Aboriginal Stone Tools

Australia: The Land Where Time Began

Aboriginal Stone Tools - Most stone tools observed being used were unrecognisable as tools - what are the implications?

In the book (Source 1) Hayden discusses the attitude of the Aboriginals of the Western Desert to the making and using of stone tools. This aspect of Aboriginal life in the Western Desert has also been studied by a number of other authors (Gould, 1969: 81-3; Gould et al., 1971: 163). Hayden observed a number of Western Desert men and women making and using stone tools as they remembered from their youth when such tools were being made and used in normal daily life.

When occupation sites are excavated the numbers and types of stone tools found are used to add to the knowledge of aspects of the lives of the people who made them, as well as to determine when the site was first occupied. A finding of Hayden and others who have observed the traditional making and use of stone tools is that many of the stone implements being used by the people would not be recognised as being a tool, most of the implements having not been altered for the intended purpose, simply being picked up from the ground because they were already of a type suitable for their use, such as a sharp edge, that would probably be blunted by any attempts at reworking, and they fitted the hand sufficiently well. With the exception of hafted adzes, the working of stone implements was rarely practiced, though in special cases they could make such tool when they needed to. One such rare case was known among the Warramunga and Walbiri, where knives with prismatic blades were made for trade.

Hayden found that there were no craftsmen specialising in making particular tools, everybody made tools as the needed them. Flakes were generally chosen from what was apparently a random group of flakes, the only criterion appearing to be their usefulness for the task at hand. A similar casual attitude to the selection of tools has been observed among the Nakako and Pitjantjatjara (Tindale, 1965: 146,160) and the Ngatatjara (Gould et al., 1971: 160) where 'cores' were often seen to be 'flaked' by simply throwing them on the ground or 'smashed' in some other manner, or flakes were 'randomly' detached by a hammer stone. The resulting flakes would then be examined to choose a suitable one, the process being repeated if necessary until a suitable one was found. A similar method of flake production was also observed in the Kimberley (Hardman 1888: 59) and Tasmania (H. Ling Roth, 1899: 151; Hambly, 1931: 91). The variability in craftsmanship among individuals making wooden bowls (Thompson, 1964: 407) is suggested by Hayden to indicate frequent modest control over the working medium.

Hayden has stated that the biggest surprise was the lack of or rarity of tools that would be recognised as implements in archaeological excavations of occupation sites. At the beginning of his observations he saw Aboriginal people using only primary flakes for shaving and wood scraping, and chopping wood with unmodified stone blocks, none of which would be recognised as tools. This lack of retouched tools led Hayden to wonder if the knowledge of the making of retouched tools had been lost following epi-culture contact phenomenon. But a similar lack of unretouched stone tools in traditional use had been observed earlier (Mountford, 1941: 316; 1948; Tindale, 1941; Gould, 1969: 81-3; Gould et al., 1971: 163). The result was the proposition that one of the most primitive technologies in the world was used by the Pitjantjatjara, as they did not use retouched tools, exclusively using natural forms, apart from the hafted adze. After observing for some time Hayden found that retouched tools were indeed being used, but rarely. He suggests this may have been because of the reason suggested by Bordes, that the stone flakes are sharpest immediately after being removed from the core, any retouching tending to make them blunter. A similar reason was implied by Basedow (1925: 365). According to all informants, instead of retouching a primary flake they sorted through the flakes for a more suitable one, removing more flakes if necessary. They would often try several primary flakes before finding one they were satisfied with.

A similar behaviour had been observed among groups around Lake Eyre (1924: 91) "Casual stones are any that have a sharp edge. They are used for scraping. Directly they are blunt they are thrown away and another picked up. Sometimes they are chipped if the stone will keep its edge long enough to warrant chipping, but usually they are not kept". These observations substantiate those of later authors among groups from the Western Desert. The effectiveness of working edges were not always easy to gauge by visual examination, as there were occasional small, subtle variations of the stone surfaces of the working edge, several being tried before one was selected.

Hayden found it difficult to determine the criteria used to decide if a flake could be retouched, believing the decision was based on a subjective evaluation. Some retouched flakes were discarded immediately because the retouched working edge was unsuitable, though there were times when a resharpened piece would continue to be used. When the tool was retouched it was to restore the sharpness of a dulled working edge. He could see no indication that there was an ideal shape, classic form or perfect specimen, that are often spoken of by archaeologists and collectors. Among the traditional people of the Western Desert the most important attributes of a tool was its working edge, that displayed a variable morphology, and whether or not it would fit comfortably in the hand so pressure could be applied. He suggests these attributes might become more patterned by habit and tradition, but admits there is no quantitative data to support his suggestion.

After viewing stone-using activities for a short time, another author has stated that the "Pitjantjatjara might vanish and no trace of them would be left behind" (Mountford, 1941,1948). A similar situation has been reported from New Guinea, where it was suggested that the edge in hand held tools is the important variable for the makers and users of stone tools. His observations have been supported by Strathern (1969). Another factor that has been suggested to be involved in the decision of whether or not to retouch a tool is the availability of raw material. Hayden says for the technological projects he asked the people in the Western Desert to work on there were always more raw material available. He suggests it could be expected that where raw material was a limiting factor retouching would probably occur. At most camps visited by Hayden there were many flakes and a number of cores, (also Basedow, 1925: 364; Thompson, 1964: 406). It is known that in traditional societies Aboriginal people carried primary flakes and blocks of raw material when they travelled (Thompson, 1964: 405, Plate 34; Basedow, 1925: 364; Hayden, in press (1998)).

Another factor involved in the decision to retouch appears to be the availability of the raw material. Hayden found that less than 20 % of the rocks used for chopping wood were modified, in areas where there were hard metamorphic rocks or igneous rocks were available that had naturally acute edges. When flint nodules or opal were used for chopping wood, that usually lack acute, sharp edges, they were often flaked to get a sharp working edge, 90 % of such opal or flint were modified by the removal of flakes to get a sharp working edge. Of the flakes he observed being used for 'scraping' or shaving wood, less than 25 % had been retouched or otherwise modified, the primary flakes being sharp with no modification. The implements he observed being modified did not always display 'scraper' retouch. He suggests it might possibly be the case that retouching was carried out only when no suitably edged rock was immediately available, and with primary flakes, only when none of the flakes were suitable without retouching. He also suggests there may have been reasons for retouching that he was not aware of, as there is not much difference in the effort required to retouch or search through the primary flakes present.

Nearly all adzes had been retouched, probably because much more effort and time is involved in unhafting and rehafting than retouching the working edge. Some adzes were removed with no retouching, but in these cases it was because they broke or the particular implement was made of a rock that was difficult to resharpen. With regard to the resharpening and retouching of stone tools there appears to be a dichotomy between hand-held and hafted tools, the hafted ones being retouched but the hand-held ones only rarely being retouched because of special circumstances, such as unavailability of raw material with sharp surfaces. In general, hand-held tools appear to correspond to Binford's class of non-curated artefacts, the hafted tools corresponding to his curated class. Hayden suggests caution is required with making a one-one equivalence, as the transport and curation of unhafted flakes and 'scrapers' in bags has been well documented (W. Roth, 1904:20; Horne & Aiston, 1924: 109; Spencer & Gillen, 1927: 26).

Hayden has estimated the number of retouched tools used by a nuclear family group living a traditional life in the Western Desert in a year. According to Hayden, though it is inherently risky to make such an estimate, it was calculated to give some idea of the the magnitude of the problem faced by archaeologists. The numbers are presented in Table 7a.1 in Source 1. The estimation is based on statements of Pintupi people and on his own estimates based on a family that were previously based a bit to the south of Lake Macdonald. They had a limited variety of wooden tools, and he believes many of the estimates may be on the generous side, in particular with the replacement of spear throwers, hardwood bowls of the women, and 25 spears on a yearly basis. He has calculated the number of tools that would be recognised archaeologically resulting from maintenance tasks, based on the actual retouched tools produced at Papunya settlement. If unmodified hand-held flakes are incorporated the numbers would be doubled, at least. He has assumed that all chopping tools are modified, though he says he has been liberal as in most cases this would be unrealistic.

He has calculated the number of tools produced per person per week at camps where informants were living 30 years previously, a time when they used only stone tools. They were using 2.5-10 flake tools per person per week (Hayden, in press 1998), totalling 130-520 tools per person per year. This agrees well with the estimates, and one of the sites was adjacent to a quarry where old adzes were probably discarded, and at other sites several spear throwers were manufactured by young men, which would have inflated the number of tools produced and used over a short period of time, so the estimates can be considered to be on the high side. The average production rate of retouched tools used by a husband and wife, their parents and non-producing children, about 150 per year (± 50) and about 40 (± 10) chopping tools being made, seems reasonable, assuming all chopping tools were modified. Many of these tools would be scattered at various places around the landscape, and the estimates could be reduced, in some cases considerably, by the lack of raw materials available.

