Twarth? This English word is similar to the Serbian verb
'odvratiti' (avert; turn aside; divert!;-); Russian отвращать, Czech
odvrátit;
Are these words inter-related, including the possible relation between
twarth and divert?
Also interesting, Dutch verstrooien; Serbian 'prestrojiti' avert,
divert; Russian 'Perestroyka'; of course, in Serbian it comes from
'struja' (curent, stream; German Strom; cf. river Strumica /FYR of
Macedinia/).
Serbian 'prestrojiti' means to change one for another stream/row/line
(Serbian 'strujanje' streaming, flowing); it shows that the history of
the name of ancient river Strumica (Strimon) is of clear Slavic origin
and there is no place for any speculations that it belongs to so-
called Shkip-Illyrian (invented and nonexistent) vocabulary.
DV
In addition, there is the Serbian word 'strmina' (precipice, steep,
ascent, escarpment, slope); from the verb 'strme-knuti' se (fall
down); stream and storm; Serbian sručiti (plump; from sHrugniti =>
strugnuti (shuffle, run away); sHrugniti => skrenuti (divert) =>
skrajnuti (turn aside) - Sur-Hor basis; cf. Serbian zaokret (swerve,
slew)
DV
No.
Why don't you do yourself a favor and look up the origins of the words
that catch your eye yourself instead of hoping that if you excitedly
shout "Eureka" here every day about something or other, one day you'll
hit on an amazing discovery? At least notice the most obvious things
such as the fact that in "divert", "di-" is a prefix, and I'm guessing
that "ot-", "od-" is also a prefix.
Oh--by the way: blood is a red liquid and tomato juice is a red liquid.
They must be related!
He, he... Blood and tomato! Very picturesque.
Prefixed? OK. What do you say about athwart; meaning seems to be close
to divert (at right angles to the centre line of a ship; at an oblique
angle; it means that something diverts from the "basic position").
DV
> > Twarth? This English word is similar to the Serbian verb
> > 'odvratiti' (avert; turn aside; divert!;-); Russian отвращать, Czech
> > odvrátit;
> > Are these words inter-related, including the possible relation between
> > twarth and divert?
>
> No.
>
> Why don't you do yourself a favor and look up the origins of the words
> that catch your eye yourself instead of hoping that if you excitedly
> shout "Eureka" here every day about something or other, one day you'll
> hit on an amazing discovery? At least notice the most obvious things
> such as the fact that in "divert", "di-" is a prefix, and I'm guessing
> that "ot-", "od-" is also a prefix.
It is. Slavic prefix 'ot' meaning 'from' or 'out of'.
I hate to say it but, (bez)Dushan has tripped upon a likely cognate of
Latin 'vert'...
Not that it has a direct genetic relationship to 'verto'.. but that's
another story.
Perhaps. But if you look back at the history of "thwart" you reach a
form that wouldn't lead anyone to expect a connection, making it clear
that any superficial resemblence one might see with the modern word is
only the result of later evolution.
Superficial thinking ;-)
Try again; I know, you can do it much better. ;-))
DV
Clarification..
I only meant DV's Serbian verb 'odvratiti'.
Parsed out it's 'ot' + 'vrat'.
The 'vrat' I see as cognate with Latin 'vert/o'.
I find additional support for such a common IE interpretation in
Baltic Latvian where 'verst' (long e) also means 'to turn'.
'Thwart', as you say, looks too long a stretch for any direct south
Slavic 'otvrat'.
Myself, I would suspect some German transitional form indicating
evolution from Baltic 'tvert' and resultant English 'thwart'.
Here it is.. simple enough:
"thwart (adv.)
c.1200, from O.N. þvert "across," originally neut. of thverr (adj.)
"transverse, across,"
cognate with O.E. þweorh "transverse, perverse, angry, cross," ...
Goth. þwairhs "angry"), altered (by influence of *thwer- "to turn")
from *therkh..."
Quod demonstrandum est
>> >> No.
<Thwart> is cognate with Latin <torqueo>, I believe, (PIE *terk(w)-).
