Readers of recent issues of ANTIQUITY will have doubtless been struck by the passions which can be aroused by discussions concerning concepts of prehistoric as well as present-day ethnicity. Here the Megaws round off a Celtic trilogy of papers with what will surely not be the last word in an important debate.

John Oakley defends the legacy of Beazley in response to James Whitley's recent analysis of Beazley and his followers (ANTIQUITY 72: 40–47). The debate demonstrates how influential some past figures of archaeological methodology still are, and how diverse are the approaches of modern scholars in interpreting and using their legacy.

When ANTIQUITY published the historical article by Clarke, I was a 20-year-old student, deeply engaged in field activities and substantially torn away from the ‘theoretical’ debate.

My archaeological loss of innocence happened only in the early 1980s, when I discovered (thanks to people like Maurizio Tosi and Anna Maria Bietti Sestieri) the enormous explanatory potential of processual theories.

It would be absurd to label the whole of Italian archaeology as ‘atheoretical’; as a matter of fact, a powerful theoretical machine, the Marxist theory, had operated from the late 1960s, thanks to the group of Dialoghi di Archeologia. The problem was in the idealistic roots of our (academic) culture, characterized by a programmatic divorce between humanistic and scientific studies and from a substantial lack of interest for the anthropological theories.

In December 1997 we published Fekri Hassan's comments on Ian Hodder's ‘Reflexive excavation methods’ (ANTIQUITY 71: 1020–25). Ian Hodder responds here to the criticisms and defends his position.

It is the best of times and it is the worst of times. On one hand, there are more resources and people involved in archaeology than ever before; there is considerable public and media interest in the subject; and there have been exciting developments in archaeologists’ uses of social theory. On the other, competition is intense for locally scarce funding; most field research is constrained by non-archaeological considerations; and fragmentation, insecurity and disenchantment are rife. The split between theory and practice has certainly widened since David Clarke’s day, whilst theory has become not so much Clarke’s unifier within the morass of empirical detail but its own basis for division and often bitter disagreement within the profession.

Knowledge of rice domestication and its archaeological context has been increasing explosively of late. Nearly 20 years ago rice from the Hemudu and Luojiajiao sites (FIGURE 1) indicated that rice domestication likely began before 5000 BC (Crawford 1992; Lin 1992; Yan 1990). By the late 1980s news of rice from the south-central China Pengtoushan site a thousand years older than Hemudu began to circulate (Bellwood et al. 1992; Hunan 1990; Pei 1989). Undocumented news of sites having a median date of 11,500 BP with domesticated rice has recently made the rounds (Normile 1997). In addition, the first domesticated rice in Southeast Asia, once thought to be to be older than the first rice in China, is not as old as once thought (Glover & Higham 1996: 422; Higham 1995). Finally, wild rice (Oryza rufipogon) was reported to be growing in the Yangzi valley, well outside its purported original range, making domestication there plausible (Yan 1989; 1990; 1997). Significant progress continued to be made in the 1990s and unlike research on other major crops, the literature is generally not accessible to western scholars, with some exceptions (Ahn 1993; Crawford 1992; Glover & Higham 1996; Higham 1995; MacNeish et al. 1997; Underhill 1997).

Trawling through old, dust-covered folders I found out that I first read ‘Archaeology: the loss of innocence’ as a 2nd-year undergraduate for an essay on whether the New Archaeology was as theoretically sophisticated as it claimed to be. My notes of the time emphasize the beginning and end of the article; suggesting that Clarke’s purpose was just to argue that

1 there had been a sea-change in the nature of archaeology leading to the development of a critically self-conscious entity in the New Archaeology; and

2 to discuss what a general theory of archaeology might look like.

Domesticated rice (Oryza sativa) is one of the five major crops in the world and a staple food for more than 30% of the world population. Yet the question of where, when, why and how the domestication of rice originated has been, and still is, a question under debate. However, as more archaeological and archaeobotanic discoveries have recently come to light, the question of the origin of rice cultivation now seems less elusive than it was a few decades ago. To date, both archaeological and archaeobotanic discoveries seem to indicate that rice cultivation first began in the middle Yangzi Valley by 8500–8000 years BP, and subsequently expanded to south China and Southeast Asia.

