The process of distinguishing and categorising the languages of the world has been a task undertaken by many over the centuries, and in modern times, mostly by both historical linguists and geolinguists. On the one hand, it is a study that can be exercised with a measure of scientific precision; on the other, however, there seems to be so much subjectivity attached to the demarcation of languages by their speakers that the authors of a recent publication, Constructing languages (Feliu and Nadal 2016), identified in their analysis three main features evident from such demarcations that do not seem to be compatible with an objective, scientific approach, namely norms, myths and emotions.
While elements such as myths and emotions already cast some suspicion on the supportability of language classification, the first feature, namely the application of norms, is a concept that enables one to look in a nuanced way and with a measure of objectivity at how languages in the world are classified. In evaluating the way in which a specific language, such as Afrikaans, or a group of languages, such as Germanic languages, are classified, the question relating to the norms applied will be most prominent.
How Afrikaans can be classified
A thorough exposition of this theme is provided in Carstens and Raidt (2017:117–227), and this is not intended to recapitulate or supplement its contents. What is of importance here is to indicate the underlying motivation for the various classification possibilities of Afrikaans, and arrive at a possible evaluation, since there exist various classification systems to categorise the more than 6 000 languages of the world (Gunnemark 1991:102), from Abkhaz to Zulu. In addition, languages are often also grouped together on the basis of extraneous factors, to which reference will be made at the end.
An overarching approach is merely to group together geographically the languages that have found a home in particular parts of the world, with reference to the geographic appellation of the relevant area. Consequently, one could refer to European languages, African languages, South American languages, Asiatic languages, and so forth. It is indeed so that the geographic norm can be used as a rough categorisation criterion to bring languages of indigenous origin together under one denominator. One problem with this approach is that there are many languages that are not confined to a particular area or continent (and hence of indigenous origin), such as Arabic, Spanish, Portuguese, English, Dutch, Mandarin, etc. Therefore, this approach has limited value in application and is but a coarse sieve. By applying the geographic criterion, English (which is spoken in various parts of Africa), as well as Afrikaans, for more reasons, could be classified as African languages.
A second (and equally problematic) approach is the ethnographic one, according to which ethnic derivations (which can include both race and nationality) are linked to languages, in terms of which reference is made, for instance, to Indian languages (in South and North America), black languages, Caucasian languages, etc. This approach likewise represents an oversimplification and is not based on any linguistic or other scientific grounds.
The earliest scientific classification was the outcome of a historical-comparative approach, according to which particular genealogical relationships between languages were described metaphorically, especially on the basis of lexical similarities (but also the morphology and syntax) that could be indicated systematically. Hence the denomination of sister and daughter languages, etc, which lay at the foundation of the concept “language family”. Within this paradigm, about 200 language families can be identified worldwide, and reference is made to, for instance, an Indo-European (or Indo-Germanic) language family, a Semitic-Hamitic language family, the Paleo-Siberian (super)family, the Niger-Congo family, etc. Consequently, it is called a genealogical classification. The diachronic (or historical) dimension then determines which languages developed from which primitive languages (Ursprachen).
Creolisation and pidginisation
Intensive contact between (speakers of) languages from different language families led to the formation of contact languages, such as pidgins and creoles. Such contact languages are normally based on the vocabulary and structure of a dominant language, or matrix language, but deviate from such a language in different ways and to varying degrees. While pidgins are used in addition to a home language, particularly in situations of contact with speakers of other (unrelated) languages, and also deviate considerably from the matrix language, creoles normally form a continuum from strongly deviating from (basilect) to strongly corresponding with (acrolect) the matrix language. Because of a limited range of contexts of usage, pidgins are structurally (and lexically) relatively simple. A creole, on the other hand, is also the mother tongue of speakers who have adopted the matrix language as such and use it with retention of characteristics from the language of their forebears, often as a comprehensive extension of a pidgin language. In certain creole contexts, one also finds relexification (the importation of vocabulary items and grammatical features) from the matrix language, something which happened in the case of Jamaican English and Afrikaans (from Dutch). This could lead to a situation where such languages have undergone creolisation, but are also much closer to the matrix language than before relexification. At the same time, it must be noted that a language community which (has) experienced creolising contact often includes different varieties that reflect the creole continuum referred to above. Whatever the result of the relevant language contact, such languages do not fit snugly into the genealogical model anymore.
