Some thoughts on Indo-Europeans the Beaker Culture

maybe because this happened many thousands years ago. And somehow those Centum folks Centumized the Satem loanwords. Or maybe Satem folks Satemized Centum loanwords etc. Everything is possible.
You are misunderstanding how sound laws work: They, as Taranis has said in other threads, "have no memory". For example, English is a Germanic language, and was thus affected by Grimm's Law, such as p>f (for exapmple, pater>father), why then, did French pasture not turn into fasture when it was loaned into English? Because Grimm's Law was no longer affecting English. The same goes for Centum-Satem.
 
Hmm, lexicostatistics has been widely discredited by mainstream linguistics, and not for good reason. For example, a study by Quentin Atkinson (which this article mentions) suggests Proto-Celtic split from PIE 8000-10000 BP...
I am aware of the flows of this methodology, that's why it would be nice to see it polished and more adequate. My objection to mainstream linguistic thought is claimed but not proved existence of universal rules to the world languages.
 
If the words in question were all loanwords, why do they, without exception, obey to the sound laws of the respective language family?



Not everything is possible. The neogrammarian hypothesis states that sound laws have no exceptions. If they seemingly have exceptions, these are conditioned by their own set of rules. The comparative method has, in around 130-150 years or so of usage, shown that the neogrammarian hypothesis is correct.

Granted, there are some Centum/Satem loanwords in individual languages (for instance, Proto-Slavic borrows from Germanic), but you don't adopt them for every word across the board. That is not possible. If we take the word for "horse", you get the idea of this:

PIE ek´wos

Latin - Equus

Linear-B - "I-Qo"
Modern Greek - Hippo

Old Irish - Ech
Gaulish - Epos
Welsh - Ebol
Breton - Ebeul

Anglo-Saxon - Eoh
Gothic - Aihws

Tocharian - Yakwe

Sanskrit - Ashwa
Avestan - Aspa-

Italic, Celtic, Germanic and Tocharian are all Centum languages and have *k´ merged with *k. Germanic, in addition has shifted *k > *h according to the First Germanic Sound Shift. Sanskrit and Avestan are both Satem languages and *k´ has been shifted into a fricative sound in both cases.
Huh, I don't get you. According to you at the beginning there was just only 1 proto-Indo-European language. And all PIE words were the same. But the fact is that the sound shift occurred among different groups and we got different languages.

Why could not the same happen to the ancient loanwords?

Even many 'modern' French words got Satemized (Slavicnized) by the Russians.

like: cinématographique - kinomatografia

shift from 'S' sound to 'K' sound!
 
Huh, I don't get you. According to you at the beginning there was just only 1 proto-Indo-European language. And all PIE words were the same. But the fact is that the sound shift occurred among different groups and we got different languages.

Why could not the same happen to the ancient loanwords?

It could not happen because depending when they were adopted from elsewhere, they would not be subjects to previous sound laws, because, as I Asturrulumbo (and before that, myself) said, sound laws have no memory. If we pick the example of "horse" and assume for a moment that the IE languages had diverged at that point already, then we would find the word "ek´wos' attested like that in all branches because it could not be subject to individual sound laws.

Because the Neogrammarian hypothesis is correct, it is possible to relatively date when a word did enter the vocabulary of a language due to it's adherence/non-adherence to the sound laws of the language. Asturrulumbo brought an excellent example of this with "pasture": we know because of the cognate in French that the word must have entered the English language after the First Germanic Sound Shift (Grimm's Law).

Even many 'modern' French words got Satemized (Slavicnized) by the Russians.

like: cinématographique - kinomatografia

shift from 'S' sound to 'K' sound!

You are wrong. That is a particularly poor example because the word is a Greek loan in itself (cinema = movement, graphikos = picture).
 
It could not happen because depending when they were adopted from elsewhere, they would not be subjects to previous sound laws, because, as I Asturrulumbo (and before that, myself) said, sound laws have no memory. If we pick the example of "horse" and assume for a moment that the IE languages had diverged at that point already, then we would find the word "ek´wos' attested like that in all branches because it could not be subject to individual sound laws.

Because the Neogrammarian hypothesis is correct, it is possible to relatively date when a word did enter the vocabulary of a language due to it's adherence/non-adherence to the sound laws of the language. Asturrulumbo brought an excellent example of this with "pasture": we know because of the cognate in French that the word must have entered the English language after the First Germanic Sound Shift (Grimm's Law).



You are wrong. That is a particularly poor example because the word is a Greek loan in itself (cinema = movement, graphikos = picture).
Ok.

But my example is a good one. Becasue we do have about loanwords & soundshift in these loanwords between Centum and Satem languages!

Another example of Satemization of a French (Centum) word:

chance - shans (шанс)

Sound shift from CH-sound to SH-sound.

or

meuble - mebelj (ме́бель)

Sound shift from EU-sound to E-sound

etc.
 
