Language Families :: Classification Language Family :: Similarities Among Languages

The end of the 18^th century saw the first scientific attempts at discovering the history of world languages. Comparisons between groups of languages were undertaken in a systematic and detailed manner to determine whether correspondences could be found. If similarities could be demonstrated then language relationships could be inferred. They were investigating whether the world's languages had developed from a common source.

Previous investigations had determined there was a common origin for groups of languages within Europe: French, Spanish and Italian, the Romance languages, were descended from Latin. By the beginning of the 19^th century convincing evidence supported the hypothesis that many Eurasian languages had an originating language: Proto-Indo-European. The same techniques were then applied to the study of other groups of languages.

August Schleicher (1821-1868), a German linguist, introduced the phrase Stammbaumtheorie (family tree theory). The Romance family has Latin as the 'parent' language with French, Spanish and Italian as 'daughter' languages. French and Spanish are sister languages. The Indo-European family has Proto-Indo-European as the 'parent' language with Latin, Greek, and Sanskrit as 'daughter' languages. In a large language family it is necessary to distinguish between a number of 'branches'. Each may contain 'sub-families' of languages or individual languages.

Languages can converge as well as develop new branches so the tree analogy should not be taken too literally. Linguistic changes occur unevenly, responding to changes in social groups in different ways. The term family is still used where there is a clear and close linguistic relationship. Where the relationship is less well-defined or wider then the term phylum is referred. Even macro-phylum can be used for even less-defined relationships. An example of a macro-phylum is the Aboriginal languages of Australia where there is no clear-cut historical or typological evidence for relationships even though it is evident that they must be related.

The comparative method is a systematic linguistic comparison of languages to prove historical relationships. Formal similarities and differences are identified between languages and earlier stages of development of the languages are reconstructed. This is internal reconstruction . If an earlier ancestor is shown for languages then they are cognate .

Where a parent language exists then this is a good example for sister languages being cognate. Where reconstructed forms are derived there is no relevance associated to the pronunciation of that form and phonetic values are assigned by scholars.

August von Schlegel (1767-1845) propounded three main linguistic types based on language constructs of words:

1. Isolating, analytic, or root languages - there are no endings to words. Relationships are determined by word order. Chinese, Vietnamese and Samoan are examples.
2. Inflecting, synthetic, or fusional languages - the internal structure of words is changed usually by inflectional endings. Latin, Greek and Arabic are examples.
3. Agglutinative or agglutinating languages - words are built up from sequences of units. Turkish, Finnish, Japanese and Swahili are examples.
4. Polysynthetic or incorporating languages - a fusion of agglutinating and inflectional features. Eskimo, Mohawk and Australian languages are examples.

The two main classification methods for languages are:

1. Genetic (or genealogical) - a historical classification following the assumption that languages have derived from a common ancestor language. This is evidenced by written remains or by deductions using the comparative method. This method has been widely used since the 18^th century and is the framework for all world-wide linguistic surveys. It does rely on written remains so Eurasia has been its proving ground. Where there is little written evidence classification has been more tentative.
2. Typological - based on formal comparison of similarities between languages. Languages are grouped into structural types. The basis for this is phonology, grammar and vocabulary. Languages can be grouped by their use of sounds: vowel availability and use; usage of clicks; tonal uses. They can also be grouped by their use of word order.

Some early classifications have been superseded by modern interpretation. No one now evaluates languages as if they were points in evolution - that isolating languages are not as well developed as inflecting languages. There has been no evidence obtained that particular types of language are associated with particular geographical area, or with peoples of a particular ethnic or cultural group.

A typological classification is all down to the way the variables are interpreted. Morphology is just one of the variables. The other features of a language: syntax, phonology, discourse and language use all need to be taken into consideration. These criteria can give a vast array of classifications. What importance must be given to each of the criteria? Linguistic theory has this problem to overcome.

Cultural impacts are also not fully taken into account. Sometimes languages converge thus clouding the issues. Sometimes languages are strongly influenced by non-related languages such that differences are more pronounced than similarities. These are the problems that present themselves to those that wish to classify languages.

Some linguists now propose to rank languages in terms of individual structural criteria rather than general classifications.