R. V. S. WRIGHT
Department of Anthropology,
University of Sydney
The flaking of stone is virtually a lost art. It is no longer part of our domestic and industrial material culture. Whereas a student of prehistoric pottery may transfer everyday experience to the study of prehistoric sherds, the student of stone artefacts is a novice with no knowledge of the various processes that produce finished implements from lumps of stone.
Not everyone interested in stone artefacts bothers to acquire a knowledge of the processes that produce finished implements. Many collectors interest themselves in silhouetted shapes. Consequently, the study of stone artefacts is prey to the wishful thinking of the uninstructed, who points wonderingly at a curious shape in stone, a shape that on proper analysis can be explained as a lusus naturae (a whim of nature). Every Australian archaeologist will be invited to look at innumerable whims of nature, and it is almost invariably stone artefacts that these whims of nature simulate. Of course the more spectacular stone implements, such as Kimberley points, are self-evidently the work of humans and can thus be identified as artefacts without the aid of the type of formal analysis presented in this chapter. But prehistorians must go beyond intuition and wrong information from pieces of stone that are not self- evidently artefacts.
Of course, most of these artefacts would have vanished - ground to sand in beaches, swept from long vanished land surfaces into alluvium, or drowned on the continental shelf by rising sea levels of the late Pleistocene. Nevertheless, these crude calculations show why there are still so many stone artefacts in Australia.
Contrasting with isotropic rocks is shale, a rock with a pronounced plane of cleavage.
Sometimes the artisan does not flake a sharp edge but seeks to achieve an exceptionally even edge by grinding. A rock that lends itself to sharpening by grinding is greenstone. Greenstone is not much used for flaking, being low in silica and somewhat soft. However, lumps of greenstone are often trimmed by flaking into the rough shape of an axe, before the finished product is ground out.
It should be noted that flaked stone artefacts from sites that are independently known to be archaeological are almost invariably made from silica rocks with the properties outlined above. Any 'interesting' stone, supposedly flaked, should be examined to see if it is made from a silicious rock, homogeneous and with absence of cleavage. If it is not, it should be treated with suspicion.
Silicious rocks are hard, a property that appealed to the artisan. Hardness, of course, resists abrasion and so it is a property which aids the survival of prehistoric artefacts in sediments and soils. An additional property of silicious rocks that has aided their survival is their resistance to chemical corrosion. Note how many corrosive chemicals, both acid and alkali, may be stored in glass. The resistance of stone artefacts to physical and chemical destruction is why they are often the only evidence of humans to have survived from remote prehistoric periods. The fieldworker may therefore expect to find artefacts looking as fresh as the day they were made and dating from the remotest periods of the human settlement of Australia.
The methods of excavating stone artefacts do not differ from the methods used for the excavation of 'small finds' in general. Surface collecting, however, does offer special problems, chiefly because of the sheer numbers of stone artefacts that often lie around awaiting collection. Collect only when there is a specific reason for collecting. Do not collect in the vague hope that someone, some day, might find your collection of scientific value. Stone artefacts are antiquities and therefore are subject to the laws of each state and territory. Check what these laws are before collecting.
Above all, take advantage of ploughed paddocks; after rain, artefacts may be found washed clean, perched on little pedestals of soil that have been protected from erosion by the cover of the artefact. If all else fails, look at the stones carried up by ants and dumped on the mounds above their nests. The essential principle, in looking for surface artefacts, is to look for natural disturbances of the soil.
Do not be too anxious to clean off all soil or adhering concretions. Important questions may arise about the exact stratigraphic provenance of a particular artefact. If this happens, there is nothing like adhering matrix to help settle questions.
As well as collecting artefacts the fieldworker may wish to find the sources (perhaps even quarries) from which the raw material for making the artefacts came. A geological map of the surroundings may help in locating silicious rocks in situ. Remember, however, that many artefacts are made from river pebbles, and these may have been transported kilometres from their source as part of the river's bed load. So the archaeologist who wishes to locate sources is faced with a task that does not coincide with a geologist's usual interests. It is therefore important to try and locate sources while in the field and not rely on later geological advice.
It may be both desirable and possible to collect all the stone artefacts exposed on a surface site. But every fieldworker will some day have to face the methodological problems of sampling a site for its stone implements. Sampling is often forced on a fieldworker by the sheer numbers of artefacts present.
Archaeologists may want to use samples of stone artefacts to discover patterns of implement distribution within sites, examine size differences of artefacts between sites, compare proportions of implement types at different sites, and compare the uses of different raw materials at different sites. None of these comparative analyses has validity unless a proper method of sampling is used. No fieldworker can collect a representative and unbiased sample unless special methods of collecting are resorted to.
In brief, if your project requires the collection of a sample of stone artefacts from a surface scatter do not collect any artefacts until you have worked out a sampling strategy.
A true random sample requires that each artefact in the scatter has an equal and independent chance of being incorporated into the sample collected. This means that not only must each artefact have an equal chance of being collected, but in addition the collection of any particular artefact in the scatter must not influence the collection of any other artefact.
To be rigorous, the only reasonable method of drawing a true random sample from a scatter of artefacts is to assign to each artefact in the scatter a unique number and then draw the sample by drawing a set of such numbers at random. The artefacts that are then collected are those whose numbers were drawn. For the purpose of making up the sample random numbers can be taken from tables or generated on a programmable calculator.
In practice it may be most inconvenient, or even impossible, to assign numbers to artefacts as they lie in a disordered scatter on the ground. An acceptable, but less rigorous method, is to make the unit of collection not the individual artefacts but the quadrats or transects within which they lie.
The method of quadrats does not yield a random sample of the scatter. Most importantly, the principle of independence is violated. So it is no substitute for a true random sample. The method of quadrats will, however, allow examination of questions of technological homogeneity within sites.
The sample to be collected will, therefore, have to be censored by setting a size limit below which pieces of flaked stone will not be collected. A reasonable limit is to exclude all pieces that do not achieve 5 mm in at least one dimension. The censoring should be recorded and reported to anyone who may wish to study the collection.
It should, furthermore, be obvious that objects of special interest should be salvaged only after the more formal samples have been collected. If they are collected first they are precluded from appearing in the sample.
The key to understanding flaked stone technology is to understand the flake and the core. Fortunately, the flake is easy to understand and has attributes that are easily described. Yet the flake's technical simplicity often masks what a portentous invention it was. There is nothing sharper and harder in nature, and once you master the simple technique of making flakes you can easily produce them at a rate of one a second.
Simple though a flake is, many misunderstandings arise from not knowing how to interpret its attributes, and those of the core from which it is detached. If you can understand the attributes of flakes and cores you will avoid confusing them with natural rocks or with finished implements.
There are two ways of learning the attributes of flakes and cores.
Flaking rock yourself is the best way of learning. After such self- instruction you are most unlikely to forget the properties of flakes and cores, and in particular you will learn to diagnose those special features that set finished implements apart from the waste products of flaking. Those who have never done any flaking continue to run the risk of being distracted by 'interesting' shapes that the practiced stone flaker knows to be produced fortuitously.
If you are going to flake stone yourself remember that lumps of industrial slag or glass may be used but that you will produce dangerously sharp splinters. The more granular products of flint or silcrete are safer (the flint from the southern coastline of Australia is ideal material). Whatever material you use, do wear leather gardening gloves to protect the fingers and goggles to protect the eyes. Remember that there is nothing sharper than a primary flake. And do not have anyone sitting near you; a flake does not fall off a struck core, it shoots off.
(1) The potential core. Fig. 1 represents a potential core from which flakes have not yet been struck. Note that the angle between the top surface (striking platform) of this potential core and its sides is less than 90'; it is very difficult to detach flakes from cores where the angle between the striking platform and the sides is greater than 90'. The arrow represents force, in this case force applied by a hammerstone. The hammerstone is being directed at a point just in from the actual edge of the striking platform. The core rests on an anvil. Both the hammerstone and the anvil will become pitted with use.
Fig. 1
(2)The flake at detachment. Fig. 2 shows the flake at the moment of detachment from its core, it is at this stage an unmodified or primary flake. On the top of the flake is the striking platform, carried away from the striking platform of the core. In addition to carrying away part of the platform, the flake has carried away part of the side of the core and this surface of the flake is called the dorsal surface. The other surface of the flake, that was 'inside' the core, is called the bulbar surface (sometimes the ventral surface).
Fig. 2
(3)The bulbar surface. Fig. 3 illustrates the bulbar surface, which though smooth shows several features that are characteristic of a conchoidal fracture, the sort of fracture induced by a hammerstone working on silicious rock. The point of impact of the striking platform. From this point the shock waves, that detached the flake, radiated out through the core and left characteristic features on the bulbar surface, namely;
Fig. 3
(c) Over the bulbar surface are found waves or ripples which represent the dispersal of energy from the blow that detached the flake. If the waves were complete circles they would be concentric around the point of impact; on the flake they are only arcs of circles, the concavities of which point to the place of impact.
d) Fissures radiate out from the point of impact and are especially noticeable near the margins of the flake. These fissures represent minute incipient splinters, air having got behind them and changed the refraction of light reflected off the surface of the flake.
(e) It is the margin of the flake that is the sharp part. Because the specimen we are considering is an unmodified primary flake, it has no identifiable working edge on its margin. But this is not to say that part of its margin could not be used immediately as an edge for cutting-there is nothing sharper in flaked stone technology than the unmodified margin of a flake.
