In many aspects of Indigenous Australian life prior to the 1970s, exclusion from mainstream institutions was the order of the day. In the case of the census, this was built on the 1901 Constitution, which at s. 127 stated that 'in reckoning the numbers of the people of the Commonwealth, or of a State or other part of the Commonwealth, Aboriginal natives shall not be counted'. Within this constitutional provision, Aboriginal people were counted, only to be then excluded from reckoning, although in the early years there was not even an attempt at full enumeration of Aboriginal people prior to their exclusion. The Statistician's report on the 1911 Census set the tone:- 4 -
the full blooded Australian aboriginals…represent only those who were in the employ of whites at the date of the Census or were living in a civilised or semi-civilised condition in the vicinity of settlements of whites at that date. An enumeration of aboriginals living in a purely wild state was not undertaken (cited in Smith 1980: 30),
and
[f]rom time to time attempts have been made to ascertain the number of aboriginals in the various divisions of Australia, but the results have not been satisfactory, and the efforts in this direction of the Commonwealth Bureau of Census and Statistics at the taking of the Censuses in 1911 and again in 1921 proved disappointing (Commonwealth Bureau of Census and Statistics 1924: 951).
This comment on the 1921 Census is interesting: it was this census that saw the adoption of special methods designed to move towards more comprehensive and reliable population figures. This involved the cooperation of the Statisticians and Protectors of Aborigines in the States and the Northern Territory to arrive at an estimate of the total number of Aboriginal people, and their 'caste' and sex by administrative area (Smith 1980: 35). A system of annual Aboriginal censuses administered by welfare authorities was in place between 1921 and 1944, and it was claimed from 1933 onwards that coverage of the Indigenous population progressively improved with each succeeding census. In effect, though, for much of the twentieth century, the enumeration of remote Indigenous populations was more a case of mustering those administered by welfare authorities. As Smith (1980: 35) puts it, the 'practical difficulties' of enumeration were partly solved by the gradual movement of Aboriginal people into settlements—a process that was virtually complete by 1966 when the Bureau of Census and Statistics felt that, in conjunction with the Aboriginal welfare authorities, it had obtained for the first time a relatively complete enumeration of the Aboriginal population throughout Australia.
Thus, population counts essentially referred to individuals who were known to, and in regular contact with, welfare authorities. Numbers were presented by 'contact groups': those in contact with government settlements and depots, those in contact with missions, and those not in contact. This last group was classified as 'nomadic' and its numbers simply guesstimated. At the 1954 Census, for example, it was estimated that some 13 000 Aboriginal people were not contacted by census collectors (Commonwealth Bureau of Census and Statistics 1971: 7). By 1961, this figure was reduced to 4000 and, as noted, the official view was that the 1966 Census was all-encompassing.