The basis of this classification of languages is the principle of their historical kinship, that is, the initial ascent of a group of languages to a common, root language. It is far from always possible to establish this progenitor language, but, nevertheless, a clearly traceable connection gives linguistic scientists good reason to assume its existence in the distant past. In order to search for similar elements in different languages, the comparative historical method is traditionally used - it is on the data obtained with its help that the traditional genealogical classification of languages is based.
According to the explicit principle of historical kinship, or historical similarity, languages are usually divided into several fairly large groups, called families in linguistics. All languages within the same family have some similarities in the structure of words, the features of pronunciation of sounds, or in the rules of word formation. Not always these connections are noticeable at first sight - sometimes a lot of painstaking work is required to help reveal the distant kinship of certain languages. But, nevertheless, traditional linguistics, agreeing with the fundamental idea that some languages have common roots, easily finds similar points in their structure and features.
The genealogical classification of the world's languages today implies not only the division into language families - within each family there are other degrees of related languages, on the basis of which groups are distinguished. It should be noted that languages from different families do not have similarities, and this allows us to speak with confidence about the different nature of their origin. The presence of some similar aspects caused by the historical borrowing of words and word forms, only confirms the fragmentation of different language families.
Within each family, the genealogical classification of languages implies the identification of several branches (groups), languages in which have much more similarities with each other than with other members of the family. This may be due to later language processes in a given territory, which caused the fragmentation of the language into several similar groups, or to the increased isolation of a particular nationality due to historical, natural or military disasters.
Often within the linguistic subgroups, the genealogical classification of languages also distinguishes languages with the closest possible connections - they are usually called subgroups. A striking example of such a classification is traditionally considered the division of the Slavic languages belonging to the Indo-European family into East Slavic, West Slavic and South Slavic subgroups.
In some cases, the lack of historical knowledge and the disappearance of certain nationalities leads to a number of certain difficulties. So, some languages, despite a long study, cannot be assigned to one or another language family, because they do not have a pronounced similarity with the rest of its members. The genealogical classification of languages usually calls such cases languages "outside the classification."
But it is worth noting that the data obtained by linguists during the study of a particular family cannot be considered static. Very often, the appearance of new information or the finding of previously unknown texts written in one or another language forces us to revise the traditional classification, again and again analyzing the facts that were previously considered already established.
Therefore, even languages that are still outside any family may well, in a few years, as a result of obtaining new data, be assigned to an already known family or become the basis for developing a new classification.