Typological features and tracing language history

Jul 26, 2010 by

Generally speaking, historical linguistics tries to elucidate patterns of language change and to determine language relatedness. Most of this work, starting with the work of earliest philologists such as James Parsons and William Jones, focuses on the so-called cognates — words that come from the same origin. Since Franz Bopp’s work in the 19th century, patterns of grammatical similarity, such as patterns of inflectional morphology, are also taken into account. But such work can only trace language relatedness 6,000-10,000 years. And that’s in the best case scenario, for languages and language families whose past is known from written documents. But can linguistics reveal longer time depths?

This question is at the crux of a recent study by Greenhill et al. [citation below]

Here is the problem: language is constantly changing, both in the vocabulary and in grammar. And those changes are not happening along a specific trajectory. Practically, at any “fork in the road” language can take one direction or another. Sometimes this even happens with parts of a given language: take a language L and with time some of the speakers of L will adopt one type of change and another group will adopt a different change. This is indeed how dialects and languages arise. Take, for example, the disappearance of the long /a:/ in English as a result of the Great Vowel Shift (I will discuss this process in more detail in a future posting). As a result of a number of smaller shifts, the long /a:/ of Middle English turned into a diphthong /ej/, so that the former pronunciation /na:me/ turned into the modern /nejm/ (‘name’). But the lack of (long) /a/ is a very unusual situation for a language, so different dialects of English filled the gap, but in different ways. In some dialects, /l/ was deleted and the preceding short /a/ lengthened in such words as calm, palm, half, calf. In other dialects, the low back rounded vowel was unrounded, lowered and lengthened in such words as not, pot, hot, Don, and in yet other dialects, /r/ was deleted with the preceding short /a/ lengthened in such words as park, car, far, dark. Looking back it’s impossible to say how each dialect/language would approach any given “fork in the road” of language change. That’s why reconstruction methodology can only go so far, and written documents from past epochs help enormously.

So is there a way to turn a linguistic microscope with fine resolution for 6,000-10,000 years into a telescope that can “see” perhaps 50,000 years into the past? Greenhill et al. thought so.

There idea is to look for those features of language that are more stable than others, that is change less rapidly and are therefore better indicators of the language past. They write:

“If some typological features are consistently stable within language families, and resistant to borrowing, then they might hold the key to uncovering relationships at far deeper levels than previously possible.”

They applied network methodology based on biological sciences to visualize the relatedness among languages examined and came up with the following chart:

This network does group some of the languages into known families, as shown in Indo-European, Altaic and Nakh-Daghestanian. In other instances, however, the language families were not recovered — including, Sino-Tibetan, Uralic, and Trans-New Guinea. They also note that there are substantial number of conflicting signals (box-like structures), leading to an inaccurate recovery of many well-attested phylogenetic relationships within major language families. For instance, the network links German to French, when in fact German is more closely related to English.

But most importantly, Greenhill et al. failed to find any typological features that evolve at consistently slower rates than the basic lexicon or other typological features. So is it the case that, as one blogger puts it, “the current study highlights how little linguists know about the shape and tempo of language change”? Or is the problem in something else?

I think that a larger problem is that, despite appearance to the contrary, we do not have a good, exhaustive list of typological grammatical features to consider. Sure enough, the World Atlas of Linguistic Structures (WALS) lists over 140 typological features across 2561 languages. But it is not true that these features are truly the attributes that define cross-linguistic variation. Take such a well-established typological feature as the order of S(ubject), O(bject) and V(erb). Note first that alongside this feature (their #81) WALS lists two related features: the order of Subject and Verb and the order of Object and Verb (features # 82 and 83, respectively). Moreover, even though WALS presents the order of S, O and V as a feature of 1228 languages, its validity must be questioned for at least three types of languages. First, in languages with Philippine-style topic marking (Tagalog, Malagasy, etc.), one of two elements may be considered the subject: the semantic subject (aka the Agent) or the grammatical subject (aka the topic). Second, in many syntactically ergative languages the word order is determined not by whether a given element is a subject or an object, but by its case marking (ergative vs. absolutive). And third, in non-configurational languages, such as Walpiri, the predominant word order of S, O and V cannot be determined at all.

The same can be said of many other features listed in WALS, and unsurprisingly, its authors redefined some previously well-established typological features. For example, WALS authors Balthasar Bickel and Johanna Nichols chose to redefine the morphological type (isolating, agglutinative, fusional, introflexive) as a combination of three separate typological features: phonological fusion, formative exponence, and flexivity (i.e. allomorphy, inflectional classes).

Naturally, having a different set of typological features to work with would result in a different relatedness network. So recovering those typological features that truly underpin cross-linguistic variation and change is a high priority. And this is exactly what the proponents of the Parametric Theory of language are doing. More on this in future postings.

Citation: Greenhill SJ, Atkinson QD, Meade A, & Gray RD (2010). The shape and tempo of language evolution. Proceedings. Biological sciences / The Royal Society

Related Posts

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below: