Languages, genes and cultures

As you may know, in this blog I have often criticised many aspects of traditional historical linguistics, e.g. the reconstruction of Proto-Indo-European (PIE), including the imaginary set of laryngeals (one of my 'favourite' topics indeed) or the generally accepted chronology of IE expansion. I have written many posts criticizing these things, inspired by the work of some authors, like Mario Alinei and Xaverio Ballester, who oppose the traditional assumptions held in mainstream Indo-European (IE) studies. The good news is that now a major scholarly work, led by Francisco Villar, seems to support these ideas!

Francisco Villar is a renowned expert in Indo-European, and also in the languages of Pre-Roman Iberia. As we saw in this post, one of his theories is that the study of ancient toponyms, especially hydronyms, shows that the oldest languages spoken in the Iberian Peninsula were IE. Any other type of language (Basque, Iberian) appeared later (Villar, 2000). In his last research work (Villar et al 2011), carried out in collaboration with Blanca Prósper, Carlos Jordán and María Pilar Fernández Álvarez, he continues his previous research, comparing the linguistic data with the archaeological and genetic evidence that's now available. I will comment on the results in a series of posts, starting with this one. For the moment, I'll try to summarize some of the main points.

In their research, they focus on the ancient hydronyms of Europe and southwest Asia. The choice of material is relevant: hydronyms usually retain signs of archaic linguistic layers. Analysing these toponyms, they identify a series of components that are significantly present in those areas, e.g. *aisko/isko-, *ab-, or *balso-. Not only that: they also demonstrate, using phonological and lexicological criteria, that these components are IE, with no exception.

The aim of the research is to to try to correlate this set of data with the currently available theories of IE origin and expansion into Europe. The novelty is that the authors take into account Alinei's Paleolithic Paradigm  as one of the possible scenarios. Putting together linguistic, archaeological and genetic data, they reach the conclusion that the distribution of these toponyms correlates basically with two main events: the Mesolithic population expansion from the Glacial refugia of southern Europe, and the expansion of agriculture in the Neolithic. Both events involve IE languages. This is important. If the Mesolithic populations that migrated north were already carrying IE languages with them, then  those languages were there already in the Paleolithic. In order words, the Paleolithic Continuity Paradigm (PCP).

Of course, some may think: "Ok, there were IE language in Europe at that early age, but then there was another wave of IE dispersal at the bronze age which brought the IE languages as we know them today and historically". The authors admit this possibility, but also say that it is quite unlikely. As they say, and as I have insisted in this blog many times, there is no evidence of any sort of relevant population movement in the Bronze Age that could even remotely support this theory, usually known as the Kurgan theory.

As I said, I'll publish more posts getting into the details of this important research work. For example, I'll talk about their criticism of some aspects of traditional IE reconstruction, e.g. the reconstruction of PIE phonology. Let's see some excerpts (the highlighting is mine):

(p. 724-725): "Ciertas líneas de investigación han tendido a limitar el sistema vocálico indoeuropeo a dos vocales /e/ y /o/ e incluso a una sola (...). Tal reconstrucción, que no vamos a criticar aquí en detalle, desemboca en sistemas vocálicos irreales, inexistentes en las lenguas del mundo, sea cual sea la familia lingüística en la que busquemos. El testimonio de los arqueo-hidro-topónimos lleva la reconstrucción profunda del vocalismo indoeuropeo por derroteros muy diferentes. En las series vocálicas de nuestras arqueo-raíces la /e/ y la /o/ se manifiestan como variantes triviales y en parte locales de las respectivas formas básicas /i/, /u/ y /a/ (...). De ese modo, el sistema vocálico que se dibuja en el estadio cero es de tres miembros (a, i, u)".

(p. 726): "al pretender, como se ha hecho tradicionalmente, explicar la supuesta lengua común como un sistema cerrado en sí mismo, sin un origen y un devenir, se ha incurrido en simplificaciones, distorsiones e invenciones tendentes a buscar regularidades artificiales en terreno de la fonética, la morfología y la semántica".

The authors use cautious language, but this is actually a complete demolition of the many aspects of traditional PIE reconstruction, including laryngeals and other inventions.

NOTE (Apr 22, 2012): I have translated the quotes into English. See comments.