Note that if we were presented with just the broken end of a flake we would be able to make an estimate of the size and shape of the flake from which it was broken off; we would do this by using the waves and fissures as orientating guides.
(4) One possible state of the dorsal surface of a flake is shown in Fig. 4. In this example the right half of the dorsal surface is a flake scar. The flake has carried away with it, from the core, the scar of a flake that was detached previously. Looking to the left half of the dorsal surface we see the original 'skin' of the rock that was turned into the core; this skin is called thecortex.
(b) part cortex and part flake scar/scars (as in Fig. 4);
(c) all flake scars.
Fig. 4
The condition where a flake has all flake scars on its dorsal surface may be a valuable indicator of a flake being an artefact. Condition (c) is the result of repeated flaking from the same area of a core, and is characteristic of human flaking. On the other hand, haphazard natural processes of flaking (rock falling off cliffs, rolling off torrents) characteristically produce flakes with cortex.
(5) The core in Fig. 5 illustrates the points made in section 4 above. It can readily be seen that any further flake detached from this core could assume conditions (a), (b) or (c), depending on whether it was removed from the left, middle or right side of the core illustrated.
Fig. 5
Figs 6-11 illustrate the dorsal surface of primary flakes and show
attributes that can often be troublesome to interpret.
(6) Fig. 6 shows an untroublesome flake, with the remains of three primary flake scars.
(7) Fig. 7 is superficially like Fig. 6 but shows, in addition, a facetted striking platform. Cores may not have plain surfaces as their striking platforms.
| Fig. 6 | Fig. 7 |
 |  |
(8) At the junction of the striking platform and dorsal surface of Fig. 8 is to be seen some crushing. The crushing extends onto the dorsal surface. Such crushing can be produced by an earlier, misdirected, and unsuccessful attempt to detach the flake. The crushing should not be regarded as intentional and secondary modification of the flake.
| Fig. 8 | Fig. 9 |
 |  |
(9) Fig. 9 is an elaboration of Fig. 8, and shows where a mishit with the hammerstone caused crushing, then a small flake was detached and finally the flake itself was successfully detached. Again, these features are primary.
(10) In Fig. 10 some preliminary flaking was done at right angles to the direction of the blow that detached the flake.
| Fig. 10 | Fig. 11 |
 |  |
(11) The flake in Fig. 11 has carried away from the core part of an earlier striking platform.
The comments made about Figs 8 - 11 indicate that the flaking should be interpreted as primary. It is, however, possible to interpret such flaking as secondary, that is as flaking done after the flake was detached from the core. Yet, as a general rule, and unless there are compelling reasons for interpreting it as secondary, it is safer to interpret such flaking as primary. A compelling indication to the contrary would be if the flaking continued round to the primary margins of the flake.
Figs 12 - 15 illustrate the important process of modification to the primary margins of flakes.
| Fig. 12 | Fig. 13 |
 |  |
Modification may be deliberate (e.g., shaping an edge by secondary flaking) or accidental (e.g., use-wear, where an edge is fractured by a chopping action). Whether or not the modification is deliberate or accidental, it is possible to identify it unambiguously as secondary modification where, as in Figs 12 - 15, it affects the primary margin of a flake. In other words, we can be very sure that we are dealing with secondary modification because we can define secondary modification as modification of the primary margin of a flake.
(12) In Fig. 12 secondary flaking has modified the primary margin of the flake into a smooth edge. The edge is formed by a series of miniature flake scars. Note that the flake has been flaked as though it were a core; however the word 'core' is best reserved for a flaked object that served as a source for flakes. In the case of Fig. 12 the flakes produced can best be interpreted as incidental to the shaping of the edge.
(13) In Fig. 13 a concave edge has been produced by secondary flaking on the bulbar surface. Secondary flaking is found normally on the dorsal surfaces of flakes, in other words it is carried out by means of force exerted onto the bulbar surface. If secondary flaking is found either on the dorsal surface or on the bulbar surface, but not on both, it is said to be unifacial. If it is found on both surfaces it is said to be bifacial.
(14) In Fig. 14 is represented use-wear, with abrasions perpendicular to the edge and produced by a chopping action.
| Fig. 14 | Fig. 15 |
 |  |
(15) The abrasions in Fig. 15 are parallel with the edge and produced by a sawing action.
(16) Pieces of stone (that are not flakes) can be flaked implements and may also show flaked and abraded use-wear. Fig. 16 is a piece of stone with bifacial flaking, designed to produce a rough chopping edge. Such implements are often called core tools. It may be difficult to decide whether such a flaked piece of stone has been shaped for use (a core tool) or taken its shape merely because it was a source of flakes (a core proper). Indeed it may have served the purpose of a core and then been modified into a core tool.
