/Archaeological data includes those items which were manufactured by the members of a past society, those items which they collected and used, any alterations in the natural landscape resulting from man's activities, the bodies of the former inhabitants, and the relationships between any of these factors that are the result of human activity. We will discuss each of these in turn. We may further divide archaeological data into a number of categories based on other considerations. For example, stone tools are almost indestructable and therefore are commonly found, while perishable items are found only under the most favorable circumstances of preservation. Also of concern is whether the particular prehistoric society had cities, towns, isolated farmsteads, or campsites and whether they practiced irrigation, built temples or pyramids, or occupied caves. Because of these inherent differences in archaeological data, it is tempting to lose sight of the larger goals of culture history and
become interested in the nature of specific items. What we must always bear in mind is that the archaeological data we find are evidences of the past cultural practices of a particular society. No matter how simple or how complex these remains are, they can and should be utilized as a means of inferring past patterns of human behavior. This point of view, probably more than any other characteristic, serves to distinguish between the professional archaeologist and the amateur or antiquarian; the former is interested in antiquities as sources of information, while the latter is more interested in the objects themselves.
What are the typical objects, and what do they reveal about the past? Stone tools are the most common items, exceeded in frequency by pottery after it was invented. Probably next in frequency are bone tools, with wooden tools, cordage, basketry, gourds, and other perishable items less common. In most sites, even more common than the tools are bones, shells, plant parts, and other food remains. Perhaps next in importance are structural remains; these vary from the least permanent—slight indentations in the ground sometimes with an associated firepit or post holes—to multistory buildings of cut stone. The grouping of individual structures or the clustering of contiguous rooms can be segregated into types, and these types then form the basis of inferences as to family, band, or tribal size and social organization.
In addition to the objects themselves, there is another order of data which we may obtain by precisely recording the association of the objects one to another. Associational data of this type provides a means whereby we may infer that several separate objects which do not fit together, at least in form, were utilized together in a particular cultural practice. A modern example would be the finding together of a tin can and a can opener. Their occurrence together strengthens our belief that they were used together.
There are other situations in which archaeological inferences may be made. For example, large stones occurring in fine grained lacustrine sediments which obviously could not have been placed there by water action or other geological processes may be the result of human activity, even though the stones themselves show no modification, evidence of use, or pattern. Obviously, human footprints or handprints constitute another category of data.
Finally, we rely in great part on data recorded by scientists skilled in other disciplines. Geologists, paleontologists, botanists, palynologists, geochronologists, and others study the natural as well as cultural materials recovered from archaeological sites. Their results aid our understanding of the age of the sites as well as the nature of the environment that man was occupying at the time, and give us clues as to how man had modified and adapted to his environment.
All archaeological data is dependent upon the degree of its preservation in terms of weathering, corrosion or other modification of the objects. In addition, there may have been alteration in the physical association of the objects one to the other, brought about through erosion and other geological processes. Before we can make valid interpretations about the nature of archaeological sites, we must have some knowledge of geological processes in
order to properly attribute the association of objects to either cultural or natural causes.
Another generalization is that archaeological remains are not randomly distributed over the earth's surface; they are concentrated in those areas which formerly provided the optimum conditions for human occupation. In addition, with the refinements in cultural technology that man had effected through time, he was able to gradually expand into more hostile or marginal environments.
Finally, we must concede that our knowledge is in part limited by the extent of prior archaeological investigations. During the combined Prehistoric Expedition's work in 1963-1965 (part of the Aswan Dam Salvage program), we found hundreds of sites in the Egyptian Sahara in an area which today is uninhabited. Until our work the area had not been investigated for archaeological evidence. There is no doubt that many such areas which are archaeological ly unknown at present will reveal major evidence of past occupation once they are studied.
