Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

a few questions about Messapic language

45 views
Skip to first unread message

VK

unread,
Jan 5, 2008, 6:10:42 AM1/5/08
to

In my PIE studies I never dealt before with Messapic (Messapian)
language first because of a scanty amount of the available materials
and secondly because the available materials didn't point for me to IE
origin - though didn't exclude it explicitly. Suddenly I have been
asked to express my opinion on the matter by one higher education
body. I have just started on the material, so wondering if some quick
help could be obtained from the group's regulars.

1) Messapic as an Indo-European language.
How exactly it was arrived to this conclusion? I went from the current
publications back to the beginning and the only kind of analysis I
found so far is one sentence from Mommsen, "The History of Rome", Book
1 Chapter 2: "The genitive forms, -aihi- and -ihi-, corresponding to
the Sanscrit -asya- and the Greek --oio--, appear to indicate that the
dialect belongs to the Indo-Germanic family."
AFAICT ever since the middle of XIX century no further attempts had
been made to identify the language family. There is a great amount of
works where the authors are trying to identify the exact position of
Messapic among IE languages and/or to analyze language features: but
with the IE nature of the Messapic pre-postulated as some "common
knowledge" fact: see for instance R.M.Bechtel, "The Messapic Klaohizis
Formula", 1937 or R.Giacomelli, "Written and Spoken Language in Latin-
Faliscan and Greek-Messapic", 1979.
Am I missing some important publication on the subject?

2) Does anyone have "Corpus Inscriptionum Messapicarum" by Ribezzo?
This book is out of print with 3 weeks reprint delivery the earliest
in the available bookstores. I could expedite my work if I could take
a look at a scanned copy of the infamous Vaste Inscription first
reported by Momssen (Corpus Inscriptionum Messapicarum 149)

3) Does anyone have: Carlo De Simone, "Iscrizione messapiche della
grotta della poesia", 1988 - I have the same 3 weeks delay for the
same reason as above.

Dušan Vukotić

unread,
Jan 5, 2008, 6:33:38 AM1/5/08
to

Mission Impossible! I cannot remember exactly which part. :-)

Why don't you try Masovian/Mazovian or Sacher Masochian instead?

DV

VK

unread,
Jan 5, 2008, 7:14:03 AM1/5/08
to
On Jan 5, 2:33 pm, "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:
> Mission Impossible! I cannot remember exactly which part. :-)

...? You mean I cannot get those books I have mentioned? Sure I can,
they are already ordered, I just wanted to expedite the process a
little bit: with final check when the books will be received.

> Why don't you try Masovian/Mazovian or Sacher Masochian instead?

Sorry, but this comment escapes me as well. Maybe you did not
understand: I am not asked to decrypt Messapian, with the current
amount of material it is pretty much impossible. I am just asked to
summarize the known facts and paleographies.


grap...@www.com

unread,
Jan 5, 2008, 7:22:04 AM1/5/08
to
On Jan 5, 12:10 pm, VK <schools_r...@yahoo.com> wrote:

Seems to me that you are trying to reinvent the wheel!.. The opinion
that Messapic is an IE language is very general. For instance, E.
Pulgram in his book "The Tongues of Italy" wrote : "The Messapic
language is no doubt IE , and we may also accept the judgment that it
is of Illyrian type as long as we are cognizant of the uncertain and
limited real content of the term "Illyrian".." Same opinion of R.S.
Conway, J. Whatmouth and S.E. Johnson in "The Prae-Italic Dialects of
Italy". Etc.
As for the "Vaste Inscription", you may find its transcription in
Wikipedia "Massapian Language".
Regards
grapheus

VK

unread,
Jan 5, 2008, 7:50:00 AM1/5/08
to
On Jan 5, 3:22 pm, "graph...@www.com" <graph...@www.com> wrote:
> Seems to me that you are trying to reinvent the wheel!.. The opinion
> that Messapic is an IE language is very general. For instance, E.
> Pulgram in his book "The Tongues of Italy" wrote : "The Messapic
> language is no doubt IE , and we may also accept the judgment that it
> is of Illyrian type as long as we are cognizant of the uncertain and
> limited real content of the term "Illyrian".." Same opinion of R.S.
> Conway, J. Whatmouth and S.E. Johnson in "The Prae-Italic Dialects of
> Italy". Etc.

I am not denying IE nature of Messapic - nor taking it as a given
fact. I am just asking what exact structural, lexical, morphological
and phonetical studies have been conducted, by what authors and in
what publications. Having such works pointed out would indeed save my
time from reinventing the wheel. I am in the possession of a micro
copy of
Conway, Whatmought & Johnson, The Prae-Italic Dialects of Italy, Vol.
2, The Raetic, Lepontic, Gallic, East-Italic Messapic and Sicel
Inscriptions, 1933

In the language family appertaining aspect it is the same problem as
spelled in my original post: the IE nature is pre-postulated as some
common knowledge fact and all further conclusions are going from that
point onward. In some publications the original Mommsen's work is
referred: but again Mommsen himself gave just a single sentence I
already quoted based on his own word division and interpretation of
the Vaste Inscription. This way so far I am in a kind of circular
reference loop between the involved authors and I'd like to break it
with a reputable source.

> As for the "Vaste Inscription", you may find its transcription in
> Wikipedia "Massapian Language".

I need to see the original drawing as made by Mommsen as the
inscription itself alas lost. The Wikipedia copy is altered by the
publisher. The original was consisting out of eight lines without word
division. All available online copies I could find so far are having
proprietary word breaks with line breaks removed.

grap...@www.com

unread,
Jan 5, 2008, 11:48:54 AM1/5/08
to
On Jan 5, 1:50 pm, VK <schools_r...@yahoo.com> wrote:
> On Jan 5, 3:22 pm, "graph...@www.com" <graph...@www.com> wrote:
>
> > Seems to me that you are trying to reinvent the wheel!.. The opinion
> > that Messapic is an IE language is very general. For instance, E.
> > Pulgram in his book "The Tongues of Italy" wrote : "The Messapic
> > language is no doubt IE , and we may also accept the judgment that it
> > is of Illyrian type as long as we are cognizant of the uncertain and
> > limited real content of the term "Illyrian".."  Same opinion of R.S.
> > Conway, J. Whatmouth and S.E. Johnson in "The Prae-Italic Dialects of
> > Italy". Etc.
>
> I am not denying IE nature of Messapic - nor taking it as a given
> fact. I am just asking what exact structural, lexical, morphological
> and phonetical studies have been conducted, by what authors and in
> what publications. Having such works pointed out would indeed save my
> time from reinventing the wheel. I am in the possession of a micro
> copy of
> Conway, Whatmought & Johnson, The Prae-Italic Dialects of Italy, Vol.
> 2, The Raetic, Lepontic, Gallic, East-Italic Messapic and Sicel
> Inscriptions, 1933
>
> In the language family appertaining aspect it is the same problem as
> spelled in my original post: the IE nature is pre-postulated as some
> common knowledge fact and all further conclusions are going from that
> point onward.

"Pre-postulated" ?.. Surely not ! There are very good motives to
classify Messapic as IE, or better said for the Kretschmerians as
"proto-IE in a kretschmerian sense". You may find a list of these
motives in any good encyclopedia, for instance in :
<http://www.search.com/reference/Messapian_language>

> In some publications the original Mommsen's work is
> referred: but again Mommsen himself gave just a single sentence I
> already quoted based on his own word division and interpretation of
> the Vaste Inscription. This way so far I am in a kind of circular
> reference loop between the involved authors and I'd like to break it
> with a reputable source.
>
> > As for the "Vaste Inscription", you may find its transcription in
> > Wikipedia "Massapian Language".
>
> I need to see the original drawing as made by Mommsen as the
> inscription itself alas lost.

Well, I'm afraid that you will have some trouble !.. Generally,
scholars in the field rely upon the "Corpus Inscriptionem
Messapicorum" (n° 149), for this inscription as for the others.

grapheus

Dušan Vukotić

unread,
Jan 5, 2008, 4:41:39 PM1/5/08
to
On Jan 5, 1:14 pm, VK <schools_r...@yahoo.com> wrote:

Don't take me serious... I was joking. What one can say when one hears
that "Mesapic is close to Illyrian" or even a "dialect of Illyrian"
and, at the same time one knows that no one has ever (in an exact
scientific way) determined what that "Illyrian" really represents.

Unlike so-called Illyrian (which has not left any written trace),
Messapic left a few inscriptions; but, as far as I know, those
inscriptions has not yet been deciphered. It means that one existed
but undeciphered portion of language (called Messapic) is a member of
a larger ghost family.

It is ridiculous! Messapic was written in one uninterrupted row; there
is no signs of interpunction - http://www.telemaco.unibo.it/rom/italia/img05/messapi2.jpg
- and no one can be sure how exactly to read this inscription.

"klohi zis anthos thotorridas ana aprodita apa ogrebis"

Look at the above Messapic sentence. If it were an IE language, we
would decipher it without many difficulties. Maybe it is an IE
language but we have wrongly separated the words. For instance (I
suppose it is a grave inscription?), the two words at the end could be
read as one word “apaogrebis” and we would have received the Greek
word απόκρυφος (hidden; κρύπτη crypt underground burial chamber);
cf. Serbian grob (grave), pogreb (burial);
In addition, the second name name of Ἀθηνα was Παρθένος (Virgin); cf.
Latin partus (birth), Serb. poroditi/ porod (give birth), Eng. birth.
Then, why the following string of words "ana aprodita apa ogrebis"
would not be translated as "a buried maiden"?
Of course, this is just a small "experiment", but my translation could
be equally valid as any other else, because no one knows for sure what
kind of message is on that table.

DV

Richard Wordingham

unread,
Jan 5, 2008, 5:26:37 PM1/5/08
to
"Dušan Vukotić" <dusan....@gmail.com> wrote:

> In addition, the second name name of Ἀθηνα was Παρθένος (Virgin); cf.
> Latin partus (birth), Serb. poroditi/ porod (give birth), Eng. birth.
> Then, why the following string of words "ana aprodita apa ogrebis"
> would not be translated as "a buried maiden"?

Aphrodite = Venus, not a virgin.

Richard.


grap...@www.com

unread,
Jan 5, 2008, 5:27:01 PM1/5/08
to
On Jan 5, 10:41 pm, "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:
> On Jan 5, 1:14 pm, VK <schools_r...@yahoo.com> wrote:
>
> > On Jan 5, 2:33 pm, "Du¹an Vukotiæ" <dusan.vuko...@gmail.com> wrote:
>
> > > Mission Impossible! I cannot remember exactly which part. :-)
>
> > ...? You mean I cannot get those books I have mentioned? Sure I can,
> > they are already ordered, I just wanted to expedite the process a
> > little bit: with final check when the books will be received.
>
> > > Why don't you try Masovian/Mazovian or Sacher Masochian instead?
>
> > Sorry, but this comment escapes me as well. Maybe you did not
> > understand: I am not asked to decrypt Messapian, with the current
> > amount of material it is pretty much impossible. I am just asked to
> > summarize the known facts and paleographies.
>
> Don't take me serious... I was joking. What one can say when one hears
> that "Mesapic is close to Illyrian" or even a "dialect of Illyrian"
> and, at the same time one knows that no one has ever (in an exact
> scientific way) determined what that "Illyrian" really represents.
>
> Unlike so-called Illyrian (which has not left any written trace),
> Messapic left a few inscriptions; but, as far as I know, those
> inscriptions has not yet been deciphered. It means that one existed
> but undeciphered portion of language (called Messapic) is a member of
> a larger ghost family.
>
> It is ridiculous! Messapic was written in one uninterrupted row; there
> is no signs of interpunction -http://www.telemaco.unibo.it/rom/italia/img05/messapi2.jpg

> - and no one can be sure how exactly to read this inscription.
>
> "klohi zis anthos thotorridas ana aprodita apa ogrebis"
>
> Look at the above Messapic sentence. If it were an IE language, we
> would decipher it without many difficulties. Maybe it is an IE
> language but we have wrongly separated the words. For instance (I
> suppose it is a grave inscription?), the two words at the end could be
> read as one word “apaogrebis” and we would have received the Greek
> word απόκρυφος (hidden; κρύπτη crypt underground burial chamber);
> cf. Serbian grob (grave), pogreb (burial);
> In addition, the second name name of  Ἀθηνα was Παρθένος (Virgin); cf.
> Latin partus (birth), Serb. poroditi/ porod (give birth), Eng. birth.
> Then, why the following string of words "ana aprodita apa ogrebis"
> would not be translated as "a buried maiden"?
> Of course, this is just a small "experiment", but my translation could
> be equally valid as any other else,

This statement is not "scientifically" correct, because in Science,
one has to consider PROBABILITIES, not just POSSIBILITIES. Well, in
this case, there is of course a POSSIBILITY to cut the words in
another way than "ana aprodita apa ogrebis", but the PROBABILITY is
obviously in favour of the reading "ana aprodita apa ogrebis" a)-
because writings -aa- and -ao- are VERY RARE, whatever the language
b)- the historical data are in favour of a language "Proto-IE in the
kretschmerian sense".

Regards
grapheus

Dušan Vukotić

unread,
Jan 5, 2008, 5:42:38 PM1/5/08
to
On Jan 5, 11:26 pm, "Richard Wordingham" <jrw0...@yahoo.co.uk> wrote:

You misunderstood me. I said that it might be Parthenos (Athena;
Perthenos - Partus - Porod/iti - Birth - A/Prodit) and not Aphrodita
at all.
Hmmm... If we look those words more carefully it seems that Aphrodita
and Parthenos may have the same place of origin. ;-)

All this is just a guessing (most probably far away from truth) but
equaly valid to any other hypothesis anyone else have made on this
subject.

DV

Dušan Vukotić

unread,
Jan 5, 2008, 6:07:03 PM1/5/08
to
> grapheus- Hide quoted text -
>
> - Show quoted text -

There are no probabilities without possibilities. I do not see that in
this specific case anyone has arrived near any possible probability. ;
-)
I think diphthongization is not unusual in Greek, not even
reduplication (παά all, the whole)

DV

Brian M. Scott

unread,
Jan 5, 2008, 6:08:04 PM1/5/08
to
On Sat, 05 Jan 2008 22:26:37 GMT, Richard Wordingham
<jrw...@yahoo.co.uk> wrote in
<news:xuTfj.28235$KC3....@newsfe6-gui.ntli.net> in
sci.lang:

> "Dušan Vukotić" <dusan....@gmail.com> wrote:

>> In addition, the second name name of Ἀθηνα was Παρθένος
>> (Virgin); cf. Latin partus (birth), Serb. poroditi/
>> porod (give birth), Eng. birth.

<Parthénos>, <partus>, <poroditi>, and <birth> are mutually
unrelated. <Parthénos> has not been established to have any
relatives; <partus> is from a PIE *perh3- 'to produce,
procure'; <birth> is from PIE *bHer- 'to carry'; and
<poroditi> is a prefixed derivative of PSl *rodìti 'give
birth, bear (fruit)', for which Derksen reconstructs PIE
*(h)rodH-eie-.

>> Then, why the following string of words "ana aprodita apa
>> ogrebis" would not be translated as "a buried maiden"?

> Aphrodite = Venus, not a virgin.

Brian

grap...@www.com

unread,
Jan 5, 2008, 6:22:09 PM1/5/08
to

Of course !.. But I wanted to distinguish between a "possibility which
has a very small probability to be correct" and "a possibility having
a great probability to be correct".

> I do not see that in
> this specific case anyone has arrived near any possible probability. ;

Well, I am exactly of the contrary opinion !..

> I think diphthongization is not unusual in Greek, not even
> reduplication (παά all, the whole)

But the problem is once more a QUESTION of PROBABILITY. How many words
present in Greek a reduplication like -aa- ?.. And how many times you
get in a Greek sentence a word with a final -a followed by another
word beginning with an a- ?..
I pretend that the second possibility is at least 50 times more
frequent than the first. So the PROBABILITY is a lot higher for the
second hypothesis than for the first.

grapheus

Dušan Vukotić

unread,
Jan 5, 2008, 6:51:33 PM1/5/08
to
On Jan 6, 12:08 am, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
> On Sat, 05 Jan 2008 22:26:37 GMT, Richard Wordingham
> <jrw0...@yahoo.co.uk> wrote in
> <news:xuTfj.28235$KC3....@newsfe6-gui.ntli.net> in
> sci.lang:

>
> > "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:
> >> In addition, the second name name of  Ἀθηνα was Παρθένος
> >> (Virgin); cf. Latin partus (birth), Serb. poroditi/
> >> porod (give birth), Eng. birth.
>
> <Parthénos>, <partus>, <poroditi>, and <birth> are mutually
> unrelated.  <Parthénos> has not been established to have any
> relatives; <partus> is from a PIE *perh3- 'to produce,
> procure'; <birth> is from PIE *bHer- 'to carry'; and
> <poroditi> is a prefixed derivative of PSl *rodìti 'give
> birth, bear (fruit)', for which Derksen reconstructs PIE
> *(h)rodH-eie-.
>
> >> Then, why the following string of words "ana aprodita apa
> >> ogrebis" would not be translated as "a buried maiden"?
> > Aphrodite = Venus, not a virgin.
>
> Brian

Thanks Brainy, but I do not need anyone to read the books for me.
Derksen is wrong and therefore you are wrong too. None of the IE stem-
words begins with the initial "r"; it is absolutely impossible!

Originally, Serbo-Slavic roditi comes from poroditi. I hope you are
going to understand it better if I say that Serbian word bremenit/a
means pregnant. Both words are derivatives from the secondary Br-Gon-
Gon basis; i.e. from primeval basis Bel-Hor-Gon; bregnenit (Serb.
bremenit; from brenenit; cf. Serb. prinos crop). Also, Serbian breme
(weight or burdon! ;-)

Compare now burdon, bear, birth, pregnant, fruit and Serbian breme.
bremenit, porod, prinos; all words from the same Br-Gon basis.
Finally, what do you think about Serbian word brat (brother) and
English brother? Are they related or not? Serbian porodica (family),
porod (ofdpring), brat (brother), bratičina (brother's sister),
porodična (of family, attached to the family).

What are we going to do with your book reading: Slavic and Germanic
brothers are obviously related, Serbian brat (brother) is the word
related to the Serbian words porod (birth), and porodica (family).
You are a mathematician, use the given syllogism!

DV

Dušan Vukotić

unread,
Jan 5, 2008, 7:03:53 PM1/5/08
to

OK I did not say you were wrong. Nevertheless, science is neither
possibility nor probability. Linguistic is most similar to
mathematics: it means it is the same if you are thousands miles away
or just one milimetar close if you can not see or touch the object you
are looking for. ;-)

DV

Brian M. Scott

unread,
Jan 5, 2008, 7:34:09 PM1/5/08
to
On Sat, 5 Jan 2008 15:51:33 -0800 (PST), Dušan Vukotić
<dusan....@gmail.com> wrote in
<news:cb97c7f1-e107-4cf0...@v67g2000hse.googlegroups.com>
in sci.lang:

> On Jan 6, 12:08 am, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:

>> On Sat, 05 Jan 2008 22:26:37 GMT, Richard Wordingham
>> <jrw0...@yahoo.co.uk> wrote in
>> <news:xuTfj.28235$KC3....@newsfe6-gui.ntli.net> in
>> sci.lang:

>>> "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:

>>>> In addition, the second name name of  Ἀθηνα was Παρθένος
>>>> (Virgin); cf. Latin partus (birth), Serb. poroditi/
>>>> porod (give birth), Eng. birth.

>> <Parthénos>, <partus>, <poroditi>, and <birth> are mutually
>> unrelated.  <Parthénos> has not been established to have any
>> relatives; <partus> is from a PIE *perh3- 'to produce,
>> procure'; <birth> is from PIE *bHer- 'to carry'; and
>> <poroditi> is a prefixed derivative of PSl *rodìti 'give
>> birth, bear (fruit)', for which Derksen reconstructs PIE
>> *(h)rodH-eie-.

[...]

> Thanks Brainy, but I do not need anyone to read the books
> for me.

Possibly not; in that case, however, you need someone to do
your thinking for you.

> Derksen is wrong and therefore you are wrong too. None of
> the IE stem- words begins with the initial "r"; it is
> absolutely impossible!

There are quite a few, actually. This, however, may not be
one: Derksen's reconstruction allows for the possibility of
an initial laryngeal.

> Originally, Serbo-Slavic roditi comes from poroditi.

Apparently time runs backwards in your world.

[...]

> What are we going to do with your book reading: Slavic and
> Germanic brothers are obviously related,

True, though not because the relationship is 'obvious'.

> Serbian brat (brother) is the word related to the Serbian

> words porod (birth), [...]

No, it isn't.

It's entertaining, if a bit mind-boggling, to find a Slavic
speaker with no understanding of prefixes.

Dušan Vukotić

unread,
Jan 5, 2008, 8:09:40 PM1/5/08
to
On Jan 6, 1:34 am, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
> On Sat, 5 Jan 2008 15:51:33 -0800 (PST), Dušan Vukotić
> <dusan.vuko...@gmail.com> wrote in
> speaker with no understanding of prefixes.- Hide quoted text -

>
> - Show quoted text -

Porod is not originally prefixed as well as the Serbian word Brat
(brather) and Bratičina (brather's sister) are not prefixid as the
word Porodica (family) is not prefixed as the word Porodična is not
originally prefixed. The other thing is that today there are Slavic
prefixes po-, pri- sa- etc. If we were following your and Kriha's
logic then none of the Slavic words begining with pa-, pre-, sa-, iz,
od-, do-ka-, raz-, uz-, pod-, za- could be a cognate of some of the
Germanic or Romance words? Do not be sily!

DV

Brian M. Scott

unread,
Jan 5, 2008, 9:17:50 PM1/5/08
to
On Sat, 5 Jan 2008 17:09:40 -0800 (PST), Dušan Vukotić
<dusan....@gmail.com> wrote in
<news:30f75274-5b5c-41c0...@j20g2000hsi.googlegroups.com>
in sci.lang:

[...]

> If we were following your and Kriha's logic then none of
> the Slavic words begining with pa-, pre-, sa-, iz, od-,
> do-ka-, raz-, uz-, pod-, za- could be a cognate of some
> of the Germanic or Romance words? Do not be sily!

I couldn't possibly be as silly as your misunderstanding of
what Paul and I have said.

mb

unread,
Jan 5, 2008, 10:54:08 PM1/5/08
to
On Jan 5, 3:22 pm, "graph...@www.com" <graph...@www.com>
...

> > > b)- the historical data are in favour of a language "Proto-IE in the
> > > kretschmerian sense".

> Of course !.. But I wanted to distinguish between a "possibility which


> has a very small probability to be correct"

As a first example, the possiblity of such a monstrosity as ""Proto-IE
in the kretschmerian sense" comes to mind. Chances: nil.
And, of course, no definition available.

Paul J Kriha

unread,
Jan 5, 2008, 10:56:17 PM1/5/08
to
"Dušan Vukotić" <dusan....@gmail.com> wrote in message
news:30f75274-5b5c-41c0...@j20g2000hsi.googlegroups.com...

It seems that Serbian women deliver their babies first ("porodi")
and then go into labour (start "rodit").

How come Dusan doesn't propose as ur-form the verb "zroditi"
with his favoured "z-". Or perhaps "uroditi", "uroda" (harvest),
or "priroda" (nature). These must surely be old, old ur-forms.
Obviously! Nu? :-)

>> [...]
>>
>> > What are we going to do with your book reading: Slavic and
>> > Germanic brothers are obviously related,
>>
>> True, though not because the relationship is 'obvious'.
>>
>> > Serbian brat (brother) is the word related to the Serbian
>> > words porod (birth), [...]
>>
>> No, it isn't.
>>
>> It's entertaining, if a bit mind-boggling, to find a Slavic
>> speaker with no understanding of prefixes.- Hide quoted text -
>>
>> - Show quoted text -
>
>Porod is not originally prefixed as well as the Serbian word Brat
>(brather) and Bratičina (brather's sister) are not prefixid as the
>word Porodica (family) is not prefixed as the word Porodična is not
>originally prefixed. The other thing is that today there are Slavic
>prefixes po-, pri- sa- etc. If we were following your and Kriha's
>logic then none of the Slavic words begining with pa-, pre-, sa-, iz,
>od-, do-ka-, raz-, uz-, pod-, za- could be a cognate of some of the
>Germanic or Romance words? Do not be sily!

