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 Historical content remains viewable.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

The Albanization of the PIE dictionary

35 views
Skip to first unread message

Dušan Vukotić

unread,
Oct 14, 2007, 7:37:10 AM10/14/07
to Serb.lang
The new electronic Pokorny’s PIE dictionary, “revised” and crippled by
G. Starostin and A. Lubotsky, says that the root *bhegh- is changed to
bhedh-2 through the “common Illyr. -gh- => -dh- phonetic
mutation”(www.dnghu.org; Page 339). On the other side, these “experts”
are taking for granted that the Albanian word bindem (comply, obey) is
a clear-cut example of the imagined “twelf-words-Illyrian” sound
changes. It is hard to believe that these two “respected” linguists
have never heard for the Latin word oboedio (to obey, comply with;
from ob + audio).

Albanian 'bind' is used usually in sense of "persuasion" or
"obedience":
bind dikë (win over)
bind me diskutime (talk over)
bind në të kundërtën (out-argue)
bind të mos bëjë (dissuade from)
që bind (persuader)
që të bind (persuasive)
bindës (cogent, conclusive, convincing, decisive, forcible,
persuasive, plausible)
bindje (amenability, assurance, assuredness, cogency, conformity,
conviction, docility, expostulation, obedience, persuasion, politics,
suasion, submission)
bindem (answer, bend, comply, follow, make sure, obey)
bindem diçkaje (resign oneself)
nuk i bindem (disobey)

Of course, we can argue is Latin oboedio a prefixed word, constituted
of the preposition ob- ([in front of, before; in return for; because
of) and the verb audio, -ire (to hear , listen), because there is a
Serbian word (ubediti, ubedio persuade; adjective ubedjen convinced;
from be-gen <= bel-gen, with the same velar to dental sound changes
for which G.S. and A.L. are claiming to be of [their fabricated]
Illyrian origin).

Logically, if you are able to persuade (Serb ubediti) somone to do
something that person will OBEY your demand.

Middle English obeien (obey) and Gothic baidjan (coerce; Greek πεθω
convince, persuade; Lat fido trust, rely) is very close to Serbian
ubedjen (convinced), but also, it is close to the OSlav бѣдити/bѣditi
coerce, compel (Serb bediti, po-bediti win).

DV

Message has been deleted

Dušan Vukotić

unread,
Oct 14, 2007, 7:53:58 AM10/14/07
to
> from be-gen <= bel-gen), with the same velar to dental sound changes

> for which G.S. and A.L. are claiming to be of [their  fabricated]
> Illyrian origin.

>
> Logically, if you are able to persuade (Serb ubediti) somone to do
> something that person will OBEY your demand.
>
> Middle English obeien (obey; cf. Gothic baidjan coerce; Greek πεθω

> convince, persuade; Lat fido trust, rely) is very close to Serbian
> ubedjen (convinced), but also, it is close to the OSlav бѣдити/bѣditi
> coerce, compel (Serb bediti, po-bediti win).
>
> DV

I am crossposting this mesage to soc.culture.russian because someone
there maybe knows more about G.Starostin and A. Lubotsky and their
motives (political or pure lucrative; Soros and other so-called NGO
organizations) to propagate a fictional and nonsensical Illyro-
Shkipetarian theory
about the origin of IE languages.

DV

phog...@abo.fi

unread,
Oct 14, 2007, 3:37:22 PM10/14/07
to
On Oct 14, 2:53 pm, Dušan Vukotić <dusan.vuko...@gmail.com> wrote:
> On Oct 14, 1:37 pm, Dušan Vukotić <dusan.vuko...@gmail.com> wrote:
>
>
>
> > The new electronic Pokorny’s PIE dictionary, “revised” and crippled by
> > G. Starostin and A. Lubotsky, says that the root *bhegh- is changed to
> > bhedh-2 through the “common Illyr. -gh- => -dh- phonetic
> > mutation”(www.dnghu.org;Page339). On the other side, these “experts”

I think I am going to learn Albanian as soon as possible. It must be a
great language of a great people.

Christopher Culver

unread,
Oct 15, 2007, 3:53:19 AM10/15/07
to
On Oct 14, 10:37 pm, phogl...@abo.fi wrote:
> I think I am going to learn Albanian as soon as possible. It must be a
> great language of a great people.

And the girls are sexy too. I've got enough languages on my plate in
graduate school, but some of the Kosovar waitresses here in Helsinki
were enough to convince me to buy Routledge's Colloquial Albanian.

phog...@abo.fi

unread,
Oct 15, 2007, 5:37:14 AM10/15/07
to
On Oct 15, 10:53 am, Christopher Culver

Not to mention the great contribution that the Kuqi brothers have made
to Finnish football.

heliogabalus

unread,
Oct 15, 2007, 9:51:24 AM10/15/07
to

"Dusan Vukotic" <dusan....@gmail.com> wrote in message
news:1192362838....@t8g2000prg.googlegroups.com...
On Oct 14, 1:37 pm, Dusan Vukotic <dusan.vuko...@gmail.com> wrote:
-I am crossposting this mesage to soc.culture.russian because someone
-there maybe knows more about G.Starostin and A. Lubotsky and their
-motives (political or pure lucrative; Soros and other so-called NGO
-organizations) to propagate a fictional and nonsensical Illyro-
-Shkipetarian theory about the origin of IE languages.

Incidentally, this article or section appears to contradict itself:

http://en.wikipedia.org/wiki/Albanian_language

Why ?


Abdullah Konushevci

unread,
Oct 15, 2007, 11:02:06 AM10/15/07
to

Look, you missed the best derivative of this root, i.e., prefixed and
suffixed form për-bind-sh 'monster' that really fits to you and all
other chauvinists.

Konushevci

Dušan Vukotić

unread,
Oct 15, 2007, 5:59:24 PM10/15/07
to
On Oct 15, 5:02 pm, Abdullah Konushevci <akonushe...@gmail.com> wrote:
> On Oct 14, 1:37 pm, Dušan Vukotić <dusan.vuko...@gmail.com> wrote:
>
>
>
>
>
> > The new electronic Pokorny’s PIE dictionary, “revised” and crippled by
> > G. Starostin and A. Lubotsky, says that the root *bhegh- is changed to
> > bhedh-2 through the “common Illyr. -gh- => -dh- phonetic
> > mutation”(www.dnghu.org;Page339). On the other side, these “experts”
> Konushevci- Hide quoted text -
>
> - Show quoted text -

On Oct 15, 5:02 pm, Abdullah Konushevci <akonushe...@gmail.com> wrote:
> On Oct 14, 1:37 pm, Dušan Vukotić <dusan.vuko...@gmail.com> wrote:
>
>
>
>
>
> > The new electronic Pokorny’s PIE dictionary, “revised” and crippled by
> > G. Starostin and A. Lubotsky, says that the root *bhegh- is changed to
> > bhedh-2 through the “common Illyr. -gh- => -dh- phonetic

> > mutation”(www.dnghu.org;Page339). On the other side, these “experts”

> Konushevci- Hide quoted text -
>
> - Show quoted text -

Abdullah,
I do not understand your hysteria... what for?
http://vukotic.atspace.com/illyr.htm

Do you know why verdhë means yellow (not green) in Albanian?
Why gjelbër is green - not yellow (gelb, gilvus)?
Why motër is sister - not mother?
Why njëzet (1x10) is twenty?
Why Comnena is talking about SO-CALLED Albanians?
What about the Greek Melas Oros (Black Mountain), Serbian Crna Gora
(Black Mountain) and Albanian Malësor (highlander)?
Gr MELAS-ORO(S) (black mountain) => Alb MALESOR (highlander)
Can you tell us why Albanian have taken the Greek word MELAS (μελας
black) to "make" their MOUNTAIN (mal)?

Can you imagine why Albanian re means 'cloud' and e re (i ri) has the
meaning 'new'?
Use the IE logic SKY (CLOUD) = NEW (Serb nebo => novo; Gr νεφελη =>
νεφος => νεοφατος; Ita nuvolo => novello).

Of course, anyone who is able to understand what I am talking about is
welcome.
Who can solve the above riddle?

DV


Abdullah Konushevci

unread,
Oct 15, 2007, 7:12:48 PM10/15/07
to
> I do not understand your hysteria... what for?http://vukotic.atspace.com/illyr.htm

>
> Do you know why verdhë means yellow (not green) in Albanian?
> Why gjelbër is green - not yellow (gelb, gilvus)?
> Why motër is sister - not mother?
> Why njëzet (1x10) is twenty?
> Why Comnena is talking about SO-CALLED Albanians?
> What about the Greek Melas Oros (Black Mountain), Serbian Crna Gora
> (Black Mountain) and Albanian Malësor (highlander)?
> Gr MELAS-ORO(S) (black mountain) => Alb MALESOR (highlander)
> Can you tell us why Albanian have taken the Greek word MELAS (μελας
> black) to "make" their MOUNTAIN (mal)?
>
> Can you imagine why Albanian re means 'cloud' and e re (i ri) has the
> meaning 'new'?
> Use the IE logic SKY (CLOUD) = NEW (Serb nebo => novo; Gr νεφελη =>
> νεφος => νεοφατος; Ita nuvolo => novello).
>
> Of course, anyone who is able to understand what I am talking about is
> welcome.
> Who can solve the above riddle?
>
> DV

First, I suggest to you and all other interested on Albanian language
to read "Uber das Albanesische in seinen verwandtschaftlichen
beziehunge" by Franz Bopp, Berlin 1855, where the question of Albanian
origin and its realtion to other IE languages was solved once forever,
so I have no time to deal with your stupidity and chauvinism. You
claim that so-called Albanian are not Albanians, but Dalmatian, for
sure, are Serbians?! Really your lack of basic logic is pitiful. As
fare as I know we have discussed thoroughly Alb. zet 'twenty'. Again I
am not guilty that you are so stupid and you can't understand it. Alb
rê 'cloud', as you can see has nasal vowel /ê/ and Alb i/e re 'young'
has oral vowel /e/, for they are derived from different roots.
So, learn the basic of linguistics and leave me alone, because I am
sick with your constant stupidity and illiteracy.

Konushevci

Message has been deleted

Dušan Vukotić

unread,
Oct 16, 2007, 3:38:30 AM10/16/07
to
According to your logic, shoqe (friend-s) and shoqëri (friendship) are
derived from different bases? As a well educated linguist you should
have known that vowels are the first to be changed in order to aquire
a different meaning of a word: for instance, Alb kuqërremë (red) and
s/ kuqem (blush). Sometimes such changes are not conditioned at all,
like in Alb strehë, strehim - both with the meaning 'eaves',
'shelter', strehoj 'provide shelter' - from Serbian streha 'eaves',
strana 'side'). There are a lot of ablaut and umlaut examples in
Albanian as well as in any other language: çuditëri (eccentricity),
çuditem (wonder; loanword from Serbian čudo, čuditi wonder); patë
'goose', pata 'geese'. What would you say for the Albanian words
pendesë 'penitence' and pendese 'penitential'? Are they related or
not?

According to your "nasalized and vocalized" "wisdom", these words are
born from different roots?

What do you mean by saying that Bopp has solved (once for ever!) the
question of the Albanian origin? AFAIK, Bopp just has proven that
Albanian is an Indo-Europen language. Of course, he was right because
Albanian is constituted of above 90% of lonewords from Romance
(mostly), Greek and Slavic. Bopp could not have concluded otherwise,
because he had not bothered himself with the rest of less than 10% of
Non-Indo-European words in Albanian, which would have proven that
Albanian was not original but compiled IE language.

DV

Comnenna strictly differentiates "SO-CALLED Albanians" and "NATIVES of
Dalmatia sent by Bodinus".

In other words, she thinks that the Albanians are NOT NATIVE to Balkan
and suggests that Albanians is not their true name (today we know that
real and native name of Albanians is Shqipetars and that they were
named Albanians by foreigners - Greeks and Romans).

The New Monthly Magazine By Thomas Campbell. Samuel Carter Hall.
Edward Bulwer Lytton Lytton. Theodore Edward Hook. Thomas Hood.
William Harrison Ainsworth:
[...Another hypothesis holds that the Albanians derive their origin
from Alba, in Italy, and that tey are the descendants of a colony of
tbe Pretorian gnards, dismissed from Rome, by the 'Emperor Septimus
Severus, for having been accessory to the assassination of Pertinax.
Their dress, the words coming from Latin roots which are to be found
in their language, and a vague tradition prevalent among themselves,
support this idea.

"Chalcocondyles thinks that the Albanians came from the other side of
the Adriatic." The Albanians - Henry Skene Journal of the Ethnological
Society

Dušan Vukotić

unread,
Oct 17, 2007, 3:41:48 PM10/17/07
to
"The Illyrians", John Wilkes states on page 219:

"NOT MUCH RELIANCE SHOULD PERHAPS BE PLACED ON ATTEMPTS TO IDENTIFY AN
ILLYRIAN ANTHROPOLOGICAL TYPE AS SHORT AND DARK SKINNED SIMMILAR TO
MODERN ALBANIANS."


John Wilkes - The Illyrians p. 38)

[...Just as ancient writers could discover no satisfactory general
explanation for the origin of lllyrians, so most modern scholars, even
though now possessed of a mass of archaeological and linguistic
evidence, can assert with confidence only that Illyrians were not an
homogeneous ethnic entity, though even that is today challenged with
vigour by historians and archaeologists working within the perspective
of modern Albania. Notions once widely canvassed of a 'proto-Illyrian'
people of Indo-European origin once widespread across Europe were
always at best unproven and have now been generally discarded. In the
twenties and thirties of the twentieth century, belief in the
existence of a European sub-stratum of now vanished lllyrians was an
attractive idea to those bent on emphasizing the pure Aryan origin of
the nordic peoples of Europe, in the other direction there seems now
to be increasing scepticism towards theories of direct Illyrian
involvement in the movement of" new peoples into Greece at the end of
the second millennium BC. Similarly, many have now discarded the
simple identification of European Illyrians with the Urnfield Culture
of Late Bronze Age Central Europe. This particular equation stemmed
from the 'pan-lllyrian' theories propounded early this century by
philologists who could discover traces of lllyrians scattered across
the linguistic map of Europe...]


Stem sells

unread,
Oct 26, 2007, 12:32:04 PM10/26/07
to
>From what I understand on Wikipedia, they substitute phrases for
words. An example is "wolf". The Albanian word for this is, something
to the effect, "may he not open his mouth."

What are other examples of phrase-substitution for words?

Vladimir Makarenko

unread,
Oct 26, 2007, 4:19:22 PM10/26/07
to

Russian for a bear - medved, which is "the one who is taking care of
honey".

VM.

Dušan Vukotić

unread,
Oct 27, 2007, 5:06:27 AM10/27/07
to
On Oct 26, 10:19 pm, Vladimir Makarenko <makar...@popmail.med.nyu.edu>
wrote:

medved is a compound word, not a phrase
med(v)-jed = Honey_Eat/er (med + jed-)
cf. Latin ossifraga "bone-braker" (vulture, osprey); meso-jed = carni-
vore etc...

DV

lora...@cs.com

unread,
Oct 29, 2007, 4:37:10 AM10/29/07
to
On Oct 27, 1:06 am, Dušan Vukotić <dusan.vuko...@gmail.com> wrote:
> On Oct 26, 10:19 pm, Vladimir Makarenko <makar...@popmail.med.nyu.edu>
> wrote:
>
> > Stem sells wrote:
>
> > > >From what I understand on Wikipedia, they substitute phrases for
> > > words. An example is "wolf".  The Albanian word for this is, something
> > > to the effect, "may he not open his mouth."
>
> > > What are other examples of phrase-substitution for words?
>
> > Russian for a bear - medved, which is "the one who is taking care of
> > honey".
>
> > VM.
>
> medved is a compound word, not a phrase
> med(v)-jed = Honey_Eat/er (med + jed-)

That's if you want to risk a substistute 'jed' for 'ved'.. which may
not be correct.
Alternatively and more directly, the Baltic root 'med' = 'honey' +
'ved' means 'guide' or 'honey guide'.

