Saturday, 15 June 2013
Assimilation II - 'Introduction to Historical Linguistics' by Terry Crowley and Claire Bowern, Part VII
This is the seventh part in a series on historical linguistics, using Crowley and Bowern's Introduction to Historical Linguistics as a base. Part I is here. Part II, here. Part III, here. Part IV, here. Part V, here. Part VI, here.
So, last time we dealt with several aspects of assimilation, finishing on one of the most famous sound changes of them all, palatalisation. This time, we're going to look at assimilation-at-a-distance, final devoicing, and vowel and nasal harmony.
First, final devoicing, a common phenomenon best illustrated by changes in German. 'Devoicing' refers to the loss of voice - the change from a voiced consonant to a voiceless one - and 'final' refers, of course, to the end of a word. So the loss of voice at the end of a word. Modern German has devoiced final consonants as a regular phenomenon, and this is a good enough example that I can't not use it:
*ba:d > ba:t 'bath'
*ta:g > ta:k 'day'
*hund > hunt 'dog'
Why is this assimilation? Good question. It could easily be thought of as fortition, going backwards along the sonority hierarchy discussed in part II of this series. The voiced consonant becomes a voiceless one; that's fortition, a strengthening of the sound. But you'll remember that fortition is rare, and that lenition is a far more common type of sound change. Lenition tends to make things 'easier' to say in some way. So we need an explanation for why fortition like this occurs.
Strangely, final devoicing is an assimilatory change because the voiced consonant assimilates the voicelessness of the silence that follows it. It seems that humans accommodate silence into their phonemic systems quite readily and that its presence can have an effect on sound changes within words.
Anyway, onto assimilation at a distance. Assimilation at a distance is a sound change that occurs due to the presence of other sounds elsewhere in the word or phrase. Immediate assimilation is the inverse - that's like all of the examples above, and involves assimilation due to the sounds immediately next to the sound that changes. A good example, and one that Crowley uses, is a common pronunciation of orangutan in English ('orang hutan' in Malay - 'man of the forest'), where [n] at the end of the word becomes [ŋ], a velar nasal. This is to match the velar nasal in the first part of the world. Instead of [oraŋutan], we get [oraŋutaŋ]. It's assimilation even though the two velar nasals are not next to one another.
A fairly common kind of assimilation at a distance is harmony, which you'll encounter if you study Turkic or Mongolic languages in the form of vowel harmony. That's where affixes to words have to assimilate the front/back vowel in the root word. Front vowels are those like [i], [e], and so on, and back vowels are those like [a], [o], and so on. Front vowels sound higher and back vowels sound deeper and stronger. In Mongolic and Turkic languages, this is an important distinction, as both families are strongly agglutinative languages (see wiki) and rely on the attachment of affixes to create meaning. Those affixes have to obey vowel harmony. Crowley's example comes from Turkish - the plural suffix, in fact, which is -lar or ler depending on the front or back vowel in the root word:
tavuk - 'chicken'
tavuklar - 'chickens'
ev - 'house'
evler - 'houses'
You can also see this in Mongolian - for example, in -a:s and -e:s, a suffix meaning 'from':
ula:nba:tar - Ulaanbaatar, the Mongolian capital
ula:nba:tara:s - 'from Ulaanbaatar'
erdenet - Erdenet, Mongolia's second city
erdenete:s - 'from Erdenet'
That's vowel harmony. But there's also nasal harmony, where nasalisation can spread throughout a whole word. A good example is found in Yanomami, a large of the Yanomaman family, which has a very low number of speakers and is found in a very limited distribution in Venezuela. The Yanomami have been at the centre of a major academic controversy, one that still bubbles over now and then, and as a result the different transcriptions of the word 'Yanomami' have different supporters ('Yanoama', Yanomamo, etc). In fact, no one seems to pronounce it according to the correct orthography. It is entirely full of nasal vowels, due to a quirk of the language whereby the presence of a single nasal vowel in a word turns all of the other vowels of the word into nasal vowels. So, it should be pronounced something like [jãnõmãmɨ], according to this rule.
Crowley's example of nasal harmony comes from Enggano, a language of Sumatra. Here, plain vowels following nasal vowels become nasalised as well.
*honabu > honãmũ 'your wife'
Finally, we come to another kind of vowel harmony known as Umlaut. This is the name for a phenomenon common in the Germanic languages whereby back vowels become front vowels, and low vowels are raised, before a front vowel. That sounds complicated but isn't, and it helps account for some of the irregularities of English verbs and plurals.
For instance, the original word for 'foot' was [fo:t], and the plural was [fo:ti]. The [o:] in the plural became a front vowel, [ø:] due to the presence of the [i]. This was raised again to [e:] and the final [i] was dropped, giving us [fe:t] in the plural and [fo:t] in the singular, the forms preserved in English spelling ('feet' and 'foot' respectively).
There are many more examples of Umlaut in other Germanic languages, as well as in English.
Now we've covered the types of assimilation, we can then easily understand dissimilation, whereby sounds becomee less like one another. So we'll cover that next time, in addition to the adoption and changes in tone.