many of these archaeological tools will be only slightly superficially modified, many of the choppers having a single flake removed, and others may be 'used up', making them closer to the 'classic' type as considered by some archaeologists (e.g Fig. 7a.2, Source 1). Archaeological tools from the Western Desert are generally not of predetermined form, being the mechanical results of needing to create or sharpen a cutting edge one or more times. The factors that determines when any given tool is discarded, whether unmodified, 1 flake removed, 1 resharpening or multiple resharpenings, are the nature of the material, the potential of the particular tools to be resharpened, availability of raw material, and the point at which a particular task is finished. Archaeological 'tools' are the results of attempts to rejuvenate and pass through several stages. Hayden believes it is worth restating that habit or tradition may be an important factor in determining which implements have the best prospects of being resharpened, and the most frequently resharpening mode used.

Functions

One of the observations of Hayden's technological projects was that almost all use of retouched stone tools was for woodworking. This is in agreement with observations of Pintupi life 'certainly the main use of stone here was for shaping and maintaining wooden tools and weapons' (Long, 1971: 269). Hayden suggests there is little doubt that all or nearly all retouched stone tools were used for woodworking, though there may have been rare occasions when retouched stone tool were used, even made, for purposes other than woodworking, such as cutting meat or skins. No retouched stone tools were made in any of Hayden's technological projects, that included plant food procurement and processing. A similar conclusion was reached from information provided by older members of the Pintupi. Skins were not used for any purpose among the people of the Western Desert with whom Hayden worked. He suggests it is possible that retouched stone tools may have been used for purposes associated with skins in other parts of the country where the people used skins. Study of other hunter-gatherer groups around the world has found a similar association of retouched stone tools with woodworking. In other parts of Australia there are 4 references to the use of chipped stone tools for the gathering and processing of plant food.

The chopping up of fern roots with a unifacial implement, referred to by Jackson (1939). It has since been claimed that this is almost certainly is unreliable (see Bancroft, 1894; W. Roth, 1901: 10).

It has been claimed that chopping implements might be used for digging holes for roots among peoples of the Western Desert. This observation is unique, the digging stick being the tool of choice for this purpose. Hayden suggests this was either a fortuitous occurrence or the detachment of roots, that is a woodworking function, with a chopping tool was confused with the excavation part of the process.

Tindale observed a crude hand chopper being used to cut the husks from pandanus fruit (Hale & Tindale, 1933: 114; 1934; 131). According to Hayden, this is ambiguous as it is not clear if the implement was made specifically for this purpose, or if any stone would have served the same purpose, the use of the chopper being a matter of convenience.

O'Connell has reported the former use of retouched blades to make spoons for eating tubers (O'Connell, 1977).

Hayden has said he has no knowledge from anywhere else in the world to indicate that chipped stone was being used for procuring or processing plant foods. Sickles were used by some cultures, but from the Palaeolithic that pebble tools were being used almost invariably for the chopping up of plant food (Deevey, 1668: 286). He suggests the generalisation that retouched hand-held tools were used for woodworking, and possibly in areas where the people used animal skins, for skin working, is a good place to start analysis. Microliths are excluded from consideration.

When comparing woodworking and the cutting of meat, he suggests that the cutting of meat requires a sharp blade, as is best found in unretouched flakes, while sharpness if not usually important in woodworking. In practice, when an animal is being cut up any waste flake is habitually used to gut and open the skin. Among the Pintupi and Ngatatjara these flakes were rarely retouched. According to Gould,

'These knives are discarded after only a few uses, and no effort is made to resharpen them. Thus they rarely show much in the way of secondary trimming and could be difficult for an archaeologist to recognise once the gum handle has decomposed.' (Gould et al., 1971: 156).

The making of several retouched 'knives' among the Pntupi has been illustrated and described (Tindale, 1965: 114-19). Several factors make consideration of the representativeness of these examples difficult.

The larger flakes were never observed being used, and no specific use for them was given. The smaller 'knives' were described as being used to inscribe lines on wooden implements.

The original article was written in 1933, the 1965 article being a reconstruction of the field notes and memory. Hayden considers it puzzling, from a functional viewpoint, that the "knives' were retouched before they were used. In the 1965 article it is implied that these knives often had one 'blunt, thick margin and a sharper, somewhat more arcuate one opposite' (Tindale, 1965: 141). These pieces may well be morphologically distinguishable, though they were retouched in a similar manner to scrapers and adzes that had not been used much. With the exception of instances where unmodified flakes were used for butchering, every major description of chipped stone being used in Australia, some form of woodworking has been mentioned much more frequently than any other activity. Evidence from Tasmania supports this conclusion.

