> > I think we can exclude that the Ario- segment in the Latinized
> > anthroponym Ariovistus may be related to Proto-Indo-Iranian *ari-
> > *arya-/a:rya-.
>
> I don't.
>
> > Ario- (< *Hario- according to Pokorny p. 67) is most likely from
> > Germanic *harjaz 'army'.
>
> 'Most likely' because?
Other possibilities (1. Gaulish; 2. Germanic mediated by Gaulish):
http://tech.groups.yahoo.com/group/cybalist/message/11759
"*arjo-, not *arya-. Well, Eremon/Airem is one well-known example,
and Gaul. Ariios is another (cf. Old Irish aire 'nobleman'). The
elements *arjo- (note the -o-, Germanic would show -a- instead) and
*wisto- are directly attested in Celtic. It is highly likely that
Ariovistus was known under a Celtic name even if he spoke Suevic
Germanic."
http://www.archivum.info/sci.lang/2006-05/msg01663.html
"Several ancient West Germanic compound names are known in Latinized
form, and the expected -a- usually appears as -o-, e.g.
<Aistomodius> 'Hot-mooded' (one expects *Aista-mo:ðjaz)...... I
suspect that Germanic names which were recognized by Latin authors
as compounds were treated as if they were Gaulish, where -o- is
retained. Indeed some Germanic names, such as <Ariovistus>,
probably first reached Latin by way of Gaulish."
Regards,
Francesco