It's apparently not related to Latin <verto> (PIE *wert-) or Slavic
*vrat- (Russian <вращать> or Lithuanian <versti> at all.
According to my Lithuanian dictionary, <tverti> means "sieze, grab". I
don't know anything about its etymology. Lithuanian <verti>, "thread a
needle", is cognate with Russian <verat'>, "prick", from PIE *(h2)wer-.
John.
Right, Latin <aperio>, Got. <warjan>, Pruss. <etwerreis>.
The Slavic <-vrat-> is a highly productive word stem. There are
zillions of related words, such as <vrata> gates. Semantically,
they usually have something to do with turning/rotating/flipping.
However, I agree <-vrat-> words are not cognates of E. <thwart> or
Lit. <tverti>.
pjk
>
> John.
>
>
>
>
>
>
>
>
What can you say about Ulfila's andavaurd (Ger. Antwort answer; að
verða a., to come to words with...)? Now we see that the Serbian
'odgovoriti'(answer), 'odvratiti' (side track, divert, repulse,
answer) are the cognates to the Gothic 'and-waúrdyan' (answer) - (Gon-
Bel-Hor basis; cf. Serbian 'ogo-varati' gossip, 'pre-go-varati'
negotiate).
Latin torqeo (torquere, torsi, tortus turn, twist; torsion) is
generated from the Hor-Gon basis (Latin circo, circare; Serbian
kruženje /circulate/, krug /circle/; German Kreis; Italian cerchia,
circolo, girone, cerchio /circle/; Greek θεωρείο, θεάτρου; Serbian
ćeranje => teranje, terati /driving, drive/ - velar to dental change -
obterati /drive in circle/, Latin orbit ;-)
Latvian tverti is the cognate of the Serbian word tvrd (hard), utvrda,
tvrđava, utvrđenje (stronghold); Latvian tvert has the meaning
"grasp"; tveršana means "capture" and it is an equivalent to the
Serbian words <utvrditi> (to harden, ascertain, establish), "utvrditi
znanje" (to fix the knowledge); Latvian 'aptvert' (conceive, realize,
comprise, grip, apprehend, grasp etc.) is a counterpart to the Serbian
word 'potvrditi' (to confirm, acknowledge, support its claim, assert,
ratify)...
In reality, all the following words are cognates: Latin vertex,
Serbian 'breg' (hill; German Berg), English word, preach, German
predigen (Serbian pridika (prich), Serbian obratati se (address,
refer), obrt (turnover, turn), Latin orbit, Serbian vrteti (spin,
rotate), vrh (peak, apex)...
DV
Sorry for multiple messages. I lost my temper because Google
"swallowed" a few of my previous (unsaved) answers.
DV
'Right' in what sense?
What is right?
'Aperio' means 'open' - a complete reversal of the meaning of
'thwart', 'bar', or 'prevent'.
'Etwerreis' means `open' .
Only 'warjan' might be inferred as 'guard (against)'
> The Slavic <-vrat-> is a highly productive word stem. There are
> zillions of related words, such as <vrata> gates. Semantically,
> they usually have something to do with turning/rotating/flipping.
>
> However, I agree <-vrat-> words are not cognates of E. <thwart> or
> Lit. <tverti>.
> pjk
Well... it seems rather obvious that posting opposites in meaning of
any given term would produce dissimilar termis leading to less than
acceptable cognates..
Perhaps it was late when you posted (?)
PS: Cognates anyone? Here's some more:
Latin <aperio> means 'to open'
Latvian <atvert? means 'to open'
Got. <warjan> means 'to guard'
Latvian <vara> means 'power' (military)
Pruss. <etwerreis>. means `open'
Latvian <atvert? means 'to open'
Slavic <vrata> means 'gates' (?)
Lstvian <varta> means 'gate'
It's a figure of speech, like, you know, "okay", "I hear you", "right on".
>What is right?
She'll be.
>'Aperio' means 'open' - a complete reversal of the meaning of
>'thwart', 'bar', or 'prevent'.
>'Etwerreis' means `open' .
>
>Only 'warjan' might be inferred as 'guard (against)'
I have no idea what you are on about.
