Exploratorium Magazine Online: The Evolution of Language
Where Do Languages Come From?, page 3 of 5


Merritt Ruhlen discusses explanations for linguistic similiarity, including the "borrowing" that occurs when two languages are in frequent contact.


If we examine words other than "hand," we find many additional instances where each of these three families is characterized by different-looking roots, just as in the case of "hand." But we also find, from time to time, roots that seem to be shared by these three families; that is, the same root is found in all three families. What is the meaning of such roots? In fact, similarities among language families such as Romance, Germanic, and Slavic have the same meaning as similarities among languages in any one family — they imply that these three families are branches of an even more ancient family. In other words, a language that existed long before Latin, Proto-Germanic, or Proto-Slavic first differentiated into these three languages and then they, in turn, diversified into the modern languages of each family. This larger, more ancient family is known as the Indo-European family and it includes almost all European languages (but not Basque, Hungarian, or Finnish), and many other languages of Iran, Afghanistan, Pakistan, and India. The Indo-European family has, in fact, thirteen branches; in addition to Romance, Germanic, and Slavic, there are also Baltic, Celtic, Iranian, Indic, Tocharian, Anatolian, and three single languages that are by themselves separate branches of the family: Armenian, Greek, and Albanian.

The thirteen branches of Indo-European are connected to one another by numerous words and grammatical endings. One example is the word for "mouse," which exhibits striking similarities among languages from different branches of the family: Greek muus, Latin muus, Old English muus, Russian msh', and Sanskrit (Indic) muu-. Not surprisingly, scholars believe that the original Proto-Indo-European word was *muus- (the * indicates a hypothetical reconstructed form, rather than an actually attested written form). Another root shared by different branches is the word for "nose": Latin naas-, Old English nosu, Lithuanian (Baltic) nos-, Russian nos, and Sanskrit naas-. All of these words are thought to have evolved from Proto-Indo-European *naas-. The precise time and place that Proto-Indo-European was spoken remains a matter of some dispute even today. The two most popular hypotheses postulate it was spoken in Ukraine around six thousand years ago, or Anatolia (modern Turkey) around eight thousand years ago.

The story does not end here, for Indo-European is but one branch of an even larger (and more ancient) family known as Eurasiatic. In addition to Indo-European, this family also includes the Uralic family (Finnish, Hungarian, Samoyed); the Altaic family (Turkic, Mongolian, Tungus, Korean, Japanese); the Chukchi-Kamchatkan family just across the Bering Strait from Alaska; and the Eskimo-Aleut family that extends along the northern perimeter of North America from Alaska to Greenland. One of the words found in all five branches has a general meaning of "tongue, speak" or "call": Proto-Indo-European *gal "call," Proto-Uralic *keele "tongue," Proto-Altaic * "tongue, speak," Kamchadal (Chukchi-Kamchatkan) kel "shout," Proto-Eskimo *- "inform." The Eurasiatic family is also characterized by distinctive first- and second-person pronouns, the first based on M, the second on T. Within the Indo-European family, almost every language exhibits such forms: English me and thee, Spanish me and te, Russian menya and tebya, and so forth. This pattern is, however, characteristic of the entire Eurasiatic family, not just the Indo-European branch. In other parts of the world, different pronominal systems are found. For example, in the Amerind family, which includes most Native American languages, the most common pattern is first- person N and second-person M.


Back Exploratorium Magazine Next