L21 on the Iberian Peninsula

[Cambrius (The Red)]

Taranis,

You are criticizing Koch's work on Tartessian without any actual evidence that he is in fact doing what you say he is doing. You talked with a linguist? Who?

Koch is perhaps the foremost Celticist in the world. Do you really think he is calling Tartessian Celtic merely because of a few personal names?

Here is what this site says about Koch's work with Tartessian:



"Abundance, diversity, archaism" all merely from some personal names?

Some of Koch's work is based on proper names, but there is much else, as well, as can be seen from this 2009 paper.

The Corded Ware culture is usually connected with Proto-Germanic, not Celtic. David Anthony, in his The Horse the Wheel and Language, speculates that Italo-Celtic may have arisen from contact between Beaker Folk and elements of the Yamnaya culture on the Hungarian Plain (p. 367).

In Koch's 2009 paper that I linked above, he mentions "the iconography of the ‘warrior stelae’ " shared by Iberia, Armorica (Bretagne), and Britain (p. 1). As I mentioned before, it is possible that Indo-European was spread by sea by the "Stelae People" from the Pontic Caspian region. Anthony mentions their anthropomorphic stelae in his The Horse the Wheel and Language and their spread to western Europe by sea (pp. 336-339).

Indeed, Koch's research has gone well beyond personal names.

 

[rms2]

Two more men of Spanish ancestry have gone L21+ since I last reported: Calzada and Fernandez. Calzada's ancestor came from Bilbao in Spain, which is in the Basque country; however, Calzada is not a Basque surname and is known in neighboring Cantabria. Fernandez can't get his paper trail out of Mexico, but he belongs to a strongly Spanish-Portuguese cluster with 19=15, 459=9-9, and YCAII=19-19.

We have another member of that cluster, this one from Portugal, awaiting L21 test results. Thus far everyone in that cluster is L21+, and all of them have ancestry in either Spain or Portugal.

 

[Taranis]

Indeed, Koch's research has gone well beyond personal names.

Simply put, for Koch to prove his theories, all he has to do is demonstrate regular sound correspondences between the two languages. He has not bothered to do this. This is simply not how historical linguistics functions, not since the Neogrammarian revolution. There's a reason why; if you simply compare two words from two different languages, and allow yourself free reign as far as sound correspondences and the semantics of the words go, you can easily find false cognates in any two languages. It is also undeniable that, in the 2009 paper, the bulk of his data appears to be drawn from personal names. If he is not "calling Tartessian Celtic merely because of a few personal names" then why is he doing so? Is it a big secret?

 

[rms2]

. . . If he is not "calling Tartessian Celtic merely because of a few personal names" then why is he doing so? Is it a big secret?

I read that paper and that is not what Koch is doing. He does analyze quite a few personal names, and they are Celtic, but he analyzes some phrases, as well, and comes up with intelligible translations of them, if they are regarded as Celtic.

Apparently his work continues, and that 2009 paper is not the last word.

If Koch thinks Tartessian was a Celtic language, then it probably was. It doesn't necessarily follow that Celtic originated in the Iberian Peninsula. It simply means that Celtic is older there than we thought and cannot be attributed to Hallstatt influences.

 

[Taranis]

I read that paper and that is not what Koch is doing. He does analyze quite a few personal names, and they are Celtic, but he analyzes some phrases, as well, and comes up with intelligible translations of them, if they are regarded as Celtic.

Apparently his work continues, and that 2009 paper is not the last word.

If Koch thinks Tartessian was a Celtic language, then it probably was.

Well, my point is, if Koch does come up with demonstrable evidence for sound changes/correspondences, the case for Tartessian as a Celtic language is solid. But, from my perspective, it is too early to tell. The problem is, for the "foremost Celticist in the world" (as you called him), Koch's record is a tad patchy for my taste.

It doesn't necessarily follow that Celtic originated in the Iberian Peninsula. It simply means that Celtic is older there than we thought and cannot be attributed to Hallstatt influences.