Morphology

Choppers

The choppers for rough work were primarily unifacial, though there were also bifacial choppers. The modification of the choppers ranged from the removal of a single flake, multiple flakes removed around the periphery, to bifacial choppers, including large flake specimens. A fuller illustration of these tools can be found elsewhere (Hayden, in press (1998)). They differ little from the choppers found in abandoned camp sites throughout the Western Desert at the present (1998), and generally are of a similar size and weight to the Kartan heavy duty implements (Bauer, 1970). They were used to procure wood for all wooden implements, often being used to hollow out hardwood bowls, chopping out the spear-thrower interiors, thinning fighting spears, removing branches from the shafts of spears, starting nooks for spear barbs, and the initial spear point shaping. (Source 1, Fig. 7a.3, 7a.4). They appear to have been left at sites where the work was carried out, as also observed by Mountford (1941). They were often made of quartzites or other locally available rocks that are non-cryptocrystalline.

High quality cryptocrystalline rocks appear to have been carried around, when they were available, to be used as a source of raw material, as well as for chopping, possibly until the piece was exhausted (Thompson, 1964: 405; Hayden, in press (1998)). According to Hayden, while working with the 2 dialectical groups he worked with most, they had separate names for cryptocrystalline and non-cryptocrystalline rocks that were suitable for tool-making. The Pintupi called them kanti vs pilari, and the Yankuntjara called them kanti vs kaltjiliri. He reports witnessing on several occasions metamorphic rocks and quartzites, that were coarser grained, being preferred over fine-grained opal material. He noted that the finer grained rocks that he expected to be most sought after were not the rocks of choice, the coarser grained material being the preferred type. Some authors have regarded the Aboriginal people who had only metamorphic rocks to make their tools from as being worse off than those with cryptocrystalline rocks to work with (Stockton, 1972: 22), but the observations of Hayden and others suggest it is actually the other way around. Tools made from cryptocrystalline rocks are more aesthetically appealing, and their flaking properties are better, but they tend to shatter easier and become dulled more quickly when used for chopping hardwood. Hayden suggests the grain of metamorphic rocks may bite into the grain of the wood, being more effective in wood separation and detachment. Experiments have come to this conclusion (Crabtree & Davis, 1968: 428).

According to Hayden, there appeared to be a prohibition among the Yankuntjara against women using cryptocrystalline rocks. One of the women told him she had never used kanti (flint, chert, opal, etc.), but had used kaltjiliri, though kanti was regularly used by men for adze stones. In central Australia a similar prohibition has been reported (Spencer & Gillen,1912: 373, 376). A preference for pilari, non-cryptocrystalline rocks was also observed among Pintupi women, as well as for using choppers, as opposed to adzes, for all woodwork. When hardwood was being thinned down, men would use adzes but women would use choppers, the women were observed to be more clumsy at using an adze than the men. When making fighting sticks, digging sticks, sharpening the blades and smoothing the surfaces of bowls and fighting sticks, the women used grinding, telling Hayden they always used tjiwa (a small pounding slab made of sandstone) for these purposes. This appears to have probably been an alternative method of working with wood throughout Australia, including parts of Tasmania. The first report of its use was in the Western Desert (Basedow, 1925: 362), supported by observations (Finlayson, 1943: 79; Thomson, 1964; Horne & Aiston, 1924: 93). Murrray didn't see any men grinding their wood implements.

Grainy choppers of chipped stone were observed being used by men and women, though women tended to use only such tools. It appears women rarely used adzes, whether by proscription or preference, using grinding to finish many of the tools they use. These heavy duty stone chopping tools tended to have high edge angles, with an edge angle mode of 75^o.

An implication of this in the Western Desert is that women are not known to have made or used any type of chipped stone tool that is unique to women in the Western Desert. It would be difficult to detect the presence of women at an archaeological site, other than by the presence of small flat stones and hammerstones for pounding lizards and pitchuii leaves ash burned (a Nicotana species), made more difficult by the use of these implements by single males. Hayden suggests the presence of grinding stones is probably the best evidence of women, so the absence of grinding stones as an indication of the presence of women is negative evidence, so is unsatisfactory.

