Methods

Although research into the strengths and weaknesses of field and laboratory methods is not usually the primary objective of zooarchaeologists, most zooarchaeological results are quantified in some fashion. The scientific method requires that theories be testable, hypotheses be evaluated, and results be independently verifiable. Consequently, zooarchaeological methods should be scientifically sound, appropriate to the materials, compatible to the research questions, applied so as to avoid introducing further bias, and thoroughly described. For these reasons, critiques of archaeological and zooarchaeological methods are common in the literature.

Field Methods

Quantification requires that all of the materials which might be used to address research questions should have an equal and random (in the statistical sense) opportunity to be recovered using a consistent recovery technique and that the samples be adequate in size. Only through adherence to these principles can it be argued, for example, that domestic mammals were more common in one context than in another or that warm-water mollusks were more common than were cold-water ones. Although the number of biases that can be traced to poor field decisions appears infinite, many are related to the placement of excavation units and the size of the screen or sieve used to recover animal remains. The composition of animal remains from different activity areas (e. g., house floors, trash heaps, temple complexes, etc.) may be quite different. This bias is compounded by field protocols that use inappropriate screening methods. With few exceptions, animal remains should be recovered by passing the excavated soil through a screen with a mesh dimension suitable to sample the full range of animals formerly present at the site.

Laboratory Methods

Zooarchaeologists primarily study the bones and teeth of vertebrate animals (other than humans) and the exoskeletons of mollusks, crustaceans (crabs and lobsters), and echinoderms (sea urchins). Other vertebrate tissues and other animals are less frequently studied because they either are uncommon in most archaeological deposits or are more common in samples studied by palaeoethnobotanists and soil scientists. These include skin, feathers, hair, eggshell, horn, fur, insects, foraminifera, and endoparasites. In most cases, zooarchaeologists study the remains of animals which are members of the modern environment. However, it is common to find animals which are extirpated from their former range or which are now extinct.

The preliminary laboratory study involves recording primary data such as those in Table 1. Not all of these observations are available for all specimens and additional data may be required by some research designs. Primary data are used to estimate secondary data such as body size and conformation, age classes

Table 1 Primary data and attributes of animal remains which may be recorded during a zooarchaeological study

Taxonomic identification of the specimen Element represented by the specimen

Side (e. g., left, right, axial, unknown, or some other description)

Portion (e. g., proximal, distal, anterior, lateral, medial, shaft, unknown, or some other description)

Sex (description of morphological evidence for sex such as dental attributes, presence of sexually diagnostic features such as antlers or the shape of a turtle plastron, or other characteristics)

Age (e. g., fused or unfused long bone, degree of wear on teeth, stage of tooth eruption, or other characteristics)

Count (number of specimens referred to the taxon, often abbreviated as NISP)

Weight (weight of specimens referred to the taxon)

Minimum number of individuals (abbreviated as MNI)

Modification (description of the modification(s) including: state of preservation; gnawed by a human, rodent, carnivore, or artiodactyl; evidence for passing through a digestive system; butchering marks such as cut, hacked or chopped, sawed; evidence that the specimen was burned, worked, trampled, weathered, or pathological; description of where the mark is located and evidence that the mark was made by a metal or stone implement; other characteristics)

Measurements (definition of the dimension measured, or source of the description; actual measurement of the defined dimension) Other data as required by the research design (e. g., incremental growth patterns in dental cementum, or mollusk valves; stable isotopes;

Trace elements; DNA and molecular evidence, etc.)

Explanatory notes

And sex ratios, relative frequencies of animals, frequencies of skeletal portions, dietary contributions, and the cause and function of modifications. The success of deriving data is dependent upon sampling designs and sample size. The choice of which method to use is related to the research question and the materials being studied. Each method has serious flaws, and interpretations derived from one method should be verified with data obtained using other methods. Methods are inter-related and build upon each other. Thus, a single method is unlikely to serve all analytical needs.

Identification is a multi-faceted procedure during which the animal and the element represented by each specimen is identified in terms of the part of the animal’s skeleton or exoskeleton represented and the animal’s taxonomic classification. Identification requires professional training and access to a good reference collection of modern skeletons. Skeletons in a reference collection should be from animals whose identification was confirmed before the skeleton was prepared. During identification, each archaeological specimen should be compared to the appropriate reference materials. Only in extreme cases should illustrations or archaeological specimens be used to identify materials. In the case of rare or extinct animals, it may be necessary to consult two or more reference collections. In addition to the reference collection itself, biological information about current and former ranges and zooarchaeological reports for similar sites occupied recently or in the distant past may be helpful. Some identifications are based on, or confirmed by, DNA and molecular evidence.

Estimates of body size and conformation are based on measurements of dimensions defined in published protocols. Often one dimension of a specimen, such as greatest length, is plotted against another dimension, such as greatest width; or a ratio is established among dimensions. Body size or conformation may be estimated from measurements using formulas derived from modern animals or from descriptive biological data for animals whose size or conformation is known. Such estimates support interpretations about animals in the past and variation within populations but they require the consideration of individual, regional, and breed variations. A change in dimensions might suggest that the size or conformation of an animal, or of an entire population, had responded to changes in climate, predation, or food availability. Body size may also influence human choices about which animals to capture or avoid capturing, which habitats to exploit, or how and when to capture the animal. Changes in dimensions, body size, or conformation are the primary signs of early animal domestication.

