Stone Artefacts: Procurement and Manufacture
Stone is regarded as the most visible surviving trace of prehistoric societies or human activity. This is due to the relative indestructibility of lithics compared to organic material in the archaeological record.
Organic material, unless preserved in anaerobic (oxygen reduced) conditions deteriorates rapidly leaving no trace for the archaeologist. The oldest surviving wooden artefacts in Australia come from Wyrie swamp in south-east Australia and are about 10 000 years old.
While there is one site with wooden implements from the terminal Pleistocene, there are billions of stone artefacts scattered around Australia, stretching back to the period of initial colonisation. Undoubtedly, stone artefacts are important, but they only represent a fraction of cultural material and wooden implements were heavily utilised.
We know this because many of the stone tools found are woodworking tools, whether they are hatchets, adzes or wood-shaving implements. The abundance of stone artefacts, can present a lop-sided view of an ancient culture, especially when stone artefacts abound. This makes stone appear to be the only form of material culture. However, stone is a permeable record and much can be gleaned from the study of lithics. For instance, Peter Hiscock in a paper titled, Technological Responses To Risk in Holocene Australia, considers the development of new stone technologies as a response to environmental changes.
In Australian archaeology, as in other parts of the world, stone artefacts also contribute to developing a broad chronology for occupation, simply because they span the total period of occupation. In an excavation, stone artefacts are often found in association with fireplaces. Fragments of charcoal in hearths associated with stone artefacts are dated, and this provides a radiocarbon date for each stratified layer. By using the Law of Superposition, a premise stating that the bottom layer is the oldest, with each succeeding layer above being younger, a site chronology can be developed. This in turn is combined with information from other sites and a general chronology at local, regional, and continent levels can be created.
Australian Archaeology Raw Materials
The function of a stone tool often decides what raw material is to be used, although if a particular material is not available, a second choice may be necessary. Stone implements form two broad groups, those that are used for cutting or scraping (flaked tools) and those that are used to grind or pound.
Flaked Tools: When making a flaked tool, the raw material or core stone must have certain inherent properties, otherwise the flakes will not flake off properly, or the flakes may be malformed and useless. The first criterion is the hardness of the raw material and the second is how isotropic the material is. Essentially this means the piece (core) selected to make the flake should be hard and brittle to provide a sharp working edge, and homogenous enough not to fracture into pieces when struck by the hammer stone (Figure 1).
The most common raw materials in Australian archaeology suitable for flaked stone tools are quartz, quartzite, silcrete, chert, chalcedony and mudstone. Quartz and silcrete are abundant throughout the vast interior of Australia, and not surprisingly many flaked artefacts are made from these two materials. Quartz is milky white in colour, made up of many hexagonal crystals, and when struck tends to fracture along internal faults, breaking into many pieces.
While this provides sharp shards useful for cutting, it is not a homogenous material, and subsequently it is difficult to predict what the outcome will be. Silcrete is generally grey-brown, fairly coarse-grained, speckled throughout with quartz grains, and quite homogenous. This means a predictable outcome is possible, and a proficient knapper can control the design of the flake.
Chert and chalcedony are present in a variety of colours, both homogenous, and fine grained in texture. A proficient knapper can control the design of the flake, resulting in a functional hardwearing tool. On a regional scale, and found locally are other siliceous raw materials, such as obsidian, tektites and Darwin glass.
Obsidian is found in parts of northern Australia, and was traded over long distances. Flakes from obsidian were highly valued and renowned for an extremely fine cutting edge. Tektite nodules originate from meteorite showers and are generally only found in small quantities. Small fragments of Darwin glass are scattered over a localised area in southern Tasmania, originating from the impact of a single meteorite.
In manufacturing a flake, a knapper must decide which raw material to use. Often this dictated by the proximity, availability and suitability of the material to the task. Most raw materials are sourced from outcrops occurring in hills, ranges, or from large pebbles along river courses.
Silcrete is abundant throughout the arid region, being the remnants of Tertiary duricrust, now eroded into boulders of various sizes. The silcrete boulder pictured below (Figure 2) is from the Merton Escarpment in western Queensland. The boulder has been broken apart, creating smaller manageable pieces, and a suitable piece would be selected to serve as a core. A discarded flake is shown in the inset, possibly a result of knapper sampling.
