Most people, when exposed to a number of languages other than their own, sense that some languages are more closely related to each other than are others. Upon hearing Spanish, Italian, and Portuguese, for example, English language speakers can sense from the sounds that these three languages are related to each other. If languages seem to be closely related to each other, does this mean that they somehow branched off from an ancestral language at some time in the past?
In 1786 Sir William Jones (1746-1794) gave a speech in which he compared Sanskrit with other languages and noted that they all came from a common source. This marked the beginning of the concept of the language family. Jones was not the first to note the similarities of language, but what makes his study different in that it was based on the careful inspection of data rather than simply intuition. In his book Perspectives in Linguistics, John Waterman writes:
“Once the conservatism of tradition and literalistic theology had been overcome, the way was clear to approach the study of language in this new perspective. Scholars came to understand that language was in a state of constant flux, that it had a history, and that its genesis and development could be studied from the historical point of view.”
During the next century, scholars began to lay the foundations of historical linguistics through the direct comparison of languages. The nineteenth century linguists adopted the biological model of language development. Family trees of different language groups were drawn up, showing the relationships of the languages with one another. In his book
Masters of the Planet: The Search for Our Human Origins, Ian Tattersall writes:
“The study of comparative linguistics makes it clear that languages have evolved much as organisms have done, with descendant versions branching away from the ancestral forms while still retaining for some time the imprint of their common origins. Many scientists have accordingly used the differentiation of languages as a guide to the spread of mankind across the globe.”
Thus linguists often group languages in a kind of genealogical tree. In their book
The Tarim Mummies: Ancient China and the Mystery of the Earliest Peoples from the West, J.P. Mallory and Victor Mair report:
“Linguists are fond of drawing family trees of languages which provide a simplified genealogy of a language’s development and relationships.”
A grouping of related languages is known as a language family. All of the languages within a language family can be shown to be “genetically” related; that is, they have descended from a common ancestral language. As with a genealogical family tree, there can be subdivisions of the family which are commonly known as branches.
As in the field of paleoanthropology, historical linguistics has “lumpers” and “splitters.” The “lumpers” see fewer language families and have attempted to group families into larger super-families. For example, Joseph H. Greenberg sees only three language families in the Americas, including the Amerind family which includes most of the Native languages of both North and South America. In Language, Culture, and Society: An Introduction to Linguistic Anthropology, Zdenek Salzmann writes:
“Most specialists in Native American languages are not ready to accept the validity of Greenberg’s huge Amerind genetic unit or family. In any case, a family of this size has little in common with the earlier conservative concept of language family.”
There are a few languages for which there is no apparent or yet discovered historical relationship with other languages. These are called language isolates and include languages such as Basque (spoken in Europe) and Kootenai (an American Indian language).
The family tree for any language family presents a simplified, perhaps overly simplified, history for any language. English, for example, belongs to the Indo-European language family which means that it is related to a number of other languages which trace their roots back to Proto-Indo-European, a language which was spoken perhaps 6,000 years ago in a region north of the Black Sea. Within the Indo-European language family, English is a part of the Germanic branch, and within this it is a part of the sub-branch English-Frisian. This simplified genealogical family tree from Proto-Indo-European, to Proto-Germanic, to English-Frisian, to English is interesting, but it does not explain the relationship of English to other languages. J.P. Mallory and Victor Mair report:
“Embedded in any language is evidence of its past. The preceding sentence of nine words contains five words which are derived from Germanic (in, any, is, of, its), three which derive from Latin via French (language, evidence, past) and one which combines a Latin prefix with a Germanic root (embedded). That the most basic verbs, articles and prepositions are all Germanic would suggest a relative chronology in which English was originally a Germanic language that was later heavily influenced by a Romance language.”
In other words, the family tree model does not show the influences of other languages through conquest, diffusion, word borrowing (some would call this “word theft” or “word sharing” since there is no intention of returning the word), and other factors.
Language families do show ancient connections and even provide clues regarding migration. For example, one of the major Native American language families is Algonquian, which appears to have originated in the eastern woodlands North America and includes languages such as Penobscot, Wampanoag, and Massachusett (all spoken in the Northeast) as well as Blackfoot and Cheyenne (Plains Indian languages). In California, we find Wiyot and Yurok, both Algonquian languages, which suggests a prehistoric connection between people of the Pacific and Atlantic coasts.
The concept of the language family remains an important part of historical linguistics. Linguists Joan Swann, Ana Deumert, Theresa Lillis, and Rjend Mesthrie note in A Dictionary of Sociolinguistics:
“A central part of the discipline involves the comparison of languages with a view to ascertaining which ones are related, for example Indo-European as the common ancestor of languages such as Latin, Greek, Sanskrit, Anglo-Saxon, etc.”
Many of the nineteenth century linguists viewed languages as similar to living organisms and thus created family trees for them in a way similar to which biologists create evolutionary family trees for plants and animals. Some assumed that languages, like plants and animals, have a life cycle: birth, infancy, adulthood, and finally death. John Waterman, however, reminds us that:
“Language is not an organism; it does not decay. Language changes—that is all we can say.”