Implements of stone are one of the most common cultural items found on sites. Their study is of importance to the archaeologist in a number of ways. They occur throughout the archaeological record from early to late, and for large numbers of sites they are practically the only cultural material preserved. Stone tools may be classified as: those pieces of stone that were modified to a specific form by man prior to use and those that were modified by use. Those stones that were imported to sites by man but show no evidence of use have been termed manuports.
Tool use is not limited to man but is also a characteristic of some anthropoids as well as the sea otter and a few other animals. In addition, naturally broken stones can resemble man-made implements; thus the identification of valid early man-made tools is difficult. Toward the end of the nineteenth century this problem resulted in the formulation of the concept of an "Eolithic" or Dawn Stone Age. This idea was first published by an Englishman, Prestwich, in 1891 as a means of explaining the presence of peculiarly shaped flints found in English beach gravels. The "Eolithic” was considered to represent a phase in human cultural evolution represented by the simplest of all possible tools made from broken flints. The idea has not stood the test of time for several reasons. The eoliths were found to be indistinguishable from flints naturally broken by geologic processes such as wave action on a beach and the rolling of stones in stream beds (Fig. 2-1). Another part of the concept was the belief that the simpler the tool was in form the older it must be. This point of view has subsequently been proved erroneous through use of absolute dating methods and the comparison of stone tools made by modern aborigines with archaeological industries. Finally, "proof” that such stones of nondiagnostic
Fig. 2-1 "Eoliths" and accidents of nature: a. and c. are eoliths; b. and d. were chipped by natural agencies. (Oakley 1972:Fig. 2.)
form were used as artifacts can be demonstrated only by their occurrence on a verifiable living floor, a place where the splintered food bones and other
evidences of human occupation also occur.
While the "Eolithic" cannot be proved to have existed, we can at least examine the earliest evidence definitely attributed to man. At present the earliest stone industry known is from Olduvai Gorge in East Africa. Dated by the Potassium-argon radioactivity decay method at 1.75 million years of age, this industry has been named the Oldowan by its finders, the Leakeys. What is remarkable about the Oldowan stone industry is that it is not simple; most of the implements are battered pebbles and utilized flakes which can be divided into 10 or more distinct form categories. There may have been an even earlier, simpler use of stones which has not been located to date; so the search goes
Stone tools are frequently the only remains that have been preserved. As a result they have been subjected to detailed analysis to determine, if possible,
how they were made and what functions they performed. Critical to such studies is the distinction between naturally broken stones and those that were deliberately broken by men to be manufactured into tools. This is, I believe, the most confusing factor in archaeology for the nonarchaeologist to comprehend. The most common question people ask me is, "How can you tell that this piece of stone is a tool ?" By way of answer, I would I ike to quote from Kenneth Oakley's Man the Tool-Maker, for I believe he has clarified the situation more than any other author. Within this quotation the figure references have been changed to correspond with the figures in this chapter Oakley (1956:9-12).
Wherever it was available to him, Stone Age man made tools of flint or flint-like rock. Flint and similar hard homogeneous rocks break somewhat after the fashion of glass. A sharp blow directed vertically at a point on the surface of a slab of glass or flint knocks out a solid cone (resembling a limpet-shell in shape), with the apex or origin at the point of impact (Fig. 2-2a). Fracture of this type is called conchoidal (from Ko'yxn, Greek for shell). When a blow is directed obliquely near the edge of a slab of material which breaks conchoidally, a chip or flake is detached (Fig. 2-2b, b'). The fractured face of the flake looks like a musselshell; it has a half-formed cone of percussion at the point of impact, passing into a salient, or swelling, called the positive bulb of percussion, followed by low concentric ripples. There is a corresponding rippled hollow, or flake-scar, with negative bulb of percussion, on the parent lump, or core (Fig. 2-2c). The bulb of percussion on a large flake struck by a sharp blow commonly shows a miniature scar or eraillure, near the centre (Fig. 2—2b').
