1
The Indo-European Beginnings
The Indo-European Roots of Greek
Greek belongs to a family of related languages which are called âIndo-Europeanâ because at the time of the discovery of this family the known languages were distributed in Europe and the Indian Âsubcontinent (Indo-European languages were subsequently discovered in Asia Minor and central Asia). The existence of such a family was suggested by William Jones, a British scholar and lawyer who was appointed to the Supreme Court at Calcutta in 1783. Jones was an expert linguist who had taught himself Arabic and Persian at Oxford in addition to Greek and Latin; he was also a radical politician, who supported the American revolution and bitterly attacked the slave trade. When he arrived in India as a judge he learned Sanskrit, the ancient classical language of India and the sacred language of Hinduism, in order to understand the principles of the native Hindu legal tradition (he wrote several books on Hindu and Moslem law in India). In 1786 he delivered a paper in Calcutta to the Asiatic Society of Calcutta, which included the following famous words:
The Sanscrit language, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a Âstronger affinity, both in the roots of verbs and the forms of grammar, than could possibly have been produced by accident; so strong indeed, that no philologer could examine them all three, without believing them to have sprung from some common source, which, perhaps, no longer exists.
Throughout the nineteenth century work continued on the newly discovered family of languages, mostly in Germany, and this gave rise to the new science of linguistics in the West. In India there was a long and illustrious tradition of linguistics, going back to the late sixth Âcentury BC, when the famous grammarian PÄášini composed his exhaustive grammar of the Sanskrit language (and the tradition of systematic thought about language in India was doubtless older than PÄášini). There was no analogous âclassicâ in Greek or in Roman literature. Although in both the Greek and the Roman world there was interest in language, this was mostly related to its importance to philosophy and rhetoric in the early period; there was more technical work on language in the Hellenistic and Roman periods, but this was focused more on textual criticism and the explication of archaic and classical forms of the language for educational purposes. Europeans were still rather unsophisticated linguists in the eighteenth century. However, the kick-start given by the comparison of Greek, Latin, and Sanskrit, followed by study of Germanic and then Slavic, Celtic, and others, led to the development of what we now call historical linguistics: the study of the development of languages over time, and the reconstruction of an unattested âparent languageâ by systematically comparing the later languages which have survived in written form. This was the start of modern Western linguistics: at the end of the nineteenth century Ferdinand de Saussure, who had been trained in historical Indo-European linguistics, moved from considering the development of languages over time (historical linguistics) to the analysis of structural relations of languages at a given point in time (synchronic linguistics).
The Family Tree
Indo-European historical linguistics was, of course, a child of its time, and many of the linguistic models and metaphors which have become ingrained in our way of thinking about language reflect the intellectual environment of the nineteenth century. Part of this Âenvironment was a fascination with biological taxonomy and the evolution of species: Charles Darwinâs On the Origin of Species (1859) provoked an intellectual revolution, and it is no coincidence that much of the terminology of historical linguistics is reminiscent of biology. Languages are described as related, and form a family; one aims to reconstruct a parent language, from which the daughter languages evolve; relations between languages are set out in branching tree diagrams, like a family tree. This type of relationship between languages is called genetic. Both the model and terminology have the potential to be extremely misleading, since languages are not in fact organisms: an essential difference from the Darwinian model is that languages (or rather, their speakers) do pass on acquired changes. In addition to this, language is a sociocultural force which plays a central role in the self-definition of the speaker: these two facts have consequences for the way we think about language change and the model of the family tree.
It is true that most speakers learn a variety of the native language from parents (or older speakers in general); in this sense a language may be said to be âinherited.â But the metaphor does not bear Âpressing: for in fact a speaker learns not just one native idiom, but a variety of idioms from a variety of different speakers. In addition to grandparents, parents, and siblings, most children are exposed to different varieties of the language from the community at large. A competent native speaker is capable of recognizing a wide range of varieties (and their social connotations), and has mastery of quite a few varieties which are employed in different social situations. This reflects that fact that the notion of a language is to some extent a social construct: a language typically consists of a variety of different idioms and dialects, and in many cases is not clearly distinguishable from neighboring languages. And even when neighboring languages are in fact distinct, they may still form part of the speakerâs linguistic competence (monolingual cultures are exotic in the world, not the norm). Of course, in many cultures there is a prestigious standard language which many speakers think of as the language (and other varieties may be seen as inferior by comparison to this standard), but this perception is a cultural and political phenomenon, rather than a reflection of linguistic reality.
There are clear consequences for the genetic metaphor of language relationship and language change when we replace the idea of a uniform language inherited from parents with that of a continuum of language varieties taken over from across the language community. First, it can be seen that the native speakerâs competence has multiple sources, and is subject to continuing development, so one cannot contrast the validity or purity of a genetic relationship with âcontaminationâ or âinfluenceâ from other sources. The second point is closely related to this: a language change occurs when a majority of speakers adopt for use in a majority of situations a variant which was previously used by a minority of speakers, or in a restricted social context, or both. The reasons that prompt speakers to adopt such changes are complex: sociolinguistic research indicates that these decisions â like decisions pertaining to clothing and personal appearance â are the result of the speakerâs desire to shift his or her identity with regard to a particular section of the community. This type of behavior is easy to observe in adolescents, but research indicates that it persists in a subtler form in people of all ages. Speakers may be unconscious of many of the linguistic shifts they are making.
Since the growth of sociolinguistics enabled linguists to understand how languages change, it has become common to emphasize the importance of âarealâ factors in describing linguistic change and language relationships, at the expense of the traditional âgeneticâ family tree. This shift in emphasis offers important insights into the historical development of Greek, even though we have seen that the distinction itself is slightly dubious. âGeneticâ can be applied, metaphorically, to features of a language which were observable in an earlier stage of that language, while âarealâ covers features which have entered the language from elsewhere.
The language groups which are now derived from the ÂIndo-European parent language are: Albanian, Baltic, Anatolian, Armenian, Celtic, Germanic, Greek, Indic, Iranian, Italic, Slavic, Tocharian. Very poorly attested languages or groups include Illyrian, Phrygian, Thracian; it seems certain that many other languages have disappeared without trace. The relationships between these groups are not identical: for example, Indic and Iranian are so close that they are generally grouped together as âIndo-Iranian,â and Celtic, Germanic, and Italic show overlapping similarities which are best explained by their contiguity in the northwestern area of the Indo-European world. It is generally agreed that the Anatolian group must have split off from the parent language earlier than the others, since it has peculiarities which sets it apart from the rest of the âfamilyâ (such as lack of a separate feminine gender). There are a number of different types of reason for thinking that these languages are related.