[Note: this transcription was produced by an automatic OCR engine]

25

What should be remembered from this general chronology is that section and subsection
systems progressed in space through the mechanism of cultural diffusion, not necessarily
through migration of people, while the Western Desert language probably spread through
migration.

We shall see that the routes of diffusion of the section system and its terminology in the
Western Desert confirm McConvell’s hypothesis regarding the region of origin of the western
section system in the South-West, as well as its diffusion along the coast northwards into the
Pilbara-Kimberley regions. This study, however, will also depict a diffusion that originated in
the Pilbara-Kimberley region and then moved back southwards along the coast. The lines of
regression elaborated below demonstrate the existence of routes of diffusion, but do not allow
us to determine their cardinal direction, which has to be deduced from other historical facts.
Along a north-south route, for example, it will not be possible to decide if diffusion progressed
from north to south or from south to north, but only that it took place along such an axis.
While the south to north direction McConvell proposes is the most appropriate hypothesis
for the initial diffusion of the section system, it also seems plausible that, before entering the
Western Desert, the system moved back southwards in later stages with a slightly modified
terminology.

The section system penetrated the desert from two general regions, one the South-West and
the other the Pi1bara«Kin1berley area with its system of modified terminology. Interestingly,
these two partly distinct terminologies encounter each other in the Western Desert again,
engendering new combinations. It is the reason for, and nature of, these combinations that
give the methodology developed here its substance, for it is only because the section systems
we encounter today are the result of such recombinations of two original and distinct sets of
terms that they reveal their transformations in the past and the routes of diffusion they must
have employed.

It is difficult to determine an exact timefiame, or even an approximate starting date,
for the diffusion of the section system into the Western Desert. One of the earliest accounts
producing a general picture of the social category systems of tribes and groups in Western
Australia is Mathews’ 1903-4 paper published in the Queensland Geographical Journal.
Mathews’ distributional pattern of these systems is a summary of his correspondence with
station managers, the Police Force and, as he states, “others who might be recommended by
any of these” 1903-4:45. The author explains that he is confident of his informants’ reliable
character. While some of his statements have to be taken with much caution, and while for
some regions it seems improbable that he could have received detailed ethnographic material
and therefore was probably extrapolating from existing data, his description nevertheless
allows us to draw a picture of the likely geographical distribution at the beginning of the 20"‘
century see Map 4.