Ok.

But my example is a good one. Becasue we do have about loanwords & soundshift in these loanwords between Centum and Satem languages!

Another example of Satemization of a French (Centum) word:

chance - shans (шанс)

Sound shift from CH-sound to SH-sound.

or

meuble - mebelj (ме́бель)

Sound shift from EU-sound to E-sound

etc.

You are still wrong. If I take your example of "chance", yes it's shifted in Russian, but it does NOT adhere to Balto-Slavic or Proto-Slavic sound laws because Russian language has no memory of that.

French "chance" derives from Latin "cadentia", which in turn derives from PIE "k´ad-" ('to fall'). If the Russian language somehow had a memory of it's own development, the word should be rendered as something akin to "sady" in Russia (PIE *k´ becomes *s in Slavic). Instead we see "shans".
 
Huh, I don't get you. According to you at the beginning there was just only 1 proto-Indo-European language. And all PIE words were the same. But the fact is that the sound shift occurred among different groups and we got different languages.

Why could not the same happen to the ancient loanwords?

Even many 'modern' French words got Satemized (Slavicnized) by the Russians.

like: cinématographique - kinomatografia

shift from 'S' sound to 'K' sound!

κινηματογραφος is a Greek word not IE κινημα = motion move and Γραφος = recorder

the letter c of Latin is greek Γ Γ.Κ.Χ belong to the same family
so in a latin form should be C in place of Γ but the mistake is that letter C in west and North is sounded as S
just compare the sound of C at words like Copy Compare etc,
that mistakes many times happened when a word pass from one language to another,

A good example in that is the word potion (drink)

the Greek word is posis πωσις = drinking
but what we drink is ποτον , and the quantity dossage is ποτεριον

so in Greek is poton t= t but in west is potion t=s

same had ancient Greeks like thalatta and thalassa
first id Ionic form second is Doric form
Γλωττα and Γρουσσα Glotta Grussa
Ionic and Doric form

the case of Centum and Satem is complicated and not clear
for Example we put Greek to Centum
but the latin Con wich is a good characteristic of Difference in Greek is Συν sun pronounced sin
does that put Greek language to Satem?

no cause Greek grammar has 3 sounds for s the Sigma Σ the double ΣΣ and the quoppa S the quoppa is an s sound that leaps take form like spelling γ χ or k

to understand these just observed that sound S in word sound, you pull your leaps back and up, same moves like the word hound
then speak the word shift and watch your leaps that are pushed infront,
that is why first s is quopa S and your mouth has the form of Larygeal while second is a purely s and your mouth has a position like the ones with leaps
π β φ

to understand this just compare the Greek s in the end of words with the slavic c
like Greek michailidis and slavic michailovic

and also pronounce the common word slavic in both forms
slavic in Centum is c=k Greek σλαυικ-ος
slavic in Satem is c = ts slavits

more words to understand could be copper and chopper the copper is kopper but the chopper is tsopper

so the cinema follows the k Γ Χ sounds and is centum,

satemization and centum has also to do with climate
the more the worm the more we find big word and many vowells the more the North the less the voweels and the smaller the word,

example typical is the hundred 100
grek is Ekato Hekato and in possesive ekatontados
look at German is Hundat misses e k becomes H a+ becomes u meaning less warm air that exits body, so Nature is stronger than language,
now look the Russian form Sto, the 3 greek vowells became 1 and the need for k+t bacamt st cause leaves less warm air to exit body

so in greece people need to exit warm air and keep vowels and many syllabes
while in cold russia people save warm inside body and make 1 syllabic word while in Germany middle weather conditions is 2 syllabes

but the above does not explain why Sansqrit are satem

and that is a good question.
 
Thank you for the info Yetos!

κινηματογραφος is a Greek word not IE κινημα = motion move and Γραφος = recorder

the letter c of Latin is greek Γ Γ.Κ.Χ belong to the same family
so in a latin form should be C in place of Γ but the mistake is that letter C in west and North is sounded as S
just compare the sound of C at words like Copy Compare etc,
that mistakes many times happened when a word pass from one language to another,
This is exactly what I think. Different syntax and different alphabets make shifting of sound possible! And this happens every time even 5000 years ago!
 
Thank you for the info Yetos!


This is exactly what I think. Different syntax and different alphabets make shifting of sound possible! And this happens every time even 5000 years ago!

No, you still haven't understood it.

Also, Yetos is heavily mixing up orthography with sound laws, which are two very different things. A word as it's written will be pronounced different in various languages as these have different orthographies. Different orthographies do not automatically mean that sound laws are at work.
 
Thank you for the info Yetos!


