Celtic substrate influence

[Larry Trask]

On Sun, 18 Apr 1999, Frank Rossi wrote:

> Given that:

> 1)  in North-West Spain, Galicia and Asturias, which are considered,
> rightly or wrongly, the most Celtic areas of Spain, the two forms of
> the past tense have been reduced in normal usage to one (the simple
> past) and

> 2) a similar phenomenon can be observed in Northern France, Northern
> Italy and SOUTH GERMANY, where in the Pre-Roman period the La Tene
> Iron Age culture and presumably dialects of the Gallic language were
> prevalent if not univarsal, although in this case it is the compound
> past that has replaced the simple past,

> Question:

> Could there be a parallel influence of the Celtic substrate in both
> areas, in the sense of a rejection of two forms for the past tense,
> i.e. either the simple or the compound past, but not both? What do
> the experts on Celtic languages think? In other words, are there any
> similar phenomena in the Celtic languages, ancient and modern?

Well, I'm no Celticist, but I find it very hard to believe that Celtic
speech lasted long enough in any region to have any effect upon the
development of the Romance verbal system.

The Romance preterite directly continues the Latin perfect, which had
both preterite and perfect functions.  The Romance periphrastic perfect
did not exist in classical Latin, but seems to have become well
established by the 5th or 6th century -- by which time Celtic speech had
surely disappeared everywhere on the Continent.

The question is when the contrast between the Romance preterite and
perfect was lost, in the areas in which it has been lost.  In Parisian
French, the preterite seems to have disappeared from speech by about the
16th century, according to Price, but conclusions are difficult because
the written language (which is all that is generally recorded) has
retained the distinction down to the present day.  For Iberia, I have no
information.

But I might add that the loss of the preterite/perfect contrast is far
from rare generally.  Even in those Romance languages that retain it
today, like Castilian, the original force of the contrast has been
significantly altered.  And, of course, contemporary American English
vernacular has lost the perfect in favor of the preterite in certain
circumstances, such as `We did it!' for traditional `We've done it!'
Further, in most varieties of American Spanish, the preterite has
displaced the perfect to a greater or lesser extent.  I don't think that
the loss of this contrast is a surprising phenomenon requiring any
external explanation.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk