Ethnology of Tokelau Islands
There is no stone on the Tokelau atolls suitable for fashioning adzes and tools. Shell adzes (fasua) were made from Tridacna, and black stone adzes were brought from Samoa. These stone adzes were rare and considered very valuable. On Atafu no shell adzes were found and only two stone adzes, one of basalt and one of tufa, whose origin was unknown.
The tufa adz (fig. 24, a) is the ground quadrilateral and tanged type found on the coral atolls of Rakahanga and Nassau, but it lacks the characteristic lugs on the poll. From the front, the tanged quadrilateral adz (toki talae) presents an even trapezoidal face tapering to the shoulder. The tang tapers more sharply toward the poll and page 153 projects backward from the plane of the front of the blade at an angle of approximately 30 degrees. The front of the blade is ground smooth but the tang surface is convex and rough. The back side (fig. 24, a, 3) is concave, due mainly to the angle at which the butt sets from the blade. The ends and sides are beveled out from this face. The bevel between chin and edge is broad. The poll extends at an obtuse angle, but the sides taper sharply, the right maintaining an even face for the entire length. The left side is asymmetrical with a gradual convexity which cuts away sharply above the shoulder. The left side of the tang is less steep and shows a flaw in making.
The basalt adz (toki koko) is relatively straight and narrow and is triangular in cross section (fig. 24, b). The width at the butt equals the thickness and increases very slightly at the chin of the bevel. The sides of the front surface (fig. 24, b, 1) meet in a median ridge. The extension of the ridge to the lower end has been removed by chipping and grinding. This work accidentally took away one corner of the lower end of the front face. Less than half of the wide ground surface at the lower end remains. Evidently this was originally a smooth triangular surface bounded below by the cutting edge and continued upward by grinding away the median ridge. The back (fig. 24, b, 3) has a slight convexity longitudinally and transversely. The surface is ground but shows shallow depressions from chipping above the bevel. The bevel curves gradually to the blade. The original cutting edge was convex and narrower than the width of the chin. The poll surface inclines upward. This adz is like the triangular Samoan adzes classified by Hiroa (28) as Type VI. The Tokelau adz with the median ridge of the front partially chipped and worn away is intermediate between the adzes illustrated by Hiroa (28, figs. 195, 196).
Tokelau craftsmen classified their adzes into three types, according to size and the work for which each was used. The largest adz (toki tingi) had a blade with flat sides and straight edge. It was hafted to a very long handle. This was used for the heavier chipping, the forming of timber into planks and the hollowing of a canoe hull after the inside of the log had been burned out. The second type (toki ualoa) had an angular handle, the two arms being of nearly equal length. The angle of the handle gave an almost horizontal line to the swing when the adz head was struck and prevented it from cutting too deeply. This was a safeguard against puncturing canoe hulls while dubbing out the inside. The small adz (toki atupa) had a very short handle that just fitted the hand. It was a trimming adz to smooth down the rougher work of the larger adzes. The blade had convex back and bevel.