Most grinding stones in Australian archaeology are slabs of sandstone (Figure 3), often rounded at the edges for ease of handling. Seeds were placed on the grindstone and another stone (muller) was used to grind seeds into flour. Sandstone is granular, relatively soft, containing quartz grains that provide sharp edges to shred the seeds. The surface of the grindstone is continually being renewed, as grains are ground off the slab, exposing more sharp edges. The resultant flour would have grains of quartz mixed in, contributing to a high rate of dental wear. The muller stone is generally a hard granite-like material, and after years of use a muller develops a flat working surface and becomes polished. You can see how flat and polished a stone becomes in the broken muller below (Figure 4.).
The grindstone picture above (Figure 3) was utilised in seed grinding, however other uses for grindstones have been observed in ethnoarchaeology. The uses include; nut cracking, pounding meat, grinding ochre for pigment, and grinding plants for food. Some polished grindstone fragments were found at Cuddie Springs (lesson 4 Map 1) suggestive of seed grinding, and are and tentatively dated to 30 000 BP.
However, evidence for widespread use of seed grinding in Australian archaeology only dates to the Holocene. At Malakunanja II an ochre-stained grindstone fragment, dating to 18 000 BP was excavated, confirming Pleistocene use of grindstones for ochre preparation. Ochre was ground to provide pigments for rock art and body decorations.
Some of the earliest dated stone artefacts in Australia are edge ground hatchet heads. Whilst some people refer to these artefacts as axes , the size and weight means they were used with one hand like a hatchet, and not two hands like contemporary axes. Figure 5.1 shows a waisted edge ground hatchet head, excavated from Nawamoyn rockshelter dating back 20 thousand years. The similarity is striking to a hatchet head from the 1900s made in southern Australia (Figure 5.2)
Other hatchet heads older than 40 000 BP have been excavated from the Huon of New Guinea, and ca 32 00 BP in Queensland. Hatchet heads appear early in the archaeological record throughout northern Australia, however they are not found in southern Australia until the mid-late Holocene. The waisted part of the hatchet head is the ground groove (Fig 5.1) enabling it to be hafted onto a handle as seen in the AD 1900s example. The cutting edges were shaped and ground on large blocks of sandstone like the one below. (Figure 6).
Before we begin on the techniques and mechanics of flake production, a small side note on where stone flakes are found. Many flakes (billions Australia-wide) are found on the surface, called background lithic scatters. These can be isolated flakes discarded after usage or simply lost while travelling. Often larger scatters indicate places where people camped, and these lithic scatters are around waterholes, along rivers or other obvious watering points (Fig 7.1).
In other places knapping floors (places where people knapped flakes) occur in what appears to be unlikely surrounds. For instance Figure 7.2 shows an exposed campsite perhaps thousands of years old. Whilst it initially appears an unlikely camp area, it is within a few kilometres of an extensive quarry where high quality raw materials were quarried to make stone artefacts. The small flags dotted through Figure 6.2 denote the position of surveyed flakes.
Flakes found in stratigraphic sequences within a rockshelter (Fig 7.3) provide a better opportunity to securely date a site. This is because a rockshelter offers less chance of disturbance than does an open-air site, permitting artefacts to remain in situ while the sands of time slowly cover the deposit.
Naming the Parts
Whilst working in Australia s arid region, and conducting archaeology ecotours, one of the first things people asked was, how do you know this is a flake? Firstly, in order to identify a knapped flake or a core stone, you must know what to look for and be able to isolate specific features. Therefore, these features need a name, so let s start with the illustration, Figure 8.
1. The point where the flake is struck with the hammer stone is called, the point of force application or pfa.
2. The top of part of the flake where it is struck off the core is the proximal end. (The bottom of the flake, where it terminates is the distal end.
3. As mechanical force radiates down from the pfa, it creates a very distinctive bulb of percussion. Look at the bulb of percussion on the proximal end of the flake in Figure 9. A flake with a bulb of percussion is called a conchoidal flake.
4. The ringcrack is the tiny little protruding circle, just above the bulb.
5. Undulations are often seen on the ventral surface (surface inside the core prior to flaking) and are formed as mechanical energy travels down, striking irregularities in the core causing the downward force to deviate slightly.
6. Looking at the Core in illustration 8, you will notice the reverse shape of the flake. The overall reverse shape of the flake in the core is referred to as a negative scar.
7. Also notice the negative bulb of percussion.
In Figure 10, the rather large core stone still has some cortex attached, or outside rind on the core. In this case the core has been discarded as the core is considered to be exhausted and the knapper has decided no more flakes could be struck off.