References:
- VILLAR, Francisco (2000). Indoeuropeos y No-Indoeuropeos en la Hispania Prerromana. Salamanca, Ediciones Universidad de Salamanca.
- VILLAR, F., B. PRÓSPER, C. JORDÁN, and  M.P. FERNÁNDEZ ÁLVAREZ (2011). Lenguas, genes y culturas en la prehistoria de Europa y Asia suroccidental. Salamanca, Ediciones Universidad de Salamanca.

Witold Manczak: criticism of PIE laryngeals

It is true that the Laryngeal Theory of Proto-Indo-European is widely accepted nowadays, but with different degrees of 'faith'. Many IE linguists have expressed their doubts about some aspects of the theory and in many cases (notably Oswald Szemerényi) only accepted a weak version, with just one laryngeal sound. But of course, there are other linguists who seem to be more enthusiastic about their h1's, h2's and h3's, as we saw in this post, with nice examples like the reconstructed word for 'two PIE widows' (nom. dual). In a book by Mallory and Adams (2006), I have found a really beautiful set of laryngeals. There are nine of them (see picture below), and it's not just the normal h's with numbers, but also with little letters (x and a) and even some mysterious combinations of numbers:

Does it make sense to invent a whole set of imaginary phonemes just for the sake of reconstruction? Is it justified? There are some linguists who have noticed some of the important inconsistencies in PIE Laryngeal Theory, and in some cases are completely against it. It's not easy to find their articles, as they are generally ignored by the IE linguistics establishment. And don't try to find much about them in Wikipedia or other Internet sources, they are just neglected. One of these authors is the eminent Polish linguist Witold Manczak, who has written a series of articles with strong criticism, actually a refutation, of the Laryngeal Theory. I have recently read one of these articles (Manczak, 2006), which has a significant title: Invraisemblance de la théorie des laryngales (=The Unlikelihood of the Laryngeal Theory). As we can see in the initial remarks, the author is quite aware of the difficulties of trying to raise a critical voice in IE studies (p. 25): "Nos articles ayant passés sous silence, il nous est venu à l'esprit de présenter nos arguments dans une revue beaucoup plus connu".

And indeed, he has some arguments. First, he starts by reviewing the process that led to the invention of the theory. Let's remember it briefly: by the end of the 19th c. the Swiss linguist Ferdinand de Saussure proposed the idea that PIE had only one vowel (/e/) and a series of 'coefficients sonantiques' that could influence this vowel. Later on, the Danish linguist Hermann Möller introduced the concept of IE laryngeals, which he saw as arising from the supposed kinship between IE and Afro-Asiatic. The final impulse for the Laryngeal Theory came with the discovery oh Hittite, an IE language which showed a 'laryngeal' sound. In 1927, the Polish linguist Jerzy Kurylowicz combined Saussure's 'coefficients sonantiques' with the evidence from Hittite to produce the first version of the Laryngeal Theory.

According to Mańczak, the first problem is that Saussure's proposal is untenable, for various important reasons, among them the absurdity of proposing a language with just one vowel (/e/), when in fact the most common pattern in the world's languages is, at least /i a u/. (p. 26): "l'explication de ces alternances à l'aide des coefficients sonantiques est fausse. (...) les coefficients sonantiques n'ont existé que dans l'imagination de Saussure". This would be enough to invalidate the whole edifice of PIE laryngeals, which was based on Saussure's assumptions. But then Manczak goes on to analyse some further details of the theory, for example the fact that there are so many different versions of the set of laryngeals: (p. 29) "les laryngales n'existent que dans l'imagination de certains linguistes, la théorie des laryngales est un domaine où tout est permis, où rien ne freine la fantasie des chercheurs, où la notion de rigueur scientifique est inconnue". He quotes several authors who have expressed similar opinions, e.g . Hiersche (1964: 11): "die Laryngale sind in der Lage, bainahe jede Lautveränderung hervorzurufen oder selbst zu erfahren, was in der allgemeine Phonetik nicht seinesgleichen hat". The laryngeals were, and still are, the perfect solution to solve any possible PIE reconstruction mystery that could not be solved by other means, even if it's necessary to propose quite abnormal things, like consonants turning into vowels and other unlikely events.

