The genetic-etymological classification of languages. Cognate and non-cognate languages in comparative philology.

A language family is a group of languages related through descent from a common ancestor, called the proto-language of that family. The term 'family' comes from the tree model of language origination in historical linguistics, which makes use of a metaphor comparing languages to people in a biological family tree. From this point of view stand out some types: an insulating (amorphous) type (Chinese, Vietnamese), agglutinating (agglutinative) type (Turkic, many Finno-Ugric languages), plectrude (inflected) type (Russian language). Some scholars are incorporating (polysynthetic) languages (some paleo-Asiatic and Caucasian languages).

The question of types of languages or language families has become the most important in establishing the ontology (or variety) of contrastive studies.

The fundamental principles of describing sets of languages that derive from a common ancestor had been devised by comparative philologists in the nineteenth century. Thus, the Indo-European family tree represents the relationships between cognate (related) languages such as German, French, English, Polish, Russian, etc.

In the English this type of studies became known as comparative philology. Its primary purpose especially in the early stages was to investigate the relationships between cognate languages leading to the formation of language families (Sanskrit, Latin, Germanic, etc.). 

The phrase "cognate language" means the same thing as "sister language." The term "Cognate" means having a common ancestor. For instance, English and Danish are cognate languages, both being Germanic. Therefore, "non-cognate languages" do not have a common ancestor. For instance, English and Japanese are non-cognate languages.