Estimates of age-at-death and age classes are approached in different ways depending upon whether the animal has determinate growth (grows toward an optimal adult body size as in birds and mammals) or indeterminate growth (grows throughout its life as in mollusks, sharks, rays, bony fishes, amphibians, and reptiles). In animals with determinate growth, sequences in long bone maturation, tooth development and attrition, and growth increments such as those in dental cementum may indicate the age at which an individual animal died. In animals with indeterminate growth, age-at-death is estimated from body size and characteristics of growth increments in otoliths, vertebras, scutes, scales, and other hard tissues. If several members of a species are represented, these observations can be used to create age classes and to derive mortality or survivorship curves. As with body size, estimates of age-at-death and age classes require the consideration of other sources of variation. Shifts in age classes, particularly if they coincide with shifts in body size and sex ratios, may indicate a population-wide response to phenomena such as environmental change, predation rates, capture technologies, or domestication.

Estimates of the sex ratios are based on sexually diagnostic features and the size of specimens. Sexually diagnostic characteristics include shape and the presence of features typical of a specific sex. These include horns and antlers in some male mammals, spurs in some male birds, medullary bone in female birds, the shape of horn cores in cattle, sheep, and goats, and the shape of the pelvic girdle in some mammals. In some animals, members of one sex may be much larger than members of the other sex. Body size or conformation may suggest the presence of intact males, castrated males, and females, thereby indicating domestic animals. As with body size and age, estimating sex ratios requires the consideration of multiple sources of variation. Changes in sex ratios may signal changes in environmental variables, predation decisions, or domestication, especially if a change in sex ratios coincides with changes in age classes and body size.

Estimates of relative frequencies of animals are usually based on the number of different animals identified in the sample (richness), the number of archaeological specimens referred to a specific animal in the sample (NISP), estimate of the smallest number of individuals necessary to account for all of the specimens referred to a specific animal (MNI), or the weight of the specimens referred to a specific animal. Unlike NISP, which describes the actual number of specimens in the sample studied, MNI is an analytical product and the estimate should not be confused with actual individuals. To estimate MNI, it is necessary to consider not only the taxonomic identification and the elements represented for the animal in question, but also measurements, age, sex, and archaeological context.

These measures or relative frequency may have little to do with human behavior or the environments in which people lived. Each method assumes that all animal remains are equally influenced by chance events, human behavior, site-formation processes, field techniques, laboratory methods, and analytical decisions, an assumption which is unlikely to be true. Rarely can the interdependence or independence of the specimens be assessed. Furthermore, it is unlikely that any of these variables are uniform among animals, between sites, between temporal components of the same site, or even among excavation units. On the other hand, changes or continuities in relative frequencies are associated with environmental and cultural phenomena which often can be evaluated in no other way. Estimates of relative frequencies can be used to assess diversity, equitability, optimal foraging strategies, social complexity, social identity, belief systems, political alliances, and environmental change.

Frequencies of skeletal portions and utility indices are based on the identity of the elements represented in samples and their relationships to anatomical regions, body parts, or butchering units. Most methods quantify specimens in terms of the NISP from various parts of the skeleton, a ratio between the number of specimens observed and the number of specimens expected, rank order of body parts represented, and indices based on the utility of carcass portions. The specimens may also be evaluated in terms of the minimum number of elements (MNE). Such values distinguish between animals used for food and those that were not and between samples created by nonhuman scavengers and those created by humans. Differences in skeletal frequencies may indicate which animals were killed some distance from the excavated deposit and which were killed nearby, an important aspect of subsistence strategies and a way to distinguish between domestic and wild animals. This principle extends to the association of skeletal frequencies with distinctions between sacred and secular animals, site function, culture change (acculturation and assimilation), trade, as well as social hierarchies in complex societies and urban environments.

Estimates of dietary contribution are used to assess social and economic systems. These estimates are most frequently based on ratios of skeletal weight to body weight, estimates of weight for whole animals multiplied by the MNI for that animal in the sample, estimates of body size derived from measurements, or data from reference collections and the literature. Distinctions are made between whole animal weight, edible meat weight, and nutritional contributions measured as vitamins, minerals, protein, and fat. Dietary value guided decisions about which animals and resource areas to use most frequently, how much effort to expend in finding and capturing a particular animal, and which portions of a carcass to transport from where it was acquired to where it was used. Such studies are important when assessing human demographic and health potentials.

The causes and functions of fragmentation levels, fracture types, and other modifications provide information on many aspects of human and nonhuman behavior. In combination with skeletal frequencies, fragmentation and other modifications may distinguish between human and nonhuman uses of animal parts or between food and nonfood animals. They also provide evidence for site function, the nutritional status of the human population, skinning, the ethnic identity of the butcher, the social affiliation of the consumer, butchery for household use instead of trade, and the presence of specialist butchers producing standard cuts for a discriminating market compared to a local householder intent on maximizing the amount of food and other products obtained from the carcass. Modifications also provide evidence for tools, ornaments, ritual sacrifices, displays, and other uses.