In the final part of the article, Manczak asks himself why it is that the Laryngeal Theory has been so successful among linguists. According to him, there is a general lack of validity criteria in historical linguistics. (p. 31): "le terme "critères de verité" n'est jamais employé par les linguistes, bien que les linguistes soient unanimes pour dire que la linguistique est une science". The important thing is the 'authority' behind the theory, not the validity of the theory itself. (p. 32): "Comme les linguistes croient en l'infaillibilité des autorités, ils détestent ceux qui osent critiquer les autorités et adorent ceux qui approuvent ou développent les idées des autorités".

I find Manczak's proposals quite interesting, and I think anyone doing research in the field of IE studies should take them into account, instead of assuming the Laryngeal Theory as indisputable truth.

References:
- Hiersche, R. (1964). Untersuchungen zur Frage der Tenues aspiratea im Indogermanishen. Wiesbaden.
- Mallory, J. P. and D.Q. Adams (2006). The Oxford Introduction to PIE and the PIE World. Oxford University Press.
- Manczak, Witold (2006). «Invraisemblance de la théorie des laryngales». In Historische Sprachforschung, 119: 25-34.

Two proto-Indo-European widows

Reconstructed proto-Indo-European (PIE) forms, especially the ones produced in the context of the laryngealist tradition, have something in common: they all look quite bizarre, with many consonants followed by little superindexed w's or h's and plenty of m's and n's with vocalic value, not to mention the most bizarre of all, a set of three (sometimes more) phonemes called laryngeals (h1, h2, h3) that no-one really knows how to pronounce. I remember reading an article by Xaverio Ballester in which he talked about IE horses. In his opinion, the reconstructed word for 'horse' was so difficult to pronounce that it would have been easier for a PIE speaker to imitate a horse's whinny! I have recently come across a book with a nice selection of PIE reconstructions. It is a book by the eminent American linguist Don Ringe, a firm believer in laryngeals (Ringe 2006). Let's see some examples from his book:

This word is supposed to mean 'name' in PIE. Here, we find a couple of laryngeals and a final n with a vocalic component, written as a little circle below the consonant (sorry, my word editor does not include this symbol). According to Ringe (2006: 9): "There seems to have been very few constraints on the distribution of (...) the laryngeals (...) *h2 was perhaps the second most common [obstruent] in a lexical count". Constraints on laryngeals? What for? They can appear everywhere and they can be nearly anything, from vowels to consonants and also semi-vowels and semi-consonants. They are like jokers in a card-game.

This string of letters is supposed to be the word for 'widow' in dual nom., voc. and acc. The meaning, therefore, would be 'two widows'. Let's imagine a speaker of PIE, in prehistory, pointing at two elderly ladies and calling them that... It's hard to imagine. One peculiarity of this word is that it has two laryngeals in a row! What are they? Two vowels? One vowel and a consonant? One consonant and a vowel? What about the combination 'dhw'? Does anybody really know how this was pronounced? Was there ever anyone who used this word?

Finally, the reconstructed word for 'tongue':

... A real tongue-twister!

In the past few months I have written a series of posts about PIE laryngeals, showing some more examples of how bizarre traditional PIE reconstructions can be. I have also expressed these ideas in some Internet forums and blogs, e.g. Language Log. PIE laryngeals are only one of the many aspects of traditional comparative linguistics that I criticise. Using Angela Marcantonio's terminology, laryngeals might just be 'artefacts of the linguistic method of analysis', i.e. produced by a type of methodology based on language trees, linguistic paleontology and other unacceptable or at least dubious ideas.

Angela Marcantonio is an expert in Uralic languages. In a series of interesting articles and books (e.g. Marcantonio 2002) she has reviewed some of the traditionally held views on the Uralic family, reaching the conclusion that a Uralic proto-language cannot be reconstructed scientifically. She has also carried out research on other language groups. In a recent article, "Evidence that most Indo-European lexical reconstructions are artefacts of the linguistic method of analysis" (in Marcantonio ed. 2009), she analyses some traditionally held assumptions about PIE, with striking results.

First, she focuses on some of the laws that have been proposed for PIE, and finds that in many cases they might be examples of circular reasoning. To test her hypothesis, she applies a quantitative test to a set of reconstrucetd PIE words (the verbal roots in Rix's dictionary, 1998). The results are as follows: 66% of the recontructed verbs are based on words found in only one or two of the IE branches; only 34 % are attested in three or more branches. On the other hand, it is supposed that the laws governing phonetic change in IE, e.g. Grimm's Law, should be a useful tool to determine these reconstructions. However, these laws are usually modified with a series of secondary laws or refinements, so that there is always some kind of intricately designed new parameter to explain any apparent deviation from the norm. Marcantonio has clearly shown that, when you have a PIE verbal root with forms attested in many IE branches, a high number of laws is needed to account for the whole set. In some cases, the number of rules equals the number of forms. This is how the corpus of PIE reconstructions has grown in the last 150 years: by a cumulative amount of laws, many of them designed 'ad hoc'. What is the use of a law, e.g. Grimm's Law, if it is immediately followed by new laws, e.g. Verner's, to make it tenable? Marcantonio sees the adjustable parameters of PIE laws as an indication of circularity.

She also analyses some particular aspects of traditional PIE reconstruction, e.g. the present-perfect alternation. Again, she considers them an example of fabrication of the method of analysis.

Finally, she focuses on the laryngeal set. In her quantitaive analysis, she notices that a high percentage of PIE verb roots and morphemes have been reconstructed with the aid of laryngeals. In her opinion, the use of laryngeals adds some extra flexibility to the system: they are phonetically unspecified and can appear virtually anywhere in a word; they are the perfect solution for any reconstruction, a real magical wand that makes any prediction possible. Marcantonio does not reject the laryngeal theory (she says that this issue is beyond the reach of her present research) but I have the feeling that she is quite convinced that the set of laryngeals is, basically, one more artefact in the hands of those who believe in proto-languages.

So, what is left of PIE if we subtract the poorly attested roots (those found in less than three IE branches) and if we eliminate artificial artefacts like vowel gradation and laryngeals, based on circular reasoning? Maybe PIE is reduced to just a relatively small set of cognate words and an even smaller set of grammatical elements, if any. Maybe it's about time IE scholars started to abandon the idea of perfect proto-languages with their complete sets of phonemes, declensions and conjugations and equipped with immaculate laws that would predict every step from one proto-language to another. Maybe it's time the whole idea of a proto-form with an asterisk were replaced with a different type of notation, one that reflected the complex nature of language rather than an aspiration to immaculate, circularly-proven forms.

Bibliography:
- MARCANTONIO, A. (2002). The Uralic Language Family: Facts, Myths and Statistics. Transcations of the Philological Society. Oxford/Boston, Blackwell.
- MARCANTONIO, A., ed. (2009). The Indo-European Language Family: Questions about its Status. Journal of Indo-European Studies Monograph Series, No 55. New York, Institute for the Study of Man.
- RINGE, D. (2006). From Proto-Indo-European to Proto-Germanic. Oxford, OUP.

The Laryngeal Theory revisited

A recent article in Language Log, written as a guest author by the American linguist Donald Ringe, has triggered an interesting discussion about historical linguistics and Indo-European (IE) studies, in which I have also participated. You can read the post and the comments here: The Linguistic Diversity of Aboriginal Europe.

One of the discussion topics was the Laryngeal Theory, a stronghold of traditional IE studies. In a previous post in this blog (Laryngealists) I already gave my opinion about this theory, so it’s not necessary to repeat the same arguments again. In the recent discussion at Language Log, however, I added some more information. For example, the fact that there are some linguists who have never accepted the Laryngeal Theory. Among them, Giuliano Bonfante and Witold Manczak. In the PCT workgroup we have Ballester and Alinei.

As you can see, the IE Laryngeal Theory is not as generally accepted as some people would like to believe. The surprising thing is that it is still taken seriously by many linguists.

This discussion reminded me of an article by Xaverio Ballester that I included as reference material in my first post about the Laryngeal Theory. This is the full citation: Ballester, Xaverio, "/a/ y el Vocalismo Indoeuropeo", Alessandria 1 (2006) 3–37. In this article, Ballester criticizes many aspects of the various reconstructed phonemic systems that have been proposed for IE, especially the ones that include laryngeals. In his opinion, these ‘systems’ are rather artificial and do not correspond with the distribution of phonemes that are usually found in most human languages. They look more like the invention of some linguists in order to justify the ‘structure’ of their proposals. The most striking example is the fact that the phoneme /a/ is not present in many of the proposed IE vowel sets. It seems that this basic phoneme was a problem for some theoretical linguists, a real obstacle for the purity of their method, so they preferred to devise an a-less model. Ballester’s article provides all kinds of evidence to prove the existence of /a/ in PIE and the absurdity of the Laryngeal Theory and other proposals which were born in the context of structuralism in linguistics.