I am talking about cognates, not about words that mean the
same thing. A reversal (complete or not:-) of the meaning
of cognates occurs quite often, it's nothing unusual.
>> The Slavic <-vrat-> is a highly productive word stem. There are
>> zillions of related words, such as <vrata> gates. Semantically,
>> they usually have something to do with turning/rotating/flipping.
>>
>> However, I agree <-vrat-> words are not cognates of E. <thwart> or
>> Lit. <tverti>.
>> pjk
>
>Well... it seems rather obvious that posting opposites in meaning of
>any given term would produce dissimilar termis leading to less than
>acceptable cognates..
>
>Perhaps it was late when you posted (?)
It always is.
>PS: Cognates anyone? Here's some more:
>
>Latin <aperio> means 'to open'
>Latvian <atvert? means 'to open'
>
>Got. <warjan> means 'to guard'
>Latvian <vara> means 'power' (military)
>
>Pruss. <etwerreis>. means `open'
>Latvian <atvert? means 'to open'
>
>Slavic <vrata> means 'gates' (?)
>Lstvian <varta> means 'gate'
Slavic languages also have <varta> 'watch' or 'guardpost'
which is probably an old borrowing from Germanic.
The Latvian <varta> was possibly borrowed from the
same source.
Slavic <varta> and <vrata> are not cognates.
If two words in two languages mean the same thing
they don't necessarily have to be cognates.
Please find what is really meant by the term "cognate".
pjk
I guess we can add 'evasive in defeat' to you previous
accomplishments.
This is unfortunate as it precludes your inclusion in reasoned debate.
> >'Aperio' means 'open' - a complete reversal of the meaning of
> >'thwart', 'bar', or 'prevent'.
> >'Etwerreis' means `open' .
>
> >Only 'warjan' might be inferred as 'guard (against)'
>
> I have no idea what you are on about.
> I am talking about cognates, not about words that mean the
> same thing.
And I was talking about cognates that had the same meaning without
your (incidental and unmentioned) semantic inversions. Mine are,
therefore more likely to be true cognates/equivalents and not imagined
ones.
> A reversal (complete or not:-) of the meaning
> of cognates occurs quite often, it's nothing unusual.
See above. Comparative linguistics require comparison in form and
meaning... and not random terms that might mean this, that, or
something other.
> >> The Slavic <-vrat-> is a highly productive word stem. There are
> >> zillions of related words, such as <vrata> gates. Semantically,
> >> they usually have something to do with turning/rotating/flipping.
>
> >> However, I agree <-vrat-> words are not cognates of E. <thwart> or
> >> Lit. <tverti>.
> >> pjk
>
> >Well... it seems rather obvious that posting opposites in meaning of
> >any given term would produce dissimilar termis leading to less than
> >acceptable cognates..
>
> >Perhaps it was late when you posted (?)
>
> It always is.
>
> >PS: Cognates anyone? Here's some more:
>
> >Latin <aperio> means 'to open'
> >Latvian <atvert? means 'to open'
>
> >Got. <warjan> means 'to guard'
> >Latvian <vara> means 'power' (military)
>
> >Pruss. <etwerreis>. means `open'
> >Latvian <atvert? means 'to open'
>
> >Slavic <vrata> means 'gates' (?)
> >Lstvian <varta> means 'gate'
>
> Slavic languages also have <varta> 'watch' or 'guardpost'
> which is probably an old borrowing from Germanic.
> The Latvian <varta> was possibly borrowed from the
> same source.
Impossible.. even for someone as purposefully obtuse as yourself.
The root is 'var' and it is pervasive in Baltic - and not in so-called
'germanic' - which (with the exception of late Gothic) is not attested
anywhere. Except in the minds of impoverished germano-centrists, of
course.
> Slavic <varta> and <vrata> are not cognates.
Interesting development in Czech academia...
'Vrata' (gates) is not genetically related to 'varta' (gate)?
What are you smoking?
> If two words in two languages mean the same thing
> they don't necessarily have to be cognates.
But almost always are... prevaricator.
> Please find what is really meant by the term "cognate".
Please keep your snippety imbecilities for your usual company.