Zooarchaeology cannot produce a catalog of ethnic index fossils – the linkage between all types of social identities and material culture items is simply too complex for such a straightforward methodology. What can be done, however, is to specify the behavioural conditions under which social interaction took place and provide an understanding of the contextual constraints that structured innovations in animal symbology and their social meanings (Hesse and Wapnish 1997:238–39).
Animal bones have the potential to illuminate many aspects of past human behaviour. Coupled with their general ubiquity at most archaeological sites, it is disheartening that their study has not been a consistent feature of archaeological investigations in Australia. Fortunately, this situation is rapidly changing, with faunal analyses playing an increasingly greater role in both Indigenous and historic contexts. We would therefore like to conclude this manual by discussing the state of zooarchaeology in Australia, and delve a bit deeper into just a few aspects of human behaviour that zooarchaeology can address. Given that this is first and foremost a manual on bone identification, we have tried to refrain from adding too much analytical information. However, we feel that it is our duty to conclude even an identification manual with a fuller discussion of the power of zooarchaeology, and to advocate its application and interpretations as the next steps in faunal analyses. We have tried to pare down a large body of work by focusing on a few specific ways in which different qualitative and quantitative methods can be used to address larger theoretical (and practical) issues. So, we would like to conclude this manual with a few words about the way in which assemblages are quantified and analysed (and have included references along the way for those interested in further reading).
Historically, there has been a relative lack of emphasis on zooarchaeology in Australia (see Cosgrove 2002 for an excellent summary). This situation stems from many interrelated issues; however, the dearth of both training opportunities and funding come readily to mind. At the time of publication, there were a handful of published sources of information on the identification of Australian animals (Archer 2002, Merrilees and Porter 1979, Triggs 2004), none of which are specifically geared toward archaeologists in the field. Fortunately, the tide is turning in Australia. We believe this book in itself is ‘black and white’ evidence of this turn (excuse the pun). In the academic sphere, a few universities are building good comparative bone collections and are teaching students the art of faunal analyses. An increasing number of postgraduate students are analysing faunal material for higher degree theses, distinct faunal sessions regularly occur at the Australian Archaeological Association annual conferences, and faunal papers make significant contributions to the Australasian Society for Historical Archaeology’s conferences. So we believe the outlook is positive. In the professional sphere, it is now commonly accepted that faunal analyses will form part of every professional archaeological consulting report – be it historical or Indigenous. The largest problem facing zooarchaeology in Australia now is the lack of trained specialists.
Animal bones provide information beyond a ‘simple’ laundry list of species. They speak to ancient subsistence practices, and can be used to reconstruct everything from human–animal interactions (including domestication and animal husbandry), ancient environments, trade, inter- and intra-site economic relationships, to social hierarchies. For example, variation between faunal assemblages at different archaeological settlements can help reconstruct several aspects of social organisation. In historic sites, social differentiation can be addressed by using the faunal data to ascertain whether there were discernible economic differences between households. These differences can be seen in the types of fauna exploited, as well as in the anatomical elements present and their relative frequencies. Economic complexity can be addressed by reconstructing inter-site economic relationships involving the exchange, redistribution and trade of animals and animal products. A question commonly asked by zooarchaeologists is whether there is a recognisable economy built upon animals, and if so, how this economy can be identified and what it means.
A good example of the theoretical application of zooarchaeology abroad is Melinda Zeder’s extensive work on animal economies and the provisioning of cities (Zeder 1991, 2003). She examined animal remains in urban and rural contexts in the Near East to ascertain whether some settlements were acting as centres of animal production while others were centres of consumption. This type of differentiation between settlements speaks to the larger issue of economic specialisation and economic complexity, wherein social and economic relationships are maintained between sites based upon trade and resource acquisition. This economic specialisation can be used as evidence of social complexity, as a common feature of all complex societies is a degree of specialisation and differentiation in production. The concept of specialisation between states involves not only economic differentiation (i.e. division of labour), but also economic individuation, in which certain persons spend a significant portion of their time on particular non-subsistence activities.
At the centre of the producer–consumer dichotomy and social complexity, is the much broader concept of human–environment interaction – an environment of which animals are a part. Flannery’s Systems Theory (1968, 1971) treats humans and the environment as ‘a single complex system, composed of many subsystems which mutually influenced each other’ (Flannery 1971:345). The ecological perspective stresses this dynamic mutual relationship between culture and environment. This interaction has the ability to change, alter and restructure societies, ultimately influencing social complexity.