That, I can agree on. But, unless the Stelae people theory is correct and the Celts really arrived by a very wild migration route (possibly by sea) in Western Europe, the Celtic languages must have somehow originated in the vicinity of the former Corded Ware area.

 

[Wilhelm]

It can also mean that a group of earlier Celts splitted and went to Iberia earlier while the others Celts stayed and developed see the Hallstatt and La Tene cultures

 

[Cambrius (The Red)]

Well, my point is, if Koch does come up with demonstrable evidence for sound changes/correspondences, the case for Tartessian as a Celtic language is solid. But, from my perspective, it is too early to tell. The problem is, for the "foremost Celticist in the world" (as you called him), Koch's record is a tad patchy for my taste.



That, I can agree on. But, unless the Stelae people theory is correct and the Celts really arrived by a very wild migration route (possibly by sea) in Western Europe, the Celtic languages must have somehow originated in the vicinity of the former Corded Ware area.

The majority view seems to be that Celticity emerged with Bell Becker. The earliest Bell Beaker sites have been located in Portugal.

How is Koch's record patchy?

 

[Cambrius (The Red)]

Simply put, for Koch to prove his theories, all he has to do is demonstrate regular sound correspondences between the two languages. He has not bothered to do this. This is simply not how historical linguistics functions, not since the Neogrammarian revolution. There's a reason why; if you simply compare two words from two different languages, and allow yourself free reign as far as sound correspondences and the semantics of the words go, you can easily find false cognates in any two languages. It is also undeniable that, in the 2009 paper, the bulk of his data appears to be drawn from personal names. If he is not "calling Tartessian Celtic merely because of a few personal names" then why is he doing so? Is it a big secret?

Have you read Koch's latest book? - Tartessian: Celtic in the South-west at the Dawn of History. He deals with phrases as well as place and personal names, verbs, proverbs, nouns, prepositions...

In addition, "Celtic from the West" is about to be released. This is the first research segment of the ongoing University of Wales Celtic from the West project. It contains papers from 12 other scientists, besides Koch.

 

[Taranis]

The majority view seems to be that Celticity emerged with Bell Becker. The earliest Bell Beaker sites have been located in Portugal.

I was under the impression that there were sites of similar age (2900 BC) in southern France and northern Italy. In regard for the identity of the Bell-Beaker culture, if they really already spoke an Indo-European language, I'm personally of the opinion that at that stage labeling it "Proto-Celto-Italic" might bemore appropriate. I would also not rule out the possibility that what we regard as "Celtic languages" is actually a paraphyletic construct. Also, the term "Celticity" is extremely problematic in my opinion. In an allusion to a classic question, who is a Celt?

How is Koch's record patchy?

Well, for the "foremost Celticist in the world" he does have a patchy record. As mentioned, why does he omit sound correspondences in his paper?

Have you read Koch's latest book? - Tartessian: Celtic in the South-west at the Dawn of History. He deals with phrases as well as place and personal names, verbs, proverbs, nouns, prepositions...

I seriously hesitate to mention this, since it would seem rather malicious of me to do that, but that book on Tartessian was self-published (published under the imprint "Celtic Studies Publications", which was established by John Koch himself). No serious scholar ever self-publishes his own works, he or she subjects them to peer review, which has not happened in this case. Coincidence?

In addition, "Celtic from the West" is about to be released. This is the first research segment of the ongoing University of Wales Celtic from the West project. It contains papers from 12 other scientists, besides Koch.

Well, all I can say at this point is, we shall see. I still maintain that the spread from the west creates more problems than it apparently solves, unless I see different evidence.