Fig. 16
By using the principles established in the previous section the fieldworker can sort any stone industry into simple technological categories. A systematic approach to sorting is valuable because at least it ensures that all pieces are examined. Fig. 17, therefore, offers one of many possible initial schemes of sorting.
Fig. 17 is an analytical flow chart that aids initial sorting in the field. Its use will ensure that the fieldworker has a good idea of what sort of stone industry is presenting itself. In using the chart, each piece of stone is allowed to flow through the chart by decision of the fieldworker. For example, Fig. 12 would be described as a piece of fractured stone - a flake - a modified flake - with secondary flaking applied for the purpose of shaping. Fig. 16 is a piece of fractured stone - a flaked piece - a core tool - (and nothing can be said about use-wear).
(1) Relatively objective categories are enclosed in a box. 'Objective' means that different workers will come to the same conclusions about the categories to be assigned to the same pieces of stone. The categories that are not boxed require a higher level of interpretation, and agreements between workers will be less frequent. It has, for instance, proved difficult to get agreement in distinguishing core and core tools.
(2) Flaked pieces must show negative flake scars in the form of negative bulbs of percussion. It is not enough that a piece shows vague facets; these can be caused by such natural processes as the stress of rapid temperature change and have nothing to do with human flaking.
(3) Flakes must show bulbs of percussion. It is not enough that a piece be thin, sharp and perhaps the broken end of a flake.
(4) Waste will include pieces of broken core and flake that are not complete enough to meet the strict criteria applied to the previous two categories.
(5) Damage after discard includes such damage as being trampled on by cattle and being abraded in streams.
It may help to systematically write the stone's biography by using
clues observable on its surface. The aim of the biography is to
narrate the sequence of events that happened to the artefact starting
at the stage of raw material, continuing through the process of
flaking and ending with the modifications that happened after the
artefact was discarded.
The discussion has assumed that one is dealing with an artefact. Yet
the exercise of writing a biography may prove especially useful
when facing a stone object that could be a lusus naturae.
Indeed, unless one can provide a stone object with a coherent
narrative of events, to make up a biography, its identification as an
artefact may be wishful thinking.
| The biography | Clues used |
| A flake | bulb of percussion present |
| struck from a pebble | pebble cortex survives on dorsal surface |
| the pebble having been split first | platform of flake is not cortex |
| not the first flake from the core | dorsal surface of flake has both cortex and flake scars present |
| the artisan had difficulty detaching the flake | much battering on dorsal surface, at junction with platform |
| secondarily flaked to give an evenly curved convex edge | primary margin of flake altered by unifacial flaking on dorsal surface |
| completed artefact lay for an extended period in soil | chemical alteration of surface (patination) and staining of patination by iron salts |
| for a period was moved in soil, perhaps during soil creep | abrasion of all facets |
| the end was broken off recently | snap fracture, showing in the broken cross section a kernel of unpatinated and unstained original rock; the broken edges of this fracture are not abraded |
| artefact was placed in a bag with other artefacts and the bag crushed | irregular, bifacial flaking of both the abraded patinated and the unabraded unpatinated surfaces |
In Figs 18-26 are illustrated some finished implements and other artefacts. The photographs are stereo pairs and illustrate some of the technical points made in this paper.
| Fig. 18 A horsehoof, with intensive crushing around the striking platform | Fig. 20 A miniature core, flaked bifacially from all four edges |
 |  |
| Fig. 19 A pebble flaked all over one surface. The under surface is entirely cortex. | Fig. 21 The base of a small core |
 |  |
| Fig. 22 A large flake. Secondary flaking, on the dorsal surface, has produced a horseshoe shape | Fig. 24 A flake
trimmed to a point by secondary flaking |
 |  |
| Fig. 23 A small adze flake of the type mounted on the end of stick to work wood. Secondary flaking has obscured almost all the primary features of the flake | Fig. 25 Flake with blunted backs, the opposite edge sharp edges |
 |  |
Fig. 26 Flakes with blunted backs, the opposite edges sharp |
 |
McCarthy, F. D. (1967), Australian Aboriginal stone implements, Australian Museum, Sydney. Comprehensive in illustration and description, technical, but somewhat idiosyncratic in terminology.
Mulvaney, D. J. (1975), The Prehistory of Australia, Penguin Books, Hamondsworth. Illustrates important types of stone implements, in the context of well-rounded ethnography and prehistory.
Wright, R. V. S. (ed.) (1977), Stone Tools as cultural markers, Australian Institute of Aboriginal Studies, Canberra. Covers the uses to which stone artefacts may be put by archaeologists but also has much on manufacture, use-wear, etc.
