Book review: Nhanda Villages of the Victoria District, Western Australia by Rupert Gerritsen

If you have ever subscribed to the theory of incipient agriculture in Australia or wanted to believe that (re)planting of yams or domiculture equated with the intensive management of cultigens then this slim research paper is just what you have been waiting for. Gerritsen has (re)visited the issue of 19th century historical accounts of yam ''gardens'' and ''villages'' along the rich alluvial flats and terraces of the central west coast. A number of major drainage courses, such as the Greenough, Chapman and Irwin River valleys were observed by early explorers such as Grey and subsequent settlers to support relatively high densities of people, to host shelters of substantial size and robustness and to have evidence of extensive and well tended yam grounds.

The presence of significant yam patches is of course not restricted to the semi-arid central west coast. They are also described from the southwest of the continent, portions of the southern and western Kimberley and also from parts of Cape York and offshore islands. Equally large and robust shelters to those noted by Gerritsen have been recorded from many parts of northern Australia while aggregations of Aboriginal people running into the 100's are recorded from virtually every part of the country at one stage or another.

Overall this essay is scholarly and thoughtful in the way it collates and documents the ethnohistoric sources, tries to critique likely Eurocentric biases in recording unfamiliar structures and land-use practices and the way it tries to place the apparent anomaly of the ''agro-complex'' into some kind of comparative context. Where it falls down, however, is in its approach to examining physical evidence in the field at the assumed location of the villages. Clearly pre-contact lithic scatters suddenly become proof for previous agricultural settlements (with only some grinding material being noted), what appears to be a river terrace is assumed to have been the result of land tillage or similar activity and (illustrations) of thick-framed shelters becomes houses within a village.

Gerritsen uses historical accounts of the numbers of people witnessed at places like the "Greenough Village", where 150 individuals are noted, to calculate the size of the original settlement (after Hassan) and their permanency. The degree of sedentism is assessed (after Kelly) and new labels for this kind of community are suggested including seasonal, multi-seasonal, seasonally mobile, sequential and partially sedentary as alternatives to Murdoch's categories. Unfortunately the rationale for these new labels is not developed in any depth and thus is hard to judge their utility.

Sections of the paper tend towards pure speculation. For example at page 18 the author notes "As stated previously, the site identified as corresponding to the first village encountered by Grey at the Hutt River was situated on a terraced area that did not appear to be a natural formation. If this is the case, and it is actually the product of an extended occupational history at this site, then it could be deemed to be a thick cultural deposit". Also on page 18 he states further "Assuming that the settlement here commenced as a result of a dynamic initiated by the marooning of two Dutch mutineers in this area in 1629..." In neither example is any archaeological analysis brought to bear on the issue - despite the author making field inspections of the sites. Why did he not make at least a preliminary examination of the terrace to see if it contained any cultural materials on the surface or in section? Equally, while it is well documented that two of the Batavia mutineers were indeed left on this arid coastline there is no material evidence that they in fact survived or had any lasting impact on local populations. It should be noted that there is some material evidence for survival and possible contact for other maritime events such as the later wrecking of the Zuytdorp further north along the Western Australian coastline.

Further evidence that is canvassed as support for the sedentary status of these groups are superior wells, a burial ground and the presence of large pathways. Unfortunately, the fresh water wells he notes are no larger than those found in the interior Western Desert, the burial ground quite small compared to those recorded from South Australia and Victoria and the presence of pathways in fact common in south-west Australia (see numerous papers by Bill Ferguson).

For these and a range of other reasons I found that only one of the eight conclusions presented on page 21 was supported by the data or arguments presented in the preceding text. I will summarise my objections to each of his conclusions:

1) That the Nhanda were engaged in agriculture. There is no hard data to support this conclusion. Their intensive and likely seasonal use of yams is well within the range recorded for other parts of Aboriginal Australia.

2) That there were five substantial settlements. This is only supported by generally brief accounts from explorers and settlers.

3) That the population of the settlements ranged from 99 to 290. These figures have been calculated from estimates provided by historical witnesses, which are often noted by the author to be based on superficial impressions.

4) That two of the five settlement locations have been confirmed. The presence of a lithic scatter and purported settlement mound/terrace is not proof of settlement.

5) That the dwellings were large and robust. The apparent accuracy of several early drawings makes this conclusion reasonable.

6) That the settlements were permanent and the residents displayed a high degree of sedentism. Given no long-term observations were made in the earlier part of the 19th century this is largely speculation.

7) That a range of settlement types were in evidence including "homesteads", "dispersed settlements", "hamlets" and "villages". These are more likely to be the etic categories of the early recorders; they are not supported through independent archaeological data.

8) That there were other settlement types in the region. Again, this is only raised by early accounts and not supported through any archaeological data.

In conclusion this extended paper does service to the disparate sources and records that note the presence of these flourishing hunter-gatherer communities, who it would appear for at least part of their resource schedule made heavy reliance on yams. In making the next large jump to instill them with the status of agriculturalists is to fly in the face of comparative data for other Australian Aboriginal groups and to ignore the current lack of archaeological data supporting such a proposition. It is with this balance of merit and shortcoming that I recommend the paper to readers.