A more linguistic approach to language demarcation is the comparison of languages primarily on the basis of structural similarities and differences, something that leads to a typological classification, in which phonetic, phonological, syntactic and morphological features are used as criteria. While a typological classification leads to informative linguistic discoveries about languages that are genealogically completely unrelated (eg Zulu and Japanese, which are both regarded as agglutinative languages), it does not have much practical value for the field of study of geolinguistics, which concerns itself, among other things, with the study of language in conjunction with the political geography of the earth.
The classification of languages in South Africa
It is clear that the languages used in South Africa can be classified in different ways in the geolinguistic context. A crude measure would be simply to distinguish between so-called “black” and “white” languages, something which has no utilitarian value in the politics of language. A typological classification, on the other hand, would be quite inaccessible to language users of the country, be inapplicable in geolinguistics, and have little use for the purpose of linking language classification and policy.
Would a genealogical classification then be of any use for the language politics of this country? In a genealogical context, we can distinguish between 1 300 and 1 500 languages in Africa – languages, therefore, that are related to each other in some way or another, and which can be divided into four superfamilies, namely the Nilo-Saharan (eg Maasai, Nubian, Dinka), Niger-Congo (eg Nguni and Sotho languages), Afro-Asiatic (eg Hausa, Amharic, Somali, Arabic) and Khoisan families (Khoi and San languages). According to this taxonomic model, the superfamilies are divided into families, and subsequently into groups. The Niger-Congo superfamily, for instance, consists of seven families, among which is the Benue-Congo family, of which the Bantu group is one of four groups (and has about 550 million speakers). In South Africa, the appellation “Bantu” has obviously been stigmatised politically, and the term “Sintu group” is now preferred, being also technically (and semantically) more correct. From a genealogical perspective, Afrikaans and English are categorised as Indo-Germanic languages – as a result of the diachronic links with Dutch and British English, respectively – a family which also includes German and certain northern Scandinavian languages, such as Swedish and Icelandic.
Against the background of this exposition, which is based on historical-linguistic cognation as norm, this classification, although relatively complicated, reflects an oversimplification of the local linguistic state of affairs, because it does not take cognisance of the situation of language contact in which the languages in this country are used alongside each other.
A geolinguistic classification
A geolinguistic classification of languages within South Africa places the focus on the geographic area under discussion, and not the continent or broad language family. Languages that originated within the borders of the relevant geographical area and are generally used here are indigenous languages, independent of the genealogical context. If a further distinction has to be made, for whatever reason, the genealogical relation can be used as a lower level distinction, to differentiate, for instance, between Indo-Germanic languages (eg English and Afrikaans) and Sintu languages (such as Nguni, Sotho, etc). Languages in South Africa that are used for religious purposes and are specified in the Constitution, accordingly also form part of, say, the Semitic language family (Arabic and Hebrew), the Dravidian language family (Tamil and Telugu) or the Indo-Aryan tale, such as Hindi (which also forms part of the Indo-Germanic language family). To use continental or ethnic appellations for geolinguistic purposes would be confusing and inaccurate.
To classify Afrikaans as a language is, therefore, a matter of norm. Following a geographical point of departure, Afrikaans is an African language that originated here and reflects the name of the continent (the only language in Africa having this characteristic). In the case of a genealogical approach, it is an Indo-Germanic language (and even related remotely to Hindi), and more specifically West-Germanic, which makes mutual comprehensibility with languages such as Dutch and (to a lesser degree) German possible. From the perspective of creolisation, Afrikaans originated from a contact situation in which speakers from both genealogically related and genealogically unrelated languages left a clear imprint on what was transmitted eventually by their progeny as mother tongue. The variation reflected by Afrikaans is largely also the consequence of this process. Geopolitically, then, Afrikaans is an indigenous language, with all its varieties, and shares its living space with, among others, languages that can be genealogically described as Sintu languages.
In the discussion above, I restricted myself to the recognised frames of description for language classification, and attempted to indicate how a change of norm would influence and determine classification. A further aspect of the anthology of articles by Feliu and Nadal (2016) has not been touched on, namely the role played by myths and emotions in the appellation and labelling of languages, which is reflected in creations such as “language of apartheid”, “colonial language”, “Afro-Germanic language” and even “creole language”, from a negative perspective, although this concept has in modern times been stripped of earlier negative connotations. However, this is the subject of another discussion.
Carstens, WAM and EH Raidt. 2017. Die storie van Afrikaans – uit Europa en van Afrika. Pretoria: Protea Boekhuis.
Feliu, Francesc and Josep Nadal. 2016. Constructing languages – norms, myths and emotions. Amsterdam: John Benjamins.
Gunnemark, Erik. 1991. Countries, peoples and their languages – the geolinguistic handbook. Gothenburg: Geolingua.
- Also read