The core was found within a few kilometres of a chert outcrop, and no doubt a better piece could easily be acquired. However, in other areas where this high quality raw material is not so readily available, heavily reduced cores are more likely to be found. Studies have been conducted to show a correlation between core sizes and distance to raw material.
Many core stones are one-tenth this size, presenting no cortex, with every possible flake removed. Sometimes a core is found along with flakes and other debitage, and this constitutes a knapping floor . It is possible to refit the flakes, a process called conjoining which gives insight into the reduction sequence. By studying a reduction sequence archaeologists can retrace the steps of an ancient knapper, understanding something about their approach to stone tool manufacture.
The method of flake manufacture referred to so far is primarily one of freehand knapping. Freehand knapping is the earliest known method of flaking, a practice dating back 2 million years ago to the Oldowan Industry of Africa. In Australia and other parts of the world many more methods developed over time, such as; indirect percussion, bipolar, pressure flaking (see Kimberly Point Figure 11), chimbling and anvil techniques. Here is brief outline of each method as described by Professor Iain Davidson (2001).
Bipolarreduction: If a core becomes so small that contact with the hammerstone results inthe core moving rather than detachment of a flake, its reduction can becontinued by bipolar reduction. Bipolar is also a useful technique forthe knapping of rounded pebbles which were either too small, or did notpresent angles suitable for freehand percussion.. In bipolar reductionthe core is placed on an anvil, generally with its longest axis perpendicular to the anvil. The core is then struck at an angle of 90degrees. The Term bipolar is derived from the fact that each corereduced by this method will have two polarised (directly opposite)striking platforms or zones of percussion (Binford 1972b:356) .One ofthese zones is the point where the hammerstone comes into contact withthe core, the other is formed by the core coming into contact with theanvil.
Indirect percussion: Indirect percussion involved the use of an intermediate percussor, such as a piece of stone with a rounded point, bone, wood or antler, between the hammer stone and the core. Indirect percussion is often termed thepunch technique. The flakes removed by indirect percussion frequentlyhave very small platforms.
Pressure flaking: When the knapper requires precise control over the location of flakeremoval, flake size and shape, this was achieved by pressure flaking.In the pressure flaking technique a pointed instrument of stone,copper, bone, wood, or antler was used to exert pressure to push theflakes from the implement being produced. In Australia, pressureflaking was still being undertaken by Aboriginal groups this century.
The Rolling Pressure Technique: In the rolling pressure technique the flake is held in the palm of thehand with ventral surface uppermost and the edge to be worked restingon the pad at the base of the thumb. A rounded, generally smooth flatpebble is laid on the bulbar surface, rolled towards the margin andpressure exerted. This removed very small, semi-circular, equal sizedflakes.
The Chimbling Pressure Technique: Chimbling was first described by Dickson (1973:12-13) as a method usedto retouch the edge of blades or flakes to produce Bondi Points. In chimbling the flake is not held in the hand but placed on a piece ofbark or wood, which will give a little under pressure. This allows thefull force of the arm to be used to remove the flake rather than justthe fingers, as in the rolling technique discussed above.
Anvil Technique: In the anvil technique the core is brought down directly on the anvilwith enough force to initiate fracture. The result of this action willdepend on the type and size of the core and the amount of force withwhich it contacts the anvil.
Largerpieces of raw material can be reduced bythrowing them down on the anvil. This is quite a useful technique for producing smaller cores which may then be reduced by one of the standard techniques discussed above.
Stone tool manufacture is a large field and we really have only just touched on the subject. Many artefacts are only found in specific regions within certain time frames (e.g. tula adze in the arid region), sequence and distribution.
Davidson, I. 2001, Archaeology of Stone Artefacts: Background notes, Archaeology and Palaeoanthropology, School of Human andEnvironmental Studies, University of New England, Armidale, Australia.
Flood, J. 1997, Rock Art of the Dreamtime, Angus & Robertson, Australia.
Lourandos, H. 1997, Continent of Hunter Gatherers, Cambridge Uni Press, United Kingdom.
Morwood, M. 2002, Visions from the Past: The Archaeology of Australian Aboriginal Art, Allen & Unwin, Australia.
Mulvaney, J. & Kamminga J. 1999, Prehistory of Australia, Alllen and Unwin, Australia.