The fundamental goal of zooarchaeology is to draw conclusions about societies using the faunal evidence for human–animal interactions. To arrive at this desired interpretative endpoint, we need to have some way of counting the number of bones in an assemblage, and then drawing a meaning from them. Several methods exist that facilitate the analysis of sample size and amalgamation in faunal assemblages. These methods include the most common quantitative units of number of identified specimens per taxon (NISP) and minimum number of individuals (MNI), to more recent analyses that calculate the minimum number of elements (MNE) present and minimum number of anatomical units (MAU). Each method of quantification has its relative merits, and the choice of which to calculate will often be governed by questions, sample size and the nature of the assemblage (see Lyman 1994, 2008 for more detail).
Along with MNI, NISP is the quantitative unit most frequently encountered. It is an observational unit that simply counts the number of identified bones in a faunal assemblage. So, if 30 bone fragments were identified the NISP is 30. NISP can have a meaning identified to any taxonomic level – it may be to species or even family. Thus, a NISP of 30 may mean 30 fragments from a medium-sized mammal, or it may be more specific and mean 15 identified kangaroo fragments, 10 sheep, three emu and two wombat fragments. One of the problems with NISP-based species ratios is that they fail to take into account skeletal elements that come from the same animal. Thus, a calculated wombat NISP of 100 may reflect the bones from just one individual. Calculating MNI avoids this problem. A second problem with relying solely on NISP is that the number is greatly affected by the degree of fragmentation of the assemblage.
Unlike NISP, MNI is a derived unit. It is ‘the minimum number of individual animals necessary to account for some analytically specified set of identified faunal specimens’ (Lyman 1994:100). It is a derived measurement that may not take into account variables such as age, sex, or size. If, for example, there are three left kangaroo femurs in an assemblage and one right femur, then the MNI for kangaroos is three. That is, at least three individuals would have had to be present in the original assemblage to account for the three femurs. However, what if the right femur is from a juvenile and the left femurs are all from adults? If this is the case, at least four individuals were originally present, providing an MNI of four. To avoid this confusion, analysts state whether they are ‘maximising MNIs’ by accounting for variables such as size, age, and sex of skeletal elements. In addition, measurements of MNI need not be assigned to individual taxa; they may just be assigned to element. For example, an assemblage may have five right scapulae that cannot be assigned to individual species.
A simple way to calculate MNI is to determine a distinct MNI for each skeletal element – this would take into account left and right sides, and how many times that element occurred in a given individual. Thus, if there are four kangaroo left femurs and three right femurs, there is an MNI of four kangaroos based on femora. However, there may be two left kangaroo scapulae and one right, and thus the MNI for kangaroos based on scapulae is two. Once again, however, this method is blind to age, sex and other variation between individuals. One of the biggest problems with conclusions based on MNI calculations is the unpredictable response to the problem of aggregation. A solution is to present both MNI and NISP calculations, and when presenting MNI data, to do so by skeletal element.
While MNI and NISP measure the frequency of taxa, MNE measures the frequency of portions of skeletons of individual taxa – an analytical method tied to the more recent emphasis on taphonomy (Lyman 1994). MNE signifies the number of a particular skeletal element necessary to account for a portion of a skeletal element. Perhaps the easiest way to understand MNE is to look at two ways it can be derived. The first method involves measuring something like the complete circumference represented by a long bone shaft fragment, and then summing those proportions for each portion of a skeletal element (Lyman 1994:103). Simply calculating the portion of the bone represented by the fragment and then summing the fragments to arrive at an MNE for each skeletal portion can also derive it. So if one fragment of bone represents one half of a proximal humerus and another fragment represents one half of a distal humerus, an MNE of one humerus would be calculated. Unfortunately, there are a few different ways to calculate MNE. The best way in which to navigate this web of conflicting methodologies is to be explicit about the way in which the calculation was performed. This allows for other researchers to more accurately compare two assemblages.
One quantification method that may be particularly useful in eventual interpretation of the faunal data is the ratio between NISP and MNE. Dividing NISP by MNE can provide an idea of the degree of fragmentation in an assemblage. This can examine fragmentation of different skeletal elements, but also of different taxa. It can provide clues on the taphonomic processes that may have affected the assemblage as a whole, such as carnivore activity or extraction of marrow by humans.
MAU is a standardised variation on MNI calculations and accounts for the number of specimens in a collection. Developed by Binford (1984), MAU is calculated by dividing the MNE for each anatomical unit (e.g. a proximal femur) by the frequency of its occurrence in an animal (in this case, two). This method aims to analyse survival of different skeletal parts, and reflects how many of each of the various portions of carcasses are represented.
Once the counting is done, we can get back to the big questions by applying the data. This can be done using several main comparison methods. Here we discuss the three most common ones.
Faunal assemblages may differ in the types of species present; for example, settlements raising livestock will have a different combination of species compared with settlements consuming the animal products. Examining the range of species present in a faunal assemblage will provide an insight into the overall range of animals exploited, and ultimately address variability in diet and dietary patterns.
