Basque <ibili>

[Larry Trask]

Roz Frank writes:

>  Conclusion: if there is a relationship between the Basque root-stem <bil>
>  and the IE materials, one needs to consider the time depth involved, i.e.,
>  for determining when the "copying" or "borrowing" would have taken place.

A problem with all proposed IE loans into Basque which cannot be derived
from Latin or Romance.

So far as we know, the first IE languages to reach the Basque-speaking area
were the Celtic languages, probably in the middle of the first millennium
BC.  At the time of the Roman conquest, Aquitanian/Basque was apparently
bordered by Celtic to the north (Gaulish) and to the south (Celtiberian).
The position to the east, in the Pyrenees, is uncertain, but the neighbor
there may have been the non-IE language Iberian.  To the west, we have
clear evidence for IE speech, but of uncertain affiliation.  All I can say
is that the sparse evidence is seemingly consistent with Celtic speech,
but does not require Celtic speech.

Of course, it is conceivable that some unknown branch of IE might have reached
the area even before Celtic, but, if so, this hypothetical language seemingly
disappeared without trace -- and I don't much care for positing hypothetical
languages.

But even Celtic loans into Basque are surprisingly few, given the long
centuries of contact.  We have only two or three certain cases, plus a few
more doubtful cases.

Accordingly, very little can be concluded, except that the evidence for IE
loans into Basque before the Roman conquest is sparse to non-existent.

>  I
>  don't know whether anyone has tried to assign a time depth to Class I verbs
>  in Basque, although I believe Larry would agree with me that they can be
>  assigned to Pre-Basque with no difficulty. Perhaps Larry can add some
>  additional insights into the problems that are involved here.

Almost certainly, even though we have no contemporary records.  The reasoning
goes like this.

Early Basque had a suffix <-i> for making the participles of verbs.  This
suffix occurs in all prefixing verbs (those with the prefix *<e->).  It
also occurs in a few denominal verbs, such as <hautsi> 'break, smash', from
<hauts> 'dust, powder'.  (These last have no *<e-> because their roots are
nominal, not verbal.)

This <-i> was displaced in its function of making participles by <-tu>,
which was borrowed from Latin <-tu(m)>.  Basque borrowed Latin verbs in
the form of their participles: for example, <aditu> 'hear, understand',
from Latin <auditu(m)>.  This <-tu> has been the only productive participial
suffix in the language for a long time.

Now, the form <-tu> is conservative.  Latin <-tu> developed regularly into
Romance *<-do>, preserved today in Castilian as <-do>, but variously
reduced in other Romance varieties -- for example, there is almost nothing
left of it in French.

So, if <-tu> was borrowed early, from Latin before the major Romance
sound changes had occurred, and if it displaced <-i> as the productive
participial suffix, then <-i> must already have been present in the
Pre-Basque of the Roman era.  More than that we can't say.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk