Regarding Aboriginal languages too, it would seem that there are
other factors involved as well.
In the Pama-Nyungan family the attempts to construct "tree" have
consistently proven to be impossible. Not only does one get a
situation where language borrowing is an "art", and the idea of "song-
lines" means that many words tend to travel up and down songlines at
times when groups come to "sing country". A song-line (the path
taken by a cultural hero of the "dreaming" in which the geography of
landscape gets sung into being) may travel across 5-6 different
languages - as neighbouring groups may share a story, but not the
language it is carried in. And in which no language is dominant to
any other. With words coming from multiple directions, the problem
of English is magnified. (For example imagine that rather than
Norman French Romance words entering a language that was basically
Germanic Anglo-Saxon that it had been the case of Anglo-Saxon words
fusing with a language that was basically Norman French. Then
consider a third case in which 50% of the language was Norman French
and 50% of the language was Anglo-Saxon and you begin to get close to
the situation. (This 50:50 breakdown is close to the situation one
finds with Japanese - a language that seems to be Altaic or
Austronesian depending upon at what one looks!)
Now magnify this by 1000%, so much so that one can no longer
discriminate any which one was the parent language and which the
external influences - and you have the linguistic situation in
Australia.