Pelasgian/was Etruscans

Larry Trask larryt at
Wed Apr 25 13:55:45 UTC 2001

--On Monday, April 23, 2001 5:51 am +0000 Douglas G Kilday
<acnasvers at> wrote:

> The South European Wanderwort for 'lead' probably started as a Pelasgian
> term. Like <daphne:>, Gk. <molubdos> has several variants suggesting
> borrowing from different substratal dialects. The original phonology is
> disputed. I would guess *ml.ub- vel sim. with vocalic [l.] which resolved
> to /ol/ in Greek. In the Italian dialect of Pelasgian I hypothesize
> reduction of [l.] giving *mlub-, borrowed into Latin as <plumbum>. Similar
> prenasalized labials in substratal words occur in <sa(m)bu:cus>
> 'elder-tree' (cf. Dacian <seba> id.) and <simpuvium> 'sacrificial vessel'
> (cf. Gk. <sipue:> 'flour-bin, meal-jar'). For initial Lat. pl- borrowed
> from ml- cf. <place:re> 'to be pleasing' from Etr. <mlach> 'smooth,
> pretty, pleasing'. On its way west, *mlub- must have undergone metathesis
> to *blum-, perhaps in Ligurian. Old Basque lacked [m] and disfavored
> initial clusters, so this would have become *belun, yielding Basque
> <berun> by well-attested intervocalic [l] > [r].

Right.  Let's talk about the Basque word.

The idea that Basque <berun> 'lead' is related to the Greek and Latin words
has been around a long time.  But it is far from being the only proposal on
the table.

To begin with, looking for Latin etymologies on the Tiber, we need not
appeal to a hypothetical Ligurian *<blum-> as a source.  If this sort of
origin is considered acceptable, then Gascon <ploum>, or a related Romance
form, will do as well -- and we *know* that Basque has borrowed lots of
words from Gascon and from other Romance neighbors.  Nor should we worry
about such a seemingly late borrowing.  The eastern dialects of Basque do
not have <berun> at all, but only <plomu>.  And some central and western
varieties have <plomo> alongside <berun>, or even in place of it.  Of
these, <plomu> is recorded from 1571, and <plomo> from 1562 -- very early
by Basque standards -- while <berun> is not recorded before about 1620,
though its first attestation occurs in a proverb.  And, for 'tin', *all*
varieties of Basque have only <eztainu> or a related form.  This is a
borrowing from Romance, since Latin <stannum> would not have produced this

Moreover, a wholly native origin for <berun> is not out of the question.
First, note that the Bizkaian dialect in the west has <beraun>.  Now, the
reduction of /au/ to /u/ in a closed syllable is a commonplace phenomenon
in Basque, while the expansion of /u/ to /au/ is without parallel in any
position.  Moreover, Bizkaian happens to be the one dialect which preserves
vowel sequences better than any other.  Accordingly, Vasconists are
inclined to see western <beraun> as conservative, and the more widespread
<berun> as secondary, and the late Luis Michelena, the greatest etymologist
Basque has ever seen, espoused precisely this position.  But conservative
<beraun> does not appear to be helpful for the pan-European origin sketched

Now, there are two proposals for a native origin.  One idea is that the
word is a derivative of <bera> 'soft', with an unidentifiable second
element -- and lead is, of course, a soft metal.  The other sees the word
as built on
*<bel> 'dark', again with an unidentifiable second element -- and lead is
also dark.  The item *<bel> is nowhere recorded as an independent word, but
its former existence is assured by its presence in a number of derivatives,
both as an initial and as a final element.

Now, the otherwise categorical medieval change of intervocalic */l/ to /r/
did not occur in the easternmost dialect, and so a form from this dialect
might help us to choose between earlier *<bel-> and earlier *<ber-> -- but,
as I remarked above, <berun> is not attested in the east, so no joy.

Nevertheless, many scholars have wanted to relate <berun> to the Greek and
Latin words in some way, though the proposals differ substantially in
detail.  Some people note that lead was known in the Near East much earlier
than in Europe, and so they want to see a Near Eastern word diffusing
westward across Europe together with the metal.  Others, however, observe
that Spain was the principal source of European lead in classical and late
pre-classical times, and so they want to see a word of Iberian origin as
diffusing eastward.

These proposals are too numerous and too complicated to repeat here -- and
they not infrequently involve even more far-flung words, such as German
<Blei> 'lead', Georgian <brp'eni> 'lead', Hebrew <bdi:l> 'lead', and a
reported Berber <buldun> 'tin'.  But, as an example of the western-origin
scenario, I might note that the Lusitanian town of Medubriga was so famous
for its lead mines that Pliny called its inhabitants <plumbarii> -- 'the
lead men'.  Some have seen the initial element <medu-> in this town name as
representing the stem of a local word for 'lead', and they have proposed
that Basque <berun> derives from the same source -- not unreasonably, since
variation between /d/ and /r/ between vowels is commonplace in Basque.

Finally, to Douglas Kilday's proposal that <berun> derives from a possibly
Ligurian word of the form *<blum->.  There are some serious phonological
problems with this.

First, as noted above, it seems likely that <beraun> is the more
conservative form of the Basque word -- not helpful.

Second, that /m/.  It is true that Pre-Basque had no */m/.  But, in early
borrowings from Celtic and Latin, an original /m/ was usually retained in
Basque, and not converted to */b/.  Of course, if the source /m/ was
word-final, then this could not have happened, since Basque has never
permitted word-final /m/.  In this case, we don't know what would have
happened to
/m/, since there are no parallels.  (Latin word-final /m/ always appears in
Basque as zero, but then final /m/ was probably already gone from popular
Latin speech, or at least reduced to nasalization of the preceding vowel.)

Third, the initial cluster */bl-/.  Pre-Basque didn't merely "disfavor"
initial clusters: it prohibited them absolutely.  In borrowed words,
initial clusters were invariably eliminated in one way or another, as
indeed were plosive-liquid clusters in all positions.

In borrowings from Latin, initial clusters were resolved as follows.  (Bear
in mind that Latin initial /b p f/ were all rendered as Pre-Basque */b/.)
If the cluster was /Cl-/, then the C was lost, and only the /l/ remained.
Examples: Latin <flammam> 'flame' > Basque <lama> 'flame'; Latin <plumam>
'feather' > Basque <luma> 'feather'; Latin <florem> 'flower' > Basque
<lore> 'flower'; Latin <planum> 'flat' > Pre-Basque *<lanu> > modern Basque
<lau> 'flat', by the regular medieval loss of intervocalic /n/; Latin
<placet> 'it is pleasing' > Basque <laket> 'pleasing'; Latin <plantatum>
'planted' > Basque <landatu> 'planted', with a regular Basque voicing.  The
sole certain exception known to me is Latin <cletam> 'rustic gate' >
Pre-Basque *<geleta> > modern <gereta> 'rustic gate'.  This is not helpful,
since we would expect a *<blum-> to be borrowed as something like *<luN->.

In contrast, with an initial /Cr-/ cluster, both consonants were retained,
but a vowel was inserted between them.  However, the inserted vowel was
*always* an echo of the following vowel.  Examples: Latin <granum> 'grain'
> Pre-Basque *<garanu> > modern <garau>, <garaun> by other changes; Latin
<frontem> 'forehead' > Basque <boronde> 'forehead', again with a regular
Basque voicing; and see the case of <gereta> above.

This strategy continued well into the Romance period.  For example, early
Romance *<frecato> 'rubbed' (replacing classical <frictum>; note Castilian
<fregado>) > Basque <berekatu> ~ <perekatu> ~ <ferekatu> 'rubbed,
caressed', 'annoyed'.

So, the phonology is not right for a "Ligurian" *<blum-> yielding Basque
<ber(a)un>.  Of course, if the borrowing was very early, then Basque might
have been employing different strategies at the time for resolving
impermissible clusters, but there is no evidence for such a thing, and only
special pleading is available.

In sum, then, Douglas Kilday's proposal is not impossible, but it faces
serious difficulties, and I cannot see that it should be preferred to any
one of the several other proposals on the table.  At least all of those
proposals but one must be wrong, and very likely they are all wrong.  We
cannot tell, because we lack adequate evidence.  This, I think, is what
Joat Simeon was talking about.

Larry Trask
University of Sussex
Brighton BN1 9QH

larryt at

Tel: (01273)-678693 (from UK); +44-1273-678693 (from abroad)
Fax: (01273)-671320 (from UK); +44-1273-671320 (from abroad)

More information about the Indo-european mailing list