin edge. Slotters are thinner for grooving, while burin planes are thicker for higher pressure. Planes are wider because they are unspalled. A full burin family comparison with figures is in Gordon (1996:175-196). Burin family tools are databased as follows, with important functional traits and variables:
Striae: light to heavy.
Angle1 & Angle2: strial angles of burin application that are on each side of burin or spall (*needs checking in January).
Spallangle: interpretation of original burin angle of application as determined from spall (*needs checking in January).
Tipangle: Angle where burin tip joins.
Spallface: True or False.
Backed: True or False.
Backwear: True or False.
Retouch: True or False.
Groundhaft: True or False.
Prescars: True or False.
Carinated: True or False.
Scargroups: Actual count.
Dualtool: needs a table (e.g.s – burin family tool may be scraper, knife, etc.; Side1 & Side2).
Chitho Comparison
Chithos are sandstone discs for softening skins by abrading them over their rough edge. Most are roundish but some are rectangular, square, lanceolate, tear or bell-shaped and pentagonal. These simple tools-of-the-moment are in a dozen sites, mainly near treeline where caribou skins are prime in late autumn. By number and culture, and from earliest to latest, a total of 312 chithos include 26 NP, 8 SA, 11 ASTt and 67 ET (1 Earliest), 113 MT & 80 LT and 7 Chipewyan. Accounting for breakage, which was common, there are 227 full length chithos (42 tundra; 185 forest); 224 full width chithos (52 tundra; 172 forest) and 310 full thickness chithos (80 tundra; 230 forest). Mean chitho length is similar for all cultures except SA and Chipewyan which are larger, while NP chithos are much shorter. Very pronounced differences are in mean weight, with NP chithos less than half that of other cultures. ET, MT & LT weights are similar, while Chipewyan chithos (which evolved from LT) are noticeably heavier. Mean SA chitho weight is five times that of NP, while ASTt is the smallest of all except NP.
Sandstone chithos number 279 (89%), with quartzite 28 (9%), schist 2 (<1%), and 1 of gneiss, granite and silicious shale (<1%). 257 have bifacially retouched edges (82%), 67 LT (21%) & 103 MT (33%), and higher than average unifacially retouched edges in SA (55 at 17%). Bifacial edge retouch is higher in MT tundra & forest chithos. 183 chithos (58%) are complete or slightly chipped, 22 (7%) are edge fragments, 2 (<1%) are central fragments, 29 (9%) are halved, 16 (5%) are quartered and 60 (19%) are spalled. Fewer forest than tundra chithos are complete, perhaps due to greater use because raw material was under snowcover or on the distant tundra. Similarly, edge fragments are more numerous in forest. Halved, quartered, spalled and almost complete chithos are similar in both ranges.
18 (5%) of chithos are unidentifiable to plan, 121 (38%) are ovoid, 110 (35%) round, 41 (13%) rectangular or square, 8 (2.5%) tearshape, 3 (1%) lanceolate, 4 (1%) bellshaped, 3 (3%) triangular & 4 (>1%) pentagonal. 51 (16%) & 167 (53%) of tundra and forest chithos are even and flat, the rest uneven or having wavy side profiles. Less than half of NP and MT and all of SA chithos are uneven or wavy. Edgeworn chithos number 252, unworn 59. Worn chithos number 51 in tundra and 201 in forest. For comparison, a chitho attribute key follows (with important traits and variables):
Material: 1=sandstone; 2=quartzite; 3=schist; 4=gneiss; 5=granite; 6=silicious shale.
Faciality: 0=unknown/undeterminable; 1=bifacial; & 2=unifacial.
Fullness: 0=unknown/undeterminable; 1=complete; 2=edge fragment; 3=central fragment; 4=halved; 5=quartered; 6=spalled; ca. full.
Plan: 0=unknown/undeterminable; 1=ovoid; 2=round; 3=rectangular or square; 4=tearshape; 5=triangular; 6=lanceolate; 7=bellshaped; & 8=pentagonal.
Section: 0=unknown/undeterminable; 1=even & flat; 2=uneven/wavy.
Edgewear: 0=unknown/undeterminable; 1=worn; 2=unworn.
Cortex: 0=unknown/undeterminable; 1=yes; 2=no.