Linguistic Reconstruction

One of the most important tasks of historical or comparative linguistics is to try to establish a relationship between or among two or more contemporary languages based on a supposed common ancestry. The tool most often used to do this is linguistic reconstruction.

In 1786, Sir William Jones, a British judge and swashbuckling scholar who some more recently have compared to his namesake Indiana (in the Indiana Jones films), made a startling discovery: “The Sanskrit language … is of wonderful structure; more perfect than Greek, more copious than Latin … yet, bearing to both of them a stronger affinity, both in the roots of verbs and forms of grammar, than could have possibly been produced by accident; … [they] sprung from a common source which, perhaps, no longer exists.” What Jones was suggesting then was that Greek, Latin, and Sanskrit all shared a common ancestor in the distant past, just as three sisters might share a common grandfather. The reconstructed language is called a protolanguage, and reconstructed words or parts of words are called protoforms. (A protoform is usually preceded by an asterisk to distinguish it from actual speech.)

In doing a comparative reconstruction, word lists of the languages thought to be related are carefully compared to find cognates—words that share similar structures and meanings. For example, consider the following words for sun in four Native American languages: hasi, hashi, haasi, and hasi in Koasati, Choctaw,Hitchiti, and Creek, respectively. Ignoring the “long” a vowel in Hitchiti, what might the protoform for sun have been in the parent language? All of these words begin with the same sound, h, and use the low central vowel, a, so the word in the protolanguage probably contained these as well. But what about the s andsh sounds? By the “majority rules” strategy, we would argue that s was the original sound and that Choctaw underwent a change to sh for some reason. Thus, we would probably posit the protoform in this [Page 1480]protolanguage—actually called Muskogean by most Americanists—to be *hasi rather than *hashi.

But there is another reason why *hasi is the more likely candidate. The “phonetic plausibility” strategy says that the sound changes we hypothesize must make sense phonologically based on our knowledge of how the speech organs operate. There is much worldwide linguistic evidence where an s sound becomes palatized (i.e., the tongue moves back toward the hard palate in the roof of the mouth) when it comes before a high vowel ee (the last vowel in all of the Muskogean words mentioned previously). There are few, if any, examples of the opposite; in other words is it more likely that an s becomes a sh than an sh becomes an s in this context. Thus, if a native English speaker is tired or drunk and wants to say the sentence “I didn't see she was home,” the speaker would be more likely to slur it as “I didn'tshe she was home” than to slur it as “I didn't see see was home.”

Although phonological reconstructions are the most publicized, reconstructions can also be made at the level of morphology (units of words) and syntax (grammar). In terms of linguistic reconstruction in general, the Indo-European language family probably has received the most attention. Besides shifts in sound changes due to formal regularities, more historical linguists are also investigating the sociolinguistic motivations behind them. Linguistic anthropologists, from the time of Edward Sapir onward, have used linguistic reconstructions to explain the movement, contact, and history of non-Western peoples and will no doubt continue to do so.

...