 

[Cambrius (The Red)]

I was under the impression that there were sites of similar age (2900 BC) in southern France and northern Italy. In regard for the identity of the Bell-Beaker culture, if they really already spoke an Indo-European language, I'm personally of the opinion that at that stage labeling it "Proto-Celto-Italic" might bemore appropriate. I would also not rule out the possibility that what we regard as "Celtic languages" is actually a paraphyletic construct. Also, the term "Celticity" is extremely problematic in my opinion. In an allusion to a classic question, who is a Celt?
Well, for the "foremost Celticist in the world" he does have a patchy record. As mentioned, why does he omit sound correspondences in his paper?
I seriously hesitate to mention this, since it would seem rather malicious of me to do that, but that book on Tartessian was self-published (published under the imprint "Celtic Studies Publications", which was established by John Koch himself). No serious scholar ever self-publishes his own works, he or she subjects them to peer review, which has not happened in this case. Coincidence?
Well, all I can say at this point is, we shall see. I still maintain that the spread from the west creates more problems than it apparently solves, unless I see different evidence.

1) What do you mean by "omit sound correspondences in his paper"?


2) I don't see a problem in the Tartessian book being self published. The work has been in the public audience for about a year and I haven't run across any compelling peer criticisms concerning the findings and ideas presented.


3) Can you explain how the old Central European Celtic origination theory is more palatable than what Koch and the Atlantic School has posited thus far?


The Tartessian (Celtic) script on the stone tablets that Koch and others have deciphered predates anything known in Central Europe by over 500 years. Much of what we have thus far seen from the University of Wales project (Celtic from the West) suggests that Tartessian is indeed a Celtic language. Therefore, if the Tartessian language continues to be confirmed as Celtic by the philology and linguistic communities (likely a slow process), how can the Central European origin theory be defended? Of course, there will always be those in opposition to Tartessian as Celtic because of academic investment in competing theories or various odd axes to grind. Old paradigms die hard, don't they? Even the faulty ones.

 

[Taranis]

1) What do you mean by "omit sound correspondences in his paper"?

Alright, in layman terms, sound correspondences are one of the most fundamental principles of comparative linguistics. By it, the very existence of the Indo-European language family was established, and it's been part of the modus operandi of linguistics ever since. It would appear strange to me that Koch discards this when he is trying to prove something big like Tartessian being a Celtic language), as this takes away a lot of credibility for him. If Koch really did not make any sound correspondences at all (instead of just omitting them, for whatever reason), then his work on Tartessian is, from the perspective of linguistics, literally worthless.

Also, what I mentioned earlier, there is the focus on personal names, which is also very unusual and problematic for linguists to do.

2) I don't see a problem in the Tartessian book being self published. The work has been in the public audience for about a year and I haven't run across any compelling peer criticisms concerning the findings and ideas presented.

It is not a problem per se, however, with the problem I described above, one really has to get suspicious about the sincerity of Koch! It comes about as if Koch tries to hide something (ie, that his work is faulty).

3) Can you explain how the old Central European Celtic origination theory is more palatable than what Koch and the Atlantic School has posited thus far?

The Tartessian (Celtic) script on the stone tablets that Koch and others have deciphered predates anything known in Central Europe by over 500 years. Much of what we have thus far seen from the University of Wales project (Celtic from the West) suggests that Tartessian is indeed a Celtic language. Therefore, if the Tartessian language continues to be confirmed as Celtic by the philology and linguistic communities (likely a slow process), how can the Central European origin theory be defended?

Ah, that is very simple: you are simply making a too big conjecture there. You say that because this makes the Celtic languages attested for longer elsewhere, it cannot have originated in Central Europe. Conversely, I would argue exactly the opposite is the case. If we establish that Tartessian is actually a Celtic language, and an archaic one at the same time, attested in an early time slice, then we firmly establish that the innovations that the other Celtic language share must have occured and spread from elsewhere, and that Tartessian was left out of these innovations. If we look at the pattern at which innovations exist in the Celtic language, we precisely get the spread of Hallstatt/La-Tene influence. This means the "old" theory is essentially correct (with respect for the pattern by which linguistic innovations spread), but incomplete, as it doesn't tackle the origin of the Celtic languages as a whole.

Of course, there will always be those in opposition to Tartessian as Celtic because of academic investment in competing theories or various odd axes to grind. Old paradigms die hard, don't they? Even the faulty ones.

