For more than two centuries linguists have noticed two basic facts about language: (1) languages are related and can be grouped into language families which show historical connections, and (2) all living languages—those with actual, living speakers—change through time. Within each language, language change is not equal. Some parts of language, such as vocabulary relating to relationships and body parts, are fairly resistant to change, while other parts of language may change rapidly. The parts of language which are most resistant to change are known as the core vocabulary. Linguist Morris Swadish originally proposed a list of either 200 or 100 words which he felt would not be subject to changes brought about by migration, social change, or technological change. This core vocabulary includes pronouns (I, you, we) as well as words such as big, long, small, many, dog, ear, eye, red, green, and soon.
David Anthony, in his book The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World, writes:
“Although English has borrowed more than 50% of its general vocabulary from the Romance languages, mainly from French (reflecting the conquest of Anglo-Saxon England by French-speaking Normans) and Latin (from centuries of technical and professional training in courts, churches, and schools), only 4% of the English core vocabulary is borrowed from Romance.”
Assuming that core vocabularies change very slowly and are fairly resistant to outside influences and the fact that languages are related, is it possible to compare the core vocabularies from two languages from the same language families and from this determine how long ago they were the same language? Using a standard word list of core concepts, Morris Swadish felt that the average rate of change in this core vocabulary might yield a reliable standard measurement of the speed of language change. This, then, is the basic idea behind glottochronology. Linguist Johanna Nichols, in her chapter in The Origin and Diversification of Language, explains glottochronology this way:
“An average rate of vocabulary loss from a standard word list of 100 or 200 items has been computed at about 20% per millennium, and dates of separation can be calculated for pairs of languages by determining the number of cognate items they share from that list.”
In his book Language, Culture, and Society: An Introduction to Linguistic Anthropology, Zdenek Salzmann writes:
“Glottochronological dating is based on the assumption that in all languages there are certain words that tend to be replaced over long periods of time. This core vocabulary consists of words that designate things, qualities, and activities most likely to be named in all languages of the world.”
Glottochronology is simply a mathematical method for calculating when two related languages split apart. It is generally felt that glottochronology has a practical range of about 5,000-6,000 years. In other words, glottochronology can provide a relative date for the separation of two languages as long as this occurred within the past 5,000-6,000 years.
In his book Genetic Linguistics: Essays on Theory and Method, Joseph Greenberg describes glottochronology this way:
“A chronological time scale is provided by comparisons of vocabulary from different time periods of the same languages in areas of recorded history. The results so far indicate an average ca. 81 per cent retention of basic vocabulary over one millennium. Thus, by comparing two related languages for which no earlier recorded material is available, the percentage of basic vocabulary differences will allow for an approximation of the date of separation of the two forms of speech.”
Swadish’s initial studies suggested that changes to the core vocabulary came about slowly. In a paper published in the 1952 Proceedings of the American Philosophical Society, he writes:
“A language is a highly complex system of symbols serving a vital communicative function in society. The symbols are subject to change by the influence of many circumstances, yet they cannot change too fast without destroying the intelligibility of language. If the factors leading to change are great enough, they will keep the rate of change up to the maximum permitted by the communicative function of language. We have, as it were, a powerful motor kept in check by a speed regulating mechanism.”
In this view, language change is largely constant and varies little over time. However, recent studies have shown that the rate of language change is not constant. In his chapter in The Routledge Handbook of Historical Linguistics, Michael Dunn writes:
“The problem with this whole line of research is that the constant rate of change turned out to be illusionary. The rate of language change varies due to a number of factors.”
Like genetic drift, the rate of change is faster in smaller communities.
Not all linguists were enthusiastic about glottochronology. As more data was collected, there appeared to be some errors in the method. David Anthony writes:
“This direction in the errors suggested that real language change often was slower than Swadish’s model suggested—less than 19% per thousand years.”
Then, in 1962, there was a devastating critique of the mathematics used by Swadish in his glottochronology. It appeared that glottochronology was a mistaken idea. But, as sometime happens in science, it was found that the critique itself was in error. Glottochronology was reborn. By the 1980s, new models were being used which included equations of such critical values as borrowing rates, the number of geographic borders with other languages, and a similarity index between compared languages.
Zdenek Salzmann summarizes glottochronology this way:
“In summary, if applied to related languages whose history is not known and for which written records do not exist, glottochronology may provide some preliminary estimates of their closeness. But careful linguistic anthropologists would look for supporting evidence from archaeology, comparative ethnology, and linguistic reconstruction using the comparative method before accepting glottochronological results as valid.”
One example of glottochronology can be seen with regard to the Crow Indians of the Northern Plains. According to Crow oral traditions, they split off from the Hidatsa at some time in the past. The Crow language, like the Hidatsa language, belongs to the Siouan language family. Glottochronology shows that about 600 years separate Crow and Hidatsa.
Glottochronology is still not accepted by many linguists. For example, in his book The Rise and Fall of Languages, R.M.W. Dixon writes:
“…we should pay little heed to any putative links said to be justified by lexicostatistical counting, or to the putative dating of past stages of languages by the glottochronological method.”
In spite of this opposition to glottochronology, it is often a useful tool, particularly when used in conjunction with oral histories and archaeology.
It should be pointed out that while some scholars use the terms lexicostatistics and glottochronology as if they are synonyms, others prefer to distinguish the two terms. According to Michael Dunn:
“…lexicostatistics refers to the process of clustering languages based on distances calculated from meaning lists, whereas glottochronology refers to a method for using these distances to infer chronological dates.”
It should be pointed out that glottochronology is not the only linguistic dating method. In his book Before the Dawn: Recovering the Lost History of Our Ancestors, Nicholas Wade writes:
“Historical linguists are much more enthusiastic about a quite different dating technique called linguistic paleontology. The idea is to construct words for objects of material culture in a language family and date the language by noting the times at which such objects first appear in the archaeological record.”
More Language Essays
Language 101: Language Change
Language 201: The Indo-European Language Family
A Very Short History of the English Language
Language 101: "Body Language"
Language 101: The Search for Origins
Origins of Language: Grammar, the Architectural Structure of Language