What's changing: population size or land-use patterns? Archaeology of Upper Mangrove Creek, Sydney Basin By Val Attenbrow Terra Australis 21, 2004. Pandanus Books, Australian National University, Canberra. ISBN 1 74076 116 2. Pp. xxxii + 380. $49.50.
I like books which put a question in the title. The classic for archaeology is C.K. Brain's The hunters or the hunted?, and long before the days of Eats, shoots and leaves I always felt that Gordon Childe's What happened in history should have been What happened in history? I admire the courage that before you have opened the covers the author has already promised the reader an answer. In a discipline hedged by the probable and the possible, the might and the maybe, to have a clear question is both welcome and refreshing.
The question that Attenbrow poses goes to the heart of much systematic fieldwork in the past forty years. Across every continent archaeologists have trudged in ordered lines scouring ploughed and unploughed landscapes for lithic artefacts and the rock-shelters that might, once excavated, serve as chronological tent pegs to anchor the surface scatters. As a result, prehistoric field survey has replaced excavation as our primary means of recovering data. But sadly, many of these data have disappeared into the black hole of the grey literature; a global archive of unknown proportions and one to which Australia has made a major contribution, especially in unpublished PhDs and client reports.
Field survey rose to prominence as archaeologists sought not only to be systematic but also to lift their eyes from the trench and consider the region as the unit of analysis. The lasting legacy of processual archaeology will always be that to understand the local you must investigate the regional. Site histories cannot be divorced from regional narratives since cultural variants, and variation in the patterns they display in time and space, can only be adequately interpreted, by whatever theoretical standpoint, once a scale has been considered that is demonstrably larger than the unit under investigation. Regional analysis replaces local pleading and allows archaeologists to ask more interesting questions of their data.
Attenbrow's book is concerned with changes in the regional record of a small catchment in the Sydney basin. Her survey was intensive, systematic and long term. By the end of her fieldwork based on a random sampling strategy she had 80 archaeological traits (deposits, images and grinding areas) at 59 archaeological sites. In overall terms this produced densities of 6 sites and 8 traits per k[m.sup.2] in the catchment. At all stages in the analysis she pays attention to the description of terms and quantifies the data in accessible ways. The detailed presentation of results in Chapters 4 and 6, the consideration of sampling bias in Chapter 5 and the wealth of information from her subsequent excavations in the technical Appendices makes the Upper Mangrove Creek a model of regional reporting. Terra Australis are to be congratulated on their commitment to publishing such fine grained detail. It is even more commendable since Upper Mangrove Creek is occupied in the un-flashy Holocene and the major changes occur in only the last three millennia.
The great value of Attenbrow's study is how she sets about using her demonstration of change through time in artefact densities and variants to address the question in her title. Most of us have wrestled with that final interpretation. Do the quantifiable changes at the regional scale reflect more people or just a different way of distributing people around the landscape? …