Glottalic theory
From Wikipedia, the free encyclopedia
The glottalic theory holds that Proto-Indo-European had ejective stops instead of voiced ones, namely p’ t’ k’ rather than b d g, contrary to the traditional reconstruction of the Proto-Indo-European sound system.
Further, it holds that p’ was absent or very rare, so the glottalized series is sometimes written (p’) t’ k’.
The glottalic theory was developed independently in the United States by Paul Hopper and in the Soviet Union by Thomas V. Gamkrelidze and Vyacheslav V. Ivanov. In its earliest version, proposed by the Danish linguist Holger Pedersen, it did not involve the use of glottalized sounds. While earlier linguists, such as André Martinet and Morris Swadesh, had seen the potential of substituting glottalic sounds for the supposed plain voiced stops of Proto-Indo-European, the proposal remained speculative until substantial evidence for it was simultaneously published in 1973 by Hopper in the journal Glossa and by Gamkrelidze and Ivanov in the journal Phonetica.
Contents |
[edit] The Indo-European stop series
The traditional reconstruction of Indo-European includes the following stops:
| CONSONANTS | labials | dentals | palatalized velars | velars | labialized velars |
|---|---|---|---|---|---|
| voiceless stops | p | t | kʲ | k | kʷ |
| voiced stops | (b) | d | gʲ | g | gʷ |
| breathy voiced stops | bʱ | dʱ | gʲʱ | gʱ | gʷʱ |
/b/ is bracketed because it is at best very rare and perhaps nonexistent.
There are several problems with this reconstruction. From a typological point of view, if a voiced stop is missing from a phoneme inventory, it would normally be /g/ that is missing, not /b/; on the other hand, if a voiceless stop is missing, the labial /p/ would be the most likely candidate. (This is close to universal with ejective /p’/.) Furthermore, there are few languages which have breathy voice consonants but no voiceless aspirates. Finally, Proto-Indo-European did not permit a root to begin and end with a voiced stop, i.e. there are no such roots as *deg or *ged; this is typologically very odd again.
This inventory was not introduced as an independant proposal, but instead arose as a modification of an earlier (typologically more plausible) theory. In the original Proto-Indo-European proposal, there was a fourth phonation series, aspirated /pʰ, tʰ, kʲʰ, kʰ, kʷʰ/, assumed to exist by analogy with Sanskrit, which at the time was thought to be the most conservative Indo-European language. However, it was later realized that this series was unnecessary and was generally the result of a sequence of a tenuis stop such as /t/ and a laryngeal such as /h/. The aspirate series was removed, but the breathy voiced consonants remained.
The glottalic theory proposes a different phonetic inventory for Proto-Indo-European:
| CONSONANTS | labials | dentals | velars | labialized velars |
|---|---|---|---|---|
| voiceless stops | p | t | k | kʷ |
| ejective or glottalized stops | (p’) | t’ | k’ | kʷ’ |
| voiced stops | b | d | g | gʷ |
Such a system is common among the world's languages. Moreover, the revised system explains a number of phonological peculiarities in the reconstructed system.
Hopper (1973) also proposed that the aspiration that had been assumed for the voiced stops bh, dh, gh could be accounted for by a low-level phonetic feature known to phoneticians as "breathy voice." This proposal made it possible both to establish a system in which there was only one voiced stop and at the same time to explain developments in later Indo-European dialects (Greek, Latin, and Sanskrit) that pointed to some kind of aspiration in the voiced series. In their 1973 article, Gamkrelidze and Ivanov posited aspiration in both voiced and voiceless stops.
In addition to motivating the absence of a labial plain voiced stop *b in the proto-language, the Glottalic Theory illuminated a longstanding but unexplained observation of Indo-Europeanists about the distribution of consonants in word roots. It had long been noted that certain combinations of consonants were not represented in Proto-Indo-European words. In terms of the traditional system, these were:
1. No root contained a sequence of two plain voiced stops, that is, in schematic terms, there were no roots of the type *deg.
2. No root contained both a voiceless stop and a voiced aspirate, that is, roots of the type *dhek or *tegh were not attested.
3. On the other hand, the plain voiced stops were compatible with either of the other two series: *degh or *dek were both possible.
These constraints on the phonological structure of the root cannot be explained in terms of a theory of assimilation or dissimilation, since they display a radical difference in patterning between two sets of consonants — the voiced stops — that ought to behave identically. The Glottalic Theory provides a completely coherent explanation (Hopper 1973):
1. In very many languages that have glottalic consonants, there is a constraint against two such consonants in the same root. This constraint has been found in many languages of Africa, the Americas, and the Caucasus.
2. If the "plain voiced stops" were not voiced, then the "voiced aspirated stops" were the only voiced stops. The second constraint can accordingly be reformulated as: Two nonglottalic stops must agree in voicing.
3. Since the glottalic stops were outside the voiced/voiceless opposition, they were immune from the constraint on voicing agreement in (2).
Gamkrelidze and Ivanov (1973, 1995:5-70) posited that the two non-ejective series (traditional *p *t *k and *bh *dh *gh) were fundamentally aspirated (that is, *ph *th *kh and *bh *dh *gh) but had non-aspirated allophones (that is, *p *t *k and *b *d *g). According to them, the non-aspirated forms occurred in roots where two non-ejectives were present because of a rule that prohibited more than one aspirate in the same root. To express the variability of aspiration Gamkrelidze and Ivanov write it with a superscripted h, for example dʰ. Thus an Indo-European DʰeDʰ (where Dʰ represents any non-ejective stop) might be realized as DeDʰ (attested by Indic and Greek) or as DʰeD (attested by Italic). In contrast, traditional theory would trace a form attested as both DeDh and DheD to an Indo-European DheDh. The advantage of Gamkrelidze and Ivanov's interpretation is that it eliminates a very unusual or even unique feature of the Indo-European stop system, since according to Roman Jakobson all languages that have voiced aspirates also have voiceless aspirates. By identifying the voiceless non-aspirates of the traditional stop system (*p *t *k) as voiceless aspirates (*pʰ *tʰ *kʰ), Gamkrelidze and Ivanov restored the missing series.