I do not have a problem with the identification of Tartessian as a Celtic language per se (if it is proven without doubt, that is), and I do not have a problem with paradigm changes, but I do have a problem replacing a 'false' (actually, not so much false as incomplete) paradigm with an overtly false one that creates more problems than it solves. Specifically, we should see a pattern of spread from west to east, which we do not see, neither in linguistics nor in archaeology. We also must take the other language families that were influenced/related by/with Celtic (principally Italic and Germanic) into account, and we cannot claim that the Celtic languages emerged out of thin air.

 

[Cambrius (The Red)]

Alright, in layman terms, sound correspondences are one of the most fundamental principles of comparative linguistics. By it, the very existence of the Indo-European language family was established, and it's been part of the modus operandi of linguistics ever since. It would appear strange to me that Koch discards this when he is trying to prove something big like Tartessian being a Celtic language), as this takes away a lot of credibility for him. If Koch really did not make any sound correspondences at all (instead of just omitting them, for whatever reason), then his work on Tartessian is, from the perspective of linguistics, literally worthless.
Also, what I mentioned earlier, there is the focus on personal names, which is also very unusual and problematic for linguists to do.
It is not a problem per se, however, with the problem I described above, one really has to get suspicious about the sincerity of Koch! It comes about as if Koch tries to hide something (ie, that his work is faulty).
Ah, that is very simple: you are simply making a too big conjecture there. You say that because this makes the Celtic languages attested for longer elsewhere, it cannot have originated in Central Europe. Conversely, I would argue exactly the opposite is the case. If we establish that Tartessian is actually a Celtic language, and an archaic one at the same time, attested in an early time slice, then we firmly establish that the innovations that the other Celtic language share must have occured and spread from elsewhere, and that Tartessian was left out of these innovations. If we look at the pattern at which innovations exist in the Celtic language, we precisely get the spread of Hallstatt/La-Tene influence. This means the "old" theory is essentially correct (with respect for the pattern by which linguistic innovations spread), but incomplete, as it doesn't tackle the origin of the Celtic languages as a whole.
I do not have a problem with the identification of Tartessian as a Celtic language per se (if it is proven without doubt, that is), and I do not have a problem with paradigm changes, but I do have a problem replacing a 'false' (actually, not so much false as incomplete) paradigm with an overtly false one that creates more problems than it solves. Specifically, we should see a pattern of spread from west to east, which we do not see, neither in linguistics nor in archaeology. We also must take the other language families that were influenced/related by/with Celtic (principally Italic and Germanic) into account, and we cannot claim that the Celtic languages emerged out of thin air.

As I mentioned in an earlier post, Koch, in his "Tartessian" book, deals with much more than just personal names. There is significant space devoted to phrases, verbs, preverbs, nouns, pronouns, prepositions, syntax and word order. I'm certain that the "Celtic from the West" publication, due out next month, will provide quite a bit of additional linguistic material.

 

[rms2]

Taranis,

How is Koch's record "patchy"? You never really answered that, except to talk about "sound correspondences", after admitting that you, unlike Koch, are no linguist. What sort of "sound correspondences" would you like? A comparison to Urdu? Koch is not trying to demonstrate that Tartessian is merely Indo-European, but that it's Celtic. Celtic already has the "sound correspondences" that show it is part of the Indo-European language family.

Koch and others are working from inscriptions when it comes to reconstructing Tartessian, a language from a region we know from historical documents was inhabited by Celts at least as early as the 6th century BC. Personal names are part of that work but they are not all there is. Koch and his compatriots have translated some of the Tartessian inscriptions, and they make sense as a form of early Celtic.

The only "sound correspondences" that are needed are those that make Tartessian make sense when seen as a form of Celtic.

 

[Taranis]

Taranis,

How is Koch's record "patchy"? You never really answered that, except to talk about "sound correspondences", after admitting that you, unlike Koch, are no linguist. What sort of "sound correspondences" would you like? A comparison to Urdu? Koch is not trying to demonstrate that Tartessian is merely Indo-European, but that it's Celtic. Celtic already has the "sound correspondences" that show it is part of the Indo-European language family.