One of the chief accidental agencies by which stones are flaked is thermal change. Rapid changes of temperature cause unequal expansion or contraction of the surface of the stone or rock relative to its interior. In deserts, for example, the exposed surfaces of some types of rock are continually flaking as a result of the difference between the day and night temperature. In cold regions flakes are commonly split off by frost—the outer layer of the stone expanding through the freezing of absorbed water. A flake or flake-scar due to frost or other thermal fracture is easily recognized, for the surface of fracture has either a roughish, blank appearance, or shows ripples concentric about a central point (the stresses set up by the thermal change being concentrated within the outer layer). On the other hand, the surface of a fracture due to a sharp external blow appears clean-cut, and shows a definite bulb of percussion with faint radial fissures and ripples originating at a point on the edge of the flake or flake-scar Flakes split from a lump of flint by frost action are often round in outline and are commonly referred to as "potlids" (Fig. 2-2d). When the residual frost-pitted lump (Fig. 2-2e) happens to be of appropriate shape it is easily mistaken for an implement or weapon; but its natural origin is usually obvious on closer inspection. Thermal changes sometimes cause flint to break like starch, for example into prisms resembling bladecores (Fig. 2—2f—h), but the lines of this form of thermal fracture have probably been determined in advance by strains set up by slow internal shrinkage. Stones splintered by fire, or faceted by sandstorms (ventifacts or dreikanters) are occasionally mistaken for the work of man (Fig. 2-2j).
Flakes struck by man show a well-defined bulb of percussion. The geologist H. B. Woodward remarked in 1878: "I was astounded to pick out of the stone bed of the Norwich Crag a flake containing a good bulb of percussion." He evidently believed that e was on the track of Pliocene Man. However, heavy stones hurled by the sea against
pjg_ 2—2 Characteristics of humanly worked flints and natural forms sometimes mistaken for artifacts: a. complete cone of percussion in flint; b., b'. flint flake struck by man (two views); c. flint handaxe (Palaeolithic core tool), and one of the waste flakes (at * = flake scar with negative bulb of percussion, x' = waste flake with positive bulb); d. rounded spall of flint ("potlid ) split from nodule by frost action; e. lump of flint pitted by the intersecting scars (at /3), of frost spalls (at /?'); f. shrinkage prism of starch, g. flint showing prismatic, or starch fracture; h. prismatic core of volcanic glass from which blades (such as /.) have been struck, Chalcotithic, Crete; /. ventifacts of dreikanter type, pebbles of jasper faceted by windblown sand, Carnac, Brittany. (Oakley 1972:Fig. 4.)
flints firmly fixed in a beach commonly detach flakes showing bulbs; while bulbar flakes are produced also by the powerful pressure of one stone against another in a gravel which is disturbed by an overriding glacier, or by subsidence on a disintegrating bedrock, or by cliff-falls. But usually the flakes produced in such ways show flatter and more diffuse bulbs of percussion than those produced by purposeful blows.
A fair proportion of the flakes struck by man, even in primitive industries, have their edges dressed. But the edges of thin pieces of stone are very liable to become chipped through friction against other stones, such as occurs in soil-creep (solifluxion), in torrent action, or when stony deposits are caught up in the bottom layer of an icesheet As a general rule naturally chipped flints are easily distinguished from the works
Fig. 2-3 Eoliths' and accidents of nature: a. "eolith" from river gravel at Piltdown, Sussex; b. diagram to illustrate how an "eolith" such as a could have been produced by soil-creep under periglacial conditions; c. flint flake from Sub-Crag Stone Bed, West Runton, Norfolk; d. flake of siliceous rock from glacial desposits of Permian Age in South Africa. (Oakley 1972:Fig. 3.)
of man, for they lack logical design, flake-scars occur in uneconomical profusion, the edges have a bruised appearance, and the flake-surfaces are usually scratched.