Flake Tools

Hayden notes he was surprised at the morphology of retouched hand-held flake tools. Instead of the scraper-type of retouching he expected for the smoothing of spear shafts, only about 50 % were of this type, the remainder being of an alternative type that achieved the same goal. An example is a single flake removed from the edge that created an archaeologist's 'notch' (the larger types, not the minute denticulations). The only difference between the use of the original flake and this is that there is a new edge to cut with that is sometimes more effective than the original edge (Source 1, Figs. 7a.5, 7a.6). Denticulates result from the repeating of such modifications. In one instance a flake was first resharpened with a notch, then flaked back into a scraper. There didn't appear to be a regular pattern to the decision to use either the notch or the scraper retouch, though he suggests this could be because of his limited sample, they appeared to occur in free variation. A very small burin was found in an ethnographic excavation, and an old Pintupi man, Ngayuwa, claimed he had made and used it about 30 years earlier, saying he had used it for shaving down his spears. Hayden noted that the tone of Ngayuwa's reply suggested he was questioning why anyone would use a hand-held flake for any other use. The side edge of the burin was used in this case. In Australia, true burins are occasionally found (Mulvaney, 1969). According to Hayden, Ngayuwa claimed he had thought of it himself, not being taught by anyone, and it was about the only one he had made. While the credibility of the story was difficult to assess, he was observed to pick out a flake with a broken edge with a cross-section like a burin-blow edge from flakes he had been knapping for finishing spears. After carefully examining the piece he put it aside, saying it was a good one. When he used it as a flake shaver it proved to be very effective. It was the only flake he had picked out. At Cundeelee and Papunya, many of the primary flakes used had edges close to right angles. These flakes were never retouched. Prior to his observations the author had believed hunter-gatherers used real 'tools' for woodworking, now he was seeing broken flake edges and accidental right-angled edges being used for the same purpose.

He found that edges that were slightly less than right angles, with a flake body that was strongly buttressed, proved to be excellent, very efficient shaving edges that are slow to dull. The same results have been shown experimentally (Crabtree & Davis, 1968: 46). The same burin cross-sectional characteristics have been found to occur in many flakes with right-angled breaks or edges, characteristics recognised by the Western Desert people for use in shaving spears. The author suggests the scraper, notch, denticulate and burin may all be stylistic variations of the same functional type. These implements can all be used for finishing wooden shaft implements, especially spears, throwing sticks, digging sticks, adze shafts, and in parts of the Western Desert, spear throwers. The author suggests the different type of retouch may have been used for different activities, such as notches being used to sharpen spears, saying there is no ethnographic evidence, the frequency of hand-held retouched tools being too low in his study to reach any firm conclusion.

He observed flakes being used in a sawing motion in the making of barbs on 1 type of Pintupi spear, the karimpa (Source 1, Fig. 7a.8). These saw flakes were frequently changed, and of 17 that were used only 3 were retouched, 2 with notchs and 1 with a scraper retouch. A notch on the end of a flake was mainly used used for severing cross grain wood fibres in the barb noocks, not actually being used in a sawing motion, the other never being used. It has been noted that small resin-backed 'knives', 3 cm long, with an edge that was slightly serrated, was used effectively as a saw for incising decorative lines (Tindale, 1965: 147). An unexpectedly high number of saws the author observed being used carried an abrupt edge opposite the working edge, that was often cortex covered, the same characteristic being implied by Tindale for his 'knives'.

Vertically oriented rocks

The author observed non-retouched artefacts, in the form of flattish rocks, that were relatively large, that has been set on their edge in the ground, one in an ethnographic excavation and another in his technological projects. In both cases the vertical slab acted as a fulcrum, a pressure point for straightening spears. The part of the spear to be straightened was heated in ashes, then the placed on the apex of the vertical slab, the hands about 1 m apart on either side and downward pressure was applied. In the procedure observed at the settlement, the flat slab was a cinder block, while at the excavated site it was iron-rich metamorphic rock, about 20 cm long, and in cross section was approximately a scalene triangle. In other parts of the Western Desert, wooden blocks or 'Y' uprights have been observed performing the same function (Ackerman, 1974). It has been reported that at Isimila some hand axes were found set on the sides vertically in the ground, the function of which was unknown (Howell, 1961: 121). The author suggested that, based on the above observations, there is a strong possibility they were used for straightening spears.

Wear patterns

Wear patterns on all tools were examined, as the main concern with the archaeological identification of functions of stone implements. The author expected the stone tools to be abraded most by the hard wood, but it was found that the stone tools used to work on the hardwoods traditionally used by the people of the Western Desert displayed little trace of any distinctive or diagnostic micro edge wear, while the stone tools used to carve shields and bowls from the very soft wood of the bean tree (Erythrina vespertilio) appeared to dissolve the edge of the stone tools very quickly, with a high frequency of gloss and striations occurring. See Hayden & Kamminga (1973) for a fuller presentation.

Sources & Further reading