Koch and others are working from inscriptions when it comes to reconstructing Tartessian, a language from a region we know from historical documents was inhabited by Celts at least as early as the 6th century BC. Personal names are part of that work but they are not all there is. Koch and his compatriots have translated some of the Tartessian inscriptions, and they make sense as a form of early Celtic.

The only "sound correspondences" that are needed are those that make Tartessian make sense when seen as a form of Celtic.

Well then, let me elaborate. The fact that I'm no linguist doesn't diminish the authority of my arguments as long as they are valid (I consulted a linguist, remember?). However, I admit that I genuinely need to further elaborate things, and I need to provide you with some background here. The hypothesis of the Neogrammarians states that sound changes are regular, systemic, and purely phonetically conditioned (there are correlaries to address the effects of things like analogy). Already in 1880, the Neogrammarian linguist Hermann Paul stated on the issue:

"Wenn wir daher von konsequenter Wirkung der Lautgesetze reden, so kann das nur heissen, dass bei dem Lautwandel innerhalb desselben Dialektes alle einzelnen Fälle, in denen die gleichen lautlichen Bedingungen vorliegen, gleichmässig behandelt werden. Entweder muss also, wo früher einmal der gleiche Laut bestand, auch auf den späteren Entwickelungsstufen immer der gleiche Laut bleiben, oder, wo eine Spaltung in verschiedene Laute eingetreten ist, da muss eine bestimmte Ursache und zwar eine Ursache rein lautlicher Natur wie Einwirkung umgebender Laute, Akzent, Silbenstellung u. dgl. anzugeben sein, warum in dem einen Falle dieser, in dem andern jener Laut entstanden ist."

approximate translation (please excuse that it's somewhat awkward translated :innocent: ):

"If we hence talk about a consistent effect of the sound laws, so can this only mean that in the sound changes within the same dialect all single cases, in which the same sound conditions exist, have to be treated equally. Hence, either has to, where earlier the same sound existed, also stay the same sound in later stages of development, or, when a split into different sounds occurs, there a certain cause (and a cause of purely phonetic nature, such as the exposure of surrounding sounds, accent, hyphenation and similar) must be specified why in place of it a different sound was developed."

So, this essentially means that Koch is apparently operating at a pre-1880 stage of linguistics. To give some examples, first Tartessian Teeaiona equals a reconstructed Celtic name *Deiwonā but later Teasiioonii corresponds to Tascouanos. Well, which one is it? Initial T = D or Initial T = T? What about the vowels? Does #Ce (initial Consonant + the vowel e) = #Ce or does #Ce = #Ca?

Likewise, Tartessian leoine corresponds to a reconstructed Celtic name *Līwonāi. Then, all of a sudden, meleśae corresponds to Gaulish Meliđđus. Well, which is it? #Ce = #Cī or does #Ce = #Ce? What about #Ci = #Cī? Sure, why not? Tartessian -ris = Celtic -rīχs but then on the same page Tartessian -riś suddenly equals the same thing. Well, which is it? They're written with different characters, presumably they represent different sounds. Does Tartessian s or ś correspond to Celtic s, or does it perhaps correspond to đđ as in Meliđđus?

Later, Tartessian niiraboo is equivalent to Welsh ner, so I guess we can add #Cii = #Ce to the confusion. There is absolutely no effort made to address these correspondences systematically and explain how the differences are conditioned. He just takes a bunch of Tartessian words of unknown meaning, compares them to vaguely similar words cherry-picked from a dozen different Celtic languages spanning thousands of years (Celtiberian, Gaulish, Old Irish, Welsh, etc.), and hopes nobody will notice. The only guiding principle seems to be superficial similarities.

I hope that clarifies it.