Frequently, also, there are signs (such as varying degrees of weathering or patination) that the flake-scars have been produced at several different dates. Nevertheless, under exceptional conditions naturally flaked stones occur which, if seen out of their geological context, might be mistaken for artifacts. For example, large numbers of broken flints with neatly chipped scraper-like margins occur in the Bullhead Bed under nearly 30 feet of Eocene sand at Grays in Essex (Fig. 2-1b, d). The flint nodules in this bed touch one another and they have been crushed together during the slow subsidence of the Eocene for mation, which rests on a surface of Chalk which is undergoing solution. Again, stones flaked by glacial action into forms showing a remarkable resemblance to artifacts have been found in Pleistocene boulder clays, and in the Permian glacial beds of South Africa, some 180 million years old (Fig. 2-3d).
Thus geologists are inclined to adopt a cautious attitude with regard to crudely chipped stones resembling artifacts, particularly if they occur in situations where natural flaking cannot be ruled out. By contrast even the most crudely chipped pieces of quartz found with the remains of Peking Man in the Choukontien caves can be accepted as implements merely on account of their situation.
Identification of a stone fragment as being man-made is only the first step in the analysis. Subsequent steps include their classification into the major categories of tools and debitage or waste. The tools are then divided into form classes core tools and flake tools—which are then subdivided in terms of morphology: overall shape, shape of edge, type and amount of retouch, and so forth. In some cases such forms are given specific names which imply a function, for example, knife, scraper, and point. In other cases the name is descriptive, for example, truncated flake or denticulate, without function being implied. The classification of stone tools by type provides one means whereby we can describe specific assemblages, for example, 14.4 percent of the tools from a particular site are side scrapers. Comparison of the tool types from other sites helps determine the degree of similarity between site collections.
Another procedure is to identify the specific manufacturing techniques that were used and to determine their order. These technological studies
provide another method of describing ancient cultural practices. Through such studies archaeologists attempt to reconstruct every manufacturing step from the quarrying of the original raw material to the completion of the finished artifact.
Pottery appears much later in the cultural record than tools of stone, wood, bone, or shell. Where pottery is present, no other artifacts are as informative, with the exception of written records; even in this field the first writing was on clay tablets. Different kinds of information provided by pottery include the details of the manufacturing processes and the sources of clay utilized. Design styles provide insight into cultural beliefs as well as a means of chronological ordering, for once invented, pottery was such a flexible media that its form and decoration were constantly being modified. Since continuous changes through time are characteristic of pottery, we can identify archaeological periods on the basis of these changes. Another outstanding characteristic of pottery is its near indestructabiIity. Whole vessels are fragile and easily broken, but once broken, the smaller potsherds are likely to be preserved, since they can be destroyed only by severe erosion. Fire changes only their color and they are almost impervious to chemical decay.
The characteristics of pottery are numerous; in some cases as many as 60 or 70 separate attributes can be identified within a specific assemblage of potsherds. The basic features include shape, method of manufacture, and base color. Either the coiling method, the paddle and anvil method, or the wheel is used in the manufacture. Base color is determined by the clay and whether it was fired in a reducing or oxidizing atmosphere. The base color may be covered with a thin wash of clay, called a slip, or it may be painted with a variety of colors. Decoration may further be added by pinching, punching, incising, engraving, and the addition of applique (small pieces of clay pressed on). Pottery may also be cast in molds. In some cases, such as the Mochica culture of Peru, most of what we know about their past cultural practices has been learned from studying their pottery.
