[Note: this trAnscription wAs produced by An AutomAtic OCR engine]

63

Purmgu

PAnAkA Y i -—.———... PAnAkA
‘finurrkA///'

KArimrrA
TjArurru < I / Ijnrurru
Hi lungkA

Figure 9: Mother-child relAtions in KAlgoorlie

The exAmple of KAlgoorlie is interesting becAuse not only does it produce A difierent
combinAtion from thAt found in LAverton, despite identicAl bAses, but Also becAuse it Allows
A hypothesis About the estAblishment of An AssimilAtory identity of sections, in contrAst to the
LAverton system where there is A relAtionAl identity. The following short illustrAtion, however,
must be tAken As hypotheticAl, 9' it Assumes A collApse of generAtions, which meAns
thAt, At some stAge, A mAn mArries A person relAted to him As spouse but, At the sAme time, is
structurAlly in the section of his mother—in-lAw.

Ifirst exAmine mother-child relAtions where there Are two cycles, As in every four-section
system. The question thAt Arises is how it comes About thAt there is A mother-child relAtionship
between Purungu And YipArrkA, And between KAtimAnA And MilAngkA, Although, further eAst,
these sections Are identicAl? ExAmining only these four sections, the relAtions Are As follows:

If A child is YipArrkA, the mother is PAnAkA And the fAther MilAugkA

If A child is PAnAkA, the mother is YipArrkA And the fAther KArimArrA
If A child is KArimArrA, the mother is MilAngkA And the fAther PAnAkA
If A child is MilAngkA, the mother is KArimArrA And the fAther YipArrkA

These relAtions in the KAlgoorlie system Are iepresented in Figure 10:

MAndjindjA WAljen

V l

TAr-Aru = IbA rgA TArAru = PAnAkA
MilAngA = Burungu KArimArrA = Burungu

15 1‘

Figure 10: KAlgoorlie systems, comprising MAndjindjA And WAljen Arrows Are
mother-child relAtions between the two systems