Analyses of skeletal element frequencies centre on the portions of an animal that are most valuable as food (the prime meat-bearing bones). For example, pigs typically serve a sole function as meat producers. This means that, unlike sheep, goat and cattle, they offer no secondary products. However, we still may want to know whether the site in question was raising pigs or procuring meat from elsewhere. A predominance of head and foot bones (low utility elements) may indicate a place engaged in animal husbandry. If the prime meat-bearing bones, such as upper limbs (humerus and femur) are also present, the conclusion might be that the animals were being raised and consumed in the same place. If, however, there is an absence of prime meat-bearing bones, it may be that the animal was raised on site, butchered, and the higher valued portions were traded or exchanged with another site – indicating an inter-site economic relationship. It is also possible that live animals were traded or exchanged at the consumer site itself. In this case, the skeletal element frequencies would mirror a self-sustained site. This is where analysis of butchery patterns and cut marks becomes vital. If cut marks are present on some of the elements, they can be analysed as having resulted from either primary or secondary butchery (see Binford 1981). The placement and distribution of cut marks can then be viewed against skeletal element frequencies, and hypotheses drawn as to the place of butchery of the animal in question. We can also turn to the relative distribution of skeletal elements on the site itself. Perhaps one area is ‘industrial’ in nature, and used for butchery (yielding a lot of head and feet), while another is domestic, and therefore using the higher valued parts of the animal. This approach could also be used to address intra-settlement social differentiation, with higher utility elements appearing in wealthier contexts and vice versa.
The relative ages at which individual domestic animals were slaughtered, in conjunction with the range of species present, also offers insight into inter-site economic systems. At historic sites, many animals, such as sheep, goat and cattle, are used not only for their primary products (meat), but also for secondary products, such as milk, cheese and wool. Some animal products may be collected at intervals (wool), while others may be collected only once – at slaughter. For example, younger individuals tend to have served a primary function as meat producers. If most individuals at an assemblage are younger individuals, the conclusion may be that they were exploited for their primary product – meat. If however, the assemblage is dominated by older individuals that offer secondary products, we can ask whether there is a possible trade or specialisation in secondary products. These same analyses can provide important information on hunting strategies at prehistoric sties. Hunters often aim for a ‘sure’ kill, and may therefore kill either old or young individuals who are too slow or naive to escape. Hunting with dogs, in groups, or with specialised technology can alter this generalisation.
Ultimately, the analysis of faunal remains is dependent on a large variety of factors. As already discussed in the introduction of this manual, taphonomy plays a large role in the preservation and eventual excavation of different skeletal elements and species. Recovery methods play an equally important role. Thus, the best way in which to approach a faunal assemblage is to combine all analyses to create a full picture of the economy in which the animals played a key role. This way, empirical studies and taphonomy play a key practical role in the more abstract theoretical applications of faunal analysis.
The analytical potential of butchery marks on bone is a useful way in which to illustrate the marriage of concrete and abstract elements of zooarchaeology. In most societies, animal production and consumption plays a pivotal role in reconstructing inter-site economic relationships. The eventual analysis of skeletal element frequencies and their variability between sites is intricately related to butchery practices and the analysis of cut marks on animal bone. The past few decades have seen a great deal of research on bone surface modification caused by both humans and carnivores. In particular, discussion and research has been directed toward discerning the differences between human-modified bone and other agents of bone surface modification, such as carnivores, trampling, root etching, etc. (Blumenshine et al. 1996, Capaldo and Blumenshine 1994, Marean and Frey 1997, Pickering 2002, White 1992). Extensive studies of hominid butchery patterns exist in the literature on early hominid behaviour, with a few Australian studies on both butchery (O’Connell and Marshall 1989) and scavenging (Brown et al. 2006; Fillios 2011; Marshall and Cosgrove 1990; Reed 2001, 2009; Solomon and David 1990).
There is a fundamental relationship between humans and animals that is at the heart of reconstructing past human behaviour. We hope we have illustrated the absolute relevance of faunal remains to the study of any site – historic or Indigenous. A model of human behaviour based on rational decision-making theoretically underpins zooarchaeology and suggests that individuals act universally to increase benefits and minimise costs in any actions they undertake. Animals can be this buffer and certainly played a vital role in human actions through time.
Zooarchaeology is a natural fit with Australian archaeology as a discipline. There has been, and continues to be, a solid emphasis placed on the integration of biology and environmental science in Australian prehistory. Australian archaeologists are skilled in the use of interdisciplinary methodologies and collaborations. Unlike archaeologists in many other countries, most Australian archaeologists remain with a foot in both the professional and academic realms, and great benefits can come from specialised training. An ideal scenario is to have an archaeologist who is also trained in faunal analysis at every site. This will help to elevate and maintain a high standard of analysis (and interpretation) in both the professional and educational realms – benefiting the discipline of archaeology, and our understanding of Australia’s valuable past.