This is exactly what I think. Different syntax and different alphabets make shifting of sound possible! And this happens every time even 5000 years ago!
It is indeed strange that IE words such as those for copper, metal, silver, gold, wagon, yoke, reins, sword, wine, rye, castrated (cattle), etc., all happened to be loaned among all the Indo-European speakers. It is also strange that mythological motifs such as that of the "cattle raid" were all loaned among the Indo-Europeans...
 
No, you still haven't understood it.

Also, Yetos is heavily mixing up orthography with sound laws, which are two very different things. A word as it's written will be pronounced different in various languages as these have different orthographies. Different orthographies do not automatically mean that sound laws are at work.
I do understand you!

According to you sound laws have no memory. And that shifting in sounds occurs randomly / arbitrarily. That sound (tones) of consonants & vowels in one language are not fixed in other language. That there're no 'fixed patterns' of sound if you convert a word from one language to another one.

But according to me these 'laws' don't rule out my thoughts!
 
It is indeed strange that IE words such as those for copper, metal, silver, gold, wagon, yoke, reins, sword, wine, rye, castrated (cattle), etc., all happened to be loaned among all the Indo-European speakers. It is also strange that mythological motifs such as that of the "cattle raid" were all loaned among the Indo-Europeans...
Lol, good point..
 
Hmm, I wonder... Could R1b-U106 be attributed to the expansion of the "eastern group" of the Beaker culture northwards?
BwRaP.jpg
Haplogroup-R1b-S21.gif
 
Hmm, I wonder... Could R1b-U106 be attributed to the expansion of the "eastern group" of the Beaker culture northwards?

The problem is, we are still not quite sure where exactly U106 originated. Some people have suggest Austria (Maciamo did that a while back, but I'm not sure if that holds up), and rms2 suggested the Baltic Sea shore. And frankly, I'm not sure what the latest of the discussion is. I also haven't seen good data on U106 subclades yet to make a picture there.

What is peculiar certainly is the high ratio in Austria. However, I suspect personally that this may be a bias from a low number of samples (at the same time, you may notice that U152 is extremely rare in Austria). From the studies I've seen, there has been a ludicrously low number of samples from Austria, so U152 may actually be more common and U106 may be rarer.
 
The problem is, we are still not quite sure where exactly U106 originated. Some people have suggest Austria (Maciamo did that a while back, but I'm not sure if that holds up), and rms2 suggested the Baltic Sea shore. And frankly, I'm not sure what the latest of the discussion is. I also haven't seen good data on U106 subclades yet to make a picture there.

What is peculiar certainly is the high ratio in Austria. However, I suspect personally that this may be a bias from a low number of samples (at the same time, you may notice that U152 is extremely rare in Austria). From the studies I've seen, there has been a ludicrously low number of samples from Austria, so U152 may actually be more common and U106 may be rarer.
You are quite right, only 18 Austrians as per the compilation from Myres et al's article. But U106 need not have originated/expanded in the "Eastern Group" to be the result of the Beaker Migrations: it could be that U106 originated from L11* in the "Northern Group" (which coincides well with the sizable frequency of L11 in that area)and expanded. The reason I believe this could have been the case is that it is in my opinion unlikely that R1b arrived with the Corded Ware culture (for reasons such as aDNA and frequencies in other areas the culture occupied).
R1b-L11.jpg
 
I gather that L-11* spread pretty uniformly across Europe ("light and scattered" as one investigator put it) beginning from ca. 3500 BCE (Nordtvedt MRCA calculating system). So it might have been in place and U-106 began its expansion after the arrival of Corded Ware, sometime in the late 3rd millennium BCE. We cannot rule out, on this scenario, that it is at this time that R1b (as U-106) became a carrier of IE speech.
 
Last edited:
I gather that L-11* spread pretty uniformly across Europe ("light and scattered" as one investigator put it) beginning from ca. 3500 BCE (Nordtvedt MRCA calculating system). So it might have been in place and U-106 began its expansion after the arrival of Corded Ware, sometime in the late 3rd millennium BCE. We cannot rule out, on this scenario, that it is at this time that R1b (as U-106) became a carrier of IE speech.
Yes, L11 probably emerged c. 4000 BC from what I have read, I would say it arose somewhere between the Steppe and the Lower Danube basin and spread from there...
 
Why must we automatically link an aspect of a "culture", i.e. earthenware, to a language or for that matter a specific "people"? That "beaker" shaped pottery spread throughout Europe could be easily explained by intermarriage (after all it was the women folk who made their kitchenware), traders and nomads. I don't agree with the people who tie the Beaker Culture to a specific DNA either. Sure there was beaker ware use by R1b, or more appropriately, their wives, but I use a Japanese computer, does that make me Japanese? Many different people used beaker shaped kitchenware, it was a bit like Tupperware today:)
 

This thread has been viewed 36336 times.

Back
Top