> Seems that route was via Basque
> Latin ss > Basque sj > PCastilian sj etc...
>
Not extremely likely a priori. Instead, bear in mind that
passer/sparrow behaves suspiciously (the search function is out of
order at this time or I would have referred you to previous
discussons), and so does bassus (late appearance, vocalisme a,
geminate), so I'd consider the possibility that they are both loans
from some West European substrate.
Torsten