|
|
1. Language diversification: The map below compares the way people in what are now Morocco, Egypt, and Saudi Arabia may have said the numbers 'two' through 'four' prior to about the year 600 and how they say them now. We don't know exactly how the people in what is now Morocco and Egypt would have said these words, but the former would have been speaking a Berber language and the latter would have been speaking a later variety of Egyptian as spoken by the Pharoahs. People in what is now Saudi Arabia would have been speaking something like what we now call "Classical Arabic".
The year 622 marks the beginning of Islam, which soon began to expand from its Arabian homeland across North Africa and around the Middle East. With this in mind, explain differences between the "then" and "now" numbers in the three areas above and the differences between the "now" numbers in the three areas.
There are two kinds of differences between the "then"and "now" of Morocco and Egypt:
- Different languages: The original languages of modern Morocco and Egypt have been replaced by Arabic. Actually, Berber languages are still spoken in Morocco, though virtually everyone in Morocco also speaks Arabic, and many people speak only Arabic. The last form of spoken Egyptian was Coptic, which died out around the 10th century except as a liturgical language in the Coptic Church. In most of Egypt, the only natively spoken language is Arabic.
- Varieties of Arabic: The varieties of Arabic in Morocco, Egypt, and Saudi Arabia are different from each other--Moroccan Arabic is actually a different language from the other two--and they all differ from Classical Arabic. Reason: Once people in the various regions adopted Arabic, innovations began to develop in each region, but the innovations differed from each other. After several centuries of different accumulated innovations, the varieties of Arabic became differentiated not only from the original Arabic, but from each other.
2. Innate vs. learned properties: The table below shows a few differences between one variety of British English and one variety of American English. How do these represent changes in learned aspects of language? Give some examples of changes that we would predict could never develop between varieties of English.
Learned aspects
(aspects which could change)Impossible innovations
(prevented by innate constraints)Pronunciation The dialects differ in the pronunciation of individual words, clearly something learned from exposure in a particular community. Note, however, that both dialects draw from the same set of sounds to form their words, e.g. "sh" exists in both dialects, but only British E. happens to use it in 'schedule'. No dialect of English would substitute a hand clap for the beginning sound of 'schedule'. An innate property of ALL spoken human languages is to use only sounds produced by certain parts of the upper respiratory tract, which have evolved as part of the evolution of the language ability. Vocabulary Because the relationship between sound and meaning is arbitrary, the words used to refer to particular concepts must be a learned feature of language. The reason that British and American English are considered to be the "same" language is that the vast majority of "word/ concept" pairings are the same in the two dialects, but there are some differences such as those here. It is unlikely that some dialect of English would use a different name for an "elevator/ lift" when it was going up from when it was going down. It seems to be an innate property of vocabulary formation that things which occupy a certain space and behave as a unit get unit words. Transitory states of such entities are generally igorned. Grammar Grammatical differences across English dialects turn out to be few and minor. In the case of "...might do" vs. "...might", the learned feature seems to have to do with whether an understood word can be omitted or not, a common difference across languages. Some languages allow omission of pronouns, which, by definition, refer to understood elements, e.g. to answer, "Did you buy the book?", many languages would allow the answer, "Yes, I bought," omitting the counterpart of English "it". No English dialect would innovate by substituting a gesture for every verb, saying only the non-verbs orally. An innate property of every language is that sentences have internal structure independent of their meaning. Gestures substituted for verbs might well be clear in meaning, but language is a system that has properties beyond just the meanings that sentences might convey.
QUESTIONS 3-5 HAVE BEEN CHANGED SINCE THESE ANSWERS WERE ORIGINALLY POSTED. YOU CAN DOWNLOAD A PDF FILE WITH THE ANSWERS TO THOSE QUESTIONS AS WELL AS THE REST OF THE QUESTIONS:
Revised answer key to "Diversification" discussion questions.
6. Mass Comparison as a method of discovering relationship between languages: Use the vocabulary items to assign the languages to genetic groups.
In each line, the words in the different colors look enough alike that we might reasonably suggest that they come from a single ancestral word. For example, in the word for 'two',
Farsi do, Hindi do are identical Uzbek ikki, Turkish iki differ only in one vs. two k's Telegu rendu, Tamil irandu share r, n, d in that order Moreover, if we look at the whole table, the color coding shows that more often than not the words that resemble each other are in the same sets of languages (the red question marks in the cells for some items suggest that maybe those words could be coded as grouping with the words in red). In fact, linguists have classified these languages as follows:
Farsi, Hindi Indo-Iranian branch of the Indo-European family Uzbek, Turkish Turkic branch of the Altaic family Telugu, Tamil Dravidian family
a. There should be several fairly clear groups. The words for 'coffee' in all the languages are similar. Would this resemblance provide evidence that at a very deep historical level these languages are probably all related? Why or why not?
Not evidence: Coffee has spread around the world in only recent times on the scale of human history. Along with this new product came its name. The fact that the words for 'coffee' look alike does result from their having a single ancestral source, but that source is not in any of the groups here. (Coffee originated in northeast Africa. The word 'coffee' itself probably comes from Arabic qahwa, which originally meant 'dark' or 'brown'.)
b. Though we can identify some sound correspondences among the apparently related languages, identifying systematic sound correspondences don't play much role in arriving at reasonable groupings. What allows us to claim with some confidence that our groupings are correct even without establishing systematic sound correspondences?
Mass comparison relies on comparing lots of words from lots of languages. If we find a reasonably large number of vocabulary resemblances across languages, especially in items of everyday vocabulary like body parts, small numbers, words like 'sun' and 'moon', common actions like 'eat' or 'drink', etc., it becomes improbable that those resemblances could all be a result of chance. There may be a few accidental resemblances which we could have ruled out by finding regular sound correspondences, but if we find enough resemblances, a few "mistakes" in our identification of resemblances will not ruin our overall grouping.
The word paadam 'foot' in Telugu was borrowed from Sanskrit, which was more or less the ancestor of Hindi (maybe more like an "aunt" of Hindi than the "mother" of Hindi). I don't have much to say about the Uzbek and Turkish words for 'tooth' and their possible relation to the Indo-European words--it may just be a chance resemblance. Note that the three groups here (Indo-European, Altaic, Dravidian) have been grouped into the more ancient Nostratic family. If the Nostratic hypothesis is valid, then resemblances across the three groups could be traced back to this more ancient ancestor