One objection to this reconstruction is that the voiced consonants are frequently voiceless in the daughter languages; aspirates in Greek and voiceless fricatives in Latin, for example. While it is common for aspirates to become tenuis and then voiced, as pʰ → p → b (lenition), the reverse is rare. Thus more recent versions of this hypothesis do not have voiced consonants at all, or treat voicing as non-distinctive. Such an inventory is:
| CONSONANTS | labials | dentals | velars | uvulars | labialized velars |
|---|---|---|---|---|---|
| voiceless stops | p | t | k | q | kʷ |
| ejective or glottalized stops | (p’) | t’ | k’ | q’ | kʷ’ |
| aspirated stops | pʰ | tʰ | kʰ | qʰ | kʷʰ |
(Here the traditional palatalized vs. plain velar dichotomy is treated as a velar-uvular contrast, as posited by Hopper 1981. This is not required for the glottalic theory, and may have been allophonic at an early stage in the proto-language.)
[edit] Decem and Taihun
In 1981 Hopper proposed to divide all Indo-European languages into Decem and Taihun groups, according to the pronunciation of the numeral 10, by analogy with the Centum-Satem isogloss, which is based on the pronunciation of the numeral 100. The Armenian, Germanic, Anatolian and Tocharian subfamilies belong to the Taihun group because the numeral 10 begins from the voiceless t there. All other Indo-European languages belong to the Decem group because the numeral 10 begins from the voiced d in them.
[edit] Objections
The primary objection to the glottalic theory is the difficulty in explaining how the sound systems of the attested dialects were derived from a parent language in the above form. If the parent language had a typologically unusual system, like the traditional p-b-bh, then it might be expected to collapse into more typical systems, possibly with different solutions in the various daughter languages, which is what one finds. For example, Indo-Iranian added an unvoiced aspirate series, gaining an element of symmetry; Greek and Italic devoiced the murmured series to a more common aspirate series; Balto-Slavic deaspirated the murmured series to modal voice; and Germanic and Armenian chain-shifted all three series. In each case, the attested system represents a change that could be expected from the proposed parent.
Now if the system were typologically common, as proposed by the glottalic theory, then it might be expected to be stable and therefore to have been preserved in at least some of the daughter languages, which is not the case: no daughter language preserves ejective sounds where the glottalic theory postulates them. However, if Proto-Indo-European did not have true ejectives but rather some less stable kind of glottalic consonant, their loss would be more understandable. However, even "stable" systems change, and that an objection based on what "should" have happened cannot really overturn a serious and otherwise well-motivated reconstruction. In all reconstructions of phonological systems one proceeds by comparing the evidence of the daughter languages and projecting them back to a common proto-form, not by first declaring this or that change to be a priori implausible.
Opponents of the glottalic theory have objected that it is not based on any direct evidence. Although murmured consonants are uncommon, they are at least directly attested in the Indo-Aryan languages (which is, of course, why they were postulated in the first place). Roman Jakobson's assertion that no language is known which has murmured consonants unless it also has voiceless aspirates is disputed by some linguists who oppose the glottalic theory; for example, Robert Blust showed that a system of voiceless, voiced and murmured stops, as postulated in the traditional reconstruction of Proto-Indo-European, exists in Kelabit, a language of the Sarawak highlands in Borneo. But unlike the revised system of Indo-European stops, the traditional system is at best a typological rarity. As to the alleged lack of "direct evidence," the reconstruction of proto-languages is often based on indirect indications. Even in the traditional reconstruction of Proto-Indo-European, for example, there is no direct evidence for a voiced aspirated labiovelar stop gʷʱ.
Some have assumed that the glottalic theory represents an earlier stage in the history of Proto-Indo-European, which developed into the traditional system in later Proto-Indo-European. This would explain both the root restrictions in Proto-Indo-European and the universal loss of glottalic consonants in the daughter languages, but would leave us with a proto-language phonological system identical to the one that has been criticized, and also assumes a long period of internal evolution within which Proto-Indo-European would have been otherwise uniform before branching out into the daughter languages.
Although controversial at first, some variant of the glottalic theory is generally accepted today. The reason is that it neatly resolves a number of problems that it was not designed to solve, in effect giving it some empirical support. For example, in both Latin (Lachmann's law) and Balto-Slavic (Winter's law), vowels are lengthened before a "voiced" consonant. This had always been somewhat puzzling. It is the same behavior that vowels exhibit before laryngeals, which are assumed to include a glottal stop. It may be that the glottalic consonants were preglottalized, or that they were ejectives that became preglottalized in Italic and Balto-Slavic before losing their glottalization and becoming voiced. It is very common in the world's languages for glottal stops to drop and lengthen preceding vowels. In Quileute, for example, VC’V, VʔC’V, and VːC’V (such as ak’a ~ a’k’a ~ āk’a) are allophones in free variation.
[edit] Sources
- Paul J. Hopper, "Glottalized and murmured occlusives in Indo-European." Glossa 7:2:1973, 141-166.
- Thomas V. Gamkrelidze and Vjacheslav V. Ivanov, Indo-European and the Indo-Europeans, translated by Johanna Nichols, 2 volumes. Berlin and New York: Mouton de Gruyter, 1995.
- Robert S.P. Beekes, Comparative Indo-European Linguistics. John Benjamins, 1995.
- Anthony Fox, Linguistic Reconstruction. Oxford, 1995.