> cf. Latin ossifraga "bone-braker" (vulture, osprey); meso-jed = carni-
> vore etc...
> DV

Baltic Latv. 'miesa' = 'flesh' + 'edais' = 'carnivore'


Dušan Vukoti

unread,
Oct 29, 2007, 7:09:21 AM10/29/07
to
On Oct 29, 9:37 am, lorad...@cs.com wrote:
> Baltic Latv. 'miesa' = 'flesh' + 'edais' = 'carnivore'

On Oct 29, 9:37 am, lorad...@cs.com wrote:


> On Oct 27, 1:06 am, Dušan Vukoti <dusan.vuko...@gmail.com> wrote:
>
>
>
>
>
> > On Oct 26, 10:19 pm, Vladimir Makarenko <makar...@popmail.med.nyu.edu>
> > wrote:
>
> > > Stem sells wrote:
>
> > > > >From what I understand on Wikipedia, they substitute phrases for
> > > > words. An example is "wolf". The Albanian word for this is, something
> > > > to the effect, "may he not open his mouth."
>
> > > > What are other examples of phrase-substitution for words?
>
> > > Russian for a bear - medved, which is "the one who is taking care of
> > > honey".
>
> > > VM.
>
> > medved is a compound word, not a phrase
> > med(v)-jed = Honey_Eat/er (med + jed-)
>
> That's if you want to risk a substistute 'jed' for 'ved'.. which may
> not be correct.
> Alternatively and more directly, the Baltic root 'med' = 'honey' +
> 'ved' means 'guide' or 'honey guide'.

This word (medved) has been discussed earlier
http://groups.google.com/group/sci.lang/browse_thread/thread/b11f6a1b41c2f7cd/b868c3c5f7610e9a?lnk=gst&q=medved+vukotic#b868c3c5f7610e9a

Yes, it might be possible that 'ved' (Serbian 'voditi' lead, guide)
or, more plausable, the Serbian verb 'vaditi' (Serb 'on vadi' he takes
out, he scoops) is a part of the compound word MED-VED (honey-
scooper).
In Serbian we can see that the verb 'eat' (Slavic j/esti, j/edenje,
Lithuanian sti, valgyti; Latvian est eat, diens meal, Latin edo;
Greek edanon eatable);

Now, if we compare the Lithuanian valgyti (eat), Latin ambedo (consum)
and Serbian obed (meal) I believe that everyone with a normal mental
capacity will be able to understand why I am constantly talking about
the BEL-GON Ur-Basis, which gave the birth to the hundred of thousands
of the IE words (voda-water, vlaga-wetness, voditi-lead (from v/
ladati, V-LADA government; V-LADAR (ruler) = LEADER; Serbian POJITI
(to drink) and POJEDEN (eaten); POLITI (splash); PITI (drink)
VODA (water), VLAGA (wetness), VODJA (leader)

DV

Stem sells

unread,
Oct 29, 2007, 10:31:40 AM10/29/07
to
In Hindi, "hathi" is the word for "elephant", and it literally
translates to "Horse with hands."

Are you familiar with the Albanian word for 'wolf" ("may he not open
his mouth").

lora...@cs.com

unread,
Oct 29, 2007, 1:51:21 PM10/29/07
to

Don't know what Albanian descriptor might be applied to 'wolf'...
But I do know that the Baltic Latvian 'vilks' (from which the
'germanic' term evolved into the English 'wolf) means 'puller' or
'one who pulls'. (!)

.. a clear cut meaning that points to it's primacy in any linguistic
genetic relationship.


lora...@cs.com

unread,
Oct 29, 2007, 1:54:02 PM10/29/07
to
On Oct 29, 3:09 am, Dušan Vukoti <dusan.vuko...@gmail.com> wrote:

> Yes, it might be possible that 'ved' (Serbian 'voditi' lead, guide)
> or, more plausable, the Serbian verb 'vaditi' (Serb 'on vadi' he takes
> out, he scoops) is a part of the compound word MED-VED (honey-
> scooper).
> In Serbian we can see that the verb 'eat' (Slavic j/esti, j/edenje,
> Lithuanian  sti, valgyti; Latvian est eat,  diens meal, Latin edo;
> Greek edanon eatable);

Thanks for seeing that ('voditi' lead, guide).. I am fairly certain
that 'medved' does indeed hark back to 'honey guide'.. or 'guide to
the honey' as it were.

> Now, if we compare the Lithuanian valgyti (eat), Latin ambedo (consum)
> and Serbian obed (meal) I believe that everyone with a normal mental
> capacity will be able to understand why I am constantly talking about
> the BEL-GON Ur-Basis, which gave the birth to the hundred of thousands
> of the IE words (voda-water, vlaga-wetness, voditi-lead (from v/
> ladati, V-LADA government; V-LADAR (ruler) = LEADER; Serbian POJITI
> (to drink) and POJEDEN (eaten); POLITI (splash); PITI (drink)
> VODA (water), VLAGA (wetness), VODJA (leader)
> DV

To tell you the truth.. I really haven't given your BEL-GON theory
much credence.
Make it less ambiguous for me..

How do "the IE words (voda-water, vlaga-wetness, voditi-lead (from v/


ladati, V-LADA government; V-LADAR (ruler) = LEADER; Serbian POJITI
(to drink) and POJEDEN (eaten); POLITI (splash); PITI (drink)

VODA (water), VLAGA (wetness), VODJA (leader) "... relate to your Bel-
Gon theory. IE vocabulary generation and chronological reality?

Harlan Messinger

unread,
Oct 29, 2007, 2:41:40 PM10/29/07
to
Dušan Vukoti wrote:
> Now, if we compare the Lithuanian valgyti (eat), Latin ambedo (consum)
> and Serbian obed (meal) I believe that everyone with a normal mental
> capacity will be able to understand why I am constantly talking about
> the BEL-GON Ur-Basis,

You understand that a reference to "normal" understanding implies the
*opposite* of a theory not accepted by anyone by its creator?

Dušan Vukoti

unread,
Oct 29, 2007, 4:34:53 PM10/29/07
to

Sanskrit gaja, hastin (elephant), Ganesha (Elephant-God); also Hindi
hathi, hasti, gaja, gajaraja (elephant);
And yes, hasta means hand, but in this case hand seems not to be
directly related to elephant; to say that elephant is "the one with
hand" does not sound plausible at all.
Nevertheles, the words hati (hasti), as well as English hand, are
coming from the same paleo-basis (reduplicated GON syllable); cf.
king, hunt, Serb. gonich (hunter), gazda (host, master, boss); Serb.
konj (horse), jahati (ride), jasati (walk aimlessly), gaziti (tread,
stump, tramp)
Elephant in India was understood either as "king" among animals
(similar to lion; Serb. knez, gazda, king, host) or as a heavy tramper
(Serbian gaziti tramp)

On the other side, hasta could be related to the English words hand
and hunt (Serb. goniti, Czech honit drive, hunt); cf. Serb. prsti,
from prezati, na-prezati (press)...

DV

Dušan Vukoti

unread,
Oct 29, 2007, 4:39:19 PM10/29/07
to

Serb. prst = finger

Dušan Vukoti

unread,
Oct 29, 2007, 5:00:40 PM10/29/07
to

Lorad is right here; Serb. vuk (wolf, from volk), vu i, vlachiti;
vlkao/vukao
(pull), vlak/voz (train);
Albanian ujk (wolf) can be explained only through the Slavic
vocabulary
(Serb. volk => vuk => vujko => Alb. ujk wolf).
There is no Albanian word with the meaning "pull", which would be
close to the Albanian word name for wolf (ujk); Alb. tërheq, shkul (v.
pull).
It means that their (Konushevic's, G.Starostin's and Lubotsky's)
Shqip-Illyrian theory is a product of a mere (sick) imagination.

DV


Message has been deleted

Dušan Vukoti

unread,
Oct 30, 2007, 4:44:39 PM10/30/07
to

BEL-GON (the sun god BEL + GON paleo-syllable that denotes any kind of
movement)
GON-BEL (reversed order)
o/blak (cloud; from GON-BEL-GON basis; gn/oblak, g/noble, noble; Lat.
nebula /cloud/ from GON-BEL)
Serb. nebo /sky/; from g/n/obla/k; adj. nebeski heavenly (Latvian
debesis /sky/ from GON-BEL; welkin from BEL-GON; heaven from GON-BEL-
GON; Lat. polus /sky/, from BEL-GON
Slavic voda (water) is coming from the BEL-GON basis (Serb vlaga /
wetness/ from BEL-GON;
The stream of water was compared with the power of leader/ruler
Serbian OBLAK /cloud/, VLAGA /wetness; cf. river VOLGA/, VODA /water/,
VODJENJ/leading/, VODJA /leader/;
Slavic VODA originated from BEL-GON => BOLDA => VOLDA; i.e. from the
word OBLITI => POLITI (suffuse, splash); Serb. BLJUNUTI (gush),
PLJUNUTI (spit), BLJUZGA (slush), PLJUSAK (rain-shower) BLATO (mud).

It woud be impossible to understand the etymology of the English word
LEAD without the knowledge that this word stemmed from the BEL-GON Ur-
Basis.
For instance, what is the relation between the English words LEAD (to
guide) and LEAD (heavy metal)? Is there any relations among the two
above words and Latin fluo -ere (flow; flumen river) _FLUID?!

Whoever is able to understand the relations between CLOUD and CLOTH,
LEAD and FLUID, FETTER and WATTER or among REX, REGIS, Serb. REKA
(river) and German REGNEN (to rain) that one can say that he is
beginning to understand one of the biggest secrets of the human
existence - the human speech development.

DV

phog...@abo.fi

unread,
Oct 31, 2007, 6:25:57 AM10/31/07
to
On Oct 29, 7:51 pm, lorad...@cs.com wrote:
> On Oct 29, 6:31 am, Stem sells <gestureofresp...@yahoo.com> wrote:
>
> > In Hindi, "hathi" is the word for "elephant", and it literally
> > translates to "Horse with hands."
>
> > Are you familiar with the Albanian word for 'wolf" ("may he not open
> > his mouth").
>
> Don't know what Albanian descriptor might be applied to 'wolf'...
> But I do know that the Baltic Latvian 'vilks' (from which the
> 'germanic' term evolved into the English 'wolf)

Latvian is not the parent language of English.

John Atkinson

unread,
Nov 1, 2007, 3:20:09 AM11/1/07
to
<lora...@cs.com> wrote...

> On Oct 29, 6:31 am, Stem sells <gestureofresp...@yahoo.com> wrote:
>> In Hindi, "hathi" is the word for "elephant", and it literally
>> translates to "Horse with hands."

What's the Hindi for "horse"? What's the Hindi for "hand"?

>> Are you familiar with the Albanian word for 'wolf" ("may he not open
>> his mouth").
>
> Don't know what Albanian descriptor might be applied to 'wolf'...

The Albanian for wolf is <ujk>.

> But I do know that the Baltic Latvian 'vilks' (from which the
> 'germanic' term evolved into the English 'wolf) means 'puller' or
> 'one who pulls'. (!)

This is apparently a folk entymology. Latvian <vilkt>, to pull, comes
from PIE *h4welk-, to pull. But <vilks> and <wolf> and <ujk> (and
<lupus> too) all evolve from another, different, PIE word, namely
*wlkwos, wolf. It has been suggested that this might be derived from
PIE *wel-, to tear. If so, "wolf" originally meant "tearer", or "one
who tears".

> .. a clear cut meaning that points to it's primacy in any linguistic
> genetic relationship.

I've no idea what you mean by that.

John.

Dušan Vukotic

unread,
Nov 1, 2007, 6:09:24 AM11/1/07
to
On Nov 1, 8:20 am, "John Atkinson" <johna...@bigpond.com> wrote:

> This is apparently a folk entymology.  Latvian <vilkt>, to pull, comes
> from PIE *h4welk-, to pull.  But <vilks> and <wolf> and <ujk> (and
> <lupus> too) all evolve from another, different, PIE word, namely
> *wlkwos, wolf.  It has been suggested that this might be derived from
> PIE *wel-, to tear.  If so, "wolf" originally meant "tearer", or "one
> who tears".

Slavic vlk and Germanic *wulfaz, Latin lupus were not born from the
same basis: volk comes from Bel-Gon basis (Serb. vuk /wolf/, bak; bik
from blk/bull/; vol, volina /ox/) and wolf/lupus/belua from Bel-Bel-
Gon; cf. Lat. bos, bovis ox , bullock, cow). Greek lukos is related to
Slavic vlk (Serbian vuk, vuchina from vlkina/v-luchina and not to
Latin lupus (b/e/lu/p/a): compare words for lion (Bel-Gon), elephant
(Bel-Bel-Gon) and you will probably be able to understand that all
these words were related to the sun god Bel; i.e these animals
symbolized one of the supreme deities in the past.

If you can understand the history of the name owl (Lat. bobo, bubonis,
Serb. buljina, ger. Eule, Dan. ugle from b/ugle) you are ready for a
beginner's course in basic linguistics :-)

DV

Brian M. Scott

unread,
Nov 1, 2007, 12:20:08 PM11/1/07
to
On Thu, 01 Nov 2007 07:20:09 GMT, John Atkinson
<john...@bigpond.com> wrote in
<news:J6fWi.7485$CN4....@news-server.bigpond.net.au> in
sci.lang,soc.culture.russian:

> <lora...@cs.com> wrote...

[...]

>> But I do know that the Baltic Latvian 'vilks' (from which the
>> 'germanic' term evolved into the English 'wolf) means 'puller' or
>> 'one who pulls'. (!)

> This is apparently a folk entymology. Latvian <vilkt>, to pull, comes
> from PIE *h4welk-, to pull. But <vilks> and <wolf> and <ujk> (and
> <lupus> too) all evolve from another, different, PIE word, namely
> *wlkwos, wolf. It has been suggested that this might be derived from
> PIE *wel-, to tear. If so, "wolf" originally meant "tearer", or "one
> who tears".

>> .. a clear cut meaning that points to it's primacy in any linguistic
>> genetic relationship.

> I've no idea what you mean by that.

Just his usual linguistic chauvinism.

Brian

lora...@cs.com

unread,
Nov 2, 2007, 1:47:54 AM11/2/07
to

No one said it was.. other than you by faulty inference.


lora...@cs.com

unread,
Nov 2, 2007, 1:55:33 AM11/2/07
to
On Oct 31, 11:20 pm, "John Atkinson" <johna...@bigpond.com> wrote:
> <lorad...@cs.com> wrote...

> > On Oct 29, 6:31 am, Stem sells <gestureofresp...@yahoo.com> wrote:
> >> In Hindi, "hathi" is the word for "elephant", and it literally
> >> translates to "Horse with hands."
>
> What's the Hindi for "horse"? What's the Hindi for "hand"?
>
> >> Are you familiar with the Albanian word for 'wolf" ("may he not open
> >> his mouth").
>
> > Don't know what Albanian descriptor might be applied to 'wolf'...
>
> The Albanian for wolf is <ujk>.
>
> > But I do know that the Baltic Latvian 'vilks' (from which the
> > 'germanic' term evolved into the English 'wolf) means 'puller' or
> > 'one who pulls'. (!)
>
> This is apparently a folk entymology.