Artifacts of bone are less informative than pottery or stone because usually they are utilitarian objects of simple form. However, bone is a durable material, and because of man's early reliance on game animals for food, it was easily available. The earliest presumed bone tools are those termed the Osteodontokeratic culture. These are objects of bone, teeth, and horn that, according to ** their describer, Raymond Dart, were used as simple tools by the earliest hominids, the Australopithecines of South Africa. In most cultures items of bone are of relative unimportance; however in a few cultures, such as the Eskimo, where other materials suitable for tool manufacture are rare, bone was
The Nature of Archaeology
32
widely used, replacing stone or wood as a manufacturing material. Bone, of course, can also be carved, and it is in bone that we perceive one of man's earliest artistic expressions, small carvings of humans and animals in the Upper Paleolithic period some 20,000-30,000 years ago. Bone tools can also be studied to determine what functions they performed and the manufacturing methods employed—cutting, grinding, engraving, and so on. Tools of bone are frequently combined with portions of wood or stone to form composite tools. In norma! instances bone tools are small, usually of a size to be conveniently hand held. Large bones are used rarely; although, for example, whalebones were used for construction materials in Eskimo houses. The most common occurrence of bone is as food debris.
Wooden Objects
Tools of wood were undoubtedly among the very first implements to be used by men. We know, for example, that the simplest form of tool using by modern apes consists of the picking up or tearing off of limbs, stripping off the leaves and twigs, and then using the stick as a club, missile, or item to poke into holes or dig. There can be no doubt such tools have been in use since the dawn of culture. Unfortunately, wood is a perishable material and these early tools have not been preserved. To my knowledge, the earliest wooden implements found to date are from Kalambo Falls in Rhodesia and are about 55,000 years of age. They have been identified as digging sticks, a throwing stick, and the point of a spear (Cole 1965: 158). Those specimens were recovered from a waterlogged level where bacterial decay was inhibited. The other major depositional environment where wood may be preserved are dry caves. Any environment which is alternately subjected to wetting and drying will result in decay of wood and other perishable materials. Such decay is a major limitation to archaeology because most environments are of this type and few environments are either waterlogged or extremely dry. One other technique for recovering the form of wooden objects is to make a plaster cast. In rare circumstances while digging, you encounter holes in the soil which are the result of a piece of wood having rotted away. Usually such holes are postholes, but occasionally they are the result of decay of wooden artifacts. The most famous such find was the ancient Sumerian harp which Sir Leonard Woolley recovered by pouring plaster into the holes (Fig. 2-4). Imagine his delight after excavation of the hardened plaster cast to find attached to the plaster the metal ornaments that had originally decorated the harp.
Most wooden items were either simple tools or the hafts of tools such as spears, knives, and axes. It is only within the last few thousand years that major constructions of wood, such as ships, forts, houses, and pile villages, were built. In rare instances, examples of these have been recovered archaeologically. The Sutton Hoo ship (Fig. 2-5) can be reconstructed from its impression in the earth, and the warship Vasa, which sunk on its maiden voyage, has been
Fig. 2-4 Reconstruction of Sumerian harp found at Ur achieved by pouring plaster of paris down holes left from the rotting away of the wooden harp frame. (Hole and Heizer 1973:105.)
salvaged. The Swiss Lake dwellings are another example of preserved structures of wood. The most outstanding recovery of ancient furniture is from the tomb of the Egyptian Pharoah Tutankhamen (Fig. 2-6).
One of the most important uses of wood in archaeology is in dating. Radiocarbon analysis is possible with both wood and charcoal. Even more impressive is the fact that in certain areas pieces of wood or charcoal with more than 25 annual rings can be correlated with a master chronology, and thereby the individual specimens can be dated to the exact year in which the last ring was deposited. The method has its limitations because only certain species of trees produce distinctive annual rings. A further qualification is that the regional climate must possess fluctuations in annual rainfall which are reflected in the differential widths of growth in the annual rings. If these conditions are present, then it is possible to work out a master chronology. The
Fig. 2-5 Excavation of the ancient ship at Sutton Hoo, England. The ship was found in 1 939 and dates to the seventh century A.D. (Trustees of the British Museum (Natural History) 1959.)
best known area for tree ring studies is the American Southwest where it originated. However, it has also been, attempted in the American Plains, the North American Arctic, Egypt, Turkey, New Zealand, and Chile.