Linguistics discussion (split from new factions thread)

[Elmetiacos]

2. Lusitanian
First an apology because I don't know quite what ended up happening in EB1 with Lusitanian. It looks as if the project was abandoned before it was finished and Celtiberian names used instead.

I got a bit obsessed with this a while ago, particularly with a possible link between Lusitanian and Ligurian or Lepontic. The reason for this is the word PALA. It appears on an inscription, probably a memorial, usually classified as Lepontic. The inscription is written in an old Italic alphabet similar to Etruscan, on two staves with round heads; one reads slaniai : uerKalai : Pala and the other Tisiui : PiuoTialui : Pala. The two staves look like elongated Basque Hilarri.

We've got more writing in Lusitanian than in Dacian or Thracian, but still very little and we don't have as many place or personal names because it looks as if the Lusitanian rulers may have spoken Celtic. There are a whole three and a half Lusitanian inscriptions which are fairly easy to look up - they're even on Wikipedia. One says OILAM TREBOPALA INDO PORCOM LAEBO COMAIAM ICCONA LOIMINA OILAM USSEAM TREBARUNE INDI TAUROM IFADEM REUE which is usually translated as something like "A sheep (or ewe) for Trebopala and a pig for Laebo one of the same age for Iccona Loimina, a sheep of one year of age for Trebarune and a breeding bull for Reue." The inscription makes it hard to say Lusitanian is Celtic, because it's kept Indo-European P (porcom) which Celtic didn't (Irish orc "piglet") and the word for bull has the root tauro- instead of *tarwo- we'd expect from a Celtic language. Also there isn't a word for "and" like indi in any known Celtic language. In On The Indo-European Origin of Two Lusitanian Theonyms, Witczak thinks Trebopala is a goddess whose name means "Protectress of the Home" or "Protectress of the Tribe" and that Laebo should be *Lahebo, a dative plural of Laho, the equivalent of the Roman Lares. He also thinks (as do others) Iccona is a similar goddess to the Gaulish Epona.

But what if pala doesn't mean "protectress"? what if it's some sort of sacred stone like in the Ligurian/Lepontic inscriptions? Katia Maia-Bessa suggested this in her paper Syncrétisme Religieux dans la Lusitanie Romaine (I'm afraid it's in French) but without actually citing Ligurian or Lepontic. It got me thinking about a link and the possibility that both languages are "Old European" - Indo-European languages which in the early Iron Age got replaced by Celtic.

Apart from the inscriptions the other source for Lusitanian is dedications to local deities from the Roman period and usually in Latin. Unfortunately none of them say "pala" anywhere. However, Ligurian was said to have been much more widely spoken before the Etruscans, Gauls and Romans moved in, and sometimes a sign of a place having once been Ligurian is said to be a name ending with -sc-. There are only two -sc- names in Roman Iberia; Metallum Vipascense and Magasca and they're both in Lusitanian territory. This is getting a long post and I should start talking about what use it could be in EB2...

[Elmetiacos]

For deities, we've got Iccona as a possible horse goddess and following Witczak the Laho as a local spirit and Reue as the Lusitanian Zeus, then there's the two most popular dedications to the god Endovellicus (94 dedications) and the goddess Bandua (28), plus two probable war gods (possibly Celtic introductions) Coronos and Netonos. Can we get anything useful from the other inscriptions?

We might glean some Lusitanian grammar: o-stem accusatives seem to end in -om: taurom, porcom, angom. doenti probably means "they gave" (Witczak says not, but IMHO he's just tying himself in knots trying to say all I-E initial Ds become Rs - no problem if only Dy- becomes R) so you've got 3rd person plural -nt- as you'd expect (not that EB needs to conjugate verbs... even I'm not mad enough to try and do a Lusitanian voicemod) a-stem datives seem to end in -a (Iccona, Trebopala) and if Witczak's correct, dative plurals in -ebo and i-stem accusatives in -e. On the Arroyo de la Luz stone, carlae may be a locative "at Carla". But that's not much good - we need nominatives, genitives and plurals for a naming language. On Lamas de Moledo, doenti is followed by veaminicori which is a good candidate for a plural - if -cori is a cognate with Celtic *korio or just a borrowing it should be an o-stem noun. We've also got those -oi endings in caelobricoi, reaicoi petranoi... are those genitive plurals? In the Latin dedications there are lots of -aico and -aeco names Bandi Oilienaico - Bandua of the Sheep?

It might be possible to attempt more translations... As a piece of baseless speculation, Lamas de Moledo's REAICOI PETRANOI could contain a pre-Basque loan *lercoin and PIE *pet- to mean "of the cranes with spread wings" while ANGOM is probably a lamb. In Arroyo de la Luz I & II the inscription says: AMBATVS SCRIPSI CARLAE PRAISOM SECIAS ERBA MVITIE AS ARIMO PRAESONDO SINGEIETO INI AVA INDI VEA VN INDI VEDAGA ROM TEVCAECOM INDI NVRIM INDI VDEVEC RVRSENCO AMPILVA INDI LOEMINA INDI ENV PETANIM INDI ARIMOM SINTAMOM INDI TEVCOM SINTAMO. The first two words are Latin. The rest has the usual -OM endings, presumably accusatives. PRAISOM and PRAESONDO might be from PIE kwreiH- and be to do with buying and selling, but that spoils Iccona being Epona because it means labialisation of kw (a feature of Ligurian). SECIAS might derive from PIE segh- or seHg- so it might mean "fixed", "held" or "sought out". ARIMO suggests "highest". If SINGEIETO is related to Celtic Singidunon it might be something to do with falcons, so the next part INI AVA might mean "one bird", but this is guesswork and it doesn't provide anything much useful for EB2 except a possibility of a unit name - Arimi for elite cavalry. Maybe. But Arimo could just as easily refer to a deity. So there aren't really any useful words to be got apart from a possible Celtic loan net- for "warrior" that could form a basis for unit names, and a name for an infantry barracks Trebo Netenaicoi. If only the Lusitanians had sacrificed some horses, things would be that much easier...

a Lusitanian o-stem declension:
sing. / plu.
N Tauro Tauri
V *Taure *Tauroe
A Taurom *Tauro
G *Tauri Tauroi
D Tauro Taurebo
L *Tauroe *Tauroi

[Elmetiacos]

It might be possible to attempt more translations... As a piece of baseless speculation, Lamas de Moledo's REAICOI PETRANOI could contain a pre-Basque loan *lercoin and PIE *pet- to mean "of the cranes with spread wings" while ANGOM is probably a lamb.

Pshaw! I didn't look up enough Indo-European. Following Witczak, REAICOI is another of those collective genitive -aicoi plurals formed from *diewo- with the shift from d > r; PET- could well come from PIE *peit- so it probably means "of the gods of plenty". The whole inscription I'd read as something like:
"The Veaminicori give a shorn upland lamb on the great mound of all the gods of plenty (and) a pig to Jupiter Caelobrigae."
or:
"The Veaminicori give a shorn lamb from the pens to the great vessel of all the gods of plenty (and) a pig to Jupiter Caelobrigae."
I'll explain why if anyone actually cares

Here's something more useful; sound changes between proto-Indo-European and Lusitanian collected from various authors (these) or evident from the corpus of inscriptions [these], which will make it easier to make new words for things with "reasonable" accuracy:
PIE > Lusitanian
p > p [porcom]
b > b [Laebo]
bh > f (Gorrochategui)
t > t [taurom]
d > d [doenti]
dh > θ? (García Alonso)
k' > c [porcom]
g' > g by analogy; L. is centum
g'h > hj? (García Alonso)
k > c [-aicoi]
g > g/c [Attaegina/Attaecina?]
gh > h/χ? (García Alonso)
kw > p [praesondo] labialisation
gw > b as above
k'w > cc [Iccona]
g'w > g? ng?
gwh > w?v? [vea(un)/veaminicori?] <gwhedh- = keep*
s > s [praisom, secias]
w- > b- (García Alonso)

dy > r (Witczak) (he actually thinks all Ds become Rs, I think not)
CrC > CurC (Prósper)
VrV > VhV [La(h)ebo] (Witczak)
e- > i- [Iccona, ifadem]

*which makes VEA-MIN- (<*veath-min-) equivalent to Celtiberian kombalkez, usually translated as "decreed". VEAMINICORI is possibly "the host of the proclamations", some sort of decision making assembly.

[Ibrahim]

interesting work-I hope some EB member will take a look.

[Elmetiacos]

A new piece of writing in Lusitanian was discovered at Ribeira da Venda in 2008 and a paper on it (only in Portuguese, alas) has been published on it:http://ifc.dpz.es/recursos/publicaci...rneiroetal.pdf.
The inscription reads:
[- - - - - - - -] XX • OILAM • ERBAM
HARASE • OILA • X • BROENEIAE • H
OILA • X • REVE AHARACVI • T • AV [...]
IFATE(or IEATE) • X • BANDI HARACVI AV [....]
MVNITIE CARIA(or CARLA) CANTIBIDONE •
APINVS • VENDICVS • ERIACAINV[S]
OVOVIANI [?]
ICCINVI • PANDITI • ATTEDIA • M • TR
PVMPI • CANTI • AILATIO

It's being discussed on the Continental Celtic list. I've also been told by someone who knows a lot more than me that Iccona probably isn't Epona, and that k'w becoming cc and not p is unlikely. Also, this new inscription has harase and (a)haracui, so -r- between vowels (above) can only become -h- before E.

[cmacq]

and that k'w becoming cc and not p is unlikely.

Well then...

there goes my surname.


CmacQ

[Elmetiacos]

Well then...

there goes my surname.

CmacQ

It's only Lusitanian problem and I doubt your surname is Lustanian. The problem for this language is that we've got plenty to suggest labialised velars becoming labials (kw > p and gw > b, in other words) as they did in Greek and 'P-Celtic' Gaulish, but didn't in Latin, 'Q-Celtic' Irish or Celtiberian, such as PRAESONDO and PVMPI. A goddess called Iccona who's the same as the Gaulish Epona appears to go against this, with the only explanation being that Lusitanian changed kw into p but not k'w (an palatalised labialised velar... did I just say that?) However... apparently that is unlikely because most centum Indo-European languages merged the k'w and kw sounds a long time before the kw > p shift (which took place quite late - 1st half of the 1st millennium BC) so Lusitanian would have to have been weird and either kept the sounds separate or undergone the shift very early.