No.. absolutely no study of insects was involved.
Perhaps you should relocate to sci.bugs?

> Latvian <vilkt>, to pull, comes
> from PIE *h4welk-, to pull. But <vilks> and <wolf> and <ujk> (and
> <lupus> too) all evolve from another, different, PIE word, namely
> *wlkwos, wolf.

Absolutely not.
Both 'viks' and 'vilkt' come from standard everyday Latvian.
Prove that they don't if you dare to leave your bug collection.

> It has been suggested that this might be derived from
> PIE *wel-, to tear. If so, "wolf" originally meant "tearer", or "one
> who tears".

'Pull' or 'tear'.. much the same meaning.
You are trying to distort the obvious, aren't you?

> > .. a clear cut meaning that points to it's primacy in any linguistic
> > genetic relationship.
>
> I've no idea what you mean by that.

You wouldn't... best you get back to your musae domesticae.

lora...@cs.com

unread,
Nov 2, 2007, 2:09:12 AM11/2/07
to
On Nov 1, 2:09 am, Dušan Vukotic <dusan.vuko...@gmail.com> wrote:
> On Nov 1, 8:20 am, "John Atkinson" <johna...@bigpond.com> wrote:
>
> > This is apparently a folk entymology.  Latvian <vilkt>, to pull, comes
> > from PIE *h4welk-, to pull.  But <vilks> and <wolf> and <ujk> (and
> > <lupus> too) all evolve from another, different, PIE word, namely
> > *wlkwos, wolf.  It has been suggested that this might be derived from
> > PIE *wel-, to tear.  If so, "wolf" originally meant "tearer", or "one
> > who tears".
>
> Slavic vlk and Germanic *wulfaz, Latin lupus were not born from the
> same basis: volk comes from Bel-Gon basis (Serb. vuk /wolf/, bak; bik
> from blk/bull/; vol, volina /ox/) and wolf/lupus/belua from Bel-Bel-
> Gon; cf. Lat. bos, bovis ox , bullock, cow).

'Lops'.. Baltic Latvian for 'beast'


lora...@cs.com

unread,
Nov 2, 2007, 2:17:50 AM11/2/07
to
On Nov 1, 8:20 am, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
> On Thu, 01 Nov 2007 07:20:09 GMT, John Atkinson
> > <lorad...@cs.com> wrote...

>
> [...]
>
> >> But I do know that the Baltic Latvian 'vilks' (from which the
> >> 'germanic' term evolved into the English 'wolf) means 'puller' or
> >> 'one who pulls'. (!)

> > This is apparently a folk entymology.

> > I've no idea what you mean by that.

> Just his usual linguistic chauvinism.
> Brian

Just because I present cognates (or genetically original forms of IE
words) from a language that you are seemingly oblivious and
chauvinistically averse to, doesn't mean that you have to spite the
messenger.

Comment reasonably or go study flies with Atkinson.. preferably with
your mouth shut.


Brian M. Scott

unread,
Nov 2, 2007, 12:00:18 PM11/2/07
to
On Thu, 01 Nov 2007 23:17:50 -0700, <lora...@cs.com> wrote
in sci.lang,soc.culture.russian:

> On Nov 1, 8:20 am, "Brian M. Scott" <b.sc...@csuohio.edu> wrote:
>> On Thu, 01 Nov 2007 07:20:09 GMT, John Atkinson
>>> <lorad...@cs.com> wrote...

>> [...]

>>>> But I do know that the Baltic Latvian 'vilks' (from which the
>>>> 'germanic' term evolved into the English 'wolf) means 'puller' or
>>>> 'one who pulls'. (!)

>>> This is apparently a folk entymology.
>>> I've no idea what you mean by that.

>> Just his usual linguistic chauvinism.

> Just because I present cognates (or genetically original forms of IE


> words) from a language that you are seemingly oblivious and
> chauvinistically averse to, doesn't mean that you have to spite the
> messenger.

> Comment reasonably or go study flies with Atkinson..
> preferably with your mouth shut.

It was a perfectly reasonable comment: I was merely
answering John's implied question, and you've demonstrated
repeatedly over the years that you're an ignorant crank with
a chauvinistic bee in his bonnet about the Baltic languages.
If you don't like it, quit posting crap.

[...]

Paul J Kriha

unread,
Nov 3, 2007, 2:08:49 AM11/3/07
to
<lora...@cs.com> wrote in message news:1193982933.3...@19g2000hsx.googlegroups.com...

> On Oct 31, 11:20 pm, "John Atkinson" <johna...@bigpond.com> wrote:

[...]



> > It has been suggested that this might be derived from
> > PIE *wel-, to tear. If so, "wolf" originally meant "tearer", or "one
> > who tears".
>
> 'Pull' or 'tear'.. much the same meaning.

As in "the engine slowly torn our train out of the station" ?
Is that it?

pjk

Message has been deleted

Dušan Vukotic

unread,
Nov 3, 2007, 6:17:27 AM11/3/07
to
On Nov 3, 7:08 am, "Paul J Kriha" <paul.nospam.kr...@paradise.net.nz>
wrote:

> As in "the engine slowly torn our train out of the station" ?
> Is that it?
>
> pjk

Of course, tear and pull are two different words with different
meanings from different paleo-bases.

Nevertheless, Kriha, could you be able to explicate the historical
relation between the words train and tear (tore, torne)?
In addition, there are the Slavic cognates - Serb. trgati (tear),
Czech roz-trhnout (tear), terka (pestle), trhac shredder; Czech
trení, trenice, Serb. trenje (friction); Serb. teranje (driving,
chasing)...

DV

Heidi Graw

unread,
Nov 3, 2007, 4:19:43 PM11/3/07
to

>"Stem sells" <gestureo...@yahoo.com> wrote in message
>news:1193416324.6...@57g2000hsv.googlegroups.com...

> >From what I understand on Wikipedia, they substitute phrases for
>> words. An example is "wolf". The Albanian word for this is, something
>> to the effect, "may he not open his mouth."

>Stem sells wrote:
> What are other examples of phrase-substitution for words?

Check out skaldic "kennings" for words.

http://skaldic.arts.usyd.edu.au/db.php?table=kenning&val

A she-wolf is also known as "the bitch of wounds."

There are numerous ones listed for wolves. One example:

wolf ... "steed of the troll woman."

Lots more phrase substitions for words listed in the
abovementioned link.

Some are actually quite humorous:

eyes: "his sight paths." ;-)

Heidi

John Atkinson

unread,
Nov 4, 2007, 1:02:26 AM11/4/07
to
<lora...@cs.com> wrote...

> On Oct 31, 11:20 pm, "John Atkinson" <johna...@bigpond.com> wrote:
>> <lorad...@cs.com> wrote...
[...]

>>
>> > But I do know that the Baltic Latvian 'vilks' (from which the
>> > 'germanic' term evolved into the English 'wolf) means 'puller' or
>> > 'one who pulls'. (!)
>>
>> This is apparently a folk etymology.

> [snip irrelevancies, here and below]

>> Latvian <vilkt>, to pull, comes
>> from PIE *h4welk-, to pull. But <vilks> and <wolf> and <ujk> (and
>> <lupus> too) all evolve from another, different, PIE word, namely
>> *wlkwos, wolf.
>
> Absolutely not.
> Both 'viks' and 'vilkt' come from standard everyday Latvian.
> Prove that they don't

There's no such word as "viks" in Latvian, if the dictionary is to be
believed. Quod erat demonstrandum.

However, both "vilks" and "vilkt" are indeed everyday Latvian words.
That doesn't mean they're related, any more than English "man" and
"manage" are related.

>> It has been suggested that this might be derived from
>> PIE *wel-, to tear. If so, "wolf" originally meant "tearer", or "one
>> who tears".
>
> 'Pull' or 'tear'.. much the same meaning.

Indeed, that is so, for a rather loose interpretation of "much the
same". Nevertheless, different words are used for these two meanings --
in English, in Latvian (vilkt vs ple:st), and also in PIE (*ten- vs
*wel-)

> You are trying to distort the obvious, aren't you?

No, anything but. You're saying the modern Latvian word for "wolf"
comes from the modern Latvian word for "pull". I'm pointing out that it
doesn't, that it comes from the PIE word for "wolf", and that _that_ may
come from a word that existed in PIE -- Latvian's ancestor of five or
six thousand years ago -- a word that, whatever it meant (probably
"tear"), certainly _wasn't_ the ancestor of the modern Latvian word for
"pull".

Get it? It's you that's distorting the situation, by at least five
thousand years.

>> > .. a clear cut meaning that points to it's primacy in any
>> > linguistic
>> > genetic relationship.
>>
>> I've no idea what you mean by that.
>
> You wouldn't...

Since you're unable or unwilling to explain what it means in English,
perhaps you could repeat it in Latvian -- it might make more sense.
Again, it might not.

John.

Dušan Vukotic

unread,
Nov 4, 2007, 6:21:44 AM11/4/07
to
> John.- Hide quoted text -
>
> - Show quoted text -

It would be interesting to see if the English word 'fox' could be
related to the Slavic volk/vuk (wolf)? In Slavic languages fox was
named as an animal that "sneakingly enters" the human's yard (Serbian
lisica, Czech liška /fox/; Serb. ulaziti /enter/; Czech. vlozhit /
enter/; also Serbian dialectal uljegnuti /enter in a sneaking manner/
and hence the Serbian word uljez /interloper/; also Serb. uvlachiti,
uvuci se, uvlakach and ulizica /sycophant/). Now we have a clear
connection among the Serbian words ulaz (entrance), uljez (intruder)
and lisica/lija (fox) - fox is interloper that "illegally" visits
human's property.
Serbian adjective lukav (foxy, tricky) has been derived from the
earlier verb uljegnuti (enter) and the noun uljez (interloper).

Let us now concentrate on the word 'inter-loper', which appears to be
close in utterance and meaning to the Serbian words 'unutar' (inside;
cf. Eng. enter; Greek entera intestines; Serb. unutrica entrails) and
lopuža/lopov (burglar, thief). It seems that Serbian word 'lopov/
lopuža' (thief) comes from the Latin word volpes (fox) or lupus
(wolf). I've been trying to solve this enigma for years but without
success, although it is quite clear that Serbian lopuža is
unequivocally related to lupus/lupinus/volpes (Serb. lopuža, lopov,
lopina - all with the meaning thief, stealer, burglar).

Of course, both animals (wolf and fox; cf. Latin feles cat and hence a
thief) are thieves or stealers (Serb. lopuža and lopina; lopovi; cf.
Latin volpe- and Serbian lopov); the problem is, there is no reliable
etymology neither for the Latin words volpes and lupus (probably from
belua; i.e. belu/p/a) nor is there a clear historical origin of the
Serbian words lopov, lopina and lopuža.

Any idea?

DV

John Atkinson

unread,
Nov 5, 2007, 6:33:11 AM11/5/07
to
"Dusan Vukotic" <dusan....@gmail.com> wrote:
[...]

> It would be interesting to see if the English word 'fox' could be
> related to the Slavic volk/vuk (wolf)?

Apparently not. Running the sound-changes backwards, <fox> must come
from something like *puk-os, whereas <wolf> and <volk> come from PIE
*wlkw-os. Cf Sanskrit <puccha>, and Tocharian B <pa:ka:>, both meaning
"tail". This suggests that a fox was originally "the tailed one". Note
that the Indian language Torwali has <pu:S> for "fox", presumably from
Sanskrit <puccha>. It seems to me that naming the animal after this,
his most visible feature, probably happened independently in Germanic
and Indic.

[...]

John.

Franz Gnaedinger

unread,
Nov 5, 2007, 12:41:32 PM11/5/07
to
On Nov 5, 12:33 pm, "John Atkinson" <johna...@bigpond.com> wrote:

So we have here a test case for Magdalenian.
I claim that fox and dachshund and dog come from
late Magdalenian DhAG for able, also the PIE root
*ukos for wolf, which turned into ylkos and ancient
Greek lykos, Middle Helladic ylkios (Tiryns disk,
ca. 3650 BP). Also Slavic vuk for wolf would have
this root. Wolf is akin to Old German welf German
Welpe for a wild young animal, especially a canine.
The word goes back to PIE *kwel or *kel for to
make a howling and yelping sound. If so, the still
earlier root may be KAL for the Underworld, with
many derivatives, among them German Hall for
sound and hallen for to resound. Consider the
howling hellhound Kerberos in the Underworld
called Hades. PAC was a horse, might also
have been used for a cow in later times,
and PUC as a lateral association may have
become the words you mention above with
the meaning of tail. -- Can you nail me?
can we turn this into a test of my Magdalenian
hypothesis?

John Atkinson

unread,
Nov 5, 2007, 9:39:04 PM11/5/07
to
"Franz Gnaedinger" <fr...@bluemail.ch> wrote...

> On Nov 5, 12:33 pm, "John Atkinson" <johna...@bigpond.com> wrote:
>> "Dusan Vukotic" <dusan.vuko...@gmail.com> wrote:
>>
>> [...]
>>
>> > It would be interesting to see if the English word 'fox' could be
>> > related to the Slavic volk/vuk (wolf)?
>>
>> Apparently not. Running the sound-changes backwards, <fox> must come
>> from something like *puk-os, whereas <wolf> and <volk> come from PIE
>> *wlkw-os. Cf Sanskrit <puccha>, and Tocharian B <pa:ka:>, both
>> meaning
>> "tail". This suggests that a fox was originally "the tailed one".
>> Note
>> that the Indian language Torwali has <pu:S> for "fox", presumably
>> from
>> Sanskrit <puccha>. It seems to me that naming the animal after this,
>> his most visible feature, probably happened independently in Germanic
>> and Indic.
>
> So we have here a test case for Magdalenian.
> I claim that fox and dachshund and dog come from
> late Magdalenian DhAG for able, also the PIE root
> *ukos for wolf, [...]. -- Can you nail me?

Yes. *ukos isn't the PIE root for wolf (or anything else, as far as I
know).

> PIE *kwel or *kel for to make a howling and yelping sound

Where did you drag that one up from?

J.

Brian M. Scott

unread,
Nov 5, 2007, 10:18:50 PM11/5/07
to
On Tue, 06 Nov 2007 02:39:04 GMT, John Atkinson
<john...@bigpond.com> wrote in
<news:ctQXi.9094$CN4...@news-server.bigpond.net.au> in
sci.lang,soc.culture.russian:

> "Franz Gnaedinger" <fr...@bluemail.ch> wrote...

[...]

>> PIE *kwel or *kel for to make a howling and yelping sound

> Where did you drag that one up from?

Heaven only knows; the gloss suggests *gHel-.

Brian

John Atkinson

unread,
Nov 6, 2007, 12:36:52 AM11/6/07
to
"Brian M. Scott" <b.s...@csuohio.edu> wrote...

> John Atkinson <john...@bigpond.com> wrote:
>> "Franz Gnaedinger" <fr...@bluemail.ch> wrote...
>
> [...]
>
>>> PIE *kwel or *kel for to make a howling and yelping sound
>
>> Where did you drag that one up from?
>
> Heaven only knows; the gloss suggests *gHel-.

*gHel- gives English yell (but not, apparently, yelp), as well as
Russian nagalit', to sing (according to Mallory and Adams; the word
doesn't exist in my dictionary; possibly galka, jackdaw, is related?).

*gHel- has nothing to do with German Welpe (English whelp, Norwegian
hvalp), like the OP suggests. These don't go back to PIE AFAIK (though
if they did it would be to **kwelb-, if such a root existed).

John.

Dušan Vukotic

unread,
Nov 6, 2007, 3:27:48 AM11/6/07
to
On Nov 6, 6:36 am, "John Atkinson" <johna...@bigpond.com> wrote:
> "Brian M. Scott" <b.sc...@csuohio.edu> wrote...
>
> > John Atkinson <johna...@bigpond.com> wrote:
> >> "Franz Gnaedinger" <f...@bluemail.ch> wrote...

>
> > [...]
>
> >>> PIE *kwel or *kel for to make a howling and yelping sound
>
> >> Where did you drag that one up from?
>
> > Heaven only knows; the gloss suggests *gHel-.
>
> *gHel- gives English yell (but not, apparently, yelp), as well as
> Russian nagalit', to sing (according to Mallory and Adams; the word
> doesn't exist in my dictionary; possibly galka, jackdaw, is related?).
>
> *gHel- has nothing to do with German Welpe (English whelp, Norwegian
> hvalp), like the OP suggests. These don't go back to PIE AFAIK (though
> if they did it would be to **kwelb-, if such a root existed).
>
> John.

It seems Lat. volpe/s is coming from the same basis as Ger. Welpe,
Wolf (English whelp, wolf). Of course, there is an OE variant 'hwelp'
that suggests that this word has started with the initial "h" ( Dan.
hvalp, OHG hwelf).
I would compare these words with the Serbian noun 'voda' (water) and
the verb 'kvasiti' where I am getting an impression that this Serbian
verb has been prefixed by velar - k/voda (Lat. aq/vatio /fetching of
water/; Serb. iz/vaditi; iz/vadio extract, fish out, take out).

I hope you grasped what I wanted to say; it seems that the PIE root
for aqua was not *aqkWa- but *wed-(r)- or my Bel-Gon Ur-Basis; i.e.
the same one as for water (cf. Serbian kvashenje /soaking/ and German
waschen /wash/).

DV

Dušan Vukotic

unread,
Nov 6, 2007, 3:36:29 AM11/6/07
to
> DV- Hide quoted text -

>
> - Show quoted text -

To be more precise, I think that aqua is an ak- prefixed water (Lat.
vadum -i water, river, sea; also unda -ae is nothing else but a
nasalized vadi => vnda)

DV

Franz Gnaedinger

unread,
Nov 6, 2007, 3:36:51 AM11/6/07
to
On Nov 6, 6:36 am, "John Atkinson" <johna...@bigpond.com> wrote:
> "Brian M. Scott" <b.sc...@csuohio.edu> wrote...
>
> > John Atkinson <johna...@bigpond.com> wrote:
> >> "Franz Gnaedinger" <f...@bluemail.ch> wrote...

>
> > [...]
>
> >>> PIE *kwel or *kel for to make a howling and yelping sound
>
> >> Where did you drag that one up from?

I must correct myself, it means to howl and whimper. Google
for wörterbuch-netz and klick on Deutsches Wörterbuch
von Jacob Grimm und Wilhelm Grimm and look out for WELF.
One Person followed by Fick, Falk, Torp and Hellquist give
*hwel *(s)quel and the alternative root *kel as origin of welf
Welpe English whelp. It was the howling and whimpering animal,
then the young animal.

> > Heaven only knows; the gloss suggests *gHel-.
>
> *gHel- gives English yell (but not, apparently, yelp), as well as
> Russian nagalit', to sing (according to Mallory and Adams; the word
> doesn't exist in my dictionary; possibly galka, jackdaw, is related?).
>
> *gHel- has nothing to do with German Welpe (English whelp, Norwegian
> hvalp), like the OP suggests. These don't go back to PIE AFAIK (though
> if they did it would be to **kwelb-, if such a root existed).

Look up the source I told you. Then go for WOLF in the same
Wörterbuch, and you'll find *ukos.

Trond Engen

unread,
Nov 6, 2007, 3:39:29 AM11/6/07
to
John Atkinson skreiv:

> *gHel- gives English yell (but not, apparently, yelp), as well as
> Russian nagalit', to sing (according to Mallory and Adams; the word
> doesn't exist in my dictionary; possibly galka, jackdaw, is related?).
>
> *gHel- has nothing to do with German Welpe (English whelp, Norwegian
> hvalp), like the OP suggests. These don't go back to PIE AFAIK
> (though if they did it would be to **kwelb-, if such a root existed).

Good, old Hellquist quotes three different suggestions for an IE origin.

The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",
<skãlyti> "yell constantly" and <skalìkas> "constantly yelling hound",
Gk. <skýlax> "whelp". The same root is supposed to have yielded Sw.
<väll>, Icel. <hvellr> "strong echo".

The second is an irregular development from */gWelbh-/, a well attested
root for "dome, ceiling; womb" that was the origin of Eng. <calf>, Gk.
<délphos>, Sanskrit <gárbha>, etc., Icel. <hvalfr> "dome", and the likes
of Eng. <whelm>.

The third points at Slav. <kolĕbati> "becoming unsteady" without going
further back.

It seems to me that the "echo" words could belong to the "dome" root.

--
Trond Engen
- with a whelping hand

Franz Gnaedinger

unread,
Nov 6, 2007, 4:02:04 AM11/6/07
to

And all these meanings fit well into the wide range of one of the
five ur-words postulated by Richard Fester (Prof. Dr. Richard
Fester, if you please), namely KALL, which I shortened to KAL,
for the Underworld, also for the womb of the goddess. Just
picture living in a cave, in a subterranean dome, with all the
sound reverberating from the walls ... Person followed by Fick,
Falk, Torp and Hellquist give *hwel *(s)quel and alternatively
*kel as origin of welf Welpe English whelp, it was the howling
and wimpering animal, then the young animal. I like especially
the Lithuanian kale 'bark' you mention above, as it remained
close to the origin I find in KAL.

Trond Engen

unread,
Nov 6, 2007, 4:06:41 AM11/6/07
to
Hei.

Trond Engen skreiv:

> John Atkinson skreiv:
>
>> *gHel- gives English yell (but not, apparently, yelp), as well as
>> Russian nagalit', to sing (according to Mallory and Adams; the word
>> doesn't exist in my dictionary; possibly galka, jackdaw, is related?).
>>
>> *gHel- has nothing to do with German Welpe (English whelp, Norwegian
>> hvalp), like the OP suggests. These don't go back to PIE AFAIK
>> (though if they did it would be to **kwelb-, if such a root existed).
>
> Good, old Hellquist quotes three different suggestions for an IE origin.
>
> The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",

That's not "bark", it's "bitch (female dog)".

Sw. <hynda> "bitch" isn't present in Norwegian. Interestingly, I took it
intuitively as a verb(ing) I thought I knew, but it seems not to be
attested in Norw. dictionaries either.

--
Trond Engen
- wrong again

Franz Gnaedinger

unread,
Nov 6, 2007, 4:47:08 AM11/6/07
to

In that case, the origin is again KAL for the Underworld
and for the womb of the goddess. There is a curious
parallel with words for woman and dog in many languages,
which must have its origin in the way wolves had been
domesticated. The assumtion is that a woman adopted
a wolf whelp and gave it her breast. The young wolf was
then like a child of hers, ancient Greek gynae for woman,
kyn- for dog, German Kind for child, English kin and
kinship. German Hund English hound are derivatives
of the kyn- word, and so is Swedish hynda. This relation
in language testifies to the high esteem the early farmers
had for dogs, which again supports my hypothesis of
the words dog and fox and Dachs and dachshound
coming from DhAG for able, good in the sense of able
-- and able they were as helpers of the early farmers
in everyday life, and as guiders of the souls through
the vast realm of the Underworld in the afterlife: leading
a worthy soul back to the surface of the earth, wherefrom
it rose to a heavenly abode in the Milky Way (see the foxes
on the pair of central pillars of temple D, if memory serves,
at Göbekli Tepe, and the shakal dog Anubis of ancient Egypt).
As for the Milky Way as heavenly abode of a worthy soul:
Michael Janda confirmed this idea also for Sanskrit
(see the Proceedings of the Seventeenth Annual UCLA
Indo-European Conference).

John Atkinson

unread,
Nov 6, 2007, 8:13:23 AM11/6/07
to
"Trond Engen" <tron...@engen.priv.no> wrote ...

> John Atkinson skreiv:
>
>> *gHel- gives English yell (but not, apparently, yelp), as well as
>> Russian nagalit', to sing (according to Mallory and Adams; the word
>> doesn't exist in my dictionary; possibly galka, jackdaw, is
>> related?).
>>
>> *gHel- has nothing to do with German Welpe (English whelp, Norwegian
>> hvalp), like the OP suggests. These don't go back to PIE AFAIK
>> (though if they did it would be to **kwelb-, if such a root existed).
>
> Good old Hellquist quotes three different suggestions for an IE
> origin.
>
> The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",
> <skãlyti> "yell constantly" and <skalìkas> "constantly yelling hound",

According to my Lith dictionary, <kalè>, bitch, <skalikas>, hound
(nothing about 'constantly yelling'), <skalyti>, to yelp, bay. BTW,
English "yell" isn't used for canine vocalisations; are you sure
Hellquist doesn't say "yelping"?.

Mallory and Adams quote the root as PIE *(s)koli-, 'young dog'

> Gk. <skýlax> "whelp".

Also Albanian <kelysh>, young dog.

> The same root is supposed to have yielded Sw. <väll>, Icel. <hvellr>
> "strong echo".
>
> The second is an irregular development from */gWelbh-/, a well
> attested root for "dome, ceiling; womb" that was the origin of Eng.
> <calf>, Gk. <délphos>, Sanskrit <gárbha>, etc., Icel. <hvalfr> "dome",
> and the likes of Eng. <whelm>.

Hmm. M&A say that OE <hwealf>, vault, (which is surely cognate with
Icelandic <hvalfr>, Norwegian <hvelv>) comes from PIE *kWelp-, arch,
which also gives Gk <kolpos>. This would be more regular: *kW- >
Germanic wh, while *gW- >k (and *gWh- > w).

I have my doubts about "whelm" too, which AFAIK doesn't have attested
cognates in other Germanic languages (?)

The others you mention all mean either "womb", or by extension, "young
animal" (<calf>, and Av <g@r@bus>). I bet "womb" was the actual meaning
of *gWelbh-, not "dome".

> The third points at Slav. <kolĕbati> "becoming unsteady" without going
> further back.

In Russian, <kolebanie> is oscillation, fluctuation, hesitation.

Lithuanian has <kalb>, language, tongue, <kalbèti> to speak, which at
first glance looks cognate to this. (Late PIE short *a and *o both > /a/
in Baltic, /o/ in Slavic.)


>
> It seems to me that the "echo" words could belong to the "dome" root.

Yes, but which one?

John.

John Atkinson

unread,
Nov 6, 2007, 8:21:09 AM11/6/07
to
"Franz Gnaedinger" <fr...@bluemail.ch> wrote...

> On Nov 6, 6:36 am, "John Atkinson" <johna...@bigpond.com> wrote:
>> > John Atkinson <johna...@bigpond.com> wrote:
>> >> "Franz Gnaedinger" <f...@bluemail.ch> wrote...
>
> > [...]
>
>> >>> PIE *kwel or *kel for to make a howling and yelping sound
>
>> >> Where did you drag that one up from?

> I must correct myself, it means to howl and whimper. Google
> for wörterbuch-netz and klick on Deutsches Wörterbuch
> von Jacob Grimm und Wilhelm Grimm

Date of publication?

[...]

J.

John Atkinson

unread,
Nov 6, 2007, 8:52:40 AM11/6/07
to

"Dusan Vukotic" <dusan....@gmail.com> wrote...

> On Nov 6, 6:36 am, "John Atkinson" <johna...@bigpond.com> wrote:
>> "Brian M. Scott" <b.sc...@csuohio.edu> wrote...
>>
>> > John Atkinson <johna...@bigpond.com> wrote:
>> >> "Franz Gnaedinger" <f...@bluemail.ch> wrote...
>>
>> > [...]
>>
>> >>> PIE *kwel or *kel for to make a howling and yelping sound
>>
>> >> Where did you drag that one up from?
>>
>> > Heaven only knows; the gloss suggests *gHel-.
>>
>> *gHel- gives English yell (but not, apparently, yelp), as well as
>> Russian nagalit', to sing (according to Mallory and Adams; the word
>> doesn't exist in my dictionary; possibly galka, jackdaw, is
>> related?).
>>
>> *gHel- has nothing to do with German Welpe (English whelp, Norwegian
>> hvalp), like the OP suggests. These don't go back to PIE AFAIK
>> (though
>> if they did it would be to **kwelb-, if such a root existed).
>>
>> John.
>
> It seems Lat. volpe/s is coming from the same basis as Ger. Welpe,
> Wolf (English whelp, wolf).

No, <volpes> comes from PIE *wlop-, meaning fox. So does Lith <lape>,
Gk <alopex>, Arm <alues>, Hit <ulippana> (wolf), Av <urupis> (dog), Skt
<lopaSa> (jackal).

While <wolf> comes from PIE *wlkWos, meaning wolf. So does Lat <lupus>,
Lith <vilkas>, Rus <volk>, Alb <ujk>, Gk <lukos>, Av v@hrka, Skt <vrka>,
Toch B <walkwe>.

And <whelp> is different again (see Trond's post).

[...]

> it seems that the PIE root > for aqua was not *aqkWa- but *wed-(r)-

Right. PIE indeed had *wodr for water. The origin of <aqua>, namely
*HekWeH- > *akWa may not have existed in PIE itself, since cognates are
found only in the north-western dialects Italic and Germanic (OE <ieg>,
isle)

John.

Peter T. Daniels

unread,
Nov 6, 2007, 9:21:04 AM11/6/07
to
On Nov 6, 3:27 am, Dušan Vukotic <dusan.vuko...@gmail.com> wrote:

> It seems Lat. volpe/s is coming from the same basis as Ger. Welpe,
> Wolf (English whelp, wolf). Of course, there is an OE variant 'hwelp'
> that suggests that this word has started with the initial "h" ( Dan.
> hvalp, OHG hwelf).

OE <hw> is a digraph representing the voiceless labial approximant,
spelled in ModE <wh>. It is not a sequence of /h/ and /w/, and more
than <wh> represents a sequence of /w/ and /h/..

Trond Engen

unread,
Nov 6, 2007, 9:21:41 AM11/6/07
to
John Atkinson skreiv:

> "Trond Engen" <tron...@engen.priv.no> wrote ...
>
>> John Atkinson skreiv:
>>

>>> [...]


>>
>> Good old Hellquist quotes three different suggestions for an IE
>> origin.
>>
>> The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",
>> <skãlyti> "yell constantly" and <skalìkas> "constantly yelling
>> hound",
>
> According to my Lith dictionary, <kalè>, bitch, <skalikas>, hound
> (nothing about 'constantly yelling'), <skalyti>, to yelp, bay. BTW,
> English "yell" isn't used for canine vocalisations; are you sure
> Hellquist doesn't say "yelping"?.

It's becoming evident that my previous answer was all too quickly
waritten. Hellquist (<http://runeberg.org/svetym/>) doesn't write
English at all, so any mistake is mine. What he says is "oupphörligt
skällande jakthund". Sw. 'skälla' is a verb used about both dogs and
people. Is that an s-mobile version of 'yell' -- or 'yelp'?

> Mallory and Adams quote the root as PIE *(s)koli-, 'young dog'
>
>> Gk. <skýlax> "whelp".
>
> Also Albanian <kelysh>, young dog.
>
>> The same root is supposed to have yielded Sw. <väll>, Icel. <hvellr>
>> "strong echo".
>>
>> The second is an irregular development from */gWelbh-/, a well
>> attested root for "dome, ceiling; womb" that was the origin of Eng.
>> <calf>, Gk. <délphos>, Sanskrit <gárbha>, etc., Icel. <hvalfr>
>> "dome", and the likes of Eng. <whelm>.
>
> Hmm. M&A say that OE <hwealf>, vault, (which is surely cognate with
> Icelandic <hvalfr>, Norwegian <hvelv>) comes from PIE *kWelp-, arch,
> which also gives Gk <kolpos>. This would be more regular: *kW- >
> Germanic wh, while *gW- >k (and *gWh- > w).

'Vault' is fine. That's the word I was trying to remember before writing
'dome'. I did notice the irregular development, too, but I didn't try to
sort it out.

> I have my doubts about "whelm" too, which AFAIK doesn't have attested
> cognates in other Germanic languages (?)

Sw. <volm> < <hvolmer> "roof made of hay", also known from Da. and No.

> The others you mention all mean either "womb", or by extension,
> "young animal" (<calf>, and Av <g@r@bus>). I bet "womb" was the
> actual meaning of *gWelbh-, not "dome".

Or "vault". I thought of "cave" if we where to connect both the "echo"
and the "womb" word. But I'll leave that.

>> The third points at Slav. <kolĕbati> "becoming unsteady" without
>> going further back.
>
> In Russian, <kolebanie> is oscillation, fluctuation, hesitation.
>
> Lithuanian has <kalb>, language, tongue, <kalbèti> to speak, which at
> first glance looks cognate to this. (Late PIE short *a and *o both >
> /a/ in Baltic, /o/ in Slavic.)

Agreed. It seemed a far stretch from "calf" anyway.

>> It seems to me that the "echo" words could belong to the "dome" root.
>
> Yes, but which one?

Now we have one "vault" root with */kW-/ and one "womb" root with
*/gW-/. That would place Icel. <hvellr> with the vault -- which still
makes sense. And we have a "whelp" root that doesn't really fit "whelp".
I'm still too quick, though. I'll have a closer look this evening.

--
Trond Engen
- standing to be yelled at

Franz Gnaedinger

unread,
Nov 6, 2007, 9:58:56 AM11/6/07
to
On Nov 6, 2:21 pm, "John Atkinson" <johna...@bigpond.com> wrote:
> "Franz Gnaedinger" <f...@bluemail.ch> wrote...
> > On Nov 6, 6:36 am, "John Atkinson" <johna...@bigpond.com> wrote:
> >> > John Atkinson <johna...@bigpond.com> wrote:
> >> >> "Franz Gnaedinger" <f...@bluemail.ch> wrote...
>
> > > [...]
>
> >> >>> PIE *kwel or *kel for to make a howling and yelping sound
>
> >> >> Where did you drag that one up from?
> > I must correct myself, it means to howl and whimper. Google
> > for wörterbuch-netz and klick on Deutsches Wörterbuch
> > von Jacob Grimm und Wilhelm Grimm
>
> Date of publication?
>
> [...]
>
> J.

1971. You can get at it in a simpler way than I told
you. Just google for grimm trier and klick on
Deutsches Wörterbuch von Jacob Grimm and
Wilhelm Grimm. The online version has only few
entries, compared with the work on paper, yet
what is online is excellent.

Dušan Vukotic

unread,
Nov 6, 2007, 2:24:38 PM11/6/07
to
On Nov 6, 2:52 pm, "John Atkinson" <johna...@bigpond.com> wrote:

> > It seems Lat. volpe/s is coming from the same basis as Ger. Welpe,
> > Wolf (English whelp, wolf).
>
> No, <volpes> comes from PIE *wlop-, meaning fox. So does Lith <lape>,
> Gk <alopex>, Arm <alues>, Hit <ulippana> (wolf), Av <urupis> (dog), Skt
> <lopaSa> (jackal).


> While <wolf> comes from PIE *wlkWos, meaning wolf. So does Lat <lupus>,
> Lith <vilkas>, Rus <volk>, Alb <ujk>, Gk <lukos>, Av v@hrka, Skt <vrka>,
> Toch B <walkwe>.

Irish 'faoilleach' and Welsh blaidd could be an evidences that the
root for wolf couldn't be *wlkWos but, eventually, these words might
have been derived from the Bel-Gon basis or from the PIE root *bhelg-

I think, there is a big problem to fit some Germanic wolf-names, for
instance, OHG 'wulpa' (she-wolf) into the PIE *wlkWos (as you can see,
OHG 'wulpa' clearly corresponds to the Latin word 'vulpes'); if I
remember well I also found somewhere the Germanic noun 'vulba' (or
vulbi) for she-wolf. Simple I cannot imagine f => p shifting; for me
it could be more natural if p (ph <= bh) were shifted to f.

Apollo Lykeios (wolf-Apollo) is an important evidence that wolf was
respected as a deity symbol (Egyption gods with wolf heads
http://www.egyptianmyths.net/images/wepwawet.jpg - Anubis I think,
which is the same god as Greek Nebo; my Gon-Bel basis). Apollo is no
other god but the ancient, Akkadian-Babylonian god Bel. Now, if we
take in a serious consideration the names of deamons (dog-like
creatures; The Testament of Solomon ) - Belial and Belbel - maybe we
will be able to understand that Latin 'belua' (beast, monster;
probably from beluba/belupa or belbula) "belongs" to the above-mention
Bel-Bel or Beli/b/al (reduplicated Bel syllable) dog-like deamons.

DV


> And <whelp> is different again (see Trond's post).

> [...]


> > it seems that the PIE root > for aqua was not *aqkWa- but *wed-(r)-
>
> Right. PIE indeed had *wodr for water. The origin of <aqua>, namely
> *HekWeH- > *akWa may not have existed in PIE itself, since cognates are
> found only in the north-western dialects Italic and Germanic (OE <ieg>,
> isle)
>

Dušan Vukotic

unread,
Nov 6, 2007, 2:56:38 PM11/6/07
to

What about Danish hvalp? I am not sure, but I would say that whelp and
calf (Ger. Kalb) could be somehow related. There are words like
Serbian 'kobila' (mare), Latin caballus (horse) which appeared to be
also related to Kalb/calf; metatheses from caba/b/llo => kavablu =>
kvalup => hvalp.
Just guessing :-)

DV

Dušan Vukotic

unread,
Nov 6, 2007, 3:49:27 PM11/6/07
to
On Nov 6, 9:39 am, Trond Engen <trond...@engen.priv.no> wrote:
> John Atkinson skreiv:
>
> > *gHel- gives English yell (but not, apparently, yelp), as well as
> > Russian nagalit', to sing (according to Mallory and Adams; the word
> > doesn't exist in my dictionary; possibly galka, jackdaw, is related?).
>
> > *gHel- has nothing to do with German Welpe (English whelp, Norwegian
> > hvalp), like the OP suggests. These don't go back to PIE AFAIK
> > (though if they did it would be to **kwelb-, if such a root existed).
>
> Good, old Hellquist quotes three different suggestions for an IE origin.
>
> The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",
> <skãlyti> "yell constantly" and <skalìkas> "constantly yelling hound",
> Gk. <skýlax> "whelp". The same root is supposed to have yielded Sw.
> <väll>, Icel. <hvellr> "strong echo".
>
> The second is an irregular development from */gWelbh-/, a well attested
> root for "dome, ceiling; womb" that was the origin of Eng. <calf>, Gk.
> <délphos>, Sanskrit <gárbha>, etc., Icel. <hvalfr> "dome", and the likes
> of Eng. <whelm>.
>
> The third points at Slav. <kol bati> "becoming unsteady" without going

> further back.
>
> It seems to me that the "echo" words could belong to the "dome" root.
>
> --
> Trond Engen
> - with a whelping hand

Greek 'epibolaion' covering, wrapper; Serb 'obavijen' wrapped;
English babe; Serb beba (baby), Serb. pelena (baby wraper; diaper;
from above-mentioned 'obavijen/povijen' wrapped (this word comes from
reduplicated Bel syllable; cf. Serbian 'obljubljen/o' snuggle, nestle,
cuddle;
As you can see vulva (womb) is understood as a kind of wrapper; or
German wölben (to arch, camber); Serbian uvala; English valley

DV

Trond Engen

unread,
Nov 6, 2007, 7:27:51 PM11/6/07
to
Trond Engen skreiv:

> John Atkinson skreiv:
>
>> "Trond Engen" <tron...@engen.priv.no> wrote ...
>>
>>> John Atkinson skreiv:
>>>

>>>> ['whelp']


>>>
>>> Good old Hellquist quotes three different suggestions for an IE
>>> origin.
>>>
>>> The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",
>>> <skãlyti> "yell constantly" and <skalìkas> "constantly yelling
>>> hound",

This root is also found as */(s)kWol-/ and */(s)kol-/. Is there anything
else than 'whelp' that suggests /kW/ in the root?

>> According to my Lith dictionary, <kalè>, bitch, <skalikas>, hound
>> (nothing about 'constantly yelling'), <skalyti>, to yelp, bay. BTW,
>> English "yell" isn't used for canine vocalisations; are you sure
>> Hellquist doesn't say "yelping"?.
>
> It's becoming evident that my previous answer was all too quickly
> waritten. Hellquist (<http://runeberg.org/svetym/>) doesn't write
> English at all, so any mistake is mine. What he says is "oupphörligt
> skällande jakthund".

... and <skãlyti> "ideligen skälla".

Moving up:

>> Mallory and Adams quote the root as PIE *(s)koli-, 'young dog'
>>
>>> Gk. <skýlax> "whelp".
>>
>> Also Albanian <kelysh>, young dog.

So it's */(s)koli-/ for the dog. Moving to the sound:

> Sw. 'skälla' is a verb used about both dogs and people. Is that an
> s-mobile version of 'yell' -- or 'yelp'?

Trying to answer my own question here:

Bjorvand and Lindeman don't list No. 'skjelle'. Hellquist tells that Sw.
'skälla' is a weak verb derived from the Gmc. strong verb */skellan/.
This is from a an IE root */(s)kel-/. The root is found also in Ger.
<hell> "clear" and Gk. <kélados> "noise" and <kaléo> "call (v.)". Some
dialect forms suggest that the original is */(s)kWel-/. There's a series
of sound verbs starting with skvVl-, though, and most of them deal with
making a noise like water. Hellquist thinks they may be onomatopoetic.

>>> The same root is supposed to have yielded Sw. <väll>, Icel.
>>> <hvellr> "strong echo".

*/(s)koli-/ for the "whelp", */(s)kel-/ for the "angry shouting",
*/(s)kWel-/ for the "echoing noise"? Do I remember correctly some
speculation about the /(s)-/ as an intensive/durative marker? That would
fit nicely here.

>>> The second is an irregular development from */gWelbh-/, a well
>>> attested root for "dome, ceiling; womb" that was the origin of Eng.
>>> <calf>, Gk. <délphos>, Sanskrit <gárbha>, etc., Icel. <hvalfr>
>>> "dome", and the likes of Eng. <whelm>.
>>
>> Hmm. M&A say that OE <hwealf>, vault, (which is surely cognate with
>> Icelandic <hvalfr>, Norwegian <hvelv>) comes from PIE *kWelp-, arch,
>> which also gives Gk <kolpos>. This would be more regular: *kW- >
>> Germanic wh, while *gW- >k (and *gWh- > w).
>
> 'Vault' is fine. That's the word I was trying to remember before
> writing 'dome'. I did notice the irregular development, too, but I
> didn't try to sort it out.
>
>> I have my doubts about "whelm" too, which AFAIK doesn't have
>> attested cognates in other Germanic languages (?)
>
> Sw. <volm> < <hvolmer> "roof made of hay", also known from Da. and No.

I read him wrong again. This time 'hötak' "hay roof" for 'höstack'
"haystack". I must stop doing these things when I don't have time. That
doesn't change the fact that Hellquist sees them as related, from
*/hwelb-ma-/ or */hwalb-ma-/, and thus from your root above.

I repeat:

>>> The same root is supposed to have yielded Sw. <väll>, Icel.
>>> <hvellr> "strong echo".

It's not easy to derive */hwel-/ from */kWelb-/. Maybe it's derived
directly from IE */kWelH-/ "revolve". "Surrounding sound"?

>> The others you mention all mean either "womb", or by extension,
>> "young animal" (<calf>, and Av <g@r@bus>). I bet "womb" was the
>> actual meaning of *gWelbh-, not "dome".

Returning to Hellquist's 'valp' < */gWelbh-/, could the irregular
development of the initial sound be due to a compound? What environment
could cause */-gW-/ to develop like *-/kW-/?

I don't find */qWelbh-/ in the AHD IE root index, BTW. There's a root
*/wel-/ but 'calf' doesn't fit there. What's going on?

>>> The third points at Slav. <kolĕbati> "becoming unsteady" without
>>> going further back.
>>
>> In Russian, <kolebanie> is oscillation, fluctuation, hesitation.
>>
>> Lithuanian has <kalb>, language, tongue, <kalbèti> to speak, which
>> at first glance looks cognate to this. (Late PIE short *a and *o
>> both > /a/ in Baltic, /o/ in Slavic.)
>
> Agreed. It seemed a far stretch from "calf" anyway.

There's also a Norwegian strong verb 'skjelve' (NN conj: skjelv, skalv,
skolve) "shiver" < ON <skjalfa>. Thinking of it, if there really is no
*/gWelbh-/ and 'calf' is unexplained, then the idea of an orifin as
"shiver" or "sway" wouldn't seem too far-fetched. I've seen some newborn
calves rising.

That would leave 'whelp' still unexplained, though.

The russian group remowed.

--
Trond Engen
- revolving around the same theme

John Atkinson

unread,
Nov 6, 2007, 10:27:39 PM11/6/07
to
"Franz Gnaedinger" <fr...@bluemail.ch> wrote...

> On Nov 6, 2:21 pm, "John Atkinson" <johna...@bigpond.com> wrote:
> "Franz Gnaedinger" <f...@bluemail.ch> wrote...

[...]


>>>Deutsches Wörterbuch von Jacob Grimm und Wilhelm Grimm
>
>> Date of publication?
>

> 1971.

No, I meant the _real_ date of publication. Quote:

"Deutsches Wörterbuch (German Dictionary), 32 volumes, 1852-1960. The
Grimms themselves saw only the entries A through Forsche of this
monumental historical dictionary published during their lifetime. The
remaining parts were published by several generations of scholars over a
100 year span."

Since it appears that the volumes were published in alphabetical order,
one would guess that the volume with "Welpe" would have come out around
1940. But the obselete PIE forms that you gave look like they're of
nineteenth century vintage, possibly dating back to the Grimms
themselves.

John.

Franz Gnaedinger

unread,
Nov 7, 2007, 2:57:16 AM11/7/07
to
On Nov 7, 4:27 am, "John Atkinson" <johna...@bigpond.com> wrote:
>
> No, I meant the _real_ date of publication. Quote:
>
> "Deutsches Wörterbuch (German Dictionary), 32 volumes, 1852-1960. The
> Grimms themselves saw only the entries A through Forsche of this
> monumental historical dictionary published during their lifetime. The
> remaining parts were published by several generations of scholars over a
> 100 year span."
>
> Since it appears that the volumes were published in alphabetical order,
> one would guess that the volume with "Welpe" would have come out around
> 1940. But the obselete PIE forms that you gave look like they're of
> nineteenth century vintage, possibly dating back to the Grimms
> themselves.

I told you to look up the entry WELF, not Welpe.
Person's opinion was followed by Fick, Falk,
Torp and Hellquist, so the alternative PIE root
*kel as origin of welf Welpe whelp was approved
of by Hellquist in 1970 I guess. How can PIE forms
get obsolete? If the sound laws are good then it
can't happen that a proto-form is changing with
every generation. As 2 plus 2 being 4 in ancient
times can't be obsolete in our time, it is still four.
When I woke up this morning I came to the
conclusion that PIE is basically flawed by the
assumption that a common language was spoken
in all of Eurasia 5,000 years ago. Life was very
different by then in various parts of Eurasia,
copper was known in the eastern parts of
Eurasia while western Europe was lingering on
in the Stone Age. The common ancestral
language must belong to a deeper time level,
10,000 or 12,000 or 15,000 years ago.
The comparative method reconstructing PIE
and it's Nostratic stretcher are precious as
a platform to jump and dive into paleo-language,
now we should do it, jump and dive into a time
when live was the same everywhere from the
Franco-Cantabrian space to Britain and Czechia
and Siberia, and where life is the same, language
is the same. We should follow the approach of
Richard Fester, postulate simple words and look
how they could have evolved and developed in
the course of time, meeting with new life conditions
and with new technologies. Fester proposed KALL
as one of five ur-words. I shorten it to KAL and
I don't see any reason to restrict myself to just
five early words. KAL means the Underworld,
also the womb of the goddess. The words
mentioned in this thread can all very well traced
back to KAL. For example a vault, echo, German
Hall 'sound' and hallen 'to resound'. Also calf, since
the cave of Altamira shows how the divine hind licks
a bull into life; animals were believed to emanate
from cracks in the rock, both in southern Africa and
in Stone Age Europe. Lithuanian kalb for language
may also go back to KAL, as a painted cave
conserved messages of old, those amazing drawings
and paintings were once _speaking_ to people.
The Underworld had a positive meaning in early
times, it was the womb of the goddess of giving life
and of taking life and of giving new life in the sky
(heavenly abode in the region of the summer triangle
Deneb Vega Atair). The positive aspect becomes
clear in derivatives such as German Quelle for spring,
fountain. Then there is the aspect of *kel 'howl,
whimper', the howling animals such as the wolf,
the howling hellhound Kerberos, descendant of the
powerful animals that accompanied the sun horse
of old through the Underworld and brought it back
savely in the morning. With mining, KAL produced
an array of new derivatives, such as keltoi Kelten
Celts, Gallia, Helvetii - the Celts were miners, and
with the labor of mining, the positive Underworld
KAL turned into hell German Hoelle ... In the case
of Richard Fester's KALL simplified to KAL we can
see how a word evolved and developed in time.

Message has been deleted

Dušan Vukotic

unread,
Nov 7, 2007, 6:57:49 AM11/7/07
to

Your KAL is in fact 'kolo' (circle), Serbian
'okolina' (surroundings),
'celina' (wholeness). I understand well what are you talking about.
The basis of the above words is the same, including English whole,
hell, hollow,
Latin caelum, i- (sky, the heavens; Serb. vaseljena from vas-celjena
universe), oculus, i- (eye), colonia -ae (colony), Greek koilya
(womb,
belly)
The basis of all these words was Gon-Bel (Serbian okno /window, an
entrance into a mine, shaft/; oganj/ognjilo /fireplace, fire/; oko /
eye/); here we can see how the ancient man "equalized" human eye and
a
shaft of a mine with fire or fireplace.

XURBELGON human speech formula could help us to understand the
relation among Serbian NEBO (sky) Latin CELUM (sky) and English
HEAVEN; Serbian CELINA (wholeness) and English WHOLE


There is a Serbian word 'klipan' (obnoxious child) and this word is
related to Slavic glava (galava; cf, Latin calva -ae /the bald scalp
of the head/ Greek kefali /head/; Serbian celav /bald, Lat. calvus/;
In other words, above-mentioned 'klipan' is 'glavonja' ("big-brain")
and Slavic galava is related to Latin GLOBUS and globus is related to
celum, nebo, whole, hell and heaven; German kalben (to calf) is
related to Serbian 'tele' (calf) and teljenje (calfing;); Serb.
odvaljivanje, odeljivanje, deljenje (dividing), Greek delphos (womb)
and adelphos (brothers)

DV

John Atkinson

unread,
Nov 7, 2007, 8:25:08 AM11/7/07
to
"Franz Gnaedinger" <fr...@bluemail.ch> wrote...

> On Nov 7, 4:27 am, "John Atkinson" <johna...@bigpond.com> wrote:
>
>> No, I meant the _real_ date of publication. Quote:
>
>> "Deutsches Wörterbuch (German Dictionary), 32 volumes, 1852-1960. The
>> Grimms themselves saw only the entries A through Forsche of this
>> monumental historical dictionary published during their lifetime. The
>> remaining parts were published by several generations of scholars
>> over a
>> 100 year span."
>
>> Since it appears that the volumes were published in alphabetical
>> order,
>> one would guess that the volume with "Welpe" would have come out
>> around
>> 1940. But the obsolete PIE forms that you gave look like they're of

>> nineteenth century vintage, possibly dating back to the Grimms
>> themselves.

>I told you to look up the entry WELF, not Welpe.

Which is almost certainly in the same volume, no? (No, I didn't look
either of them up.)

> Person's opinion was followed by Fick, Falk,
> Torp and Hellquist, so the alternative PIE root
> *kel as origin of welf Welpe whelp was approved
> of by Hellquist in 1970 I guess. How can PIE forms
> get obsolete? If the sound laws are good then it
> can't happen that a proto-form is changing with
> every generation.

Why not? In fact, that's pretty much what's been happening, as more
data becomes available and more theorising is done. Remember
Schleicher's Tale, the first composition ever written in PIE? In 1868,
when he wrote it, it started:

Avis, jasmin varna: na a: ast, dadarka akvams, ....

In 1939, Hirt rewrote it as follows:

Owis, jesmin wbl@na: ne e:st, dedork'e ek'wons, ...

In 1979, Lehmann and Zgusta:

Owis, kWesyo wlhna: ne e:st, ek^wons espek^et, ...

In 1997, Adams:

h2owis, kWesyo wlh2neh4ne (h1e) est, h1ek^wons spek^et, ...

which is almost the same, except for the extra laryngeals. So things do
seem to be settling down.

The current "standard" PIE dictionary, the one referenced by most
people, is Julius Podkorny's, published in 1958. It is of course out of
date, and is expected to be superceded soon by the Leiden Indo-European
Etymological Dictionary (see http://www.ieed.nl). This will undoubtedly
be too expensive for me to buy a hard copy. I use mainly Mallory and
Adams (The Oxford Introduction to PIE..., 2006) since it's cheap, up to
date, and very convenient to use. However I don't completely trust it,
having found several errors and discrepancies.

> As 2 plus 2 being 4 in ancient
> times can't be obsolete in our time, it is still four.

That's mathematics. You'd be better comparing with the situation in
other historical sciences, such as paleontology.

> When I woke up this morning I came to the
> conclusion that PIE is basically flawed by the
> assumption that a common language was spoken
> in all of Eurasia 5,000 years ago.

Of course no common language was spoken all over Eurasia, 5000 years ago
or any other time! No one (to my knowledge) has ever assumed it was,
although some early workers came close. No Bronze Age language could
have occupied a territory greater than about 500 000 square kilometers
without breaking up into several unintelligible varieties. Before the
emergence of major state languages, almost all known languages occupied
areas less than 1000 000 sq km, most much less.

> Life was very
> different by then in various parts of Eurasia,
> copper was known in the eastern parts of
> Eurasia while western Europe was lingering on
> in the Stone Age. The common ancestral
> language must belong to a deeper time level,
> 10,000 or 12,000 or 15,000 years ago.

More likely, 100 000 years ago (proto-World).

[...]

John.

John Atkinson

unread,
Nov 7, 2007, 9:34:37 AM11/7/07
to

"Trond Engen" <tron...@engen.priv.no> wrote...

> Trond Engen skreiv:
>
>> John Atkinson skreiv:
>>
>>> "Trond Engen" <tron...@engen.priv.no> wrote ...
>>>
>>>> John Atkinson skreiv:
>>>>
>>>>> ['whelp']
>>>>
>>>> Good old Hellquist quotes three different suggestions for an IE
>>>> origin.
>>>>
>>>> The first is a root */(s)kWel-/, along with Lith. <kãlè> "bark",
>>>> <skãlyti> "yell constantly" and <skalìkas> "constantly yelling
>>>> hound",
>
> This root is also found as */(s)kWol-/ and */(s)kol-/. Is there
> anything else than 'whelp' that suggests /kW/ in the root?

Not that I can see...

>>> According to my Lith dictionary, <kalè>, bitch, <skalikas>, hound
>>> (nothing about 'constantly yelling'), <skalyti>, to yelp, bay. BTW,
>>> English "yell" isn't used for canine vocalisations; are you sure
>>> Hellquist doesn't say "yelping"?.
>>
>> It's becoming evident that my previous answer was all too quickly
>> waritten. Hellquist (<http://runeberg.org/svetym/>) doesn't write
>> English at all, so any mistake is mine. What he says is "oupphörligt
>> skällande jakthund".
>
> ... and <skãlyti> "ideligen skälla".
>
> Moving up:
>
>>> Mallory and Adams quote the root as PIE *(s)koli-, 'young dog'
>>>
>>>> Gk. <skýlax> "whelp".
>>>
>>> Also Albanian <kelysh>, young dog.
>
> So it's */(s)koli-/ for the dog. Moving to the sound:
>
>> Sw. 'skälla' is a verb used about both dogs and people. Is that an
>> s-mobile version of 'yell' -- or 'yelp'?

<yell> <OE <gellan>/<galan> <*ghel-, cry out, as already mentioned

<yelp> < PGmc *galpjan < ??

So. Does *sgh- > Gmc sk-? Can't remember. If so, I'll give you
'skälla'.

Don't know. M&A quote it as *g(W)elbhus, 'womb', with brackets around
the superscript w. As you say above, it's "well attested", in English,
Grk, Skt, and Avestan (which has both <gar@wa->, womb, and <g@r@buS>,
newly born animal). English <calf> suggests *golbhus, OE <cilfer-lamb>
suggests *gelbhus, Greek <delphis> suggests *gWelbhus, Indo-Iranian
could be either *g- or *gW-.

Thought: Since AHD is only interested in English cognates, they may
only have the proto-version without the superscript w. (?)

>>>> The third points at Slav. <kolĕbati> "becoming unsteady" without
>>>> going further back.
>>>
>>> In Russian, <kolebanie> is oscillation, fluctuation, hesitation.
>>>
>>> Lithuanian has <kalb>, language, tongue, <kalbèti> to speak, which
>>> at first glance looks cognate to this. (Late PIE short *a and *o
>>> both > /a/ in Baltic, /o/ in Slavic.)
>>
>> Agreed. It seemed a far stretch from "calf" anyway.

Closer to your *(s)kel- (M&A give it as *kelh1-). However, though M&A
give OIr <cailech>, Skt <uSa:-kala> (both 'cock'), Lat <calendae> (when
the ides and nones were announced), Grk <kaleo:>, ON <hjala>, Latv
<kaluot>, Hit <kalless>, (all call or chatter), they don't mention the
Lith <kalb>, <kalbèti> (still less the Slavic). I'll bet they aren't
cognate with the others then.

Franz Gnaedinger

unread,
Nov 7, 2007, 10:17:17 AM11/7/07
to
On Nov 7, 2:25 pm, "John Atkinson" <johna...@bigpond.com> wrote:
>
> Why not? In fact, that's pretty much what's been happening, as more
> data becomes available and more theorising is done. Remember
> Schleicher's Tale, the first composition ever written in PIE? In 1868,
> when he wrote it, it started:
>
> Avis, jasmin varna: na a: ast, dadarka akvams, ....
>
> In 1939, Hirt rewrote it as follows:
>
> Owis, jesmin wbl@na: ne e:st, dedork'e ek'wons, ...
>
> In 1979, Lehmann and Zgusta:
>
> Owis, kWesyo wlhna: ne e:st, ek^wons espek^et, ...
>
> In 1997, Adams:
>
> h2owis, kWesyo wlh2neh4ne (h1e) est, h1ek^wons spek^et, ...
>
> which is almost the same, except for the extra laryngeals. So things do
> seem to be settling down.

Thanks for mentioning Schleicher's tale, which I didn't
know. (Once I asked for a text written in PIE but got no
reply.) Are the succeeding versions really converging?
Schleicher 1868: Avis, jasmin varna na a ast, dadarka
akvams, tam, vagham garum vaghantam, tam, bharam
magham, tam manum aku bharantam. Lehman and
Zgusta 1979: (Gw&rei) owis, kwesyo wlhna ne est,
ekwons espeket, oinom ghe gwrum woghom weghontm,
oinomkwe megam bhorom, oinomkwe ghmenm oku
bherontm.

> The current "standard" PIE dictionary, the one referenced by most
> people, is Julius Podkorny's, published in 1958. It is of course out of
> date, and is expected to be superceded soon by the Leiden Indo-European

> Etymological Dictionary (seehttp://www.ieed.nl). This will undoubtedly


> be too expensive for me to buy a hard copy. I use mainly Mallory and
> Adams (The Oxford Introduction to PIE..., 2006) since it's cheap, up to
> date, and very convenient to use. However I don't completely trust it,
> having found several errors and discrepancies.

.
> That's mathematics. You'd be better comparing with the situation in
> other historical sciences, such as paleontology.

.
> Of course no common language was spoken all over Eurasia, 5000 years ago
> or any other time! No one (to my knowledge) has ever assumed it was,
> although some early workers came close. No Bronze Age language could
> have occupied a territory greater than about 500 000 square kilometers
> without breaking up into several unintelligible varieties. Before the
> emergence of major state languages, almost all known languages occupied
> areas less than 1000 000 sq km, most much less.

What is PIE then? The language of one single tribe?

> More likely, 100 000 years ago (proto-World).

I think language made a great leap in the time between
42 000 and 15 000 BP, as did art and technology, and
the center of this development was the Franco-Cantabrian
space, another center Siberia (Malta). Some words may
go farther back, for example KA for what is beyond our
reach, sky, what is inside rock, in a well, or deep within
ourselves, only accessible to a shaman in a trance,
and KU for woman. I postulate these words for the
language of the dwellers of the Blombos cave of South
Africa, Middle Stone Age, 75 000 BP. Meanwhile one
more cave in South Africa has been discovered,
Pinnacle Point, whose dwellers used red ochre and
showed modern behavior, 165 000 BP, more than
double as old. For the time being, I confine myself
to the hypothetical language in the Franco-Cantabrian
space in the last Ice Age.

Trond Engen

unread,
Nov 7, 2007, 7:17:44 PM11/7/07
to
John Atkinson skreiv:

> "Trond Engen" <tron...@engen.priv.no> wrote...
>

>> I don't find */qWelbh-/ in the AHD IE root index, BTW. There's a
>> root */wel-/ but 'calf' doesn't fit there. What's going on?
>
> Don't know. M&A quote it as *g(W)elbhus, 'womb', with brackets
> around the superscript w. As you say above, it's "well attested", in
> English, Grk, Skt, and Avestan (which has both <gar@wa->, womb, and
> <g@r@buS>, newly born animal). English <calf> suggests *golbhus, OE
> <cilfer-lamb> suggests *gelbhus, Greek <delphis> suggests *gWelbhus,
> Indo-Iranian could be either *g- or *gW-.

Could these different vowels suggest a verbal theme */gWel-bh-/ "give
birth", an extension of AHD's root */gWelH-/ "throw"?

> Thought: Since AHD is only interested in English cognates, they may
> only have the proto-version without the superscript w. (?)

Possibly, but there's no fitting */gel-/ or */gelbh-/ either. Only
*/wel-/ "to turn, roll; with derivatives referring to curved, enclosing
objects" -- including <vulva> < */wolw-a:/.

--
Trond Engen
- attesting */wel-/

John Atkinson

unread,
Nov 7, 2007, 9:27:50 PM11/7/07
to
"Franz Gnaedinger" <fr...@bluemail.ch> wrote...

[...]


> .
>> Of course no common language was spoken all over Eurasia, 5000 years
>> ago
>> or any other time! No one (to my knowledge) has ever assumed it was,
>> although some early workers came close. No Bronze Age language could
>> have occupied a territory greater than about 500 000 square
>> kilometers
>> without breaking up into several unintelligible varieties. Before
>> the
>> emergence of major state languages, almost all known languages
>> occupied
>> areas less than 1000 000 sq km, most much less.
>
> What is PIE then? The language of one single tribe?

Yes. Or something like that (depending on your definition of "tribe").

[...]

J.

Dušan Vukotic

unread,
Nov 8, 2007, 3:01:25 AM11/8/07
to
Serbian gnev (anger, fury, wrath) and bes (rage, choler), English gnaw
(Serb. gnjaviti), violence etc...From Gon-Bel; i.e. Serbian kobeljanje-
kolebanje (rolling about, fluctuation)

> Your KAL is in fact 'kolo' (circle), Serbian 'okolina' (surroundings),

> 'celina' (wholeness). I understand well what you are talking about.


> The basis of the above words is the same, including English whole,

> hell, hollow, Latin caelum, i- (sky, the heavens; Serb. vaseljena from vas- > celjena universe), oculus, i- (eye), colonia -ae (colony), Greek koilya
> (womb,

It seems that nobody caught the point. Maybe German Knabe and English
knave could help to clear your heads (Kolben bulb!!). :-)

> XURBELGON human speech formula could help us to understand the
> relation among Serbian NEBO (sky) Latin CELUM (sky) and English
> HEAVEN; Serbian CELINA (wholeness) and English WHOLE
>
> There is a Serbian word 'klipan' (obnoxious child) and this word is
> related to Slavic glava (galava; cf, Latin calva -ae /the bald scalp
> of the head/ Greek kefali /head/; Serbian celav /bald, Lat. calvus/;
> In other words, above-mentioned 'klipan' is 'glavonja' ("big-brain")
> and Slavic galava is related to Latin GLOBUS and globus is related to
> celum, nebo, whole, hell and heaven; German kalben (to calf) is
> related to Serbian 'tele' (calf) and teljenje (calfing;); Serb.
> odvaljivanje, odeljivanje, deljenje (dividing), Greek delphos (womb)
> and adelphos (brothers)

The key words here might have been Serb. glupan and tupan (dumb and
dull/dense) ;-)


DV

Franz Gnaedinger

unread,
Nov 8, 2007, 4:27:49 AM11/8/07
to
On Nov 8, 3:27 am, "John Atkinson" <johna...@bigpond.com> wrote:
> "Franz Gnaedinger" <f...@bluemail.ch> wrote...

.
> > What is PIE then? The language of one single tribe?
>
> Yes. Or something like that (depending on your definition of "tribe").

Well then, let me formulate my test case.

Comparative method: *wlkwo/wlpo was PIE for wolf.

Magdalenian hypothesis: the origin of the Eurasian words
for wolf are KAL for Underworld, and DhAG for able. KAL
has many derivatives, among them German Hall 'sound',
hallen 'to resound', English call, PIE or better EPIE (Early
PIE) *kel 'to howl, whimper', wherefrom Old German welf
German Welpe English whelp, doubled kelkel wherefrom
LPIE (Late PIE) *wlkwo/wlpo 'wolf'. Also DhAG has many
derivatives. The supreme Celtic god Dagda was the good
god in the sense of the able god. German taugen 'to be
good or fit for', and Zeug as in Werkzeug 'tool'. Now for
animals: English dog, German Dachs 'badger', English
dachshund, German Fuchs English fox, Slavic vuk 'wolf',
LPIEX (Late PIE, variant X, or from region X) *ukos 'wolf',
wherefrom ylkos (SsEyR KI PhAAiNNOS SsEyR Ai
YLKIOS ... Zeus is the shining one also when Zeus is the
Lykaion one, Tiryns Disk, Middle Helladic, ca. 1 650 BC,
commemorating Eponymous Tiryns who got represented
as lion-wolf-dog-bee-king on the gold signet ring from
Tiryns, apparently impersonating the abilities of the
respective animals) lykos lupus lupo lobo loup leuve ...
We have then a pair of deep roots, KAL for Underworld
and DhAG for able. A combination of Underworld and
able occurs in the hellhound Kerberos of Greek mythology,
descendant of the Magdalenian divine stag CER who
protected the sun horse when entering and leaving the
Underworld. In the Late Bronze Age of Armenia, the
hellhound Ker-beros, snake heads as symbols of
flames emanating from his body, was a companion of
Her-akles in the role of the Sun-Archer, protecting the
sun from evil forces. The leaping foxes on the central
pillars of temple B of Göbekli Tepe, 11 600 - 9 500 BP,
had, in my opinion, the task of guiding the worthy soul
of a dead ruler through the labyrinth of the Underworld
and back to earth, wherefrom the soul will rise in the
sky, to a heavenly abode; the foxes are shown when
emerging from the ground, leaping, their elegant back
lines, forming long arcs, indicating the trajectory of the
sun ... A charming coin of the Celtic Curisolitae,
Bretagne, shows the sun horse, under it the snout of
a fox, emerging from the ground, thus indicating the
sun horse's end of journey through the Underworld.
We may also try a linguistic combination, namely
DhAG KAL, which might have survived in Sanskrit
Srgala Persian Shag(h)al English jackal. The
Egyptian god Anubis was a jackal, a god of the
Underworld, preserving mummies from the evil forces
of the night, later on he served in weighing the hearts
under Osiris. He was an able protector and judge,
especially concerned with the sun king.

Now do the sound laws forbid any of this?


Franz Gnaedinger

unread,
Nov 9, 2007, 3:40:16 AM11/9/07
to

No reply so far, just one killrating. Hey, folks, you
have here a chance to finish off with Magdalenian.
I chose a strong case in your favor, a brillant PIE
reconstruction, and hold against it a risky hybrid
of two Magdalenian words. All you have to do now
is to point out a violation of a sound law. You have
to argue, of course. Every killrating in this question
is a small victory of mine: because it says you would
love to finish off with Magdalenian but you can't,
not even in such a case where a limping hybrid of
Magdalenian is contesting a strong PIE word.

phog...@abo.fi

unread,
Nov 9, 2007, 3:49:20 AM11/9/07
to

No use to finish off a dead body.

Franz Gnaedinger

unread,
Nov 9, 2007, 4:16:16 AM11/9/07
to
On Nov 9, 9:49 am, phogl...@abo.fi wrote:
>
> No use to finish off a dead body.

Magdalenian is alive and very well, it has a mythological
depth PIE lacks. Point out a violation of a sound law
in my message.

While waiting I may add a word about PIE. Also others
feel a need for several PIE levels. And how could IE
have spread so easily all over Eurasia when it arose
from the language of one single tribe or region?

Let us assume that some 5,500 years ago the following
languages were spoken in Eurasia: A B C D E F G H
I J K L M N O P. Now, with casting bronze in the region
of the Andronovo culture east of the Caspian Sea at
about this time, 3 500 BC, a new language arose from
the language spoken by the locals, and spread all over
Eurasia in the succeeding millenia. The new language,
IE, was absorbed almost everywhere, which indicates
that the the above languages A B C D E F G H I J K L
M N O P were quite similar, descendants of a common
earlier language. And this language would have been
- Magdalenian, of course.

Now just point out a violation of a sound law in my
Magdalenian hybrid explanation of the Eurasian words
for wolf. If you can't, and if you answer with another floppy
remark, I'll have won me one further small victory.


phog...@abo.fi

unread,
Nov 9, 2007, 5:52:17 AM11/9/07
to
On Nov 9, 11:16 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
> On Nov 9, 9:49 am, phogl...@abo.fi wrote:
>
>
>
> > No use to finish off a dead body.
>
> Magdalenian is alive and very well, it has a mythological
> depth PIE lacks.

Magdalenian is a fantasy of your own and does not exist outside your
imagination. Magdalenian has no "mythological depth", because there is
no extant mythology written or told in Magdalenian.

Peter T. Daniels

unread,
Nov 9, 2007, 8:19:00 AM11/9/07
to
On Nov 9, 3:40 am, Franz Gnaedinger <f...@bluemail.ch> wrote:

> No reply so far, just one killrating. Hey, folks, you
> have here a chance to finish off with Magdalenian.
> I chose a strong case in your favor, a brillant PIE
> reconstruction, and hold against it a risky hybrid
> of two Magdalenian words. All you have to do now
> is to point out a violation of a sound law.

Have you not been following the other thread, the one about Sanskrit?

You can't "violate a sound law" if no sound law has ever been
extracted (deduced) from the data. When you have dozens of regular
correspondences among daughter languages, then you can reconstruct
parent forms, and then you can explain how the languages might have
changed in going from the putative parent form to the various attested
daughter forms. Those explanations are the "laws."

Franz Gnaedinger

unread,
Nov 9, 2007, 12:26:35 PM11/9/07
to
On Nov 9, 11:52 am, phogl...@abo.fi wrote:
>
> Magdalenian is a fantasy of your own and does not exist outside your
> imagination. Magdalenian has no "mythological depth", because there is
> no extant mythology written or told in Magdalenian.

"Magdalenian" is just a name. I mean the language
spoken by the Lascaux people. Or would you deny
them a language? can you imagine a highly developed
visual language going along with _oog oog oog_ for
a spoken language? By the way, already the
Neanderthals must have had a developed language
as they possessed the FOXP2 gene associated
with language.

Why did you snip my argument? Here again.
IE developed from the language of one tribe
or people (see John's reply) and it was readily
adopted by almost all the tribes and peoples
all over Eurasia. This means the languages
of Eurasia were rather cloesely related,
and this means they originated from a common
ancestor language. This unknown language
is what I call Magdalenian.

Franz Gnaedinger

unread,
Nov 9, 2007, 12:40:37 PM11/9/07
to
On Nov 9, 2:19 pm, "Peter T. Daniels" <gramma...@verizon.net> wrote:
>
> Have you not been following the other thread, the one about Sanskrit?

No, I haven't.

> You can't "violate a sound law" if no sound law has ever been
> extracted (deduced) from the data. When you have dozens of regular
> correspondences among daughter languages, then you can reconstruct
> parent forms, and then you can explain how the languages might have
> changed in going from the putative parent form to the various attested
> daughter forms. Those explanations are the "laws."

PIE for wolf was *wlkwo/wlpo. Now I say this may
have been the word for wolf used 5,500 years ago
in the region of the Andronovo culture, east of the
Caspian Sea, but there were many more words
for wolf in other parts of Eurasia, and most of these
words were related, and all go back to an earlier
kelkel, doubling of *kel 'to howl', and this *kel
goes once more back to Magdalenian KAL for
the Underworld. Is this possible in the light of
the sound laws? KAL *kel kelkel *wlkwo ?

Second question. DhAG means able. It has many
derivatives, among them Latin dux 'leader' and
Slavik vuk 'wolf'. An earlier derivative was *ukos
'wolf'. This became ylkos lykos lupus lupo lobo ...
Ylkos also became yrkos wherefrom Sanskrit
vrko 'wolf'. And the two roots, KAL and DhAG
formed a hybrid. Is there any literature on hybrids
in language, in word forming?

phog...@abo.fi

unread,
Nov 9, 2007, 1:34:06 PM11/9/07
to
On Nov 9, 7:26 pm, Franz Gnaedinger <f...@bluemail.ch> wrote:
> On Nov 9, 11:52 am, phogl...@abo.fi wrote:
>
>
>
> > Magdalenian is a fantasy of your own and does not exist outside your
> > imagination. Magdalenian has no "mythological depth", because there is
> > no extant mythology written or told in Magdalenian.
>
> "Magdalenian" is just a name. I mean the language
> spoken by the Lascaux people.

Of course the Lascaux people had a language. However, as they left no
written record behind, we cannot know anything about the language they
spoke. The "Magdalenian" you are proposing is simply a figment of your
own imagination, which cannot have anything in common with the
language actually spoken by the Lascaux people, except by chance.

ranjit_...@yahoo.com

unread,
Nov 9, 2007, 1:41:56 PM11/9/07
to
On Nov 9, 1:16 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
> While waiting

We can't finish off Magdalenian, a constructed language with only one
speaker, making the generous assumption that you can say anything in
it. Since you invented it, only you can finish it off.

> I may add a word about PIE. Also others
> feel a need for several PIE levels. And how could IE
> have spread so easily all over Eurasia when it arose
> from the language of one single tribe or region?

How could English have spread so easily all over the world?

Harlan Messinger

unread,
Nov 9, 2007, 1:54:55 PM11/9/07
to

Or Arabic over a region stretching from Mauritania to Somalia and Abu Dhabi?

Franz Gnaedinger

unread,
Nov 10, 2007, 2:30:28 AM11/10/07
to
On Nov 9, 7:34 pm, phogl...@abo.fi wrote:
>
> Of course the Lascaux people had a language. However, as they left no
> written record behind, we cannot know anything about the language they
> spoke. The "Magdalenian" you are proposing is simply a figment of your
> own imagination, which cannot have anything in common with the
> language actually spoken by the Lascaux people, except by chance.

Eighty per cent of our European genome is still the same
as the one of the Lascaux people. Basically, we are still
the same hunters and gatherers. The Mesolithic lasted
very long in parts of Switzerland, until about 4,500 years ago.
Visual language is a continuum, from Paleolithic Europe
to Celtic coins. We get to the same ideas either from
cave paintings forward in time, or via linguistics backward
in time - see Michael Janda's examination of Vedic texts
that led him to the assumption of a heavenly abode in
the Milky Way, the same assumption we get from cave
paintings (interpretations by Michael Rapp, me, and others).
And how do you know that nothing has been written down
in the Lascaux cave and in other caves? I identified several
words:

PAS CA --- everywhere (pas) sky (ca), may the supreme
ruler of the lower Rhone valley toam the sky in his next life
as he roams the land in this life, written as domino five and
additional dot next to the upper right dot in the Brunel
chamber of Chauvet, 32 000 to 30 000 BP

DAI --- protected area, written as tectiform signs

SAI --- life, existence, written as lines and fields of dots,
Neolithic houses in Switzerland had been adorned with
red dots, which may have said much as: May this house
replenish with life ...

The lunar phases were noted in the rotunda of Lascaux,
as explained in my etymological thread. Some words
had been written down, I believe, in visual language
_and_ in abstract signs that hold a precise meaning.

PIE so far is a black box, all is thrown into it. About
the same situation as two hundred years ago, when
people were seriously interested in geology but were
hampered by the belief that the world is only 6,000
years old.

And now please try to falsify my reconstruction

KAL 'Underworld' *kel 'to howl, whimper' (wherefrom
welf Welpe whelp) kelkel *wlkwo/wlpo 'wolf'

Either you can point out a violation of a sound law
in this, or you can't. In the first case I have to abandon
Magdalenian, in the second case I will go on.

Peter said a sound law can't be violated. Any law can
be violated. Speakers can't violate a sound law, but
linguistic reconstructions can. As nature can't violate
a (valid) law of nature while theories can. Then they
have to be abandoned, or revised and modified.

Franz Gnaedinger

unread,
Nov 10, 2007, 2:34:18 AM11/10/07
to
On Nov 9, 7:54 pm, Harlan Messinger
<hmessinger.removet...@comcast.net> wrote:

> ranjit_math...@yahoo.com wrote:
> > On Nov 9, 1:16 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
> >> While waiting
>
> > We can't finish off Magdalenian, a constructed language with only one
> > speaker, making the generous assumption that you can say anything in
> > it. Since you invented it, only you can finish it off.
>
> >> I may add a word about PIE. Also others
> >> feel a need for several PIE levels. And how could IE
> >> have spread so easily all over Eurasia when it arose
> >> from the language of one single tribe or region?
>
> > How could English have spread so easily all over the world?

As a second language, mostly, no non-Indo-European
language was turned into an IE language by the impact
of English.

> Or Arabic over a region stretching from Mauritania to Somalia and Abu Dhabi?

Because Arabic was close to the languages spoken in those
regions. My point with IE and PIE.

ranjit_...@yahoo.com

unread,
Nov 10, 2007, 10:56:40 AM11/10/07
to

Any Bibles in the Maghreb were in Latin. If they had only a language
similar to Arabic, they would have had Bibles only in that language.

ranjit_...@yahoo.com

unread,
Nov 10, 2007, 11:03:05 AM11/10/07
to
On Nov 9, 11:30 pm, Franz Gnaedinger <f...@bluemail.ch> wrote:
> On Nov 9, 7:34 pm, phogl...@abo.fi wrote:

> Eighty per cent of our European genome is still the same
> as the one of the Lascaux people. Basically, we are still
> the same hunters and gatherers.

Even if so, so what? If 80% of the ancestors of the French or English
were Celtic speaking at some point in the past, does that make French
and English descendants of Celtic languages?

> And how do you know that nothing has been written down
> in the Lascaux cave and in other caves? I identified several
> words:

Where are these words inscribed?

Harlan Messinger

unread,
Nov 10, 2007, 12:08:29 PM11/10/07
to
Franz Gnaedinger wrote:
> On Nov 9, 7:34 pm, phogl...@abo.fi wrote:
>> Of course the Lascaux people had a language. However, as they left no
>> written record behind, we cannot know anything about the language they
>> spoke. The "Magdalenian" you are proposing is simply a figment of your
>> own imagination, which cannot have anything in common with the
>> language actually spoken by the Lascaux people, except by chance.
>
> Eighty per cent of our European genome is still the same
> as the one of the Lascaux people. Basically, we are still
> the same hunters and gatherers. The Mesolithic lasted
> very long in parts of Switzerland, until about 4,500 years ago.
> Visual language is a continuum, from Paleolithic Europe
> to Celtic coins. We get to the same ideas either from
> cave paintings forward in time, or via linguistics backward
> in time - see Michael Janda's examination of Vedic texts
> that led him to the assumption of a heavenly abode in
> the Milky Way, the same assumption we get from cave
> paintings (interpretations by Michael Rapp, me, and others).
> And how do you know that nothing has been written down
> in the Lascaux cave and in other caves? I identified several
> words:

"Imagined", not "identified".

Harlan Messinger

unread,
Nov 10, 2007, 12:18:45 PM11/10/07
to

1. In Morocco and Libya before the Arabs arrived they spoke a
Berber-related language, not closely related to Arabic.

2. The Arabs who came to these places and began to populate them already
spoke Arabic, and continued to speak Arabic. It had nothing to do with
the relationship between Arabic and the previous local languages.

Peter T. Daniels

unread,
Nov 10, 2007, 4:46:06 PM11/10/07
to
On Nov 10, 2:30 am, Franz Gnaedinger <f...@bluemail.ch> wrote:

> And now please try to falsify my reconstruction

It can't be done, since your "reconstructions" are not based on any
principle whatsoever. You invent "roots" for any set of words that
happens to strike your fancy, with no attention to any roots you
reconstructed the day before that may have involved the same sounds.

What is _falsified_ in Popperian science is _predictions_, not
"reconstructions" or fantasies.

> KAL 'Underworld' *kel 'to howl, whimper' (wherefrom
> welf Welpe whelp) kelkel *wlkwo/wlpo 'wolf'
>
> Either you can point out a violation of a sound law
> in this, or you can't. In the first case I have to abandon
> Magdalenian, in the second case I will go on.
>
> Peter said a sound law can't be violated.

I have never said any such thing. I probably mentioned long ago the
linguistic joke that the sound laws are enforced by the
Lautgesetzpolizei.

> Any law can
> be violated. Speakers can't violate a sound law, but
> linguistic reconstructions can. As nature can't violate
> a (valid) law of nature while theories can. Then they
> have to be abandoned, or revised and modified.

Evidently your understanding of "sound law" is no better than
"analyst"'s.

phog...@abo.fi

unread,
Nov 10, 2007, 7:49:17 PM11/10/07
to
On Nov 10, 11:46 pm, "Peter T. Daniels" <gramma...@verizon.net> wrote:
> On Nov 10, 2:30 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
>
> > And now please try to falsify my reconstruction
>
> It can't be done, since your "reconstructions" are not based on any
> principle whatsoever. You invent "roots" for any set of words that
> happens to strike your fancy, with no attention to any roots you
> reconstructed the day before that may have involved the same sounds.
>
> What is _falsified_ in Popperian science is _predictions_, not
> "reconstructions" or fantasies.
>
> > KAL 'Underworld' *kel 'to howl, whimper' (wherefrom
> > welf Welpe whelp) kelkel *wlkwo/wlpo 'wolf'
>
> > Either you can point out a violation of a sound law
> > in this, or you can't. In the first case I have to abandon
> > Magdalenian, in the second case I will go on.
>
> > Peter said a sound law can't be violated.
>
> I have never said any such thing. I probably mentioned long ago the
> linguistic joke that the sound laws are enforced by the
> Lautgesetzpolizei.

You indeed did, back in 2001, as an archive search shows. The joke is
actually rather funny.

Franz Gnaedinger

unread,
Nov 11, 2007, 3:07:35 AM11/11/07
to
On Nov 10, 10:46 pm, "Peter T. Daniels" <gramma...@verizon.net> wrote:
> On Nov 10, 2:30 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
>
> > And now please try to falsify my reconstruction
>
> It can't be done, since your "reconstructions" are not based on any
> principle whatsoever. You invent "roots" for any set of words that
> happens to strike your fancy, with no attention to any roots you
> reconstructed the day before that may have involved the same sounds.

The day before? the years before. I mined Magdalenian
words according to my four Magdalenian laws, now I look
what I can do with them. And the words I mined before
go on springing surprises.

> What is _falsified_ in Popperian science is _predictions_, not
> "reconstructions" or fantasies.
>
> > KAL 'Underworld' *kel 'to howl, whimper' (wherefrom
> > welf Welpe whelp) kelkel *wlkwo/wlpo 'wolf'
>
> > Either you can point out a violation of a sound law
> > in this, or you can't. In the first case I have to abandon
> > Magdalenian, in the second case I will go on.
>
> > Peter said a sound law can't be violated.
>
> I have never said any such thing. I probably mentioned long ago the
> linguistic joke that the sound laws are enforced by the
> Lautgesetzpolizei.

Here in this thread, just a few days ago, on Nov 9
3:40 am, you told me no sound law can be violated
if no sound law has been established before. Sound
laws have been established by now, and as the
Lascaux people had the same vocal tracts as the IE
speakers, these laws also hold for languages predating
Indo-European. So anwer my question: does my
reconstruction violate a sound law? KAL 'Underworld'
early PIE *kel 'to howl' (wherefrom Old High German
welf German Welpe English whelp), doubled kelkel
'howlhowl' which turned into PIE *wlkwo/wlpo 'wolf'.
Does this recontruction violate a sound law?

> > Any law can
> > be violated. Speakers can't violate a sound law, but
> > linguistic reconstructions can. As nature can't violate
> > a (valid) law of nature while theories can. Then they
> > have to be abandoned, or revised and modified.
>
> Evidently your understanding of "sound law" is no better than
> "analyst"'s.

So sound laws do not exist? they are just resembling
laws? evoke laws without being laws? The sound laws
or pseudo-laws or whatever mirror the physiological
reality of the gab, of the vocal tract, insofar they are
laws, as the laws of nature mirror features and
properties of nature.

Paul J Kriha

unread,
Nov 11, 2007, 4:17:03 AM11/11/07
to
<ranjit_...@yahoo.com> wrote in message
news:1194710585.9...@y27g2000pre.googlegroups.com...

> On Nov 9, 11:30 pm, Franz Gnaedinger <f...@bluemail.ch> wrote:
> > On Nov 9, 7:34 pm, phogl...@abo.fi wrote:
>
> > Eighty per cent of our European genome is still the same
> > as the one of the Lascaux people. Basically, we are still
> > the same hunters and gatherers.
>
> Even if so, so what? If 80% of the ancestors of the French or English
> were Celtic speaking at some point in the past, does that make French
> and English descendants of Celtic languages?

And we share more than 50% of our genome with primates
such as chimpanzees. What linguistic conclusions do you
draw from that, Franz?

pjk

Paul J Kriha

unread,
Nov 11, 2007, 4:29:35 AM11/11/07
to
<phog...@abo.fi> wrote in message news:1194742157.3...@22g2000hsm.googlegroups.com...

What happens if the LG Polizei catches some of the Lauts
failing to obey the established sound laws? Do they silence
the offending Lauts?

pjk

Peter T. Daniels

unread,
Nov 11, 2007, 7:49:10 AM11/11/07
to
> actually rather funny.-

I heard it from I. J. Gelb, and he always got a laugh from the
audience when he used it.

Peter T. Daniels

unread,
Nov 11, 2007, 7:55:41 AM11/11/07
to
On Nov 11, 3:07 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
> On Nov 10, 10:46 pm, "Peter T. Daniels" <gramma...@verizon.net> wrote:
>
> > On Nov 10, 2:30 am, Franz Gnaedinger <f...@bluemail.ch> wrote:
>
> > > And now please try to falsify my reconstruction
>
> > It can't be done, since your "reconstructions" are not based on any
> > principle whatsoever. You invent "roots" for any set of words that
> > happens to strike your fancy, with no attention to any roots you
> > reconstructed the day before that may have involved the same sounds.
>
> The day before? the years before. I mined Magdalenian
> words according to my four Magdalenian laws, now I look
> what I can do with them. And the words I mined before
> go on springing surprises.

Whether it was the day before or the years before, you have
demonstrated no consitency whatsoever in your "reconstructions."

> > What is _falsified_ in Popperian science is _predictions_, not
> > "reconstructions" or fantasies.
>
> > > KAL 'Underworld' *kel 'to howl, whimper' (wherefrom
> > > welf Welpe whelp) kelkel *wlkwo/wlpo 'wolf'
>
> > > Either you can point out a violation of a sound law
> > > in this, or you can't. In the first case I have to abandon
> > > Magdalenian, in the second case I will go on.
>
> > > Peter said a sound law can't be violated.
>
> > I have never said any such thing. I probably mentioned long ago the
> > linguistic joke that the sound laws are enforced by the
> > Lautgesetzpolizei.
>
> Here in this thread, just a few days ago, on Nov 9
> 3:40 am, you told me no sound law can be violated
> if no sound law has been established before.

And somehow you can twist that into an assertion that a sound law can
be violated?

> Sound
> laws have been established by now, and as the
> Lascaux people had the same vocal tracts as the IE
> speakers, these laws also hold for languages predating
> Indo-European. So anwer my question: does my
> reconstruction violate a sound law? KAL 'Underworld'
> early PIE *kel 'to howl' (wherefrom Old High German
> welf German Welpe English whelp), doubled kelkel
> 'howlhowl' which turned into PIE *wlkwo/wlpo 'wolf'.
> Does this recontruction violate a sound law?

Give us an example of a sound law you have discovered through
systematic correspondence of the attested data (that's how sould laws
are discovered).

> > > Any law can
> > > be violated. Speakers can't violate a sound law, but
> > > linguistic reconstructions can. As nature can't violate
> > > a (valid) law of nature while theories can. Then they
> > > have to be abandoned, or revised and modified.
>
> > Evidently your understanding of "sound law" is no better than
> > "analyst"'s.
>
> So sound laws do not exist? they are just resembling
> laws? evoke laws without being laws? The sound laws
> or pseudo-laws or whatever mirror the physiological
> reality of the gab, of the vocal tract, insofar they are
> laws, as the laws of nature mirror features and
> properties of nature.

You really are as ignorant as "analyst"!

As Ross, John, and even Ranjit have been explaining ad nauseam to
"analyst," a "sound law" is discovered by comparison of dozens of sets
of data regarding relationships among individual phonemes in a group
of putative daughter languages. If regular correspondences are
discovered, then a sound law has been discovered. If there are no
regular correspondences, then there is no sound law and probably no
relationship of filiation between the languages investigated.

For classic examples, see Grimm's Law and, to account for some seeming
exceptions, Verner's Law.

Franz Gnaedinger

unread,
Nov 11, 2007, 11:24:17 AM11/11/07
to
On Nov 10, 6:18 pm, Harlan Messinger
> the relationship between Arabic and the previous local languages.- Hide quoted text -

So your understanding is that many or several unrelated
languages had been spoken in Europe 6,000 years ago,
one of them PIE, which evolved into IE, spread, replaced
most of the other languages, and then split up into the
European languages we know today?

I have another understanding. PIE is like a folded up
tent, which I am trying to unfold. I "see" a common
language in the Ice Age, splitting up with the end of
the Ice Age and the new ways of life emerging then,
all the languages spoken in Eurasia 6,000 years ago
were related, and then it happened that one of those
languages, the one spoken in the region of the Andronovo
culture east of the Caspian Sea, turned into IE (with the
casting of bronze and the domestication of the horse
that followed from crafting bristles), spread, and influenced
most other Eurasian languages ...


Franz Gnaedinger

unread,
Nov 11, 2007, 11:39:08 AM11/11/07
to
On Nov 11, 1:55 pm, "Peter T. Daniels" <gramma...@verizon.net> wrote:
>
> Whether it was the day before or the years before, you have
> demonstrated no consitency whatsoever in your "reconstructions."

Yes, I have. My four laws of Magdalenian:

1) inverse forms are related
2) permutations yield words around the same meme
3) S-words are comparative forms of D-words
4) important words can have lateral associations

> And somehow you can twist that into an assertion that a sound law can
> be violated?

:


> Give us an example of a sound law you have discovered through
> systematic correspondence of the attested data (that's how sould laws
> are discovered).

I am no fonemishist, I study language on the level
of physiology, as I told you many times. As for the
sound laws, the ones of PIE are good enough for me.
Now tell me the followng was impossible: Magdalenian


KAL 'Underworld' early PIE *kel 'to howl (wherefrom

welf Welpe whelp) doubled kelkel 'howlhowl' which
became *wlkwo/wlpo 'wolf' in the IE homeland.


> You really are as ignorant as "analyst"!
>
> As Ross, John, and even Ranjit have been explaining ad nauseam to
> "analyst," a "sound law" is discovered by comparison of dozens of sets
> of data regarding relationships among individual phonemes in a group
> of putative daughter languages. If regular correspondences are
> discovered, then a sound law has been discovered. If there are no
> regular correspondences, then there is no sound law and probably no
> relationship of filiation between the languages investigated.
>
> For classic examples, see Grimm's Law and, to account for some seeming

> exceptions, Verner's Law.-

I learned Grimm's laws when you were a boy.
Sound laws are moulded (?) around the physiology
of the gab, of the vocal tract, of speaking, so I prefer
this physiology over sound laws, yet I am very
grateful for the PIE words that have been and are
still reconstructed with the help of sound laws.

It is loading more messages.
0 new messages