The Liyang plain, located in the northwest of Hunan province, is part of the plain on the north of Dongting Lake. It is situated at longitude 111°22′30″E to 111°51′30″E, and latitude 29°35′31″N to 29°47′30″N. It is made up of the Li River, its tributaries and the alluvial plain, and occupies about 600 sq. km in area (FIGURE 1).

The plain is saucer-shaped and surrounded by small hills on three sides, joined at the eastern part to the plain north of Dongting Lake. The area is a classic ‘plate-basin’ structure. Inside its boundaries, the land is broad and flat, with small streams winding in different directions, and lakes and ponds dotting the landscape. It is 32–45 m above sea level, with an incline of 2° to 3°.

It is interesting to reflect that only nine years separate David Clarke’s paper ‘Archaeology: the loss of innocence’ and the publication of Symbolic and structural archaeology (Hodder 1982), which may be taken to mark the beginning of a ‘post-processual’ archaeology. Many of the ideas put forward in that book were being discussed and developed at Cambridge from around 1978. David’s paper, and its publication in ANTIQUITY, may be taken as representing the highwater mark of ‘new’ or processual archaeology in the academy. Almost as soon as the ideas had been presented, and not really very well developed in the practice of doing archaeology, they were under fire and being replaced. Yet David was still attacking ‘traditional’ archaeology, fighting for his own position in the 1973 paper, and putting foward an agenda for the future of archaeology. It was a manifesto for future work. New Archaeology was then 11 years old and had already achieved a certain hegemony in Anglo-American archaeology, at least among younger academics more interested in ideas than recovering and describing evidence. In 1998 what is labelled ‘post-processual’ archaeology differs fundamentally from many of the ideas presented in the Hodder volume and it is doubtful whether anyone would still wish to follow David’s agenda or advocate early ‘post-processual’ ideas. The pace of thinking has inexorably heated up. Both David’s paper and the Hodder book are now primarily of historical interest in the development of a disciplinary consciousness in which archaeology is becoming increasingly self-reflexive, critically interrogating its intellectual presuppositions, procedures and practices.

Rice, Oryza sativa L., is one of the most important cereal crops in the world, and its emergence as a domesticated subsistence plant drives much of the interest and research in archaeology in South and East Asia. The homeland of domesticated rice has been proposed as:

1 a specific area, such as India (Vavilov 1926; Ramiah & Ghose 1951), South China (Ding 1957), Southeast Asia (Spencer 1963) and the Yangtze valley in China (Yan 1982; 1989)

2 a biogeographic region, such as the so-called ‘belt region’ with a great diversity of Oryza species (Chang 1976), or

3 an ecological zone, such as coastal swamp habitats (Higham 1995).

The dual tasks of this paper are to examine David Clarke’s ideas about the development of archaeology as they relate both to the era when ‘the loss of innocence’ was written and to what has happened since. In his treatment of the history of archaeology offered in that essay, Clarke subscribed to at least two of the key tenets of the behaviourist and utilitarian approaches that dominated the social sciences in the 1960s: neoevolutionism and ecological determinism.

Clarke viewed the development of archaeology as following a unilinear sequence of stages from consciousness through self-consciousness to critical self-consciousness. The first stage began with archaeology defining its subject matter and what archaeologists do. As its database and the procedures required for studying it became more elaborate, self-conscious archaeology emerged as a ‘series of divergent and selfreferencing regional schools … with regionally esteemed bodies of archaeological theory and locally preferred forms of description, interpretation and explanation’ (Clarke 1973: 7). At the stage of critical self-consciousness, regionalism was replaced by a conviction that ‘archaeologists hold most of their problems in common and share large areas of general theory within a single discipline’ (1973: 7). Archaeology was now defined by ‘the characteristic forms of its reasoning, the intrinsic nature of its knowledge and information, and its competing theories of concepts and their relationships’ (1973: 7). Clarke looked forward to a fourth (and ultimate?) phase of self-critical self-consciousncss, when the new archaeology would monitor and control its own development.