 

[Cambrius (The Red)]

Well then, let me elaborate. The fact that I'm no linguist doesn't diminish the authority of my arguments as long as they are valid (I consulted a linguist, remember?). However, I admit that I genuinely need to further elaborate things, and I need to provide you with some background here. The hypothesis of the Neogrammarians states that sound changes are regular, systemic, and purely phonetically conditioned (there are correlaries to address the effects of things like analogy). Already in 1880, the Neogrammarian linguist Hermann Paul stated on the issue:



approximate translation (please excuse that it's somewhat awkward translated :innocent: ):



So, this essentially means that Koch is apparently operating at a pre-1880 stage of linguistics. To give some examples, first Tartessian Teeaiona equals a reconstructed Celtic name *Deiwonā but later Teasiioonii corresponds to Tascouanos. Well, which one is it? Initial T = D or Initial T = T? What about the vowels? Does #Ce (initial Consonant + the vowel e) = #Ce or does #Ce = #Ca?

Likewise, Tartessian leoine corresponds to a reconstructed Celtic name *Līwonāi. Then, all of a sudden, meleśae corresponds to Gaulish Meliđđus. Well, which is it? #Ce = #Cī or does #Ce = #Ce? What about #Ci = #Cī? Sure, why not? Tartessian -ris = Celtic -rīχs but then on the same page Tartessian -riś suddenly equals the same thing. Well, which is it? They're written with different characters, presumably they represent different sounds. Does Tartessian s or ś correspond to Celtic s, or does it perhaps correspond to đđ as in Meliđđus?

Later, Tartessian niiraboo is equivalent to Welsh ner, so I guess we can add #Cii = #Ce to the confusion. There is absolutely no effort made to address these correspondences systematically and explain how the differences are conditioned. He just takes a bunch of Tartessian words of unknown meaning, compares them to vaguely similar words cherry-picked from a dozen different Celtic languages spanning thousands of years (Celtiberian, Gaulish, Old Irish, Welsh, etc.), and hopes nobody will notice. The only guiding principle seems to be superficial similarities.

I hope that clarifies it.

So, where are all the published or media communicated peer criticisms addressing these "deficiencies"? Can one not come up similar observations when comparing fully accepted Celtic languages such as Welsh, Celtiberian, Gaelic, etc.?

 

[Taranis]

So, where are all the published or media communicated peer criticisms addressing these "deficiencies"?

Well, if I were to criticize an article the peer-reviewed way, I would publish my criticism in the same journal as the original article. Now, the Koch article was published in Acta Palaeohispanica issue #9, and the issue #10 hasn't been released yet. That may be it, but don't hold me to it.

Can one not come up similar observations when comparing fully accepted Celtic languages such as Welsh, Celtiberian, Gaelic, etc.?

What specifically?

 

[Cambrius (The Red)]

Well, if I were to criticize an article the peer-reviewed way, I would publish my criticism in the same journal as the original article. Now, the Koch article was published in Acta Palaeohispanica issue #9, and the issue #10 hasn't been released yet. That may be it, but don't hold me to it.
What specifically?

You keep referring to Koch's 2009 paper, however, since then, "Tartessian" has been published and "Celtic from the West" is just weeks away from release (copies have been provided to a number of academic institutions already). We should focus and give greater weight to the most recent findings / constructs.

 

[Taranis]

You keep referring to Koch's 2009 paper, however, since then, "Tartessian" has been published and "Celtic from the West" is just weeks away from release (copies have been provided to a number of academic institutions already). We should focus and give greater weight to the most recent findings / constructs.

Why? Frankly, these more recent findings do not diminish the validity of the criticism about Koch's hypothesis in any ways. Well, "Celtic from the West" hasn't been released yet, obviouly. :embarassed:

 

[Cambrius (The Red)]

Why? Frankly, these more recent findings do not diminish the validity of the criticism about Koch's hypothesis in any ways. Well, "Celtic from the West" hasn't been released yet, obviouly. :embarassed:

Obviously "Celtic from the West" hasn't been released but a number of academics have already reviewed the work.

Let's not get cute... It's tacky, to say the least.

 

[Taranis]

Obviously "Celtic from the West" hasn't been released but a number of academics have already reviewed the work.

Let's not get cute... It's tacky, to say the least.

Sorry, but the existence of very stylish emoticons is tempting.

But to answer your earlier question, with the assessment I decribed above, it's hard to see how anybody could come up with a different conclusion other than that Koch's work is intrinsically flawed.