None of the Slavic words??? Who said that?

Sveti mati boga! Možda vi možete čitati, da ne razumeti!!!

pjk

>DV

Dušan Vukotić

unread,
Jan 6, 2008, 3:48:18 AM1/6/08
to
On Jan 6, 4:56 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
wrote:


> None of the Slavic words??? Who said that?

Kriha and his prefixed idée fixe :-)

Here is a simple example: is the Serbian word breg (hill; Ger. berg)
prefixed or not?
I am sure that both of you (Brian and you) would give the same answer
that that word was not prefixed. Of course, that answer is right.
Now, please, consider carefully the following Serbian string og words
"preći preko brega" (to go over the hill). Do you see anything unusual
about these words?
Is the word preći prefixed or not? Pre-ći (cross over, traverse), za-
ći (go behind), na-ići (come across), pri-ći (come closer); obviously,
following your logic this word is composed of pre (before, for, afore,
ere) + ići (go). What's happened here: the "disjoint" meaning of the
verb pre-ći doesn't make sense ("go before" is not even close to "go
across") On the other side, the Serbian word preko (over, beyond,
through, across) is the closest cognate of the verb preći (cross
over). You must be confused here because you cannot disjoin 'preko'.
In reality, the words preko and preći are coming from an earlier form
'pregoniti' (go over), which, from its side, is related to progoniti
(prosecute) or prognati (litterally "force someone to go over", "to
the other side"). Naturally, these Serbian words (progoniti and
pregoniti) can be disintegrated to pre- and pro- goniti (hunt, chase,
go)!

Now, give me a straight answer: is the Serbian word breg (hill; OCHS
брѣгъ, Russ. берег/bereg coast, brink; Avest. barǝzah mountain)
related to Germanic berg (hill) or brink (coast)?

DV

Dušan Vukotić

unread,
Jan 6, 2008, 4:44:43 AM1/6/08
to
On Jan 6, 4:56 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
wrote:

> It seems that Serbian women deliver their babies first ("porodi")


> and then go into labour (start "rodit").
>
> How come Dusan doesn't propose as ur-form the verb "zroditi"
> with his favoured "z-". Or perhaps "uroditi", "uroda" (harvest),
> or "priroda" (nature). These must surely be old, old ur-forms.
> Obviously! Nu?  :-)

Iz Bratstvo (brotherhood, fraternity) related to Porodstvo/Porodica
(family)?
Is Brother not related to brat (OSlav. братръ, Czech bratr)?
What about OCHS беременная, Serbian bremenit/a (pregnant) and Latin
praegnans (pregnant); all with the primary meaning of "carry", bear";
cf. Serb. preneti (carry over, convey).

Sveti mati boga! Možda vi možete čitati, da ne razumeti!!!

Bravo Kriha! This is the royal Serbian! You are speaking in the same
way as the (potential) Serbian king does.

Sveta Majko Božja! Možda vi možete čitati, a da ne razumete! :-)

DV

grap...@www.com

unread,
Jan 6, 2008, 4:48:34 AM1/6/08
to

I don't agree. When Linguistics is not just DESCRIPTION of
linguistical facts, but THEORY, it is like ALL the other sciences
except mathematics : BASED UPON PROBABILITIES. In etymology, for
instance, the choice between "it's a derived form" and "it's a
borrowing" is based upon PROBABILITY. A recent example: are Greek
<margarides> and Persian <marvârit> two words coming from an IE root
*marg- , or is the Greek word a borrowing from Persian ?..

grapheus

grap...@www.com

unread,
Jan 6, 2008, 4:50:13 AM1/6/08
to

You should better deal with your pizzas than intervening in matters
that you totally ignore !...
Go back to your furnace, please !

grapheus

Dušan Vukotić

unread,
Jan 6, 2008, 4:53:53 AM1/6/08
to

Sorry, Russ. беременная (gravid); OCHS брѣмѩ; also Serb. breme (load)
Another question Kriha, what do you think where the Czech word březí
(gravid) is coming from?

DV

Dušan Vukotić

unread,
Jan 6, 2008, 5:59:40 AM1/6/08
to

You found one of the hardest examples. :-) Maybe μαργαρτα (daisy) or
μαργαριτης a pearl (Persian 'marz' margine frontier); Serbian mrginja
border; German markieren; Serbian merenje (measurement), merkati,
meriti (measure): Greek μέτρον; Ger. merken (observe; to bookmark);
Eng. mark
Persian pearl is 'morvarid' (Lat. margarita) and measure is mizân
pl.mavâzin.
Obviously, Persian marvarit/morvarid is a loanword from Greek
μαργαριτης;

I hope, you caught the point? Do I need to tell you the rest?

DV

VK

unread,
Jan 6, 2008, 4:53:49 PM1/6/08
to
On Jan 6, 12:41 am, "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:
> Don't take me serious... I was joking. What one can say when one hears
> that "Mesapic is close to Illyrian" or even a "dialect of Illyrian"
> and, at the same time one knows that no one has ever (in an exact
> scientific way) determined what that "Illyrian" really represents.
>
> Unlike so-called Illyrian (which has not left any written trace),
> Messapic left a few inscriptions; but, as far as I know, those
> inscriptions has not yet been deciphered. It means that one existed
> but undeciphered portion of language (called Messapic) is a member of
> a larger ghost family.

I don't care about Illyrian for the moment - yet I knew that the word
"Illirian" will pop up somehow... A partial disclosure: in one oversee
country, in one respectable higher education establishment, one
student made a stand on the professor. The professor preferred to
double check a part of materials used for the online course. Himself
is a specialist in Caucasian ergatic languages, he knew me as a friend
and PIE specialist (Hittite primarily though), so he asked me to get
some more material for him. It is not some "joined committee
investigation" of any kind - everyone wants to keep it quiet and
private.

P.S. The offending part was in an online course largely based on
"Linguistics and the Teaching of the Less-Commonly Taught Languages"
by Brian D. Joseph
http://www.seelrc.org/glossos/issues/9/joseph.pdf
but it has no relation neither with the original course developer nor
with the university it was first used.

P.S.S. I personally already hate that day he called my and I was
stupid to say "yes".
Besides being put - indirectly - in some kind of natio-linguistical
rwars, I am spending ridiculously a lot of time - not counting money
to be reimbursed - for getting the most basic starting data. I am
really surprised how a scanty group of researchers could make such a
hideous disorder out of just 300 or less small inscriptions. I am just
glad that Co weren't on charge of Bogazkoy archives - or neither
Hittite nor Hatti could be studied. Say until now I cannot get, from
where the Vaste Inscription is coming from: from "Corpus Inscriptionum
Messapicarum" by Ribezzo, from the appendix to the Corpus by Fabretti,
or someone of them (who?) is reproducing something (what?) from
Mommsen? For a number of inscriptions no one of them can tell if it
was copied from a cave, a tombstone, a column, a pot or a la "Mene,
tekel, upharsin" had been seen as fire letters in the air... :-\
OK, sorry for bias, I guess they still did a great job for their time.

With the Vaste Inscription still pending for check out, two other most
used inscriptions seems can be confirmed as still existing, authentic
and properly transliterated:

1) from De Simone expedition to Grotta della Poesia, Melendugno,
Lecce:

klauhizis
dekiasartahias
thautouriandirahho
dausapistathivinaihi

2) Galatina inscription on a stone plate (tombstone ??) currently kept
in the museum of Bologna - thanks to Dušan for the museum link:

klohizisanthosthot
orridasanaaprodi
taapaogrebis

Odysseus

unread,
Jan 7, 2008, 1:49:17 AM1/7/08
to
In article
<263b26b1-42db-4cfc...@41g2000hsy.googlegroups.com>,
"grap...@www.com" <grap...@www.com> wrote:

<snip>


>
> You should better deal with your pizzas than intervening in matters
> that you totally ignore !...

I believe most native speakers of English will agree that intervening in
a matter and ignoring it are mutually exclusive actions.

I suspect you've been caught by an etymological fallacy, or a _faux ami_
perhaps: whatever its history, "to ignore" isn't idiomatically used with
the meaning "to be ignorant (of/about)"; it indicates absence of
attention rather than lack of knowledge. Offhand I can't think of a
transitive verb meaning "to be ignorant of".

--
Odysseus

Paul J Kriha

unread,
Jan 7, 2008, 2:25:58 AM1/7/08
to
"Dušan Vukotić" <dusan....@gmail.com> wrote in message
news:70c8d820-109e-49ea...@e4g2000hsg.googlegroups.com...

>On Jan 6, 4:56 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
>wrote:
>
>> It seems that Serbian women deliver their babies first ("porodi")
>> and then go into labour (start "rodit").
>>
>> How come Dusan doesn't propose as ur-form the verb "zroditi"
>> with his favoured "z-". Or perhaps "uroditi", "uroda" (harvest),
>> or "priroda" (nature). These must surely be old, old ur-forms.
>> Obviously! Nu? :-)
>
>Iz Bratstvo (brotherhood, fraternity) related to Porodstvo/Porodica
>(family)?

dunno.... For them to be cognates, "brother" would have to be
related to "rod". I don't feel like researching it right now.

>Is Brother not related to brat (OSlav. братръ, Czech bratr)?
>What about OCHS беременная, Serbian bremenit/a (pregnant) and Latin
>praegnans (pregnant); all with the primary meaning of "carry", bear";
>cf. Serb. preneti (carry over, convey).
>
> Sveti mati boga! Možda vi možete čitati, da ne razumeti!!!
>
>Bravo Kriha! This is the royal Serbian! You are speaking in the same
>way as the (potential) Serbian king does.
>
>Sveta Majko Božja! Možda vi možete čitati, a da ne razumete! :-)

I was just guessing here, employing a kinda panslavic extrapolation.
I have never learned any Serbian at all. I've never even looked
at Serbian grammar, so I was really guessing at all the declensions.
Even then I see could have done a better job of it.

"Sveti" is a really silly mistake, of course it should be a feminine adjective.
"Majko Boz^ja" needs more specific Serb. knowledge and at least
a Serb. dictionary which i don't have.
I should have got the "razumete", it's a run of the mill S.Slavic second
person plural instead of my silly infinitive.
And the "a"? I wasn't sure whether I should say "a" or "i".
I wasn't sure if "a" didn't have a completely different meaning as it
does in Russian.

pjk

Paul J Kriha

unread,
Jan 7, 2008, 2:38:14 AM1/7/08
to
"Dušan Vukotić" <dusan....@gmail.com> wrote in message
news:a117fd9e-1e16-4e1f...@v46g2000hsv.googlegroups.com...

It's an adjective related to nouns břéme^ and břemeno (load, weight).

It's applied only to cattle (possibly deer), never ever to people
or other animals like dogs.
(The month březen is the time of a year for the cows go gravid).

Different expressions apply to other animals and quite separate
set of expressions to humans.

pjk

Dušan Vukotić

unread,
Jan 7, 2008, 4:27:48 AM1/7/08
to
On Jan 7, 8:38 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
wrote:
> "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote in message

Is it not the same meaning as in Latin pregnans -antis? Do you really
not see that both words (prgnant & beremen) have been derived from the
same ur-basis?

> It's applied only to cattle (possibly deer), never ever to people
> or other animals like dogs.
> (The month březen is the time of a year for the cows go gravid).

It doesn't matter at all - woman, cow or some other female animal -
they are all gravid (bremenit/e or pregnant). In Serbian, the use of
the notion 'bremenita' is limited only to human being

There are Slavic words preneti/prenositi (Czech přenést, přenášet
carry across) that is also prefixed (pre-) and nositi (caryy, bear);
therefrom the Serbian word noseća (pregnant); i.e. noseća and
bremenita are the synonimous words (pregnant, gravid). Now compare
English pregnant, bring, bear, fruit, birth and on one side to the
Serbian bremenita (pregnant), prineti (bring), prinos (crop) and porod
(birth):

Finnaly, I hoped that you and Brian would be insightful enough to
grasp that the Latin pregnatus (hence pregnans) is also a prefixed
word (pre + gnatus/natus). You both disappointed me a little bit...
Nevertheless, I still respect your opinion and your capacity of sound
judgement.

DV

Dušan Vukotić

unread,
Jan 7, 2008, 4:53:07 AM1/7/08
to
On Jan 7, 8:25 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
wrote:
> "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote in message

You've done your translation amost perfectly, because we can also
say : Sveta Mati Božja!; instaed of "Sveta Majko" that is more
frequent. In addition, there is a third variant of the name Saint
Mother of God (the Holy Virgin) in Serbian - "Sveta Matero Božja!"

For one who has never read/spoke Serbian your translation is more than
ingenious. Once again: my congratulations!

DV

Douglas G. Kilday

unread,
Jan 7, 2008, 10:33:48 PM1/7/08
to
"Brian M. Scott" wrote:
>
> [...]

>
> <Parthénos>, <partus>, <poroditi>, and <birth> are mutually
> unrelated.  <Parthénos> has not been established to have any
> relatives; <partus> is from a PIE *perh3- 'to produce,
> procure'; <birth> is from PIE *bHer- 'to carry'; and
> <poroditi> is a prefixed derivative of PSl *rodìti 'give
> birth, bear (fruit)', for which Derksen reconstructs PIE
> *(h)rodH-eie-.

Eric P. Hamp, in _Homenaje Tovar_ (Madrid 1972) 177-180, derives
<parthénos> from PIE *bhr.g^hwéno- originally 'full-grown', from
*bherg^h- 'to increase, grow'. Since many of the reflexes of this
verb in other languages refer to hills or high points, it seems at
least as plausible to me that the sense of <parthénos> was something
like 'exalted' or 'protected' (cf. Eng. "virgin on a pedestal"),
before acquiring its historical meaning of 'virgin, maiden'.

Douglas G. Kilday

unread,
Jan 7, 2008, 10:46:59 PM1/7/08
to

What he means is Kretschmer's "Protindogermanisch", for an exposition
of which you will have to dig out some of the earliest volumes of
_Glotta_. It was superseded by Sturtevant's "Indo-Hittite", then by
the recognition of an Anatolian branch of Indo-European. I very much
doubt that Paul Kretschmer, who lived until 1956, would have agreed
with Grapheus's fossilized ragtime-era view of IE linguistics. G. has
repeatedly shown himself to be a Rip van Winkle on this issue.
Perhaps he guzzled too much bathtub gin, tried to dance the Charleston
on the running board of his Duesenberg, and landed on his head?

Douglas G. Kilday

unread,
Jan 7, 2008, 11:23:41 PM1/7/08
to
VK wrote:
>
> [...]
>
> I am not denying IE nature of Messapic - nor taking it as a given
> fact. I am just asking what exact structural, lexical, morphological
> and phonetical studies have been conducted, by what authors and in
> what publications. Having such works pointed out would indeed save my
> time from reinventing the wheel. I am in the possession of a micro
> copy of
> Conway, Whatmought & Johnson, The Prae-Italic Dialects of Italy, Vol.
> 2, The Raetic, Lepontic, Gallic, East-Italic Messapic and Sicel
> Inscriptions, 1933

Then see "Notes on the Grammar of the Messapic Inscriptions" on pp.
594-610. Enough material is presented that you should be able to show
skeptics that Messapic is indeed an Indo-European language.

> In the language family appertaining aspect it is the same problem as
> spelled in my original post: the IE nature is pre-postulated as some
> common knowledge fact and all further conclusions are going from that
> point onward. In some publications the original Mommsen's work is
> referred: but again Mommsen himself gave just a single sentence I
> already quoted based on his own word division and interpretation of
> the Vaste Inscription. This way so far I am in a kind of circular
> reference loop between the involved authors and I'd like to break it
> with a reputable source.

The presentation in PID is not circular. The authors disagree with
Mommsen in that they regard -ihi (with its variant spellings) as
cognate with Latin -i: (here the inserted -h- merely indicates
length), not with the other genitive suffix reflected in Faliscan -
osio, Homeric -oio, Sanskrit -asya.

mb

unread,
Jan 8, 2008, 12:18:53 AM1/8/08
to

I happen to be aware of what he does *not really mean, just because
those Kretschmer texts, then already rather obsolescent, were
compulsory part of some boring seminar in Geneva waay back in time.
Apart from the fact, as you well say, that Faucounau's "Kretschmer"
has nothing left to do with the good man of Zuerich, what I am really
objecting to is anyone's presenting his own interpretation (that
accomodates any and all etyma for anything at all) as if it were
somehow based on this obsolete and mostly forgotten speculation.

grap...@www.com

unread,
Jan 8, 2008, 5:35:28 AM1/8/08
to
On Jan 8, 4:46 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
> mb <azyth...@yahoo.com> wrote:
> > "graph...@www.com" <graph...@www.com>
> > ...
>
> > > > > b)- the historical data are in favour of a language "Proto-IE in the
> > > > > kretschmerian sense".
> > > Of course !.. But I wanted to distinguish between a "possibility which
> > > has a very small probability to be correct"
>
> > As a first example, the possiblity of such a monstrosity as ""Proto-IE
> > in the kretschmerian sense" comes to mind. Chances: nil.
> > And, of course, no definition available.
>
> What he means is Kretschmer's "Protindogermanisch", for an exposition
> of which you will have to dig out some of the earliest volumes of
> _Glotta_.

Yes. But don't count upon the IGNORANT Italo-American kook Arythos to
be able to know that !...

>  It was superseded by Sturtevant's "Indo-Hittite",

WRONG !.. Sturtevant's "Indo-Hittite" is ANOTHER definition of "proto-
IE", and this is why I am always mentionning "proto-IE in the
Kretschmerian sense" when I talk about Kretschmer's concept ...

> then by
> the recognition of an Anatolian branch of Indo-European.  I very much
> doubt that Paul Kretschmer, who lived until 1956, would have agreed
> with Grapheus's fossilized ragtime-era view of IE linguistics.

That Kretschmer's fundamental idea is outmoded is YOUR opinion. Not
the opinion of "modern kretschmerians" like J. Faucounau.
By the way, have you ever read the first chapters of his book (in
French) on "the Origins of Greeks" ?.. Very entlightening concerning
the present discussion...

grapheus

grap...@www.com

unread,
Jan 8, 2008, 5:38:29 AM1/8/08
to
On Jan 8, 6:18 am, mb <azyth...@yahoo.com> wrote:
> On Jan 7, 7:46 pm, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
>
>
>
> > mb <azyth...@yahoo.com> wrote:
> > > "graph...@www.com" <graph...@www.com>
> > > ...
>
> > > > > > b)- the historical data are in favour of a language "Proto-IE in the
> > > > > > kretschmerian sense".
> > > > Of course !.. But I wanted to distinguish between a "possibility which
> > > > has a very small probability to be correct"
>
> > > As a first example, the possiblity of such a monstrosity as ""Proto-IE
> > > in the kretschmerian sense" comes to mind. Chances: nil.
> > > And, of course, no definition available.
>
> > What he means is Kretschmer's "Protindogermanisch", for an exposition
> > of which you will have to dig out some of the earliest volumes of
> > _Glotta_.  It was superseded by Sturtevant's "Indo-Hittite", then by
> > the recognition of an Anatolian branch of Indo-European.  I very much
> > doubt that Paul Kretschmer, who lived until 1956, would have agreed
> > with Grapheus's fossilized ragtime-era view of IE linguistics.  G. has
> > repeatedly shown himself to be a Rip van Winkle on this issue.
> > Perhaps he guzzled too much bathtub gin, tried to dance the Charleston
> > on the running board of his Duesenberg, and landed on his head?
>
> I happen to be aware of what he does *not really mean, just because
> those Kretschmer texts, then already rather obsolescent, were
> compulsory part of some boring seminar in Geneva waay back in time.

Please, don't play the "informed scholar" !!!!
Anybody knows that you are a PERFECT IGNORANT in matters of
Linguistics !... And follow my advice : GO BACK to your pizzas. It
the ONLY matter in which you have some competence...

grapheus

Jim Heckman

unread,
Jan 8, 2008, 5:50:31 AM1/8/08
to

On 6-Jan-2008, Odysseus <odysseu...@yahoo-dot.ca>
wrote in message <odysseus1479-at-9F...@news.telus.net>:

Almost certainly a _faux ami_. "grapheus" is a native speaker of
French, where the verb <ignorer> can mean either 'to ignore' or 'to
be ignorant of', according to context.

--
Jim Heckman

Dušan Vukotić

unread,
Jan 8, 2008, 8:32:44 AM1/8/08
to

Where the "parent" and "pairing" (Serb. 'parenje' coupling) are coming
from? παρθένος is a marriageable maiden; it means that parthenos is a
grown up female or a girl that just reached fertility - ready for
pairing. In Serbian it would be "porodna" (fertile). Compare Serbian
'porodilja' (pregnant woman) and English adjective 'fertile' and
'parental'.

Nothing to do with 'exalted' or 'protected'

DV

VK

unread,
Jan 8, 2008, 12:23:24 PM1/8/08
to
On Jan 8, 7:23 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
> The presentation in PID is not circular. The authors disagree with
> Mommsen in that they regard -ihi (with its variant spellings) as
> cognate with Latin -i: (here the inserted -h- merely indicates
> length), not with the other genitive suffix reflected in Faliscan -
> osio, Homeric -oio, Sanskrit -asya.

I wouldn't take this hypothesis as much possible, I mean intervocalic
h as a mark of hiatus and especially as a mark of the length of the
preceding vowel. That brings way too much of modern thinking into the
orthography of ancient alphabetic writing systems. For such use of h
we have to assume several German teachers who knew the modern German
orthography so explained to Messapians how to express this or that
sound :-) If ancient person really wanted to say "longer i" he would
simply write "ii", if he wanted to say "speech break" then he just
didn't write anything. That doesn't eliminate the possibility of a
tradition, so initial "ihi" at the time of writing was already
pronounced as "ii i" or "i'i" - but it is irrelevant then to the
analysis of the core language grammar.

From the point of view of possible segmentations of available
inscriptions it is crucial to know the exact origin of each "klohisis-
ihi" inscription. I temporarily call as "klohisis-ihi" all
inscriptions with initial "klohisis" and "ihi" formant(s) in the rest
of the sentence.
If they are all from tomb/cave inscriptions then my primary guess
would be about very common tradition of curses against possible tomb
riders. The Vaste Inscription (if authentic) is especially
demonstrative in this aspect:

klohizisthotoriamartapidovasteibasta
veinanaranindaranthoavastistaboos
xohedonasdaxtassivaanetosinthitrigonoxo
astaboosxohetthihidazimaihibeiliihi
inthirexxorixoakazareihixohetthihitoeihithi
dazohonnihiinthivastima
daxtaskratheheihiinthiardannoapoxxonnihia
imarnaihi

It definitely has some pseudo-poetic rhythm, reminding me very much
the Neo-Phrygian "ios ni semoun knoumane kakon abberet..." ("whoever
anything bad does to this tomb...") and then various lists of
punishments that should follow. They also have that pseudo-poetic
rhythm because of usage of uniformed syntagmas of the kind "then gods
punish him", "then his wife stays fruitless", "then his balls
disappear" (I am not fantasying, one of real courses), "then his legs
get broken", etc.

With such interpretation it is possible that "ihi" indeed has a
genitival value, but not as a genitive ending but as postfixed
possessive pronoun with the meaning "his".

Of course if any "klohisis-ihi" inscription proven to be found on an
offering or another clearly non-burial holder then all this
speculation doesn't worth the time it took to type it in.

mb

unread,
Jan 8, 2008, 2:26:42 PM1/8/08
to
On Jan 8, 2:35 am, "graph...@www.com" <graph...@www.com> wrote:
...
> > then by
> > the recognition of an Anatolian branch of Indo-European.  I very much
> > doubt that Paul Kretschmer, who lived until 1956, would have agreed
> > with Grapheus's fossilized ragtime-era view of IE linguistics.
>
> That Kretschmer's fundamental idea is outmoded

Not outmoded, obsolete in the face of today's data

> is YOUR opinion.