Finally, a very simple experiment, for which we will use Wikipedia (I’m not very fond of Wiki but it’ll be useful for our purposes). Take a look at the article about Indo-European Sound Laws: Vowels and Syllabic Consonants. In the first column on the left we have the “old reconstruction”, that is the one that was proposed at the beginning of IE studies by linguists such as Franz Bopp (1791-1867; portrait on the left). As you can see, the letters in this column look familiar and simple. In the second column we have what is described as “New reconstruction”. Take a look at the symbols, and the combinations of vowels and laryngeals. What do you think of them? I agree with Ballester when he says that these things do not represent any human language at all. They are mere theoretical speculations.

Last Edit: 31 March 2009
Reason for edit: the inclusion of a link to Ballester's article.

Laryngealists

The following image shows a traditional example of IE (Indo-European) word reconstruction. In this case, some words from Latin (an IE language) are explained by means of the reconstructed PIE (Proto-Indo-European) forms:

( In: Philip Baldi, 1993, The Foundations of Latin; Mouton de Gruyter; p. 244).

Click on the image if you want to see it in a larger size.

What do we have here? Basically, the author is trying to apply a rule governing the change from an IE root (in this case *h3e) into its Latin equivalent (/o/). The validity of this supposed rule is confirmed by a series of examples and by comparing them to words belonging to other IE languages. Let’s take a closer look. Some of you may be wondering what these strange phonemes (h1, h2, h3) can be, for example in the reconstructed PIE form for number 8: *h3ektoh1. What is this, the formula of some kind of hydrocarbon?... No, what we have here is an example of PIE laryngeals. There are three of them (h1, h2, h3) and they are used in traditional PIE reconstruction as a means of justifying the set of systematic rules which is essential in the traditional view. Ferdinand de Saussure was the first linguist who ‘predicted’ their existence (they were necessary for the perfection of mainstream IE theories) and it was some decades later, at the beginning of the 20th century, with the discovery of Hittite, that ‘evidence’ for the laryngeal theory was found at last. That was the last piece of the jigsaw puzzle. Now the whole structure of PIE seemed perfect, based on systematic rules which governed how original PIE evolved and divided into the different IE languages. Everything in order, the theories are safe and sound! But... What are these laryngeals? In fact, nobody really knows how they were pronounced, and it is quite difficult to imagine their possible articulation. Let’s take the reconstructed word for number 8 (*h3ektoh1). Let’s go back in history and let’s imagine one of our ancestors pronouncing something like that, with one laryngeal at the beginning and another one at the end... It’s hard to imagine... Did the original speakers of PIE really have those strange phonemes in their speech? Or is the existence of laryngeals simply an invention of traditional linguists to justify their theories? These linguists’ main concern was (and still is) to find systematic rules. They assume there was an original language (PIE) which at approximately 4,000 BC started to spread around Asia and Africa, at the same time splitting into different languages. No matter how strange or illogical these rules are, the important thing is to try to find the 'perfect' explanation.- But this is all quite wrong, as the Continuity Theory has proved in the last few decades.

The Continuity Theory (CT) sees the origin of PIE at a much earlier date, at the time when Homo Sapiens migrated from Africa and started to populate the rest of the world. Therefore, the idea of reconstructing this initial language is nearly impossible. The CT tries to describe the facts, rather than organizing them in a systematic way. Languages don’t change following intrinsic, well-defined rules, but as a consequence of more general extralinguistic factors, especially one: hybridization, i.e. when speakers of different languages start to mix, or also by means of word diffusion. Rules are not the important thing any more. There’s no need to imagine our ancestors pronouncing intricate laryngeals (in fact, there’s no language in the world today with something similar to that set of laryngeals!).

Xaverio Ballester, a member of the CT workgroup, has variously criticized laryngeal theories. He has coined the term laringalista (= laryngealist) to refer to the linguists who use laryngeals in their IE reconstructions. Further reading: Xaverio Ballester: Indoeuropeización en el Paleolítico. Una Réplica, in Estudis Romànics 26 (2004) 217–32. By the same author: /a/ y el Vocalismo Indoeuropeo, Alessandria 1 (2006) 3–37; and Protofonología de las lenguas indoeuropeas, Mœnia 5 (1999) 173–87.