Simply the general consensus. The whole world in one more conspiracy
against poor little Faucounau.

> Not
> the opinion of "modern kretschmerians" like J. Faucounau.

Meaning yourself. And who the hell are you besides being the
mountebank inventor of the 120th kooky reading of a fake Phaistos
disk?

Also, you use the term "PIE in the Kretschmerian 'sense'" every time
you want to make a wildly improbable adjustment to your disc nonsense
by admitting impossible words based on the vaguest of sound
similarities without any systematic correspondences. That is the
smuggling that you use Kretschmer for.

> By the way, have you ever read the first chapters of his book (in
> French) on "the Origins of Greeks" ?.. Very entlightening concerning
> the present discussion...

Wrong group for trying to sell more of your hallucinations. Go to
alt.sci.fi.

grap...@www.com

unread,
Jan 8, 2008, 3:40:19 PM1/8/08
to
On Jan 8, 8:26 pm, mb <azyth...@yahoo.com> wrote:
> On Jan 8, 2:35 am, "graph...@www.com" <graph...@www.com> wrote:
> ...
>
> > > then by
> > > the recognition of an Anatolian branch of Indo-European.  I very much
> > > doubt that Paul Kretschmer, who lived until 1956, would have agreed
> > > with Grapheus's fossilized ragtime-era view of IE linguistics.
>
> > That Kretschmer's fundamental idea is outmoded
>
> Not outmoded, obsolete in the face of today's data

Here comes again the Italo-American kook who is playing to the
specialist in Indo-European Linguistics!.. Why don't you go back to
your pizzas, the only field in which you have some competence ?...

grapheus

mb

unread,
Jan 8, 2008, 4:00:51 PM1/8/08
to
On Jan 8, 12:40 pm, "graph...@www.com" <graph...@www.com> wrote:
...

You snipped the significant part, as usual :
===


you use the term "PIE in the Kretschmerian 'sense'" every time
you want to make a wildly improbable adjustment to your disc nonsense
by admitting impossible words based on the vaguest of sound
similarities without any systematic correspondences. That is the
smuggling that you use Kretschmer for.

===


> Here comes again the Italo-American kook who is playing to the
> specialist in Indo-European Linguistics!.. Why don't you go back to
> your pizzas, the only field in which you have some competence ?...

What do you know about it and how can you judge competence there? As
usual, invention, mountebankery and mythopoeia...

grap...@www.com

unread,
Jan 8, 2008, 4:09:59 PM1/8/08
to
On Jan 8, 10:00 pm, mb <azyth...@yahoo.com> wrote:
> On Jan 8, 12:40 pm, "graph...@www.com" <graph...@www.com> wrote:
> ...

I already told you : STOP playing the "serious linguist" !.. GO BACK
to your pizzas !...

grapheus

António Marques

unread,
Jan 8, 2008, 6:26:09 AM1/8/08
to
grap...@www.com wrote:

Let's change to a much more interesting subject. Have you actually tried
any of mb's pizzas? It's not the first time you've praised them. Is
there any one you're especially fond of? Do they have home delivery there?

--
Posted via a free Usenet account from http://www.teranews.com

grap...@www.com

unread,
Jan 8, 2008, 6:03:51 PM1/8/08
to
On Jan 8, 12:26 pm, António Marques <m...@sapo.pt> wrote:
> graph...@www.comwrote:

> > On Jan 8, 6:18 am, mb <azyth...@yahoo.com> wrote:
>
> >> I happen to be aware of what he does *not really mean, just because
> >>  those Kretschmer texts, then already rather obsolescent, were
> >> compulsory part of some boring seminar in Geneva waay back in time.
>
> > Please, don't play the "informed scholar" !!!! Anybody knows that you
> > are a PERFECT IGNORANT in matters of Linguistics !...  And follow my
> > advice : GO BACK to your pizzas. It the ONLY matter in which you have
> > some competence...
>
> Let's change to a much more interesting subject. Have you actually tried
> any of mb's pizzas? It's not the first time you've praised them. Is
> there any one you're especially fond of?

Yes, the "margarita".

> Do they have home delivery there?

Why asking you this ?.. Do you also live in California? Then, the best
thing is to take your car and go to the restaurant where Azythos
works...

grapheus

mb

unread,
Jan 8, 2008, 9:50:10 PM1/8/08
to
On Jan 8, 3:03 pm, "graph...@www.com" <graph...@www.com> wrote:
...
Marques:

> > Let's change to a much more interesting subject. Have you actually tried
> > any of mb's pizzas? It's not the first time you've praised them. Is
> > there any one you're especially fond of?
>
> Yes, the "margarita".

Stupid mountebank can't even keep his Romances straight.

> > Do they have home delivery there?
>
> Why asking you this ?.. Do you also live in California? Then, the best
> thing is to take your car and go to the restaurant where Azythos
> works...

Disc-kook logic: Asked about delivery to avoid driving, recommends
driving.

Richard Wordingham

unread,
Jan 15, 2008, 1:51:50 AM1/15/08
to
"VK" <school...@yahoo.com> wrote in message
news:116c5400-ef05-4bcb...@i12g2000prf.googlegroups.com...

> On Jan 8, 7:23 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
>> The presentation in PID is not circular. The authors disagree with
>> Mommsen in that they regard -ihi (with its variant spellings) as
>> cognate with Latin -i: (here the inserted -h- merely indicates
>> length), not with the other genitive suffix reflected in Faliscan -
>> osio, Homeric -oio, Sanskrit -asya.
>
> I wouldn't take this hypothesis as much possible, I mean intervocalic
> h as a mark of hiatus and especially as a mark of the length of the
> preceding vowel. That brings way too much of modern thinking into the
> orthography of ancient alphabetic writing systems. For such use of h
> we have to assume several German teachers who knew the modern German
> orthography so explained to Messapians how to express this or that
> sound :-)

So you would claim that the Vai practice of writing long-vowelled syllables
/CV:/ as <CV hV> (two syllabograms) in the Vai script comes from German?

Richard.

Darkstar

unread,
Jan 15, 2008, 9:02:01 AM1/15/08
to

Brian M. Scott:
> On Sat, 5 Jan 2008 15:51:33 -0800 (PST), Dušan Vukotić
> <dusan....@gmail.com> wrote in
> <news:cb97c7f1-e107-4cf0...@v67g2000hse.googlegroups.com>
> in sci.lang:
>
> > On Jan 6, 12:08 am, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
>
> >> On Sat, 05 Jan 2008 22:26:37 GMT, Richard Wordingham
> >> <jrw0...@yahoo.co.uk> wrote in
> >> <news:xuTfj.28235$KC3....@newsfe6-gui.ntli.net> in
> >> sci.lang:


>
> >>> "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:
>
> >>>> In addition, the second name name of  Ἀθηνα was Παρθένος
> >>>> (Virgin); cf. Latin partus (birth), Serb. poroditi/
> >>>> porod (give birth), Eng. birth.
>

> >> <Parthénos>, <partus>, <poroditi>, and <birth> are mutually
> >> unrelated.  <Parthénos> has not been established to have any
> >> relatives; <partus> is from a PIE *perh3- 'to produce,
> >> procure'; <birth> is from PIE *bHer- 'to carry'; and
> >> <poroditi> is a prefixed derivative of PSl *rodìti 'give
> >> birth, bear (fruit)', for which Derksen reconstructs PIE
> >> *(h)rodH-eie-.
>

> [...]
>
> > Thanks Brainy, but I do not need anyone to read the books
> > for me.
>
> Possibly not; in that case, however, you need someone to do
> your thinking for you.
>
> > Derksen is wrong and therefore you are wrong too. None of
> > the IE stem- words begins with the initial "r"; it is
> > absolutely impossible!
>
> There are quite a few, actually. This, however, may not be
> one: Derksen's reconstruction allows for the possibility of
> an initial laryngeal.
>

An off-topic: I often see many laryngeal-related reconstructions in
your posts. What is your modern primer on larygeal reconstructions
(outside Pokorny), I'm especially interested in an etymological
dictionary. Is this some kind of local school or something widely
accepted, how do you get all those H1, H2, etc. In my world, time
might indeed run backward, I have not read many western books, so I
wonder what this is all about.

VK

unread,
Jan 15, 2008, 10:29:16 AM1/15/08
to
On Jan 15, 9:51 am, "Richard Wordingham" <jrw0...@yahoo.co.uk> wrote:
> So you would claim that the Vai practice of writing long-vowelled syllables
> /CV:/ as <CV hV> (two syllabograms) in the Vai script comes from German?

Possibly. The official story tells that Dualu Bukele was inspired by a
dream as you know :-) That artificial writing system created at the
beginning of XIX century never interested me, sorry - it is just out
of field of my interests.

Peter T. Daniels

unread,
Jan 15, 2008, 5:05:58 PM1/15/08
to

Konrad Tuchscherer has shown with considerable plausibiliity that
Bukele was aware of Sequoyah's Cherokee writing system, because it was
described in a missionary magazine that the missionaries to his
community had access to.

Dušan Vukotić

unread,
Jan 16, 2008, 5:51:36 AM1/16/08
to
On Jan 15, 3:02 pm, Darkstar <darkstar...@front.ru> wrote:
> Brian M. Scott:
>
>
>
>
>
> > On Sat, 5 Jan 2008 15:51:33 -0800 (PST), Dušan Vukotić
> > <dusan.vuko...@gmail.com> wrote in
> wonder what this is all about.- Hide quoted text -
>
> - Show quoted text -

The so called Laryngeal Theory is a sort of "serious" joke or a sweet
delusion in which one can prove his linguistic (scientific) human-
hyper-knowledge. In reality, Laryngeal Theory serves only to scare the
people away of that "complicated" chemistry-like (H20 :-) "precise"
noesis. Saussure was the first who predicted the former existence and
later development of the PIE "laryngeals" (in reality mostly velars
and glotals).

Of course, Saussure was right that some "laryngeals" were lost,
especially those placed at the begining of word. It means that none of
the IE words has originally started with a vowel as an initial sound.
If a "laryngeal" is placed in the middle of word it means that that
word is a compound word; i.e. that "laryngeal is always the initial
sound of the second (or third)ˇ"member" of the compound word. Allso,
there are "laryngeals" at the end of word and they are denoting a
certain kind of motion related to subject.

For instance, German word 'enge' (density, narrowness, thight)
originally sounded as (H)enge (according to my Xur-Bel-Gon theory
'enge' comes from the reduplicated Gon basis). Farther, we could see
that German 'enge' (from henge) is related to English dense (born from
condense; gon-gen-ge; cf. dick/ness) wherefrom sprang a spectar of
Serbian words as gusto (dense), usko/uzak (narrow), sužen (narrowed),
s/taza/sokak (narrow path), nagon (impuls), gun-gu-la (crowed),
utegnut (thight).

Now let us scrutinize the English words dense, thight, dick: are those
words born from the same basis as the above-mentioned German 'enge'?
Certainly, they are! Things are getting more dense when pressured or
_kneaded (knead = Serb. gnjeti/gnječiti). Finally, at the "end" of our
narrow road from "enge" we ended in a "kitchen" with a "gusto" and
"ukusno" (tasty) "dough-doughnut".

If I say that English dough and dense are closely related to Serbian
testo and (dhough) and that 'end' is closely related to 'enge' and
Serbo-Slavic 'konac' (end) you will maybe be able to understand that
"laryngeal Theory" is nothing else but a big and vacant foam bubble,
which is going to turn into nothing with the first sound thought of an
unfettered thinker.

DV

Douglas G. Kilday

unread,
Jan 18, 2008, 10:25:48 PM1/18/08
to
VK wrote:
> "graph...@www.com" wrote:
>
> > [...]
>
> > As for the "Vaste Inscription", you may find its transcription in
> > Wikipedia "Massapian Language".
>
> I need to see the original drawing as made by Mommsen as the
> inscription itself alas lost. The Wikipedia copy is altered by the
> publisher. The original was consisting out of eight lines without word
> division. All available online copies I could find so far are having
> proprietary word breaks with line breaks removed.

Mommsen's paper, "Iscrizioni messapiche", _Annali dell'Instituto [sic]
di Corrispondenza Archeologica_ 20:59-156 + Tavv. d'agg. B-D (Roma
1848) is available on Google Books. You can find the volume by
searching for "iscrizione di vaste" or "vastae repertus est his annis
praeteritis lapis". Mommsen's drawing on Plate B (at the end of the
volume, which you can reach by going to Page 440, then clicking on the
"Next Page" arrow to access Plate A, etc.) includes the "standard"
apograph which Lepsius regarded as a fake Oscan inscription, plus
interlinear variant readings. As explained in the paper, the original
apograph was made between 1500 and 1510 by Antonio dei Ferrarii "il
Galateo", whose manuscript _De situ Japygiae_ was published in 1558
with a transcription of the apograph (given by M. on p. 77); a
separate copy of Galateo's apograph was the eventual source of the
"standard" on Plate B. Thus in order to establish the text, three
variants must be considered: the two graphic variants given by M. on
Plate B, and the transcription of 1558 given on p. 77.

Douglas G. Kilday

unread,
Jan 19, 2008, 12:20:28 AM1/19/08
to
VK wrote:

> "Douglas G. Kilday" wrote:
>
> > The presentation in PID is not circular.  The authors disagree with
> > Mommsen in that they regard -ihi (with its variant spellings) as
> > cognate with Latin -i: (here the inserted -h- merely indicates
> > length), not with the other genitive suffix reflected in Faliscan -
> > osio, Homeric -oio, Sanskrit -asya.
>
> I wouldn't take this hypothesis as much possible, I mean intervocalic
> h as a mark of hiatus and especially as a mark of the length of the
> preceding vowel. That brings way too much of modern thinking into the
> orthography of ancient alphabetic writing systems. For such use of h
> we have to assume several German teachers who knew the modern German
> orthography so explained to Messapians how to express this or that
> sound :-) If ancient person really wanted to say "longer i" he would
> simply write "ii", if he wanted to say "speech break" then he just
> didn't write anything. That doesn't eliminate the possibility of a
> tradition, so initial "ihi" at the time of writing was already
> pronounced as "ii i" or "i'i" - but it is irrelevant then to the
> analysis of the core language grammar.

Intervocalic -h- as a mark of length is found in Latin <nihil>,
<vehemens>, and <Ahala>. In <nihil> it is etymological, apparently
restored from <nihilo:minus> where the /h/ was sounded longer.
Plautus never has NIHIL, only NIL, but he has NIHILOMINVS. In
<vehemens> the -h- is unetymological (cf. <ve:sa:nus>, <ve:cors>,
<Ve:jovis>) but is retained as an "emphatic spelling", as though
<craaazy>. The cognomen <Ahala> of the gens Servilia retains the
family's traditional spelling, but the name is identical to the
appellative <a:la>. Umbrian written with Latin characters has a
similar device; in <ehe> with its compounds, and <sahata> with its
related forms, the -h- is unetymological and can only indicate
length. Thus no "Doktor Wer" is required to timewarp from Germany
back to Messapia to show them how to use -h-.

> From the point of view of possible segmentations of available
> inscriptions it is crucial to know the exact origin of each "klohisis-
> ihi" inscription. I temporarily call as "klohisis-ihi" all
> inscriptions with initial "klohisis" and "ihi" formant(s) in the rest
> of the sentence.
> If they are all from tomb/cave inscriptions then my primary guess
> would be about very common tradition of curses against possible tomb
> riders. The Vaste Inscription (if authentic) is especially
> demonstrative in this aspect:

You are making a false categorization. The inscription which you
cited from Galatina begins with <klohizis> and contains no forms in -
ihi. It is plausibly explained as a dedicatory inscription to
Aphrodite. Deecke's segmentation is as follows:

klohizis avithos thotorridas ana aprodita apaogrebis

<klohizis> appears to contain some form of the root *k^leu- 'to hear',
evidently meaning 'hear ye!' or the like, syntactically isolated from
what follows. (The belief that it means 'hear, O Zeus!' must be
rejected on morphological grounds.) <avithos thotorridas> is the
subject, the second word apparently derived from *teuta: 'civitas', so
it likely refers to some body of citizens. <ana> is probably a
preposition with the sense 'for', with <aprodita> in the dative (-a
from *-a:i) a loan-name from Greek, 'for Aphrodite'. <apaogrebis> is
the verb, with the prefix *apo-, root *ghrebh- 'to scratch, carve,
engrave', so we have (more or less): 'Hear ye! The ... body of
citizens inscribed (this monument) to Aphrodite.'

> klohizisthotoriamartapidovasteibasta
> veinanaranindaranthoavastistaboos
> xohedonasdaxtassivaanetosinthitrigonoxo
> astaboosxohetthihidazimaihibeiliihi
> inthirexxorixoakazareihixohetthihitoeihithi
> dazohonnihiinthivastima
> daxtaskratheheihiinthiardannoapoxxonnihia
> imarnaihi
>
> It definitely has some pseudo-poetic rhythm, reminding me very much
> the Neo-Phrygian "ios ni semoun knoumane kakon abberet..." ("whoever
> anything bad does to this tomb...") and then various lists of
> punishments that should follow. They also have that pseudo-poetic
> rhythm because of usage of uniformed syntagmas of the kind "then gods
> punish him", "then his wife stays fruitless", "then his balls
> disappear" (I am not fantasying, one of real courses), "then his legs
> get broken", etc.

Try this segmentation:

klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
in daranthoa vasti staboos khonedonas dakhtas sivaanetos
inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
inthi vastima dakhtas kratheheihi
inthi ardannoa po[ll]onnihiai marnaihi

I believe this to be a dedicatory inscription with a postscript
recording the names of public officials at the time of dedication. In
the first line, <klohizis> is again 'hear ye!' or the like.
<thotoria> is something like 'citizenry', modified by the adjective
<marza> (from *mardia, but I have no sure etymology; the -z- is from
Mommsen's variant reading). <pidogas> is the verb, the prefix cognate
with Greek epi-, in zero-grade as in <Pisaurum> 'city on the Isaurus'
which I discussed in another thread, and the root from *dheugh- 'to
make (something useful)'. <basta> is dative (like <aprodita> in at
least 3 inscriptions) and refers to the tutelary goddess of the place
(not the place itself, as some scholars have thought); the name Basta
(Plin. 3:100) of course survives in Vaste. <tei> is probably a
demonstrative in the dat. sg. agreeing with <basta>, cognate with
Greek <te:i>, not monophthongized because it was monosyllabic.
<veinan varan> (the second v- by collation of all 3 versions in
Mommsen) is accusative, the object of the verb. Tentatively I take
<veinan> as acc. sg. fem. of an adj. *weig-no- 'consecrated', from a
root *weig- 'to consecrate' found in Latin <victima> (from zero-grade
*wig-tm.ma:) and Umbrian <eveietu> (from *e:-weig-e:-to:d)
'consecrate!', and <waran> as acc. sg. of a noun *wora: 'enclosure'
reflected in Old Church Slavonic <vora> 'id.' from *wer- 'to enclose,
cover, protect'. Then we have for the first line (more or less):
'Hear ye! The ... citizenry constructed (this) sacred enclosure to
the (revered) Basta.'

The parallelism of the 2nd to 6th lines is easily understood if we
take <in> as the prep. 'in', with -thi an enclitic conjunction 'and',
like Lat. -que (I have no etym. for -thi). The second word of each
line is a fem. noun in the dative case (used with prep. 'in' for
locative, as in Greek), and to me they are all likely to be
magistracies, though I have no good etymology for any of them. The
remaining words, other than <vasti>, are personal names in the nom. or
gen. case, most of them known from epitaphs. Thus, these lines name
the various public officials serving at the time the sacred enclosure
was dedicated: 'In the consulship(??) were serving(?) Staboos Kh.
(and) Dakhtas S., and in the tribunate (??) Staboos (the son) of
Khonetthis the son of Dazimas, and in the ... (the son) of Kazares
Khonetthis and (the son) of Otoes Dazohonnis, and in the ... Dakhtas
(the son) of Krathehes, and in the ... (the daughter?) of Pollonnis
(the son?) of Marnas.'

> With such interpretation it is possible that "ihi" indeed has a
> genitival value, but not as a genitive ending but as postfixed
> possessive pronoun with the meaning "his".

I doubt it. Simple epitaphs show that -ihi is indeed a genitive
ending.

dakhtas moldahiaihi
'Dakhtas (the son) of Moldahias'

bizatas solahiaihi
'Bizatas (the son) of Solahias'

Genitives in -ihi may have been substantivized. The suffix -ai is
apparently feminine, hence my rendering of <pollonnihiai> as a
daughter, following this:

plastas moldatthehiai bilia ettheta hipades aprodita
'Ettheta, the daughter of Moldatthis (the son?) of Plasta, set up
(this monument) to Aphrodite.'

> Of course if any "klohisis-ihi" inscription proven to be found on an
> offering or another clearly non-burial holder then all this
> speculation doesn't worth the time it took to type it in.

If it gets further discussion going, it's well worth the time.

Dušan Vukotić

unread,
Jan 19, 2008, 9:52:05 AM1/19/08
to
On Jan 19, 6:20 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:

> > > The presentation in PID is not circular.  The authors disagree with
> > > Mommsen in that they regard -ihi (with its variant spellings) as
> > > cognate with Latin -i: (here the inserted -h- merely indicates
> > > length), not with the other genitive suffix reflected in Faliscan -
> > > osio, Homeric -oio, Sanskrit -asya.
>
> > I wouldn't take this hypothesis as much possible, I mean intervocalic
> > h as a mark of hiatus and especially as a mark of the length of the
> > preceding vowel.

Compare Latin nihil, negatio, nullus and you will maybe be able to see
that the sound h is quite natural in this case. Nullus is just a
simple assimilation of the sounds in the word nihil (nihilum =>
nullum). Serbian words nekati (deny) and neću (I won't; assimilated ne-
hoću will not) are clearly related to Latin negatio. Later on, Serbo-
Slavic nekati has transformed itself to Serbian niko (nobody; Russ,
никто/nikto; Czech nikoho, nikdo) and ništa (nothing; Russ. ничто;
Czech nic, ničeho, ničem). In addition, there is Serbian word
'ništavilo' (nothingness; Gon-Gon-Bel basis; compound word: ništa /
nothing/ + bilo /exist/). In case of Serbian ništa, German nicht and
English nothing (from nohting; cf. Serbian ništenje /annulment/) there
is a well visible loss of the vowel a (Serb. nekati /deny/) =>
Russ.nikto /nobody/ => ničto => ništa /nothing/). Above analysis
clearly shows that Latin "intervocalic" h is a normal and regular
sound, which in some "negation" words appeared as glottal, in other as
velars and in third, it disappears completely.

There is no and there has never been an inserted (non-etymological)
'h' at least among Romance, Slavic and Germanic languages. On the
other side, there are thousands and thousands of IE words where the
historical 'h' has been lost.

There is nothing unusual about such a vowel gemination as 'ii'; it is
always triggered by the loss of glottal (Cf. Eng. seing, Ger. sehen
and sight/Sicht; Eng. vehicle and Serbian vozilo /vehicle/; vozi-ko-
lo). Latin vehemens, just like veho, is the word that has been derived
from the Bel-Gon ur-basis and that can be compared to Serbian besneti
(rampage) or more closer to Serbian besomučan (amuck; be-so-mu-čan <=>
ve-he-me-nt). In reality, the history of those words is very simple:
it started from the same Bel-Gon ur-basis where the Serbian word pogon
(drive; from pol-gon; cf. Latin pulsus push, drive) sprang from.

As I already told many times, the primeval Bel-Gon basis was branched
into the three main directions:
1) form (oblo, oval; round shape of the sun)
2) fire (light, blast, heat); cf. Serbian ispaliti <=> explode
3) water (flow, flood, tide etc.)

DV

VK

unread,
Jan 19, 2008, 3:35:26 PM1/19/08
to
On Jan 19, 6:25 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
> Mommsen's paper, "Iscrizioni messapiche", _Annali dell'Instituto [sic]
> di Corrispondenza Archeologica_ 20:59-156 + Tavv. d'agg. B-D (Roma
> 1848) is available on Google Books. You can find the volume by
> searching for "iscrizione di vaste" or "vastae repertus est his annis
> praeteritis lapis". Mommsen's drawing on Plate B (at the end of the
> volume, which you can reach by going to Page 440, then clicking on the
> "Next Page" arrow to access Plate A, etc.) includes the "standard"
> apograph which Lepsius regarded as a fake Oscan inscription, plus
> interlinear variant readings.

Thank you for the directions. Well possible that I am not using Google
Book search right. I have found
"Iscrizioni messapiche" by Theodor Mommsen, 1848
http://books.google.com/books?id=DowOAAAAQAAJ&pgis=1
but as a title description only, no content. Neither of pages or links
you have mentioned are found using the keywords you had suggested.
Could you point to URL directly?

> As explained in the paper, the original
> apograph was made between 1500 and 1510 by Antonio dei Ferrarii "il
> Galateo", whose manuscript _De situ Japygiae_ was published in 1558
> with a transcription of the apograph (given by M. on p. 77); a
> separate copy of Galateo's apograph was the eventual source of the
> "standard" on Plate B. Thus in order to establish the text, three
> variants must be considered: the two graphic variants given by M. on
> Plate B, and the transcription of 1558 given on p. 77.

That is a great inside to the history of the Vaste inscription, thank
you.

Dušan Vukotić

unread,
Jan 19, 2008, 4:38:30 PM1/19/08
to
On Jan 19, 6:20 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:

> > From the point of view of possible segmentations of available
> > inscriptions it is crucial to know the exact origin of each "klohisis-
> > ihi" inscription. I temporarily call as "klohisis-ihi" all
> > inscriptions with initial "klohisis" and "ihi" formant(s) in the rest
> > of the sentence.
> > If they are all from tomb/cave inscriptions then my primary guess
> > would be about very common tradition of curses against possible tomb
> > riders. The Vaste Inscription (if authentic) is especially
> > demonstrative in this aspect:
>
> You are making a false categorization.  The inscription which you
> cited from Galatina begins with <klohizis> and contains no forms in -
> ihi.  It is plausibly explained as a dedicatory inscription to
> Aphrodite.  Deecke's segmentation is as follows:
>
> klohizis avithos thotorridas ana aprodita apaogrebis


> <klohizis> appears to contain some form of the root *k^leu- 'to hear',
> evidently meaning 'hear ye!' or the like, syntactically isolated from
> what follows.  

Possible... but klohizis could also mean "renowned, esteemed,
celebrated, famous" etc. Firstly I supposed that avithos thotorridas
is the personal name (something like Greek Αβδιας/Αβυδος Θεοδωρίδης or
Serbian Vidoje Todorić; Ovidius Theodoric etc. In that case, the whole
sentence could be translated as either "Esteemed Avidos Thotorridas on
the burial ceremony" or, "The burial ceremony of the esteemed Avidos
Thotorridas". I do not know if there is/was a picture underneath that
inscription table? In case that a certain picture existed, the above
"deciphering" might be considered as fairly correct.

>(The belief that it means 'hear, O Zeus!' must be
> rejected on morphological grounds.)  <avithos thotorridas> is the
> subject, the second word apparently derived from *teuta: 'civitas', so
> it likely refers to some body of citizens.  <ana> is probably a
> preposition with the sense 'for', with <aprodita> in the dative (-a
> from *-a:i) a loan-name from Greek, 'for Aphrodite'.  <apaogrebis> is
> the verb, with the prefix *apo-, root *ghrebh- 'to scratch, carve,
> engrave', so we have (more or less):  'Hear ye!  The ... body of
> citizens inscribed (this monument) to Aphrodite.'

> Try this segmentation:


>
> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> inthi vastima dakhtas kratheheihi
> inthi ardannoa po[ll]onnihiai marnaihi

Just in a moment when I started to believe that the problem of "Avidos
Thotorridas" was nearly solved, "thotoria mar[z]a" came in and spoiled
the "fun". Is "thotoria" a personal name similar to modern Theodora?
Maybe Douglas was right, "thotoria" must have something in common with
"teuta"?

Finally, I tried to re-segment the above inscription, but with no
success. I could not find a single word that is phonetically or
morpologically close to any word of the IE vocabulary. I wonder, who
(and on what grounds) has made a conclusion that Messapic is an
extinct Indo-European language?

DV

VK

unread,
Jan 19, 2008, 5:23:56 PM1/19/08
to
On Jan 19, 8:20 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
> Intervocalic -h- as a mark of length is found in Latin <nihil>,
> <vehemens>, and <Ahala>. In <nihil> it is etymological, apparently
> restored from <nihilo:minus> where the /h/ was sounded longer.
> Plautus never has NIHIL, only NIL, but he has NIHILOMINVS. In
> <vehemens> the -h- is unetymological (cf. <ve:sa:nus>, <ve:cors>,
> <Ve:jovis>) but is retained as an "emphatic spelling", as though
> <craaazy>. The cognomen <Ahala> of the gens Servilia retains the
> family's traditional spelling, but the name is identical to the
> appellative <a:la>. Umbrian written with Latin characters has a
> similar device; in <ehe> with its compounds, and <sahata> with its
> related forms, the -h- is unetymological and can only indicate
> length. Thus no "Doktor Wer" is required to timewarp from Germany
> back to Messapia to show them how to use -h-.

To indicate length of vowels, Romans used macrons and vowel doubling -
unless the length was positionally determined so didn't need any extra
hints (the majority of cases). In relevance to a vowel before "h" the
Latins themselves (Varro, "De Lingua Latina") tells that such vowel is
always _short_ with samples like "veho". So yes, a unregistered time
machine would help greatly to explain how an orthographic rule of
classical German came into effect in VI B.C. against any possible
local usages ;-) I have a non sci-fi related explanation though and I
tend to believe it is the right one: it just happened that the early
studies of Italic inscriptions in XIX have been done by native German
speakers like Mommsen or by people fully fluent in German. Do not
forget that at that time German together with Latin was the language
of science and scientific publications in Europe. So we are having
here a rather often case of "fake obvious": when a feature very
natural to the researcher is being "seen" in a completely different
environment despite that there is not such at all there.

> > From the point of view of possible segmentations of available
> > inscriptions it is crucial to know the exact origin of each "klohisis-
> > ihi" inscription. I temporarily call as "klohisis-ihi" all
> > inscriptions with initial "klohisis" and "ihi" formant(s) in the rest
> > of the sentence.
> > If they are all from tomb/cave inscriptions then my primary guess
> > would be about very common tradition of curses against possible tomb
> > riders. The Vaste Inscription (if authentic) is especially
> > demonstrative in this aspect:
>
> You are making a false categorization. The inscription which you
> cited from Galatina begins with <klohizis> and contains no forms in
> -ihi

Right,
klohizisanthosthot
orridasanaaprodi
taapaogrebis
is strange in this aspect. Until more firm segmentation decisions are
made nothing more than "strange" to say.

[1]


klohizisthotoriamartapidovasteibasta
veinanaranindaranthoavastistaboos
xohedonasdaxtassivaanetosinthitrigonoxo
astaboosxohetthihidazimaihibeiliihi
inthirexxorixoakazareihixohetthihitoeihithi
dazohonnihiinthivastima
daxtaskratheheihiinthiardannoapoxxonnihia
imarnaihi

[2]
klauhizis
dekiasartahias
thautouriandirahho
dausapistathivinaihi

http://images.google.com/imgres?imgurl=http://www.telemaco.unibo.it/rom/italia/img05/messapi2.jpg&imgrefurl=http://cegliemessapica.splinder.com/archive/2007-05&h=228&w=685&sz=55&hl=en&start=4&sig2=XLQmH1E9RSosC0NXmGUTbQ&um=1&tbnid=Plb-UCMs8p4QtM:&tbnh=46&tbnw=139&ei=g-d_R9jGFZeowQGpysXlDw&prev=/images%3Fq%3DMessapii%26svnum%3D10%26um%3D1%26hl%3Den%26sa%3DG
This inscription (if authentic) is intensively interesting because it
is missing the opening "klohizis" but contains "thotori" in a very
tempting for an IE interpretation spelling "THEOTORI" (too tempting
though to jump on it right away) and "-ihi" formant.
I am preparing block structure analyzer for Messapian right now on my
computer. In terms of regular expressions the category could be
quick'n'dirty defined as
[^klohizis]{1}||[thotori]{1}&&[$ihi]+
so opening klohisiz no more than one time or none, thotori no more
than one time or none, ending ihi any amount of time - where klohisiz
and thotori are in XOR relations: so either one can be missing but not
both. It is all sketchy yet of course.

<possible inscription interpretation are skept>
not because of disrespect! I will study them with a greatest
attention. Right now I want to accomplish the first and second stages
first:
1) get all more-or-less long available inscriptions, check their
authenticity and the conditions they appear (inscription bearers)
wherever it is still possible.
2) eliminate if possible impossible word breaks based on similar
structural blocks from different inscriptions.

On this stage early interpretations - if I ever get to such dare aim -
are dangerous as it may force one to see in the material what one
decided to see rather than what the material is willing to tell by
itself.

Trond Engen

unread,
Jan 20, 2008, 8:15:44 AM1/20/08
to
Douglas G. Kilday skreiv:

> VK wrote:
>
>> From the point of view of possible segmentations of available
>> inscriptions it is crucial to know the exact origin of each

>> "klohisis-ihi" inscription. I temporarily call as "klohisis-ihi" all

>> inscriptions with initial "klohisis" and "ihi" formant(s) in the
>> rest of the sentence.
>> If they are all from tomb/cave inscriptions then my primary guess
>> would be about very common tradition of curses against possible tomb
>> riders. The Vaste Inscription (if authentic) is especially
>> demonstrative in this aspect:
>
> You are making a false categorization. The inscription which you
> cited from Galatina begins with <klohizis> and contains no forms in -
> ihi. It is plausibly explained as a dedicatory inscription to
> Aphrodite. Deecke's segmentation is as follows:
>
> klohizis avithos thotorridas ana aprodita apaogrebis
>
> <klohizis> appears to contain some form of the root *k^leu- 'to
> hear', evidently meaning 'hear ye!' or the like, syntactically
> isolated from what follows. (The belief that it means 'hear, O
> Zeus!' must be rejected on morphological grounds.) <avithos
> thotorridas> is the subject, the second word apparently derived from
> *teuta: 'civitas', so it likely refers to some body of citizens.
> <ana> is probably a preposition with the sense 'for', with <aprodita>
> in the dative (-a from *-a:i) a loan-name from Greek, 'for
> Aphrodite'. <apaogrebis> is the verb, with the prefix *apo-, root
> *ghrebh- 'to scratch, carve, engrave', so we have (more or less):
> 'Hear ye! The ... body of citizens inscribed (this monument) to
> Aphrodite.'

To me, <thotorridas> looks like either a name or a title. The second
element could be *reidh- "ride, drive" or perhaps a derivation of *rei-
"(make) flow, run". Or even a misspelling or a misreading of a reflex of
*reig-? If it's a name then it's a straightforward compound with
parallels all over IE. If it's a title then it's a question of
interpretation of the compound. From *reidh- I can see "coach of the
people" and "national horseman", but I'm sure there are more. From *rei-
and *reig- I can see "leader of the people". Anyway, that would leave
<avithos> as his personal name. "Hear ye! Avithos, coach of the people,
..."

> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> inthi vastima dakhtas kratheheihi
> inthi ardannoa po[ll]onnihiai marnaihi
>
> I believe this to be a dedicatory inscription with a postscript
> recording the names of public officials at the time of dedication.
> In the first line, <klohizis> is again 'hear ye!' or the like.
> <thotoria> is something like 'citizenry', modified by the adjective
> <marza> (from *mardia, but I have no sure etymology; the -z- is from
> Mommsen's variant reading).

If we leave the -rz- < -rd-, how about *merg- "boundary, border" ->
"land"? Or the word known from ON <margr> "several, many"? Alas, I've
not been able to find it's etymology. We'd have to suppose rhotacism,
though.

> [...] Then we have for the first line (more or less): 'Hear ye!


> The ... citizenry constructed (this) sacred enclosure to the
> (revered) Basta.'

"Hear ye! Many leaders of the people constructed ..."

Holding the two together I'd say that <thotorridas> looks like "chief
horseman". The etymology of the title, of course, doesn't say anything
of the duties of its bearer at the time of the inscription.

> The parallelism of the 2nd to 6th lines is easily understood if we
> take <in> as the prep. 'in', with -thi an enclitic conjunction 'and',
> like Lat. -que (I have no etym. for -thi). The second word of each
> line is a fem. noun in the dative case (used with prep. 'in' for
> locative, as in Greek), and to me they are all likely to be
> magistracies, though I have no good etymology for any of them. The
> remaining words, other than <vasti>, are personal names in the nom. or
> gen. case, most of them known from epitaphs. Thus, these lines name
> the various public officials serving at the time the sacred enclosure

> was dedicated: 'In the consulship(??) [...], and in the tribunate
> (??) [...], and in the ... (the son) of Kazares Khonetthis and (the
> son) of Otoes Dazohonnis,

Would that be <retorikhoa> with Greek rhe- simplified to re-?

> and in the ... Dakhtas (the son) of Krathehes, and in the ...
> (the daughter?) of Pollonnis (the son?) of Marnas.'

IS a variant spelling <b/vasta> implausible? Since Dakhtas serves in
<vastimae>, apparently connected to <vasti>, could these two last lines
list leaders of circuits? And <daranthoa vasti> mean "consulship(?) of
Vasta"?

>> [...]
>
> [...]


>
>> Of course if any "klohisis-ihi" inscription proven to be found on an
>> offering or another clearly non-burial holder then all this
>> speculation doesn't worth the time it took to type it in.
>
> If it gets further discussion going, it's well worth the time.

Mind your words. I take them literally.

--
Trond Engen
- making Messapic Messapical

Italo

unread,
Jan 21, 2008, 8:57:36 AM1/21/08
to
How about "..ana aprodita apa ogrebis"
"mother Aphrodite, father Grabovius"?

Also, any possibility that Eteo-Cypriot "a-na-ma-to-ri" may
be relevant?

But wasn't the "city Basta" for "vastei basta" the only
thing in this inscription commonly agreed on?
Is the v of vastei now read as g?

mb

unread,
Jan 21, 2008, 4:24:54 PM1/21/08
to

Douglas G. Kilday

unread,
Jan 27, 2008, 12:36:37 AM1/27/08
to
VK wrote:

That is a ridiculous charge against Mommsen, who was also fluent in
Italian, and against other scholars. If you expect me to believe that
<vehemens> and <Ahala> were pronounced with sounded /h/, you will have
to find unequivocal examples from poetry requiring two short syllables
instead of one long.

The Messapic alphabet is derived from the Tarentine Doric, and
examples of Tarentine words with intervocalic -h- are known, such as
<oudehen> 'nothing, not at all'. Obviously the -e of <de> was not
apocopated in Tarentine as it was in Attic-Ionic <oudén> from
*oud'(h)én, and it is not difficult to envision loss of the
intervocalic /h/ sound with a long vowel resulting, that is /oude:n/
still written <oudehen>, like Latin /ni:l/ written <nihil>. And it is
an easy step to the usage of intervocalic -h- to denote a long vowel
when Messapians adapted the Tarentine alphabet to their own needs. I
fail to see why you should have any problem comprehending this, and
why you must resort to impugning the objectivity of 19th-century
researchers.

The texts from both Galatina and Vaste above can be reasonably well
segmented, as I already showed.

> [2]
> klauhizis
> dekiasartahias
> thautouriandirahho
> dausapistathivinaihi

That is one of the texts from Grotta della Poesia in Melendugno,
Lecce. I have not seen Carlo de Simone's paper, but Cornell's Prof.
Weiss gives what appears to be an English version of CdS's analysis in
his PDF file "Comparative Grammar of Latin"; you can find it by
googling for "artahias". I disagree with most of the segmentation and
interpretation. This is what Weiss gives:

klauhi zis 'Hear Zeus!'
dekias artahias 'Dekias Artahias'
thautouri andirahho 'to the infernal Thaotor'
daus apistathi vinaihi 'set up _daus vinaihi_'

First, there is no principled way to regard Mess. <zis> as the
vocative of 'Zeus'. If Messapic preserved vocatives, the final -eu
(cf. Greek <Zeu>) would have become -au in the Melendugno orthography
(elsewhere most commonly -ao). If nominatives were substituted for
vocatives, we would expect -aus here (elsewhere mostly -aos), but
under no circumstances -is. Therefore, we should not segment
<klauhizis>, and 'Hear ye!' remains the most plausible rendering, even
if the -zis remains an obscure ending.

I have no quarrel with the second line. <Dekias> is evidently a
Messapization of Latin <Decius> and indicates that Romanization was
already in progress here. <Artahias> recalls <Artas>, leader of
Messapian darters (Thuc. 7:33.4).

<andirahho> is supposed to be an adjective 'infernal' based on PIE
*ndher-, whence English <under>, etc. Now, Messapic may well have had
a prefix *andi- 'under'. Hesychius has a Tarentine gloss
<andiklóbolos> 'díke: he: ex hupárkho:n dikazoméne:' i.e. 'the
judgment handed down by under-commanders'; possibly this is borrowed
from a Messapic word for 'under-judgment' or the like used in that
sense. However, in the Vaste text <thotoria>, following the first
word <klohizis>, appears to be the subject of the sentence. Thus it
seems much better to take <thautourian> as the accusative of the same
word (the dialect of Melendugno being more conservative than that of
Vaste in retaining diphthongs), 'citizenry' or the like, and the
remaining <dirahho> as a verb.

In the fourth line, <thi> looks like the enclitic 'and' of the Vaste
text, joining <dausa> and <pista>, nouns in the dative case like
<aprodita> of the three dedicatory inscriptions. <vinaihi> is the
genitive of a noun <vina>. My segmentation:

klauhizis
dekias artahias
thautourian dirahho
dausa pista-thi vinaihi

Not knowing what the nouns <dausa>, <pista>, and <vina> mean, I can
offer no translation. If <thautouria> indeed means 'citizenry', the
verb might mean 'purified', 'delivered', 'provided', or the like, but
this is just whistling in the dark. I am reasonably sure about my
segmentation and my morphology, however.

> http://images.google.com/imgres?imgurl=http://www.telemaco.unibo.it/r...


> This inscription (if authentic) is intensively interesting because it
> is missing the opening "klohizis" but contains "thotori" in a very
> tempting for an IE interpretation spelling "THEOTORI" (too tempting
> though to jump on it right away) and "-ihi" formant.
> I am preparing block structure analyzer for Messapian right now on my
> computer. In terms of regular expressions the category could be
> quick'n'dirty defined as
> [^klohizis]{1}||[thotori]{1}&&[$ihi]+
> so opening klohisiz no more than one time or none, thotori no more
> than one time or none, ending ihi any amount of time - where klohisiz
> and thotori are in XOR relations: so either one can be missing but not
> both. It is all sketchy yet of course.

I fail to see how computerized analysis can help. The total number of
inscriptions is too small, and the variation in orthography (klohizis
= klaohizis = klauhizis, etc.) means that you will spend more time
screwing around with the program than you would need to spend in order
to analyze the texts with good old brainpower.

> <possible inscription interpretation are skept>
> not because of disrespect! I will study them with a greatest
> attention. Right now I want to accomplish the first and second stages
> first:
> 1) get all more-or-less long available inscriptions, check their
> authenticity and the conditions they appear (inscription bearers)
> wherever it is still possible.
> 2) eliminate if possible impossible word breaks based on similar
> structural blocks from different inscriptions.
>
> On this stage early interpretations - if I ever get to such dare aim -
> are dangerous as it may force one to see in the material what one
> decided to see rather than what the material is willing to tell by
> itself.

That can also happen with late interpretations. If you remain
consciously objective, you should be able to revise early
interpretations when more material becomes available.

Douglas G. Kilday

unread,
Jan 27, 2008, 1:40:17 AM1/27/08
to

The main problem with seeing a binary compound here is that the
connecting vowel -o- should have become -a- in Messapic. If <avithos>
(also read as <anthos>; I have not seen an apograph, and if -n- is
correct possibly borrowed from Greek <anthoûs> 'flowery'??) is indeed
a personal name, <thotorridas> may be a Doric-style patronymic based
on the native personal name Th(e)otor, cognate with the praenomen
Tutor of the Cloelii of Alba Longa. We have another instance in an
epitaph from Ceglie:

ettis arnisses theotorres

> > klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> > in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> > inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> > inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> > inthi vastima dakhtas kratheheihi
> > inthi ardannoa po[ll]onnihiai marnaihi
>
> > I believe this to be a dedicatory inscription with a postscript
> > recording the names of public officials at the time of dedication.  
> > In the first line, <klohizis> is again 'hear ye!' or the like.
> > <thotoria> is something like 'citizenry', modified by the adjective
> > <marza> (from *mardia, but I have no sure etymology; the -z- is from
> > Mommsen's variant reading).  
>
> If we leave the -rz- < -rd-, how about *merg- "boundary, border" ->
> "land"? Or the word known from ON <margr> "several, many"? Alas, I've
> not been able to find it's etymology. We'd have to suppose rhotacism,
> though.

I like the first idea. If memory serves, all the Celtic reflexes mean
'land', 'district' vel sim., including the formant in the Gaulish
Allobroges 'Other-landers'. Then we could have a zero-grade adjective
*mrg-ja: > *margja > marza, and <thotoria marza> would be 'the
citizenry of the land' or the like. Nice!

> > [...]   Then we have for the first line (more or less):  'Hear ye!
> > The ... citizenry constructed (this)  sacred enclosure to the
> > (revered) Basta.'

I now think <tei> is probably a fossilized locative of the
demonstrative stem, 'in this place' = 'here', rather than a dative in
apposition with <basta>. Then we have: 'Hear ye! The citizenry of
the land constructed the sacred enclosure here to Basta.'

> "Hear ye! Many leaders of the people constructed ..."
>
> Holding the two together I'd say that <thotorridas> looks like "chief
> horseman". The etymology of the title, of course, doesn't say anything
> of the duties of its bearer at the time of the inscription.

An epitaph from Ostuni appears to refer to a 'charioteer', though the
morphology is peculiar, and the fellow's first name happens to be
Theotoras:

theotoras artahiaihi bennarrihino

<benna> has been explained as a loanword from the Gaulish for
'chariot'; presumably *bennarras would correspond to a Latin
*benna:rius 'charioteer', and if we could drop the -no we could read
'Theotoras, son of Artahias the charioteer'. But the pesky -no
remains. My best guess is that it represents the genitive plural in *-
o:m of an adjective in -no-, so the last word corresponds to a Latin
*benna:ri:no:rum, perhaps '(chief) of those serving the charioteers'??

> > The parallelism of the 2nd to 6th lines is easily understood if we
> > take <in> as the prep. 'in', with -thi an enclitic conjunction 'and',
> > like Lat. -que (I have no etym. for -thi).  The second word of each
> > line is a fem. noun in the dative case (used with prep. 'in' for
> > locative, as in Greek), and to me they are all likely to be
> > magistracies, though I have no good etymology for any of them.  The
> > remaining words, other than <vasti>, are personal names in the nom. or
> > gen. case, most of them known from epitaphs.  Thus, these lines name
> > the various public officials serving at the time the sacred enclosure
> > was dedicated:  'In the consulship(??) [...], and in the tribunate
> > (??) [...], and in the ... (the son) of Kazares Khonetthis and (the
> > son) of Otoes Dazohonnis,
>
> Would that be <retorikhoa> with Greek rhe- simplified to re-?

The apograph appears to have a double chi. Minervini corrected the
other double chi to alpha on the basis of a name he read as
Poaonnihi. Deecke corrected the correction to double lambda on the
basis of more legible names elsewhere, but did not correct the first
double chi. On the apograph they appear the same, and I regard both
as digraphs for double lambda, essentially two lambdas back to back.
That is, I read <rellorikhoa> here, no way to get tau instead. The -
ll- is probably from earlier -li-, but I still have no etymological
clue.

> > and in the ... Dakhtas (the son) of Krathehes, and in the ...
> > (the daughter?) of Pollonnis (the son?) of Marnas.'
>
> IS a variant spelling <b/vasta> implausible? Since Dakhtas serves in
> <vastimae>, apparently connected to <vasti>, could these two last lines
> list leaders of circuits? And <daranthoa vasti> mean "consulship(?) of
> Vasta"?

I don't like the idea that beta and digamma could be interchanged in
Messapic. I can find no evidence for this in the personal names. The
word <vasti> is problematic. The PID authors take it for a neuter
noun. I would like to read it as a numeral adverb (indicating which
time in the consulship) but cannot justify that. Or it might be a
verb, since Messapic verbal morphology is so poorly understood. I
can't explain the relation of <vasti> to <vastima> under either
hypothesis.

Douglas G. Kilday

unread,
Jan 27, 2008, 2:15:22 AM1/27/08
to
Italo wrote:
> Douglas G. Kilday wrote:
>
> [...]

>
> > You are making a false categorization.  The inscription
> > which you cited from Galatina begins with <klohizis> and
> > contains no forms in - ihi.  It is plausibly explained as
> > a dedicatory inscription to Aphrodite.  Deecke's
> > segmentation is as follows:
>
> > klohizis avithos thotorridas ana aprodita apaogrebis
>
> > <klohizis> appears to contain some form of the root
> > *k^leu- 'to hear', evidently meaning 'hear ye!' or the
> > like, syntactically isolated from what follows.  (The
> > belief that it means 'hear, O Zeus!' must be rejected on
> > morphological grounds.)  <avithos thotorridas> is the
> > subject, the second word apparently derived from *teuta:
> > 'civitas', so it likely refers to some body of citizens.
> > <ana> is probably a preposition with the sense 'for',
> > with <aprodita> in the dative (-a from *-a:i) a loan-name
> > from Greek, 'for Aphrodite'.  <apaogrebis> is the verb,
> > with the prefix *apo-, root *ghrebh- 'to scratch, carve,
> > engrave', so we have (more or less):  'Hear ye!  The ...
> > body of citizens inscribed (this monument) to Aphrodite.'
>
> How about "..ana aprodita apa ogrebis"
> "mother Aphrodite, father Grabovius"?

No way to get <ogrebis> from 'Grabovius'. The latter should become
*graboas. The only revision I would make above is that <anthos(?)
thotorridas> is probably a personal name with a Doric-style
patronymic, as mentioned in my reply to Trond.

> Also, any possibility that Eteo-Cypriot "a-na-ma-to-ri" may
> be relevant?

Unfortunately, I have no knowledge of that.

> [...]

The letter in this position was originally read as gamma, and the 1558
Basilean text of Galateo's _De situ Japygiae_ has gamma. Deecke and
Hirt read gamma here. The apograph in Mommsen's paper shows a letter
that could be gamma or pi, possibly followed by a dot. It is a real
stretch to suppose that this dot is the remnant of the crossbar of a
digamma, as those who read <vastei> do suppose. In fact <vastei> is a
forced reading by those who take the inscription as recording the
purchase of land from other districts 'for the city Basta', and there
are strong objections to such an interpretation. In Mommsen's time
the place had a population of 200, and it could hardly have been much
larger in antiquity, since only Pliny even mentions it among Latin
geographers. It was a mere village, not a city, and not deserving to
be called the equivalent of Greek <(w)ástu>. Also, a verb <pido>
cannot be harmonized with the verbs <hipades> and <apaogrebis>
(plausibly regarded as 3sg. /s/-aorists, earlier -st), but <pidogas>
can be easily so harmonized.

As I mentioned in my reply to Trond, I now prefer to read <tei> as
'here'.

Joachim Pense

unread,
Jan 27, 2008, 3:04:28 AM1/27/08
to
Am Sat, 26 Jan 2008 21:36:37 -0800 (PST) schrieb Douglas G. Kilday:

>
> The Messapic alphabet is derived from the Tarentine Doric, and
> examples of Tarentine words with intervocalic -h- are known, such as
> <oudehen> 'nothing, not at all'. Obviously the -e of <de> was not
> apocopated in Tarentine as it was in Attic-Ionic <oudén> from
> *oud'(h)én, and it is not difficult to envision loss of the
> intervocalic /h/ sound with a long vowel resulting, that is /oude:n/
> still written <oudehen>, like Latin /ni:l/ written <nihil>.

Is there an explanation why there was no lengthening of the -e- in the
Attic-Ionic ouden, compensating the -h-?

Joachim

lora...@cs.com

unread,
Jan 27, 2008, 6:47:13 AM1/27/08
to
On Jan 5, 3:10 am, VK <schools_r...@yahoo.com> wrote:
> In my PIE studies I never dealt before with Messapic (Messapian)
> language first because of a scanty amount of the available materials
> and secondly because the available materials didn't point for me to IE
> origin - though didn't exclude it explicitly. Suddenly I have been
> asked to express my opinion on the matter by one higher education
> body. I have just started on the material, so wondering if some quick
> help could be obtained from the group's regulars.
>
> 1) Messapic as an Indo-European language.
> How exactly it was arrived to this conclusion? I went from the current
> publications back to the beginning and the only kind of analysis I
> found so far is one sentence from Mommsen, "The History of Rome", Book
> 1 Chapter 2: "The genitive forms, -aihi- and -ihi-, corresponding to
> the Sanscrit -asya- and the Greek --oio--, appear to indicate that the
> dialect belongs to the Indo-Germanic family."

This of course, is a parochial and chavanistic view that has
subsequently clouded the perceptions of subsequent readers' 'indo-
germanic'.

> AFAICT ever since the middle of XIX century no further attempts had
> been made to identify the language family. There is a great amount of
> works where the authors are trying to identify the exact position of
> Messapic among IE languages and/or to analyze language features: but
> with the IE nature of the Messapic pre-postulated as some "common
> knowledge" fact: see for instance R.M.Bechtel, "The Messapic Klaohizis
> Formula", 1937 or R.Giacomelli, "Written and Spoken Language in Latin-
> Faliscan and Greek-Messapic", 1979.
> Am I missing some important publication on the subject?
>
> 2) Does anyone have "Corpus Inscriptionum Messapicarum" by Ribezzo?
> This book is out of print with 3 weeks reprint delivery the earliest
> in the available bookstores. I could expedite my work if I could take
> a look at a scanned copy of the infamous Vaste Inscription first
> reported by Momssen (Corpus Inscriptionum Messapicarum 149)
>
> 3) Does anyone have: Carlo De Simone, "Iscrizione messapiche della
> grotta della poesia", 1988 - I have the same 3 weeks delay for the
> same reason as above.

lora...@cs.com

unread,
Jan 27, 2008, 7:24:45 AM1/27/08
to
On Jan 6, 1:53 pm, VK <schools_r...@yahoo.com> wrote:
> On Jan 6, 12:41 am, "Du¹an Vukotiæ" <dusan.vuko...@gmail.com> wrote:
>
> > Don't take me serious... I was joking. What one can say when one hears
> > that "Mesapic is close to Illyrian" or even a "dialect of Illyrian"
> > and, at the same time one knows that no one has ever (in an exact
> > scientific way) determined what that "Illyrian" really represents.
>
> > Unlike so-called Illyrian (which has not left any written trace),
> > Messapic left a few inscriptions; but, as far as I know, those
> > inscriptions has not yet been deciphered. It means that one existed
> > but undeciphered portion of language (called Messapic) is a member of
> > a larger ghost family.
>
> I don't care about Illyrian for the moment - yet I knew that the word
> "Illirian" will pop up somehow... A partial disclosure: in one oversee
> country, in one respectable higher education establishment, one
> student made a stand on the professor. The professor preferred to
> double check a part of materials used for the online course. Himself
> is a specialist in Caucasian ergatic languages, he knew me as a friend
> and PIE specialist (Hittite primarily though), so he asked me to get
> some more material for him. It is not some "joined committee
> investigation" of any kind - everyone wants to keep it quiet and
> private.
>
> P.S. The offending part was in an online course largely based on
> "Linguistics and the Teaching of the Less-Commonly Taught Languages"
> by Brian D. Josephhttp://www.seelrc.org/glossos/issues/9/joseph.pdf
> but it has no relation neither with the original course developer nor
> with the university it was first used.
>
> P.S.S. I personally already hate that day he called my and I was
> stupid to say "yes".
> Besides being put - indirectly - in some kind of natio-linguistical
> rwars, I am spending ridiculously a lot of time - not counting money
> to be reimbursed - for getting the most basic starting data. I am
> really surprised how a scanty group of researchers could make such a
> hideous disorder out of just 300 or less small inscriptions. I am just
> glad that Co weren't on charge of Bogazkoy archives - or neither
> Hittite nor Hatti could be studied. Say until now I cannot get, from
> where the Vaste Inscription is coming from: from "Corpus Inscriptionum
> Messapicarum" by Ribezzo, from the appendix to the Corpus by Fabretti,
> or someone of them (who?) is reproducing something (what?) from
> Mommsen? For a number of inscriptions no one of them can tell if it
> was copied from a cave, a tombstone, a column, a pot or a la "Mene,
> tekel, upharsin" had been seen as fire letters in the air... :-\
> OK, sorry for bias, I guess they still did a great job for their time.
>
> With the Vaste Inscription still pending for check out, two other most
> used inscriptions seems can be confirmed as still existing, authentic
> and properly transliterated:
>
> 1) from De Simone expedition to Grotta della Poesia, Melendugno,
> Lecce:
>
> klauhizis
> dekiasartahias
> thautouriandirahho
> dausapistathivinaihi
>
> 2) Galatina inscription on a stone plate (tombstone ??) currently kept
> in the museum of Bologna - thanks to Du¹an for the museum link:
>
> klohizisanthosthot
> orridasanaaprodi
> taapaogrebis

Here's another:
"From the Vaste inscription (Corpus Inscriptionum Messapicarum 149):

klohi zis thotoria marta pido vastei basta veinan aran in daranthoa
vasti staboos xohedonas daxtassi vaanetos inthi trigonoxo a staboos
xohetthihi dazimaihi beiliihi inthi rexxorixoa kazareihi xohetthihi
toeihithi dazohonnihi inthi vastima daxtas kratheheihi inthi ardannoa
poxxonnihi a imarnaihi "

"Another Messapic inscription from Galatina is dated to the 2nd
century BC:

klohi zis anthos thotorridas ana aprodita apa ogrebis
The separation of the last two elements is uncertain (apa, ogrebis, as
shown here). Klohi (as klauhi in the preceding inscription) probably
means "listen, hear". Zis may be the Messapic Zeus, as in the
preceding inscription. Aprodita is a loanword from Greek Aphrodite.
Anthos Thotorridas is a Messapic anthroponym, showing a personal name
plus patronymic or nomen gentile in the genitive (-as).

The Messapian language is preserved in a scanty group of perhaps fifty
inscriptions, of which only a few contain more than proper names, and
in a few glosses in ancient writers collected by Mommsen
(Unteritalische Dialekte, p. 70). Unluckily very few originals of the
inscriptions are now in existence, though some few remain in the
museum at Taranto. The only satisfactory transcripts are those given
by:

Mommsen (loc. cit.)
John P Droop in the Annual of the British School at Athens
(1905-1906), xli. 137, who includes, for purposes of comparison, as
the reader should be warned, some specimens of the 'unfortunately
numerous class of forged inscriptions.
A large number of the inscriptions collected by Gamurrini in the
appendices to Fabretti's Corpus inscriptionum italicorum are
forgeries, and the text of the rest is negligently reported. It is
therefore safest to rely on the texts collected by Mommsen, cumbered
though they are by the various readings given , to him by various,
authorities. In spite, however, of these difficulties some facts of
considerable importance have been established.

The inscriptions, so far as it is safe to judge from the copies of the
older finds and from Droop's facsimiles of the newer, are all in the
Tarentine-Ionic alphabet (with <no font for this character> for v and
<no font for this character> for h). For limits of date 400-150 BC may
be regarded as approximately probable; the two most important
inscriptions--those of Bindisi and Vastemay perhaps be assigned
provisionally to the 3rd century BC. Mommsen's first attempt at
dealing with the inscriptions and the language attained solid, if not
very numerous, results, chief of which were the genitival character of
the endings -aihi and -ihi; and the conjunctional value of inthi (loc.
cit. 79-84 sg(1).

Since 1850 little progress has been made. The Norwegian scholar Alf
Torp (1853-1916) in Indogermanische Forschungen (1895), V, 195, deals
fully with the two inscriptions just mentioned, and practically sums
up all that is either certain or probable in the conjectures of his
predecessors. Hardly more than a few words can be said to have been
separated and translated with certainty--kalatoras (masc. gen. sing.)
"of a herald" (Written upon a herald's staff which was once in the
Naples Museum); "aran" (acc: sing. fem.) "arable land"; mazzes,
"greater" (neut. acc. sing.), the first two syllables of the Latin
maiestas; while tepise (3rd sing. aorist indic.) "placed" or
"offered"; and forms corresponding to the article (ta = Greek to) seem
also reasonably probable."
..
Note:
- The link to Illyrian appears certain - if already old and divergent
at the time of the inscriptions;

"of -ni- to -nn- (as in the Messapian praenomen Dazohonnes vs. the
Illyrian praenomen Dazonius; the Messapian genitive Dazohonnihi vs.
Illyrian genitive Dasonii, etc.)
of -ti- to -tth- (as in the Messapian praenomen Dazetthes vs. Illyrian
Dazetius; the Messapian genitive Dazetthihi vs. the Illyrian genitive
Dazetii; from a Dazet- stem common in Illyrian and Messapian)
of -si- to -ss- (as in Messapian Vallasso for Vallasio, a derivative
from the shorter name Valla)
the loss of final d (as in tepise), and probably of final t (as in -
des, perhaps meaning "set", from PIE *dhe-, "to set, put")"

..But actually not from *PIE because *PIE never existed.
More likely from some original related to a Baltic form of 'tupi' +
'sed'

- "aran" (acc: sing. fem.) "arable land" corresponds to Baltic Latvia
'arau'
"mazzes" "greater" (neut. acc. sing.), now means the opposite in
Baltic Latvia "mazais" - 'the smaller (one)'.


lora...@cs.com

unread,
Jan 27, 2008, 7:27:38 AM1/27/08
to
On Jan 6, 11:38 pm, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
wrote:
> "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote in message
>
> news:a117fd9e-1e16-4e1f...@v46g2000hsv.googlegroups.com...
>
>
>
>
>
> >On Jan 6, 10:44 am, "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:
> >> On Jan 6, 4:56 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
> >> wrote:
>
> >> > It seems that Serbian women deliver their babies first ("porodi")
> >> > and then go into labour (start "rodit").
>
> >> > How come Dusan doesn't propose as ur-form the verb "zroditi"
> >> > with his favoured "z-". Or perhaps "uroditi", "uroda" (harvest),
> >> > or "priroda" (nature). These must surely be old, old ur-forms.
> >> > Obviously! Nu? :-)
>
> >> Iz Bratstvo (brotherhood, fraternity) related to Porodstvo/Porodica
> >> (family)?
> >> Is Brother not related to brat (OSlav. братръ, Czech bratr)?
> >> What about OCHS беременная, Serbian bremenit/a (pregnant) and Latin
> >> praegnans (pregnant); all with the primary meaning of "carry", bear";
> >> cf. Serb. preneti (carry over, convey).
>
> >> Sveti mati boga! Možda vi možete čitati, da ne razumeti!!!
>
> >> Bravo Kriha! This is the royal Serbian! You are speaking in the same
> >> way as the (potential) Serbian king does.
>
> >> Sveta Majko Božja! Možda vi možete čitati, a da ne razumete! :-)
>
> >> DV
>
> >Sorry, Russ. беременная (gravid); OCHS брѣмѩ; also Serb. breme (load)
> >Another question Kriha, what do you think where the Czech word březí
> >(gravid) is coming from?
>
> It's an adjective related to nouns břéme^ and břemeno (load, weight).
>
> It's applied only to cattle (possibly deer), never ever to people
> or other animals like dogs.
> (The month březen is the time of a year for the cows go gravid).
>
> Different expressions apply to other animals and quite separate
> set of expressions to humans.

Latvian 'bremze' - 'brakes'.. as in slowing down.

lora...@cs.com

unread,
Jan 27, 2008, 7:34:34 AM1/27/08
to
On Jan 7, 7:33 pm, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
> "Brian M. Scott" wrote:
>
> > [...]

>
> > <Parthénos>, <partus>, <poroditi>, and <birth> are mutually
> > unrelated.  <Parthénos> has not been established to have any
> > relatives; <partus> is from a PIE *perh3- 'to produce,
> > procure'; <birth> is from PIE *bHer- 'to carry'; and
> > <poroditi> is a prefixed derivative of PSl *rodìti 'give
> > birth, bear (fruit)', for which Derksen reconstructs PIE
> > *(h)rodH-eie-.
>
> Eric P. Hamp, in _Homenaje Tovar_ (Madrid 1972) 177-180, derives
> <parthénos> from PIE *bhr.g^hwéno- originally 'full-grown', from
> *bherg^h- 'to increase, grow'.  Since many of the reflexes of this
> verb in other languages refer to hills or high points, it seems at
> least as plausible to me that the sense of <parthénos> was something
> like 'exalted' or 'protected' (cf. Eng. "virgin on a pedestal"),
> before acquiring its historical meaning of 'virgin, maiden'.

Good grief.. amazing constructs piled upon constructs that you people
have to mess with..
Find a genetically related word to 'thenos' and you will have your
answer..
'Par' simply means 'across' or 'over'.

lora...@cs.com

unread,
Jan 27, 2008, 7:43:57 AM1/27/08
to
On Jan 8, 5:32 am, "Dušan Vukotić" <dusan.vuko...@gmail.com> wrote:

> On Jan 8, 4:33 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
>
>
>
>
>
> > "Brian M. Scott" wrote:
>
> > > [...]
>
> > > <Parthénos>, <partus>, <poroditi>, and <birth> are mutually
> > > unrelated.  <Parthénos> has not been established to have any
> > > relatives; <partus> is from a PIE *perh3- 'to produce,
> > > procure'; <birth> is from PIE *bHer- 'to carry'; and
> > > <poroditi> is a prefixed derivative of PSl *rodìti 'give
> > > birth, bear (fruit)', for which Derksen reconstructs PIE
> > > *(h)rodH-eie-.
>
> > Eric P. Hamp, in _Homenaje Tovar_ (Madrid 1972) 177-180, derives
> > <parthénos> from PIE *bhr.g^hwéno- originally 'full-grown', from
> > *bherg^h- 'to increase, grow'.  Since many of the reflexes of this
> > verb in other languages refer to hills or high points, it seems at
> > least as plausible to me that the sense of <parthénos> was something
> > like 'exalted' or 'protected' (cf. Eng. "virgin on a pedestal"),
> > before acquiring its historical meaning of 'virgin, maiden'.
>
> Where the "parent" and "pairing" (Serb. 'parenje' coupling) are coming
> from? παρθένος is a marriageable maiden; it means that parthenos is a
> grown up female or a girl that just reached fertility - ready for
> pairing. In Serbian it would be "porodna" (fertile). Compare Serbian
> 'porodilja' (pregnant woman) and English adjective 'fertile' and
> 'parental'.
>
> Nothing to do with 'exalted' or 'protected'
> DV

Every blue moon or so, you do get things right..

eg: *(Ap + rodiit + a)


lora...@cs.com

unread,
Jan 27, 2008, 8:02:30 AM1/27/08
to
On Jan 8, 9:23 am, VK <schools_r...@yahoo.com> wrote:
> On Jan 8, 7:23 am, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
[...]

>The Vaste Inscription (if authentic) is especially
> demonstrative in this aspect:

Actually it would be informative to learn how the 'Vaste' inscription
got its name:
"- the Thracian tribal name Astai (in Strandzha) and country name
Astik and the pre Greek ásty 'a town', but there are also similar
Messapian vastei (dat.) 'a town', the Old Indian 'vástu-' 'a house',
etc. words. "

I.e The 'town' Inscription ?

> From the point of view of possible segmentations of available
> inscriptions it is crucial to know the exact origin of each "klohisis-
> ihi" inscription. I temporarily call as "klohisis-ihi" all
> inscriptions with initial "klohisis" and "ihi" formant(s) in the rest
> of the sentence.
> If they are all from tomb/cave inscriptions then my primary guess
> would be about very common tradition of curses against possible tomb
> riders.

> klohizisthotoriamartapidovasteibasta


> veinanaranindaranthoavastistaboos
> xohedonasdaxtassivaanetosinthitrigonoxo
> astaboosxohetthihidazimaihibeiliihi
> inthirexxorixoakazareihixohetthihitoeihithi
> dazohonnihiinthivastima
> daxtaskratheheihiinthiardannoapoxxonnihia
> imarnaihi
>

> It definitely has some pseudo-poetic rhythm, reminding me very much
> the Neo-Phrygian "ios ni semoun knoumane kakon abberet..." ("whoever
> anything bad does to this tomb...") and then various lists of
> punishments that should follow. They also have that pseudo-poetic
> rhythm because of usage of uniformed syntagmas of the kind "then gods
> punish him", "then his wife stays fruitless", "then his balls
> disappear" (I am not fantasying, one of real courses), "then his legs
> get broken", etc.
>

> With such interpretation it is possible that "ihi" indeed has a
> genitival value, but not as a genitive ending but as postfixed
> possessive pronoun with the meaning "his".
>

> Of course if any "klohisis-ihi" inscription proven to be found on an
> offering or another clearly non-burial holder then all this
> speculation doesn't worth the time it took to type it in.

"Klohizis" is imperative singular case. (no Zeus to be found)
It is a salutation - just as the other respondent wrote, formalized as
a 'Hear Ye' introductory anlogue.

Baltic Latvian speakers would spell it as 'Klausis!'
It mean '(You) Listen!'

PS: And thus should not be referred to as 'proto-germanic' as Gothic
only has the more distant <háusjan> 'to hear, listen'.

Klausis!


Douglas G. Kilday

unread,
Jan 28, 2008, 7:52:34 PM1/28/08
to

Yes. The aspiration was not long enough in duration to require
compensation. If it had been, an expression like <kathólou>, for
<kath' hólou>, by apocope from <kat(à) hólou>, would have its first
syllable scanned long. Indeed, we would then expect all aspirates to
make position in Greek scansion, but in fact they do not.

Douglas G. Kilday

unread,
Jan 28, 2008, 8:22:06 PM1/28/08
to
On Jan 27, 1:02 pm, lorad...@cs.com wrote:

> On Jan 8, 9:23 am, VK wrote:
>
> [...]
>
> >The Vaste Inscription (if authentic) is especially
> > demonstrative in this aspect:
>
> Actually it would be informative to learn how the 'Vaste' inscription
> got its name:
> "- the Thracian tribal name Astai (in Strandzha) and country name
> Astik and the pre Greek ásty 'a town', but there are also similar
> Messapian vastei (dat.) 'a town', the Old Indian 'vástu-' 'a house',
> etc. words. "
>
> I.e The 'town' Inscription ?

Utter balderdash. Vaste is mentioned by Pliny as Basta, home of the
Basterbini, and an earlier Greek citation is Baûsta. (Here we see the
monophthongization of -au- to later Messapic -a- in stressed
syllables; it goes to -u- in unstressed ones.) In the inscription,
<basta> is evidently in the dative as the object of a dedication, like
<aprodita> in several other inscriptions. Thus it is reasonable to
conclude that Basta was the name of a goddess who had a local shrine
here, and the village of Basta took its name after her. As for
etymology, a plausible guess is that <bausta> comes from the o-grade
of an extended form of PIE *bheu- 'to grow, produce, especially of
vegetation', the goddess Ba(u)sta then being connected with the growth
of plants.

Mess. "vastei", as I indicated elsewhere, is a forced reading not
based on the surviving apographs of this text. Moreover, Mess. beta
and digamma are not interchangeable.

Douglas G. Kilday

unread,
Jan 28, 2008, 9:08:46 PM1/28/08
to
On Jan 5, 9:41 pm, "Dušan Vukotić" wrote:
>
> [...]
>
> It is ridiculous! Messapic was written in one uninterrupted row; there
> is no signs of interpunction -http://www.telemaco.unibo.it/rom/italia/img05/messapi2.jpg
> - and no one can be sure how exactly to read this inscription.

This inscription is from Ceglie. More details and an attempt at
reading the inscription by Carlo de Simone can be found here:

http://www.storia.unibo.it/italia/apulia.htm

I read the text as follows:

anaaproditalahonatheoto
riddahipakathitheotoriddao
atoraskeokhorrihibiliv a

De Simone reads the last letter of the second line as theta, and
assumes another theta broken away from the left end of the third line,
and an omicron between the first two visible letters of that line. He
translates thus:

'To Aphrodite Ana, on behalf of Lahona Theotorridath and Hipaka.
Theotorridath, daughter of Thaotor Keosorres.'

There is plenty wrong with that. First, there is no way the same
onomastic element would be spelled both <Theotor> and <Thaotor> in the
same text. I agree that the left end of the third line is broken, but
it is much better to restore [pl]. We have the genitive <Platoras> in
another inscription of Ceglie.

The notion that <ana> is an epithet of Aphrodite is unjustifiable. It
is much more reasonable to take it as a preposition 'to' with the
dative, here and in the Galatina inscription. Unlike the Galatina
text, this dedication has no explicit verb, but it is plain silly to
regard the following proper names as "on behalf of" whomever; they are
simply nominatives. I agree that -thi is the enclitic 'and', and
since no other examples exist of personal names in -th, I consider the
omicron at the end of the second line to be in error, the first stage
of a theta carved by dittography, the engraver about to add the
enclitic -thi, then realizing the error and abandoning it. The names
of the dedicants are then Lahona Theotoridda and Hipaka Theotorrida.

The last word is evidently some form of the word 'daughter', but since
two women are the dedicants, it cannot be in the singular. I
hypothesize a collective in -oa, since there is enough room in the
image for an effaced omicron between the digamma and the final alpha.
This, then, is my reconstruction of the text, with my segmentation,
and my translation:

ana aprodita lahona theoto-
ridda hipaka-thi theotoridda<o>
[pl]atoras keokhorrihi biliv[o]a

'To Aphrodite, Lahona Theotoridda and Hipaka Theotoridda, the
daughters of Plator Keokhorras, (dedicate this monument).'

lora...@cs.com

unread,
Jan 29, 2008, 12:23:18 AM1/29/08
to
On Jan 28, 5:22 pm, "Douglas G. Kilday" <fufl...@chorus.net> wrote:
> On Jan 27, 1:02 pm, lorad...@cs.com wrote:
>
> > On Jan 8, 9:23 am, VK wrote:
>
> > [...]
>
> > >The Vaste Inscription (if authentic) is especially
> > > demonstrative in this aspect:
>
> > Actually it would be informative to learn how the 'Vaste' inscription
> > got its name:
> > "- the Thracian tribal name Astai (in Strandzha) and country name
> > Astik and the pre Greek ásty 'a town', but there are also similar
> > Messapian vastei (dat.) 'a town', the Old Indian 'vástu-' 'a house',
> > etc. words. "
>
> > I.e The 'town' Inscription ?
>
> Utter balderdash.  

What do you mean 'Balderdash'?
I was asking a question not awaiting spontaneous dramatizations.

> Vaste is mentioned by Pliny as Basta, home of the
> Basterbini, and an earlier Greek citation is Baûsta.  (Here we see the
> monophthongization of -au- to later Messapic -a- in stressed
> syllables; it goes to -u- in unstressed ones.)  

Certainly here we also see 'V' transformed to 'B'.

> In the inscription,
> <basta> is evidently in the dative as the object of a dedication, like
> <aprodita> in several other inscriptions.  Thus it is reasonable to
> conclude that Basta was the name of a goddess who had a local shrine
> here, and the village of Basta took its name after her.  As for
> etymology, a plausible guess is that <bausta> comes from the o-grade
> of an extended form of PIE *bheu- 'to grow, produce, especially of
> vegetation', the goddess Ba(u)sta then being connected with the growth
> of plants.

Zero grade imaginings. *PIE never existed.

> Mess. "vastei", as I indicated elsewhere, is a forced reading not
> based on the surviving apographs of this text.  

Thanks for that at least; apparently previous researchers made errors.
Is this: "Messapian vastei (dat.) 'a town' ".. also a forced error?

lora...@cs.com

unread,
Jan 29, 2008, 12:28:32 AM1/29/08
to
> "Klohizis" is imperative singular ****. (no Zeus to be found)

> It is a salutation - just as the other respondent wrote, formalized as
> a 'Hear Ye' introductory anlogue.
>
> Baltic Latvian speakers would spell it as 'Klausis!'
> It mean '(You) Listen!'
>
> PS: And thus should not be referred to as 'proto-germanic' as Gothic
> only has the more distant <háusjan>  'to hear, listen'.
>
> Klausis!- Hide quoted text -

>
> - Show quoted text -

No comments from the erudite gallery?
It figures.

lora...@cs.com

unread,
Jan 29, 2008, 2:06:22 AM1/29/08
to
On Jan 26, 10:40 pm, "Douglas G. Kilday" <fufl...@chorus.net> wrote:

> Trond Engen wrote:
> > If we leave the -rz- < -rd-, how about *merg- "boundary, border" ->
> > "land"? Or the word known from ON <margr> "several, many"? Alas, I've
> > not been able to find it's etymology. We'd have to suppose rhotacism,
> > though.
>
> I like the first idea.  If memory serves, all the Celtic reflexes mean
> 'land', 'district' vel sim., including the formant in the Gaulish
> Allobroges 'Other-landers'.  Then we could have a zero-grade adjective
> *mrg-ja: > *margja > marza, and <thotoria marza> would be 'the
> citizenry of the land' or the like.  Nice!

Sure, analogy is also found in Baltic 'mer|s' meaning 'measure'.
('borders').. and elsehwhere (subsequently)..
E.g "obs.) "boundary," c.1290 (in ref. to the borderlands beside
Wales, rendering O.E. *Mercia*)

> > > [...]   Then we have for the first line (more or less):  'Hear ye!
> > > The ... citizenry constructed (this)  sacred enclosure to the
> > > (revered) Basta.'
>
> I now think <tei> is probably a fossilized locative of the
> demonstrative stem, 'in this place' = 'here', rather than a dative in
> apposition with <basta>.  

Sure try Baltic Latvian demonstrative 'te' instead of 'tei'..also
means 'here'.

> Then we have:  'Hear ye!  The citizenry of
> the land constructed the sacred enclosure here to Basta.'
>
> > "Hear ye! Many leaders of the people constructed ..."
>
> > Holding the two together I'd say that <thotorridas> looks like "chief
> > horseman". The etymology of the title, of course, doesn't say anything
> > of the duties of its bearer at the time of the inscription.
>
> An epitaph from Ostuni appears to refer to a 'charioteer', though the
> morphology is peculiar, and the fellow's first name happens to be
> Theotoras:
>
> theotoras artahiaihi bennarrihino
>
> <benna> has been explained as a loanword from the Gaulish for
> 'chariot'; presumably *bennarras would correspond to a Latin
> *benna:rius 'charioteer', and if we could drop the -no we could read
> 'Theotoras, son of Artahias the charioteer'.  But the pesky -no
> remains.  My best guess is that it represents the genitive plural in *-
> o:m of an adjective in -no-, so the last word corresponds to a Latin
> *benna:ri:no:rum, perhaps '(chief) of those serving the charioteers'??

In Baltic Latv. it appears genitive but now the ending would be
rendered as 'rinu'..but still sounds very close to 'rino';
orthography...

'Daktas' appears multiple times. I therefore am predisposed to not
consider it to be a proper name.

Try 'the Baltic Latv.'valsti' instead of 'vasti'.. it means 'the
government'.

> I don't like the idea that beta and digamma could be interchanged in
> Messapic.  I can find no evidence for this in the personal names.  The
> word <vasti> is problematic.  The PID authors take it for a neuter
> noun.  I would like to read it as a numeral adverb (indicating which
> time in the consulship) but cannot justify that.  Or it might be a
> verb, since Messapic verbal morphology is so poorly understood.  I
> can't explain the relation of <vasti> to <vastima> under either
> hypothesis.

va(l)sti - dative,
va(l)stei - definitely dative
va(l)stima - genitive ?

Joachim Pense

unread,
Jan 31, 2008, 5:32:27 PM1/31/08
to
Douglas G. Kilday wrote:

> On Jan 27, 8:04 am, Joachim Pense wrote:
>> Am Sat, 26 Jan 2008 21:36:37 -0800 (PST) schrieb Douglas G. Kilday:
>>
>> > The Messapic alphabet is derived from the Tarentine Doric, and
>> > examples of Tarentine words with intervocalic -h- are known, such as
>> > <oudehen> 'nothing, not at all'.  Obviously the -e of <de> was not
>> > apocopated in Tarentine as it was in Attic-Ionic <oudén> from
>> > *oud'(h)én, and it is not difficult to envision loss of the
>> > intervocalic /h/ sound with a long vowel resulting, that is /oude:n/
>> > still written <oudehen>, like Latin /ni:l/ written <nihil>.
>>
>> Is there an explanation why there was no lengthening of the -e- in the
>> Attic-Ionic ouden, compensating the -h-?
>
> Yes. The aspiration was not long enough in duration to require
> compensation. If it had been, an expression like <kathólou>, for

> <kath' hólou>, by apocope from <kat(ŕ) hólou>, would have its first


> syllable scanned long. Indeed, we would then expect all aspirates to
> make position in Greek scansion, but in fact they do not.

Thanks, that sounds convincing.

Joachim

Italo

unread,
Feb 1, 2008, 12:05:43 PM2/1/08
to
Douglas G. Kilday wrote:

> The Messapic alphabet is derived from the Tarentine Doric

The Peuceti seem to've used also another script:
http://members.home.nl/cucaro/loomweight.html


Trond Engen

unread,
Feb 2, 2008, 5:19:20 AM2/2/08
to
I happened to stumble upon a useful idea in an earlier post. Now I don't
know how to stop.

Douglas G. Kilday skreiv:

> Trond Engen wrote:
>
>> Douglas G. Kilday skreiv:
>>

>>> The inscription [...] begins with <klohizis> and contains no forms
>>> in -ihi. It is plausibly explained as a dedicatory inscription to

> -n- is correct possibly borrowed from Greek <anthoűs> 'flowery'??) is

> indeed a personal name, <thotorridas> may be a Doric-style patronymic
> based on the native personal name Th(e)otor, cognate with the
> praenomen Tutor of the Cloelii of Alba Longa.

Could <avithos> be identical to Latin <avitus> "belonging to (by
inheritence)"? And/or could <thotorridas> be name + title <thotor ridas>?

"Hear ye! On behalf of his ancestors Tutor the Coach to Aphrodite engraved"

> We have another instance in an epitaph from Ceglie:
>
> ettis arnisses theotorres
>
>>> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
>>> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
>>> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
>>> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
>>> inthi vastima dakhtas kratheheihi
>>> inthi ardannoa po[ll]onnihiai marnaihi

I didn't notice before that the supposed names in line two are notably
different from those of lines three to six. Titles in stead of
patronyms? Or is it something like:

In <daranthoa vasti staboos> Khonedonas Dakhtas Sivaanetos
And in <tri[g]onokhoa staboos> also Khonetthis and Dazimais and Beil[l]is
And in <re[ll]orikhoa> also Kazareis and Khonetthis and [O]toeihits and
Dazohonnis
And in <vastima> Dakhtas and Kratheheis
And in <ardannoa> also Po[ll]onnia and Aimarnais

(-thi replacing -hi under some conditions, e.g. after dentals)

Could <staboos> be "town" or "village"? More on <vasti> later.

>>> I believe this to be a dedicatory inscription with a postscript
>>> recording the names of public officials at the time of dedication.
>>> In the first line, <klohizis> is again 'hear ye!' or the like.
>>> <thotoria> is something like 'citizenry', modified by the adjective
>>> <marza> (from *mardia, but I have no sure etymology; the -z- is
>>> from Mommsen's variant reading).
>>
>> If we leave the -rz- < -rd-, how about *merg- "boundary, border" ->

>> "land"? [...]
>
> [...] If memory serves, all the Celtic reflexes mean 'land',

> 'district' vel sim., including the formant in the Gaulish Allobroges
> 'Other-landers'. Then we could have a zero-grade adjective *mrg-ja:
> > *margja > marza, and <thotoria marza> would be 'the citizenry of
> the land' or the like. Nice!

Playing with my Latin dictionary, could <veinan> be "payment" or
"devotion" and <aran> be "soil" or "shrine"? Or some related verbs, of
course.

>>> [...] Then we have for the first line (more or less): 'Hear ye!
>>> The ... citizenry constructed (this) sacred enclosure to the
>>> (revered) Basta.'
>
> I now think <tei> is probably a fossilized locative of the
> demonstrative stem, 'in this place' = 'here', rather than a dative in
> apposition with <basta>. Then we have: 'Hear ye! The citizenry of
> the land constructed the sacred enclosure here to Basta.'

"Hear ye! The-people of-the-land made this to-Basta. Pay-did
for-the-shrine: in Daranthoa ..."
No, I don't get the cases to match, do I?

More along your lines, then:
"Hear ye! The-people of-the-land made here to-Basta for-devotion shrine.
In Daranthoa ..."

> [...]


>
> The word <vasti> is problematic. The PID authors take it for a
> neuter noun. I would like to read it as a numeral adverb (indicating
> which time in the consulship) but cannot justify that.

That the names are both a reference to the year of the inauguration and
a tribute to the principals? Or a simple list by rank? (Is there any way
you could get <rellorikhoa> to be "fourth ..."? Seems far-fetched: If
the supposed rho is a rare qoppa and the ligature really is a double chi
it would read <qekhkhorikhoa>, and I don't dare thinking of the sound
laws we'd have to postulate.)

> Or it might be a verb, since Messapic verbal morphology is so poorly
> understood. I can't explain the relation of <vasti> to <vastima>
> under either hypothesis.

If related to Latin <vastus>, <vasti> could simply be (a form of)
"great" and <vastima> "the biggest" (-> "the wider region"?).

--
Trond Engen
- vasting time

Douglas G. Kilday

unread,
Feb 9, 2008, 3:38:02 PM2/9/08
to
On Jan 29, 5:23 am, lorad...@cs.com wrote:

> On Jan 28, 5:22 pm, "Douglas G. Kilday" wrote:
> > On Jan 27, 1:02 pm, lorad...@cs.com wrote:
> > > On Jan 8, 9:23 am, VK wrote:
>
> > > [...]
>
> > > >The Vaste Inscription (if authentic) is especially
> > > > demonstrative in this aspect:
>
> > > Actually it would be informative to learn how the 'Vaste' inscription
> > > got its name:
> > > "- the Thracian tribal name Astai (in Strandzha) and country name
> > > Astik and the pre Greek ásty 'a town', but there are also similar
> > > Messapian vastei (dat.) 'a town', the Old Indian 'vástu-' 'a house',
> > > etc. words. "
>
> > > I.e The 'town' Inscription ?
>
> > Utter balderdash.  
>
> What do you mean 'Balderdash'?
> I was asking a question not awaiting spontaneous dramatizations.

The modern name Vaste comes from Messapic Basta, earlier Bausta, and
the initial /b/ cannot correspond to /w/ in the other words. The
reading "vastei", even if it were correct, could not be related,
because it occurs in the same text as <basta>.

> > Vaste is mentioned by Pliny as Basta, home of the
> > Basterbini, and an earlier Greek citation is Baûsta.  (Here we see the
> > monophthongization of -au- to later Messapic -a- in stressed
> > syllables; it goes to -u- in unstressed ones.)  
>
> Certainly here we also see 'V' transformed to 'B'.

We see no such thing. What we see is the Latin phoneme /w/ slowly
closing down to a voiced fricative, and eventually falling together
with /b/ in some dialects, including the one spoken here. That is why
we have the confusion of another Messapic town as Balesium (Pliny, MS
var. Balaessum), Valetium (Mela), Valentia (Itin. Hier.), Balentium
(Tab. Peut.), Baleso (Galateo, _De situ Japygiae_, pub. 1558, written
ca. 1510), and Valesio (locals reported by Mommsen in 1848). This
phenomenon did not begin until several centuries after the date of the
Messapic texts under discussion, in which there is no confusion
between beta (/b/, <b>) and digamma (/w/, <v>).

> > In the inscription,
> > <basta> is evidently in the dative as the object of a dedication, like
> > <aprodita> in several other inscriptions.  Thus it is reasonable to
> > conclude that Basta was the name of a goddess who had a local shrine
> > here, and the village of Basta took its name after her.  As for
> > etymology, a plausible guess is that <bausta> comes from the o-grade
> > of an extended form of PIE *bheu- 'to grow, produce, especially of
> > vegetation', the goddess Ba(u)sta then being connected with the growth
> > of plants.
>
> Zero grade imaginings. *PIE never existed.

Comments like that only make you sound like a nuthead. Zero-grade and
o-grade are two different things, so make that an _uninformed_
nuthead.

> > Mess. "vastei", as I indicated elsewhere, is a forced reading not
> > based on the surviving apographs of this text.  
>
> Thanks for that at least; apparently previous researchers made errors.
> Is this: "Messapian vastei (dat.) 'a town' ".. also a forced error?

It is the _same_ forced error in reading the Vaste text. The first
line of the apograph should be read and segmented as

klohizis thotoria mar[z]a pidogas tei basta

and not as

klohi zis thotoria mar[t]a pido vastei basta.

Douglas G. Kilday

unread,
Feb 9, 2008, 4:28:20 PM2/9/08
to
On Jan 29, 7:06 am, lorad...@cs.com wrote:

> On Jan 26, 10:40 pm, "Douglas G. Kilday" wrote:
> > Trond Engen wrote:
> > > If we leave the -rz- < -rd-, how about *merg- "boundary, border" ->
> > > "land"? Or the word known from ON <margr> "several, many"? Alas, I've
> > > not been able to find it's etymology. We'd have to suppose rhotacism,
> > > though.
>
> > I like the first idea.  If memory serves, all the Celtic reflexes mean
> > 'land', 'district' vel sim., including the formant in the Gaulish
> > Allobroges 'Other-landers'.  Then we could have a zero-grade adjective
> > *mrg-ja: > *margja > marza, and <thotoria marza> would be 'the
> > citizenry of the land' or the like.  Nice!
>
> Sure, analogy is also found in Baltic 'mer|s' meaning 'measure'.
> ('borders').. and elsehwhere (subsequently)..
> E.g "obs.) "boundary," c.1290 (in ref. to the borderlands beside
> Wales, rendering O.E. *Mercia*)
>
> [...]
>
> > I now think <tei> is probably a fossilized locative of the
> > demonstrative stem, 'in this place' = 'here', rather than a dative in
> > apposition with <basta>.  
>
> Sure try Baltic Latvian demonstrative 'te' instead of 'tei'..also
> means 'here'.

Good. Thank you. It appears that both Messapic and Latvian have
specialized the loc. sg. of the dem. stem *to- as 'here'. On the
other hand Messapic, unlike Latvian, seems to have lost the productive
use of the locative, and to retain only four or five productive cases
(nom., acc., dat., gen., perhaps voc.), like classical Greek.

> > Then we have:  'Hear ye!  The citizenry of
> > the land constructed the sacred enclosure here to Basta.'
>
> > > "Hear ye! Many leaders of the people constructed ..."
>
> > > Holding the two together I'd say that <thotorridas> looks like "chief
> > > horseman". The etymology of the title, of course, doesn't say anything
> > > of the duties of its bearer at the time of the inscription.
>
> > An epitaph from Ostuni appears to refer to a 'charioteer', though the
> > morphology is peculiar, and the fellow's first name happens to be
> > Theotoras:
>
> > theotoras artahiaihi bennarrihino
>
> > <benna> has been explained as a loanword from the Gaulish for
> > 'chariot'; presumably *bennarras would correspond to a Latin
> > *benna:rius 'charioteer', and if we could drop the -no we could read
> > 'Theotoras, son of Artahias the charioteer'.  But the pesky -no
> > remains.  My best guess is that it represents the genitive plural in *-
> > o:m of an adjective in -no-, so the last word corresponds to a Latin
> > *benna:ri:no:rum, perhaps '(chief) of those serving the charioteers'??
>
> In Baltic Latv. it appears genitive but now the ending would be
> rendered as 'rinu'..but still sounds very close to 'rino';
> orthography...

Yes, you have -u from PIE *-o:m. Messapic has -o from both the gen.
pl. *-o:m and the nom. sg. *-o:n in names like Moro, losing the nasal
after the long round vowel.

> [...]


>
> > > > and in the ... Dakhtas (the son) of Krathehes, and in the ...
> > > > (the daughter?) of Pollonnis (the son?) of Marnas.'
>
> > > IS a variant spelling <b/vasta> implausible? Since Dakhtas serves in
> > > <vastimae>, apparently connected to <vasti>, could these two last lines
> > > list leaders of circuits? And <daranthoa vasti> mean "consulship(?) of
> > > Vasta"?
>
> 'Daktas' appears multiple times. I therefore am predisposed to not
> consider it to be a proper name.

There is no question about Dakhtas being a praenomen. The question is
whether it is in the nom. or gen. case. In my previous postings, I
assumed that it was a nom. masc., and that Dakhta was a corresponding
nom. fem. Mommsen divided his insc. #2 from Ceglie as

dakhta morthana [v.l. morolna] aprodita hipades

and Deecke followed with

dakhta moroana aprodita hipades

which he rendered as 'Dachta Moroana stellte für die Aphrodite auf'.
I now believe that both M. and D. were incorrect, and that <ana> is
the preposition 'for' followed by the dat. sg. <aprodita>. The
dedicant is Dakhta Moro, the gentilicium being an /o:n/-stem, forms
like <laparedonas>, <baledonas>, and <krithonas> being genitives
singular of similar gentilicia. We should read 'Dakhta Moro
established (this monument) for Aphrodite'. Also, Dakhtas in the
Vaste text should be taken as gen. sg., which has important
ramifications for the interpretation, as explained below in my reply
to Trond.

> Try 'the Baltic Latv.'valsti' instead of 'vasti'.. it means 'the
> government'.

What is the etymology? What is the evidence for Messapic losing /l/
in clusters?

> > I don't like the idea that beta and digamma could be interchanged in
> > Messapic.  I can find no evidence for this in the personal names.  The
> > word <vasti> is problematic.  The PID authors take it for a neuter
> > noun.  I would like to read it as a numeral adverb (indicating which
> > time in the consulship) but cannot justify that.  Or it might be a
> > verb, since Messapic verbal morphology is so poorly understood.  I
> > can't explain the relation of <vasti> to <vastima> under either
> > hypothesis.
>
> va(l)sti - dative,
> va(l)stei - definitely dative
> va(l)stima - genitive ?

I now think that Mess. <vasti> can be understood as the dat. sg. of an
adj. 'federated' or the like, and <vastima> as the dat. sg. of a
derived thematic adj. Baltic cognates would be Lith. <vadúoti>
'redeems something pledged', Latv. <vaduo^t> 'id.'. For details see
my reply to Trond below.

Douglas G. Kilday

unread,
Feb 10, 2008, 12:37:11 AM2/10/08
to
> > -n- is correct possibly borrowed from Greek <anthoûs> 'flowery'??) is

> > indeed a personal name, <thotorridas> may be a Doric-style patronymic
> > based on the native personal name Th(e)otor, cognate with the
> > praenomen Tutor of the Cloelii of Alba Longa.
>
> Could <avithos> be identical to Latin <avitus> "belonging to (by
> inheritence)"? And/or could <thotorridas> be name + title <thotor ridas>?

Inherited *-os would become -as. I can't be sure about borrowings, so
I can't exclude the possibility that <avithos> is one, if the reading
is correct. If it is <anthos>, however, I am inclined to suspect a
name borrowed from Greek. Until I see an apograph or a photograph, I
will withhold judgment on the reading.

I also can't exclude the division <thotor ridas>, but if what precedes
it is a praenomen, a single Doric-style patronymic <thotorridas>
provides good agreement with other texts.

> "Hear ye! On behalf of his ancestors Tutor the Coach to Aphrodite engraved"
>
> > We have another  instance in an epitaph from Ceglie:
>
> > ettis arnisses theotorres
>
> >>> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> >>> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> >>> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> >>> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> >>> inthi vastima dakhtas kratheheihi
> >>> inthi ardannoa po[ll]onnihiai marnaihi
>
> I didn't notice before that the supposed names in line two are notably
> different from those of lines three to six. Titles in stead of
> patronyms? Or is it something like:
>
> In <daranthoa vasti staboos> Khonedonas Dakhtas Sivaanetos
> And in <tri[g]onokhoa staboos> also Khonetthis and Dazimais and Beil[l]is
> And in <re[ll]orikhoa> also Kazareis and Khonetthis and [O]toeihits and
> Dazohonnis
> And in <vastima> Dakhtas and Kratheheis
> And in <ardannoa> also Po[ll]onnia and Aimarnais
>
> (-thi replacing -hi under some conditions, e.g. after dentals)
>
> Could <staboos> be "town" or "village"? More on <vasti> later.

Looking at the various attestations of <dakhta(s)> and rereading
Whatmough's notes on Messapic grammar in _Prae-Italic Dialects_ forced
me to revise my analysis of the Vaste text. I now think that all the
personal names in sections 2-6 are in the genitive. The names of
offices in -oa are, I believe, collectives. The exception, <vastima>,
has only one occupant, all officers being specified in binomial form
(i.e. "praenomen" + "gentilicium", though comparison with the Roman
model should not be taken too strictly). The text must be missing at
least one personage at the end, since the office in section 6 is a
collective, but only one binomial survives there.

The difference in (gentilician) names which you notice may be an
indirect reflection of social class. That is, the officers in the
daranthoa or "consulship", Staboas Khonedo and Dakhta Sivaanetas, may
have old-line "landed gentry" surnames as opposed to the others, whose
genitives in -ihi may reflect younger surnames based on patronymic
formations, much as Wilson, Robertson, and the like are relatively
"young" English surnames and formerly indicated humble status. But
given the small sample size in this text, such speculation is
necessarily idle.

Since this text has -o- for earlier -ao-, as in <klohizis>, the
surnames Khonedo and Khonetthes can hardly fail to remind us of the
Khaones, inhabitants of Khaonia in Epirus, on the border of Illyria
proper. Mommsen, and more recently Hamp, downplayed the notion of a
close connection between Illyrian and Messapic, but here we have one
piece of onomastic evidence favoring such a connection.

All things considered, I think we must take <staboos> as the genitive
of a praenomen. The nominative <staboas> appears in a tomb-
inscription reported by Minervini, <staboas porvaides> (with Ionic-
style patronymic), and the older form of the genitive is in
<staboaus / pollonnihi> (Minervini read the <ll> as <a>, but the
gentilicium is known from other sources). Whatmough assigned the name
to the /a:w/-stems; the genitive shows that it cannot be a simple /o/-
stem like <dazimas>, gen. <dazimaihi>.

Consonant-stems have genitives in -as from *-os, so <sivaanetos> like
<staboos> must be the gen. of a diphthong-stem; I assume -os from
earlier -aos (-aus) and, again, nom. in -as in contemporary Messapic.
The genitives in cons. + ihi, following Whatmough, are from /jo/-stems
with nominatives in cons. + es.

> >>> I believe this to be a dedicatory inscription with a postscript
> >>> recording the names of public officials at the time of dedication.  
> >>> In the first line, <klohizis> is again 'hear ye!' or the like.  
> >>> <thotoria> is something like 'citizenry', modified by the adjective
> >>> <marza> (from *mardia, but I have no sure etymology; the -z- is
> >>> from Mommsen's variant reading).  
>
> >> If we leave the -rz- < -rd-, how about *merg- "boundary, border" ->
> >> "land"? [...]
>
> > [...]  If memory serves, all the Celtic reflexes mean 'land',
> > 'district' vel sim., including the formant in the Gaulish Allobroges
> > 'Other-landers'.  Then we could have a zero-grade adjective *mrg-ja:
> > > *margja > marza, and <thotoria marza> would be 'the citizenry of
> > the land' or the like.  Nice!
>
> Playing with my Latin dictionary, could <veinan> be "payment" or
> "devotion" and <aran> be "soil" or "shrine"? Or some related verbs, of
> course.

For the moment, I prefer to retain my interpretation of <veinan> as
acc. sg. 'sacred', from *weig-nam (yes, the *gn > *ñ > n is ad hoc)
and <varan> as acc. sg. 'enclosure', from *woram (the apograph makes a
simple <aran> very unlikely, though most scholars agree on it; I
consider that to be another forced reading). Above all, 'sacred
enclosure' makes good sense in this type of inscription.

> >>> [...]   Then we have for the first line (more or less):  'Hear ye!
> >>> The ... citizenry constructed (this)  sacred enclosure to the
> >>> (revered) Basta.'
>
> > I now think <tei> is probably a fossilized locative of the
> > demonstrative stem, 'in this place' = 'here', rather than a dative in
> > apposition with <basta>.  Then we have:  'Hear ye!  The citizenry of
> > the land constructed the sacred enclosure here to Basta.'
>
> "Hear ye! The-people of-the-land made this to-Basta. Pay-did
> for-the-shrine: in Daranthoa ..."
> No, I don't get the cases to match, do I?

No, everyone seems to agree that <veinan (v)aran> is accusative,
whatever it actually means.

> More along your lines, then:
> "Hear ye! The-people of-the-land made here to-Basta for-devotion shrine.
> In Daranthoa ..."

Possibly. If you can etymologize <veinan> without any ad-hoc
lenition, it might be preferable to my version.

> > [...]
>
> > The word <vasti> is problematic.  The PID authors take it for a
> > neuter noun.  I would like to read it as a numeral adverb (indicating
> > which time in the consulship) but cannot justify that.
>
> That the names are both a reference to the year of the inauguration and
> a tribute to the principals? Or a simple list by rank? (Is there any way
> you could get <rellorikhoa> to be "fourth ..."? Seems far-fetched: If
> the supposed rho is a rare qoppa and the ligature really is a double chi
> it would read <qekhkhorikhoa>, and I don't dare thinking of the sound
> laws we'd have to postulate.)

Many scholars do read double chi here, but I don't see how it could
have arisen. On the other hand only the second apparent double chi,
near the end of the insc., can be corrected into double lambda with
certainty, making the gen. <pollonnihi> of a surname with a well-known
Messapic onomastic element. The word in section 4 could be
<rettorikhoa>, for all I know. I wish I could etymologize the darned
thing.

> > Or it might  be a verb, since Messapic verbal morphology is so poorly
> > understood.   I can't explain the relation of <vasti> to <vastima>
> > under either hypothesis.
>
> If related to Latin <vastus>, <vasti> could simply be (a form of)
> "great" and <vastima> "the biggest" (-> "the wider region"?).

I hadn't thought of that. During the past week I mulled over Deecke's
connection of <vasti> to Latin <vas>, gen. <vadis> 'pledge'; if (as
many scholars, including Whatmough, presume) <vasti> is a neuter noun,
the sense becomes 'pledge', 'guarantee', or the like. Now, in my
current interpretation, <vasti> is best taken as the dat. sg. of an
adj. modifying <daranthoa>, a collective noun 'consulate' or the
like. Two other rather mutilated inscc. also appear to contain
<daranthoa>. We know that Basta was (and still is) a small village,
rating a brief mention by Pliny, too small to issue its own coinage.
But the city of Uria, considered the capital of Messapia, issued
coinage marked in Messapic <Orra>. Probably the magistrates mentioned
in the Vaste text were from Uria (or less likely, in my opinion, from
Uxentum). Presumably their city acquired hegemony over its neighbors
as Rome did, by concluding treaties (foedera) with other communities.
Thus a reasonable meaning for the /i/-stemmed adj. *vastis, dat. sg.
<vasti>, is 'federated', 'federal', or the like, indicating that the
daranthoa in question was the supreme daranthoa of Messapia, its two
officers being the highest magistrates of the land.

Justifying the morphology is somewhat involved. In Latin we have the
adj. <tri:stis> 'sad', which has been plausibly explained as a
compound of *tri:- 'worn down' (tri:vi: perf. of <terere> 'to wear
down', <tri:bulum> 'threshing sledge') and *sti- 'standing'; the sad
person is in a persistent worn-down state. We also have Lat. <postis>
'door-post' from *por-sti- 'standing in front' and <testis> 'witness'
from *ter-sti- (*tr.-sti- < *tri-sti-) 'third-standing', i.e. between
two disputing parties, with derived verb <testa:ri:> 'to testify, act
as witness', both functioning as nouns. Further we have a verb
<praesta:re> 'to guarantee' connected with <praes>, gen. <praedis>
'pledge', whose archaic acc. pl. <praevides> (CIL 1{2}:585) shows that
it is a compound of <vas>, gen. <vadis>. Now <praesta:re> has been
explained as <praes sta:re> 'to stand as pledge', but it seems better
to take it, like <testa:ri:>, as a denominative from *praestis, itself
from *praevistis, from *prai-vad-stis 'standing before as pledge'.
This in turn implies an unprefixed *vad-stis, PIE *wadh-sti-,
'standing as pledge', or the like.

On the semantic side, we find some West Germanic reflexes of *wadh-
moving naturally from 'pledge' to 'contract' or 'treaty' or the like,
Old English <weddian> 'to make a contract, promise, marry', Old
Frisian <wed> 'contract, promise, pledge'. Thus it is not
unreasonable to suppose that PIE *wadh-sti- became in Messapic
*va(d)sti- 'standing in a contract, standing under a treaty,
pertaining to a federation, federal'. This adjective could also have
been substantivized as 'federation' itself, leading to the abstract
noun <vastima>. Possibly that noun originally meant 'leadership of
the federation', its holder ruling over Messapia (or part of it), but
then deposed by a Roman-style revolution which set up annual pairs of
public officials, keeping the <vastima> only for ceremonial purposes,
as the Romans kept the office of rex sacrificulus. But again this is
veering into idle speculation, as I have no evidence for the history
of Messapic political institutions. At any rate, the connection
between <vasti> and <vastima> is no longer inscrutable.

As for <daranthoa>, my guess is that it is the collective of some
derivative of the zero-grade of *dher- 'to hold firmly, support', but
I do not have the detailed morphology worked out, and other
explanations are possible. For the other offices, so far I have
nothing. This is my revised interpretation of the Vaste text:

klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran

'Hear ye! The citizenry of the land constructed here, to Basta, the
sacred enclosure

in daranthoa vasti staboos khonedonas dakhtas sivaanetos

in the federal consulate(?) of Staboas Khonedo (and) Dakhta Sivaanetas

inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi

and in the trigonokhate(?) of Staboas Khonetthes (and) Dazimas Beilles

inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi

and in the rellorikhate(?) of Kazareas Khonetthes and Otoeas
Dazohonnes

inthi vastima dakhtas kratheheihi
and in the federal presidency(?) of Dakhta Kratheheas

inthi ardannoa po[ll]onnihi aimarnaihi ...
and in the ardannate(?) of Pollonnes Aimarnas [and ...]'

Brian M. Scott

unread,
Feb 10, 2008, 12:50:40 AM2/10/08
to
On Sat, 9 Feb 2008 21:37:11 -0800 (PST), "Douglas G. Kilday"
<fuf...@chorus.net> wrote in
<news:e1b31b06-51bf-4a5e...@i7g2000prf.googlegroups.com>
in sci.lang:

[...]

> The difference in (gentilician) names which you notice may
> be an indirect reflection of social class. That is, the
> officers in the daranthoa or "consulship", Staboas
> Khonedo and Dakhta Sivaanetas, may have old-line "landed
> gentry" surnames as opposed to the others, whose
> genitives in -ihi may reflect younger surnames based on
> patronymic formations, much as Wilson, Robertson, and the
> like are relatively "young" English surnames and formerly

> indicated humble status. [...]

If you're contrasting patronymics in general with locative
bynames, rather than these particular patronymics, then I'll
have to disagree: patronymics are as old as locatives in
post-Conquest England and probably older in pre-Conquest
England, and they don't imply an original humble status.

Brian

Trond Engen

unread,
Feb 12, 2008, 7:15:43 PM2/12/08
to
Douglas G. Kilday skreiv:

> On Feb 2, 10:19 am, Trond Engen wrote:
>> Douglas G. Kilday skreiv:
>>> Trond Engen wrote:
>>>> Douglas G. Kilday skreiv:
>>>>

>>>>> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
>>>>> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
>>>>> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
>>>>> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
>>>>> inthi vastima dakhtas kratheheihi
>>>>> inthi ardannoa po[ll]onnihiai marnaihi

[...]

>> Playing with my Latin dictionary, could <veinan> be "payment" or
>> "devotion" and <aran> be "soil" or "shrine"? Or some related verbs, of
>> course.
>
> For the moment, I prefer to retain my interpretation of <veinan> as
> acc. sg. 'sacred', from *weig-nam (yes, the *gn > *ñ > n is ad hoc)
> and <varan> as acc. sg. 'enclosure', from *woram (the apograph makes a
> simple <aran> very unlikely, though most scholars agree on it; I
> consider that to be another forced reading). Above all, 'sacred
> enclosure' makes good sense in this type of inscription.
>
> [...]
>

>> More along your lines, then:
>> "Hear ye! The-people of-the-land made here to-Basta for-devotion
>> shrine.
>> In Daranthoa ..."
>
> Possibly. If you can etymologize <veinan> without any ad-hoc
> lenition, it might be preferable to my version.

I didn't have a better suggestion -- I merely reinterpreted the "sacred"
part as related to Lat. <Venus>. But it was wrong. The "worship" word
wouldn't give a diphtong.

I do see an odd chance, though. AFAIK, your *weig- is also known as
*weik- "dedicate, make sacred" (but I suppose you prefer *weig- for the
sake of the lenition). If this and the other roots *weik- ("clan",
"bind" and "conquer") are essentially the same extension of *wei- "turn,
twist", i.e. *wei-k- with a core meaning "bind" extended to "contend"
and "fence in", maybe Messapic used another extenson *wei-n- and
<veinan> is from *wein-am or something.

>>> [...]
>>> The word <vasti> is problematic. The PID authors take it for a
>>> neuter noun. I would like to read it as a numeral adverb
>>> (indicating which time in the consulship) but cannot justify that.
>>
>> That the names are both a reference to the year of the inauguration
>> and a tribute to the principals? Or a simple list by rank? (Is there
>> any way you could get <rellorikhoa> to be "fourth ..."? Seems
>> far-fetched: If the supposed rho is a rare qoppa and the ligature
>> really is a double chi it would read <qekhkhorikhoa>, and I don't
>> dare thinking of the sound laws we'd have to postulate.)
>
> Many scholars do read double chi here, but I don't see how it could
> have arisen. On the other hand only the second apparent double chi,
> near the end of the insc., can be corrected into double lambda with
> certainty, making the gen. <pollonnihi> of a surname with a
> well-known Messapic onomastic element. The word in section 4 could
> be <rettorikhoa>, for all I know. I wish I could etymologize the
> darned thing.

I did offer you a far-fetched qoppa for rho to achieve <qe..orikhoa>
"four-something", following <tri[g]onokhoa> "three-something". For that
purpose, <tt> is good and would even make <qettorikhoa> look regular.
But that isn't enough to turn the <r> into a <q>.

Could <rettor> and <tr[g]on> be calques of -- or parallel formations to
-- the Roman titles <praetor> and <tribun>?

>>> Or it might be a verb, since Messapic verbal morphology is so
>>> poorly understood. I can't explain the relation of <vasti> to
>>> <vastima> under either hypothesis.
>>
>> If related to Latin <vastus>, <vasti> could simply be (a form of)
>> "great" and <vastima> "the biggest" (-> "the wider region"?).
>
> I hadn't thought of that. During the past week I mulled over
> Deecke's connection of <vasti> to Latin <vas>, gen. <vadis> 'pledge';
> if (as many scholars, including Whatmough, presume) <vasti> is a
> neuter noun, the sense becomes 'pledge', 'guarantee', or the like.
> Now, in my current interpretation, <vasti> is best taken as the dat.
> sg. of an adj. modifying <daranthoa>, a collective noun 'consulate'
> or the like. Two other rather mutilated inscc. also appear to
> contain <daranthoa>. We know that Basta was (and still is) a small
> village, rating a brief mention by Pliny, too small to issue its own
> coinage. But the city of Uria, considered the capital of Messapia,
> issued coinage marked in Messapic <Orra>. Probably the magistrates
> mentioned in the Vaste text were from Uria (or less likely, in my
> opinion, from Uxentum). Presumably their city acquired hegemony over
> its neighbors as Rome did, by concluding treaties (foedera) with
> other communities. Thus a reasonable meaning for the /i/-stemmed
> adj. *vastis, dat. sg. <vasti>, is 'federated', 'federal', or the
> like, indicating that the daranthoa in question was the supreme
> daranthoa of Messapia, its two officers being the highest magistrates
> of the land.
>

> Justifying the morphology is somewhat involved. [...]


>
> On the semantic side, we find some West Germanic reflexes of *wadh-
> moving naturally from 'pledge' to 'contract' or 'treaty' or the like,
> Old English <weddian> 'to make a contract, promise, marry', Old
> Frisian <wed> 'contract, promise, pledge'. Thus it is not
> unreasonable to suppose that PIE *wadh-sti- became in Messapic
> *va(d)sti- 'standing in a contract, standing under a treaty,
> pertaining to a federation, federal'.

Or "sworn", denoting the status of its holders.

> This adjective could also have been substantivized as 'federation'
> itself, leading to the abstract noun <vastima>.

Or "being sworn" -- "swornness".

> Possibly that noun originally meant 'leadership of the federation',
> its holder ruling over Messapia (or part of it), but then deposed by
> a Roman-style revolution which set up annual pairs of public
> officials, keeping the <vastima> only for ceremonial purposes, as the
> Romans kept the office of rex sacrificulus.

Or simply "service" or "term" for an office not known by a derived noun.
Perhaps as the generic term for office? Could that explain the different
form?

> But again this is veering into idle speculation, as I have no
> evidence for the history of Messapic political institutions. At any
> rate, the connection between <vasti> and <vastima> is no longer
> inscrutable.
>
> As for <daranthoa>, my guess is that it is the collective of some
> derivative of the zero-grade of *dher- 'to hold firmly, support', but
> I do not have the detailed morphology worked out, and other
> explanations are possible.

Is Messapic rhotic? Could it be a compound of e.g *dhe:s- "divine, holy,
celebrated (or the like)", known from lat. <feria> etc.? The second
element could be made from *ant- "before". "Feast-leadership"?
"God-spokesmanship"? And in that case, does your <anthos> look like the
singular of the second part of this compound?

> For the other offices, so far I have nothing. This is my revised
> interpretation of the Vaste text:
>
> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> 'Hear ye! The citizenry of the land constructed here, to Basta, the
> sacred enclosure
>
> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> in the federal consulate(?) of Staboas Khonedo (and) Dakhta Sivaanetas

in the sworn priesthood of ...

> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> and in the trigonokhate(?) of Staboas Khonetthes (and) Dazimas Beilles

and in the tribunate of ...

> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> and in the rellorikhate(?) of Kazareas Khonetthes and Otoeas
> Dazohonnes

and in the praetorate of ...

> inthi vastima dakhtas kratheheihi
> and in the federal presidency(?) of Dakhta Kratheheas

and in the term of ...

> inthi ardannoa po[ll]onnihi aimarnaihi ...
> and in the ardannate(?) of Pollonnes Aimarnas [and ...]'

and in the shrine-givership (?) of ...

--
Trond Engen
- going altarnative

Trond Engen

unread,
Feb 15, 2008, 3:57:23 PM2/15/08
to
Douglas G. Kilday skreiv:

> On Feb 2, 10:19 am, Trond Engen wrote:
>> Douglas G. Kilday skreiv:
>>

>>> The word <vasti> is problematic. [...]

[...]

>> If related to Latin <vastus>, <vasti> could simply be (a form of)
>> "great" and <vastima> "the biggest" (-> "the wider region"?).
>

> I hadn't thought of that. [...]

And you shouldn't. Lat. <vastus> is from a root *va- or *ewh2- (or some
such) meaning "flee, empty (of people)" (or some such). It's concievable
that the meaning could glide to "wide (of area)" but not to "great,
high, important (of title)".

--
Trond Engen
(or some such)

Douglas G. Kilday

unread,
Feb 15, 2008, 6:49:56 PM2/15/08
to

That looks like IZIZ (or perhaps ZIZI) in a Greek or Greek-derived
alphabet. My guess is that it is an abbreviated identification of the
woman who owned the loomweight, either Iz[...] daughter of Iz[...] or
Zi[...] daughter of Zi[...]. This would have sufficed to identify the
owner within the community.

Douglas G. Kilday

unread,
Feb 15, 2008, 6:57:46 PM2/15/08
to
"Brian M. Scott" wrote:
> "Douglas G. Kilday" wrote:
>
> [...]
>
> > The difference in (gentilician) names which you notice may
> > be an indirect reflection of social class.  That is, the
> > officers in the daranthoa or "consulship", Staboas
> > Khonedo and Dakhta Sivaanetas, may have old-line "landed
> > gentry" surnames as opposed to the others, whose
> > genitives in -ihi may reflect younger surnames based on
> > patronymic formations, much as Wilson, Robertson, and the
> > like are relatively "young" English surnames and formerly
> > indicated humble status.  [...]
>
> If you're contrasting patronymics in general with locative
> bynames, rather than these particular patronymics, then I'll
> have to disagree: patronymics are as old as locatives in
> post-Conquest England and probably older in pre-Conquest
> England, and they don't imply an original humble status.

It was a poor example on my part, then, and I withdraw it.

Douglas G. Kilday

unread,
Feb 16, 2008, 12:47:07 AM2/16/08
to

Pokorny has an alternating root, *weig-/*weik-, with *weig- required
in Italic to explain the Umbrian reflex, and *weik- in Germanic. As
you know, I do not like positing wishy-washy wanky-schwanky
consonantism in PIE roots. I prefer to regard *weig- as the original
root, with Germanic extracting a new form of the root from derivatives
of *weig-to-, *weig-tm.mo- and the like which underwent internal
sandhi to *weik-to, *weik-tm.mo- (Latin <victima>). The extraction
could have taken place before or after the consonant-shift (*weik- for
*weig-, or *weih- for *weik-).

I suppose *wei-n- as an extension of *wei- is possible, as is an
adjective *wei-no- formed directly from *wei-. Too many
possibilities, not enough context!

> >>> [...]
> >>> The word <vasti> is problematic.  The PID authors take it for a
> >>> neuter noun.  I would like to read it as a numeral adverb
> >>> (indicating which time in the consulship) but cannot justify that.
>
> >> That the names are both a reference to the year of the inauguration
> >> and a tribute to the principals? Or a simple list by rank? (Is there
> >> any way you could get <rellorikhoa> to be "fourth ..."? Seems
> >> far-fetched: If the supposed rho is a rare qoppa and the ligature
> >> really is a double chi it would read <qekhkhorikhoa>, and I don't
> >> dare thinking of the sound laws we'd have to postulate.)
>
> > Many scholars do read double chi here, but I don't see how it could
> > have arisen.  On the other hand only the second apparent double chi,
> > near the end of the insc., can be corrected into double lambda with
> > certainty, making the gen. <pollonnihi> of a surname with a
> > well-known Messapic onomastic element.  The word in section 4 could
> > be <rettorikhoa>, for all I know.  I wish I could etymologize the
> > darned thing.
>
> I did offer you a far-fetched qoppa for rho to achieve <qe..orikhoa>
> "four-something", following <tri[g]onokhoa> "three-something". For that
> purpose, <tt> is good and would even make <qettorikhoa> look regular.
> But that isn't enough to turn the <r> into a <q>.

The problem with qoppa is that we expect it to be used, if at all, to
denote the back allophone of /k/, as in archaic Greek and South
Etruscan alphabets, and not to precede /e/. Also, inherited /o/
should have become /a/, so it is very difficult to read 'four-
something' here.

> Could <rettor> and <tr[g]on> be calques of -- or parallel formations to
> -- the Roman titles <praetor> and <tribun>?

Possibly. I hadn't thought of the first one. Messapic does not, to
my knowledge, lose /p/, but that does not rule out an unrelated native
prefix, or perhaps a different word used to render 'one who goes
before'. In the second, Latin <tribus> 'tribe' and Umbrian <trifu>,
<trifo> acc., <trefi-per> abl. 'id.' require *tri-bhu-, which Messapic
should have inherited as *tribu-, like Latin. The apograph shows a
vertical line with a dot to its right about halfway up. The
transcription of 1558 has iota, neglecting the dot. The letter has
been taken as gamma (with uncertainty) apparently since Deecke. But
the dot might just as well be the remnant of beta as gamma, now that I
look again at the apograph. If so, we can read <tri[b]onokhoa> here,
and plausibly take it to be the dat. sg. of a word equivalent to Lat.
<tribu:na:tus>, formed either after *tribonas borrowed from Lat.
<tribu:nus>, or based on a native *tribos 'tribe' cognate with Lat.
<tribus>.

Yes. So perhaps the dat. sg. <vasti> is to be assumed with the other
offices.

> > This adjective could also have been substantivized as 'federation'
> > itself, leading to the abstract noun <vastima>.
>
> Or "being sworn" -- "swornness".
>
> > Possibly that noun originally meant 'leadership of the federation',
> > its holder ruling over Messapia (or part of it), but then deposed by
> > a Roman-style revolution which set up annual pairs of public
> > officials, keeping the <vastima> only for ceremonial purposes, as the
> > Romans kept the office of rex sacrificulus.
>
> Or simply "service" or "term" for an office not known by a derived noun.
> Perhaps as the generic term for office? Could that explain the different
> form?

It must have been some _specific_ office. A non-violent alternative
to the revolutionary scenario above is that it was the magistracy
responsible for contract law, like the Etruscan <zilch cechaneri>.
But again there are too many possibilities and not enough context.

> > But again this is veering into idle speculation, as I have no
> > evidence for the history of Messapic political institutions.  At any
> > rate, the connection between <vasti> and <vastima> is no longer
> > inscrutable.
>
> > As for <daranthoa>, my guess is that it is the collective of some
> > derivative of the zero-grade of *dher- 'to hold firmly, support', but
> > I do not have the detailed morphology worked out, and other
> > explanations are possible.
>
> Is Messapic rhotic? Could it be a compound of e.g *dhe:s- "divine, holy,
> celebrated (or the like)", known from lat. <feria> etc.? The second
> element could be made from *ant- "before". "Feast-leadership"?
> "God-spokesmanship"? And in that case, does your <anthos> look like the
> singular of the second part of this compound?

Kretschmer looked at place-names and determined that Messapic neither
rhotacized -s- to -r-, nor reduced it to -h-. If <anthos> is the
correct reading in that other inscription, I consider it likely
borrowed from Greek and not related to the -anth- in <daranthoa>,
which is possibly participial, reflecting *-n.t-. The personal name
[An]thidas is known from a Doric inscription, so an *Anthous is at
least possible.

> > For the other offices, so far I have nothing.  This is my revised
> > interpretation of the Vaste text:
>
> > klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> > 'Hear ye!  The citizenry of the land constructed here, to Basta, the
> > sacred enclosure
>
> > in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> > in the federal consulate(?) of Staboas Khonedo (and) Dakhta Sivaanetas
>
> in the sworn priesthood of ...
>
> > inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> > and in the trigonokhate(?) of Staboas Khonetthes (and) Dazimas Beilles
>
> and in the tribunate of ...

Yes. It works better with [b] not [g].

> > inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> > and in the rellorikhate(?) of Kazareas Khonetthes and Otoeas
> > Dazohonnes
>
> and in the praetorate of ...
>
> > inthi vastima dakhtas kratheheihi
> > and in the federal presidency(?) of Dakhta Kratheheas
>
> and in the term of ...
>
> > inthi ardannoa po[ll]onnihi aimarnaihi ...
> > and in the ardannate(?) of Pollonnes Aimarnas [and ...]'
>
> and in the shrine-givership (?) of ...

Perhaps. What I need to do is look through Hesychius for peculiar
Tarentine and Heraclean words that could be Messapic loans. Someone
has probably done this already, but I would have to be very lucky to
find a reference to such work on-line.

Trond Engen

unread,
Feb 16, 2008, 10:37:24 AM2/16/08
to
Douglas G. Kilday skreiv:

> Trond Engen wrote:
>> Douglas G. Kilday skreiv:
>>> Trond Engen wrote:
>>>> Douglas G. Kilday skreiv:
>>>>> Trond Engen wrote:
>>>>>> Douglas G. Kilday skreiv:
>>>>>>> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
>>>>>>> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
>>>>>>> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
>>>>>>> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
>>>>>>> inthi vastima dakhtas kratheheihi
>>>>>>> inthi ardannoa po[ll]onnihiai marnaihi

[...]

>>> The word in section 4 could be <rettorikhoa>, for all I know. I

>>> wish I could etymologize the darned thing.
>>
>> I did offer you a far-fetched qoppa for rho to achieve <qe..orikhoa>
>> "four-something", following <tri[g]onokhoa> "three-something". For
>> that purpose, <tt> is good and would even make <qettorikhoa> look
>> regular. But that isn't enough to turn the <r> into a <q>.
>
> The problem with qoppa is that we expect it to be used, if at all, to
> denote the back allophone of /k/, as in archaic Greek and South
> Etruscan alphabets, and not to precede /e/. Also, inherited /o/
> should have become /a/, so it is very difficult to read 'four-
> something' here.

I did think of that, actually. My idea was that if the lip rounding was
lost from /kW-/, the former back allophone of /k/ could have become an
independent phoneme /q/. But this instance alone can't support the
claim. One would have to find it in other words as well.

>> Could <rettor> and <tr[g]on> be calques of -- or parallel formations to
>> -- the Roman titles <praetor> and <tribun>?
>
> Possibly. I hadn't thought of the first one. Messapic does not, to
> my knowledge, lose /p/, but that does not rule out an unrelated
> native prefix, or perhaps a different word used to render 'one who
> goes before'.

*res-?

[...]

>>>> If related to Latin <vastus>, <vasti> could simply be (a form of)
>>>> "great" and <vastima> "the biggest" (-> "the wider region"?).
>>>
>>> I hadn't thought of that. During the past week I mulled over
>>> Deecke's connection of <vasti> to Latin <vas>, gen. <vadis>

>>> 'pledge'; [...] Thus a reasonable meaning for the /i/-stemmed

>>> adj. *vastis, dat. sg. <vasti>, is 'federated', 'federal', or the
>>> like, indicating that the daranthoa in question was the supreme
>>> daranthoa of Messapia, its two officers being the highest
>>> magistrates of the land.
>>> Justifying the morphology is somewhat involved. [...]
>>

>> Or "sworn", denoting the status of its holders.
>
> Yes. So perhaps the dat. sg. <vasti> is to be assumed with the other
> offices.
>
>>> This adjective could also have been substantivized as 'federation'
>>> itself, leading to the abstract noun <vastima>.
>>
>> Or "being sworn" -- "swornness".
>>
>>> Possibly that noun originally meant 'leadership of the federation',
>>> its holder ruling over Messapia (or part of it), but then deposed
>>> by a Roman-style revolution which set up annual pairs of public
>>> officials, keeping the <vastima> only for ceremonial purposes, as
>>> the Romans kept the office of rex sacrificulus.
>>
>> Or simply "service" or "term" for an office not known by a derived
>> noun. Perhaps as the generic term for office? Could that explain the
>> different form?
>
> It must have been some _specific_ office. A non-violent alternative
> to the revolutionary scenario above is that it was the magistracy
> responsible for contract law, like the Etruscan <zilch cechaneri>.
> But again there are too many possibilities and not enough context.

Another thought; since Messapic has /a/ < /o/, <vas-> could be from the
O-grade of the root *wes- "be, dwell". Or perhaps a borrowing from some
dialect of Greek. Among the reflexes of *wes- is Greek <astu> "town",
borrowed into Latin as the stem of <astus> "craft, skill (as practiced
in a town)". Could <vasti> mean "of the town(s)" and <vastima> be some
derivation of "town" or "trade", say, "mayorate" or "auldermanship"?

--
Trond Engen
- was here

Italo

unread,
Feb 16, 2008, 11:27:47 AM2/16/08
to

From the imprint it seems written from right to left.
I figured that the last letter (on the left) is probably a
Z. But what about the other letter. Is this just a variant
of the same Z? Afaik, it isn't known anywhere else.

Douglas G. Kilday

unread,
Feb 29, 2008, 4:11:13 PM2/29/08
to

Whatmough (PID 2:606) regarded Mess. Penkeos and Penkaheh[e] as
derived from *penkwe 'five', implying that */kw/ fell together with */
k/, instead of creating a new phoneme. (This is at odds with Mayer's
suggestion of Illyrian Kerku:ra from *perkwu-, since the same
assimilation would have given *kenke, not *penke, from *penkwe. As I
mentioned, I no longer favor Mayer's view, and I am inclined to agree
with Whatmough on the issue of */kw/.)

> >> Could <rettor> and <tr[g]on> be calques of -- or parallel formations to
> >> -- the Roman titles <praetor> and <tribun>?
>
> > Possibly.  I hadn't thought of the first one.  Messapic does not, to
> > my knowledge, lose /p/, but that does not rule out an unrelated
> > native prefix, or perhaps a different word used to render 'one who
> > goes before'.
>
> *res-?

Or perhaps *reidh-, if we take the digraph (vulgo <xx>) as <dd> rather
than <tt> or <ll>. Then we are back to one of your earlier
suggestions, if memory serves. The writing of <e> for earlier /ei/
could be justified if Divana and Divanovas are derivatives of *deiw-,
but <veinan> in this inscription itself adds a difficulty. At any
rate the praetor is 'one who goes before' (*prae-itor), so 'one who
rides before' (or perhaps 'one who goes in a raeda', this term
borrowed into Mess. from Gaulish, as was <benna> by general consensus)
is not unreasonable. Consuls, tribunes, praetors, aedile (below)?
That seems like a reasonable presentation of magistrates. Now we
still have to deal with <ardannoa>. The quaestorate, perhaps?

If derived from the o-grade of *wes-, possibly <vastima> means
'aedileship', 'magistracy responsible for public construction'. It is
not necessary for <vasti> and <vastima> to be related. The former
could still be 'federal', derived from *wadh- as outlined earlier; the
latter 'aedilician (office)', from the o-grade of *wes-. It makes
more sense that an aedile (or chief aedile) would be mentioned than a
relatively obscure magistrate in charge of contracts or whatever.

It is difficult to derive Greek <(w)ástu> from *wes-, and I prefer to
regard the word as having unknown origin. It is likewise difficult to
derive Latin <astus> from Greek <ástu>. I am inclined to regard
<astus> as a 4th-decl. abstract deverbative like <pastus> 'feeding',
though I cannot convincingly identify the verbal root. Latin
<astu:tus> 'having craft, skilled' (formed like <cornu:tus> 'horned')
is somewhat similar in sense to Greek <asteîos> 'urbane, citified',
but in my view this is coincidental.

Douglas G. Kilday

unread,
Feb 29, 2008, 4:27:37 PM2/29/08
to

It looks to me as though the engraver initially had trouble striking
the lower crossbar of the first Z (assuming IZIZ as correct) at the
desired angle, and turned the glitch into an embellishment by making
an additional stroke on the other side of the desired crossbar. I do
not think that we are dealing with a special letter here, or a
ligature (chi superposed on zeta?), but an idiosyncracy of this
particular object. I have never tried it, but I suspect that
engraving a loomweight is more difficult than engraving a fixed stone,
since the weight must be secured somehow, and may slip when struck.

Italo

unread,
Mar 2, 2008, 1:14:50 PM3/2/08
to

Actually, the thin lined X appears to've been added the last
(probably after switching the stylus). And it is impressed
in the clay, not engraved. The potter probably had to
inscribe dozens of loom weights with the same owners mark.
As for idiosyncracy, if it wasn't meant to be readable they
could've just marked it with a single sign or a seal impression.

A few more inscribed loomweights are known from the site
where my loomweight probably was found:
"The names of Messapic type inscribed on the loomweights
from Botromagno suggest that Messapic (or a closely related
dialect) continued in use on the site at least down to the
time of the Roman conquest" (Alastair Small, 1992).

The only longer text found there (on a storage jar) is in
Greek. It mentions two Messapic names, Morkos and Pyllos, as
well as an Oscan name, Gnaiwa.


lora...@cs.com

unread,
Mar 2, 2008, 5:33:21 PM3/2/08
to

Just say 'government' as I posited, and leave it at that.
Not greater resolution nor linguistic support is discernable withing
the available context.

Trond Engen

unread,
Mar 2, 2008, 5:50:55 PM3/2/08
to
Douglas G. Kilday skreiv:

> Trond Engen wrote:
>> Douglas G. Kilday skreiv:
>>> Trond Engen wrote:
>>>> Douglas G. Kilday skreiv:
>>>>> Trond Engen wrote:
>>>>>> Douglas G. Kilday skreiv:
>>>>>>> Trond Engen wrote:
>>>>>>>> Douglas G. Kilday skreiv:
>>>>>>>>> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
>>>>>>>>> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
>>>>>>>>> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
>>>>>>>>> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
>>>>>>>>> inthi vastima dakhtas kratheheihi
>>>>>>>>> inthi ardannoa po[ll]onnihiai marnaihi
>> [...]
>>
>>>>> The word in section 4 could be <rettorikhoa>, for all I know. I
>>>>> wish I could etymologize the darned thing.
>>>>

>>>> [...]


>>>>
>>>> Could <rettor> and <tr[g]on> be calques of -- or parallel
>>>> formations to -- the Roman titles <praetor> and <tribun>?
>>>
>>> Possibly. I hadn't thought of the first one. Messapic does not,
>>> to my knowledge, lose /p/, but that does not rule out an unrelated
>>> native prefix, or perhaps a different word used to render 'one who
>>> goes before'.
>>
>> *res-?
>
> Or perhaps *reidh-,

Or the almost obvious: a cognate or calque of <rector>.

> [...] Consuls, tribunes, praetors, aedile (below)? That seems like


> a reasonable presentation of magistrates. Now we still have to deal
> with <ardannoa>. The quaestorate, perhaps?

That would be a public persecutor or auditor? Since Messapic had a < 0,
could it be related to Latin <ordo>? "Giver of order"?

What do we know of the various in- and constitutions of the different
Italian states? How did they rank internally? My knowledge is largely
derived from Astérix, and I fear we're reaching a level where that might
be ... eh, insufficient.

>> Another thought; since Messapic has /a/ < /o/, <vas-> could be from
>> the O-grade of the root *wes- "be, dwell". Or perhaps a borrowing
>> from some dialect of Greek. Among the reflexes of *wes- is Greek
>> <astu> "town", borrowed into Latin as the stem of <astus> "craft,
>> skill (as practiced in a town)". Could <vasti> mean "of the town(s)"
>> and <vastima> be some derivation of "town" or "trade", say,
>> "mayorate" or "auldermanship"?
>
> If derived from the o-grade of *wes-, possibly <vastima> means
> 'aedileship', 'magistracy responsible for public construction'. It
> is not necessary for <vasti> and <vastima> to be related. The former
> could still be 'federal', derived from *wadh- as outlined earlier;
> the latter 'aedilician (office)', from the o-grade of *wes-. It
> makes more sense that an aedile (or chief aedile) would be mentioned
> than a relatively obscure magistrate in charge of contracts or
> whatever.

I see that the <aedile> had police authority ("housekeeper"?). That
would actually fit better with my suggestion for <ardannoa>. Could
<vastima> be the quaestorate in a sense of "recievership of oaths"
(=testimonies)?

> It is difficult to derive Greek <(w)ástu> from *wes-, and I prefer to
> regard the word as having unknown origin. It is likewise difficult
> to derive Latin <astus> from Greek <ástu>. I am inclined to regard
> <astus> as a 4th-decl. abstract deverbative like <pastus> 'feeding',
> though I cannot convincingly identify the verbal root. Latin
> <astu:tus> 'having craft, skilled' (formed like <cornu:tus> 'horned')
> is somewhat similar in sense to Greek <asteîos> 'urbane, citified',
> but in my view this is coincidental.

OK. I followed the AHD on this. I see that <pastus> is from <pasco>.
What about taking both <astus> and <astu:tus> from <ascio> "take up in
society; engage someone in a formal relationship"? (Or, if that's a
causative, perhaps the underlying *asco "engage oneself formally".)

--
Trond Engen
- won't ask for further etymology

Douglas G. Kilday

unread,
Mar 24, 2008, 6:27:17 PM3/24/08
to
Trond Engen wrote:
> Douglas G. Kilday skreiv:
> > Trond Engen wrote:
> >> Douglas G. Kilday skreiv:
> >>> Trond Engen wrote:
> >>>> Douglas G. Kilday skreiv:
> >>>>> Trond Engen wrote:
> >>>>>> Douglas G. Kilday skreiv:
> >>>>>>> Trond Engen wrote:
> >>>>>>>> Douglas G. Kilday skreiv:
> >>>>>>>>> klohizis thotoria mar[z]a pidogas tei basta veinan [v]aran
> >>>>>>>>> in daranthoa vasti staboos khonedonas dakhtas sivaanetos
> >>>>>>>>> inthi tri[g]onokhoa staboos khonetthihi dazimaihi beil[l]ihi
> >>>>>>>>> inthi re[ll]orikhoa kazareihi khonetthihi [o]toeihithi dazohonnihi
> >>>>>>>>> inthi vastima dakhtas kratheheihi
> >>>>>>>>> inthi ardannoa po[ll]onnihiai marnaihi
> >> [...]
>
> >>>>> The word in section 4 could be <rettorikhoa>, for all I know.  I
> >>>>> wish I could etymologize the darned thing.
>
> >>>> Could <rettor> and <tr[g]on> be calques of -- or parallel
> >>>> formations to -- the Roman titles <praetor> and <tribun>?
>
> >>> Possibly.  I hadn't thought of the first one.  Messapic does not,
> >>> to my knowledge, lose /p/, but that does not rule out an unrelated
> >>> native prefix, or perhaps a different word used to render 'one who
> >>> goes before'.
>
> >> *res-?
>
> > Or perhaps *reidh-,
>
> Or the almost obvious: a cognate or calque of <rector>.

That requires -kt- or -gt- to assimilate to -tt-. It might, but then
we could have a problem with the rather common name Dakhta.

> > [...]   Consuls, tribunes, praetors, aedile (below)?  That seems like
> > a reasonable presentation of magistrates.   Now we still have to deal
> > with <ardannoa>.  The quaestorate, perhaps?
>
> That would be a public persecutor or auditor? Since Messapic had a < 0,
> could it be related to Latin <ordo>? "Giver of order"?

There are some difficulties, but not insuperable. Lat. <o:rdo:> acc.
to Walde comes from weaving terminology, <exordior> 'I lay a warp,
begin to weave' retaining the original sense of the simplex <ordior>
and its Greek cognate <ordéo:>. But the transfer of sense from 'warp'
to 'order' could have occurred independently in Messapic as well as
Latin. And the Messapic noun, like Grk. <órde:ma> 'ball of woolen
yarn', could have retained short /o/. Then the collective <ardannoa>
could represent *ordoniowa: or the like. However, I fear we are
getting too far out onto thin etymological ice here.

> What do we know of the various in- and constitutions of the different
> Italian states? How did they rank internally? My knowledge is largely
> derived from Astérix, and I fear we're reaching a level where that might
> be ... eh, insufficient.

Oscan inscriptions give the meddix as municipal chief, and some
municipalities such as Nola had two, like Roman consuls. The meddix
tuticus in Buck's opinion was the leader of a whole confederation of
cities. The Oscan terms for 'aedile', 'quaestor', 'praetor', and
'censor' appear to have been borrowed from Latin, so the Roman
republican form of the state evidently spread to the Samnite states
before the earliest date of Oscan municipal inscriptions, and the
earlier details of political institutions can only be guessed at. Of
great importance for reading the Vaste inscription is the question of
Oscan influence on Messapic institutions. The term <meddiks> seems to
figure in the longer insc. of Carovigno, and if so, it must be
borrowed from Oscan. But of course there is no such word in the Vaste
text, whose orthography suggests that it is younger (<klohizis> Vaste,
<klaohizis> Carovigno; naturally we might be dealing with locally
precocious monophthongization at Vaste).

> >> Another thought; since Messapic has /a/ < /o/, <vas-> could be from
> >> the O-grade of the root *wes- "be, dwell". Or perhaps a borrowing
> >> from some dialect of Greek. Among the reflexes of *wes- is Greek
> >> <astu> "town", borrowed into Latin as the stem of <astus> "craft,
> >> skill (as practiced in a town)". Could <vasti> mean "of the town(s)"
> >> and <vastima> be some derivation of "town" or "trade", say,
> >> "mayorate" or "auldermanship"?
>
> > If derived from the o-grade of *wes-, possibly <vastima> means
> > 'aedileship', 'magistracy responsible for public construction'.  It
> > is not necessary for <vasti> and <vastima> to be related.  The former
> > could still be 'federal', derived from *wadh- as outlined earlier;
> > the latter 'aedilician (office)', from the o-grade of *wes-.  It
> > makes more sense that an aedile (or chief aedile) would be mentioned
> > than a relatively obscure magistrate in charge of contracts or
> > whatever.
>
> I see that the <aedile> had police authority ("housekeeper"?). That
> would actually fit better with my suggestion for <ardannoa>. Could
> <vastima> be the quaestorate in a sense of "recievership of oaths"
> (=testimonies)?

Possibly. We have arrived at the point which requires a return to the
basics, I think. Certain details of Messapic phonology demand
explanation, in particular the conditions for secondary aspiration.
Nobody seems to seriously doubt that <Theotor> and the like represent
the same root as *teuta:, but then we have a *Keokhor extractable from
an apparent patronymic, so these conditions for aspiration are not
trivially dependent on accent. I have a hunch that laryngeal residues
are involved, but checking this out will require a good deal of
homework.

> > It is difficult to derive Greek <(w)ástu> from *wes-, and I prefer to
> > regard the word as having unknown origin.  It is likewise difficult
> > to derive Latin <astus> from Greek <ástu>.  I am inclined to regard
> > <astus> as a 4th-decl. abstract deverbative like <pastus> 'feeding',
> > though I cannot convincingly identify the verbal root.  Latin
> > <astu:tus> 'having craft, skilled' (formed like <cornu:tus> 'horned')
> > is somewhat similar in sense to Greek <asteîos> 'urbane, citified',
> > but in my view this is coincidental.
>
> OK. I followed the AHD on this. I see that <pastus> is from <pasco>.
> What about taking both <astus> and <astu:tus> from <ascio> "take up in
> society; engage someone in a formal relationship"? (Or, if that's a
> causative, perhaps the underlying *asco "engage oneself formally".)

The 4th-decl. noun formed from <ascio:> is <asci:tus>. On the other
hand we have the 4th-decl. <castus> 'abstinence; rite' and the adj.
<castus> 'pure, unblemished', orig. 'free (from fault)' ("a culpa
castas", Plautus; "casta a cruore civili", Cicero) from the intr.
<careo:> 'I lack, I am without' (with abl.)., earlier *caseo: (cf.
Osc. <kasit> 'decet'). Therefore, a 4th-decl. *a:stus 'dryness' might
have been formed from the intr. <a:reo:> 'I am dry', earlier *a:seo:
(cf. Toch. A <a:sar> 'dry', Skt. <á:sah.> 'ash'). The problem now is
to justify the metaphorical use of the abl. <a(:)stu:> 'with dryness,
drily' in its attested sense 'cleverly, skillfully', almost like
<docte:>. The later metaphorical sense of <siccus> in the sense
'sound, healthy' might be cited. I am not fully satisfied with this,
but for now it is my best guess regarding <astus> and <astu:tus>.

0 new messages