Effects of attitudinal changes towards creolization in Afrikaans

Ernst Kotzé (University of Port Elizabeth, África do Sul)

0. Introduction

It is this year exactly 100 years ago that DC Hesseling published  Het Afrikaans, thereby starting the formal discourse about the genesis of this language.  In his book he investigated the un-Dutchlike characteristics of Afrikaans and, although his hypothesis contains a number of serious flaws (cf. Den Besten 1986, 1997), he made language historians aware of  the special circumstances under which this Germanic language took root in African soil.  He was the first to describe Afrikaans as a mixed language (or ”mengeltaal”), even though Th. Hahn had noticed already in 1882 that ”although phonetically teutonic, it is psychologically an essentially Hottentot idiom” (cf. Van der Merwe 1969:15).  At a later stage, however, Hesseling (1923) maintained that Afrikaans never became a true creole, because there were counteracting forces: continued immigration of speakers of Dutch, as well as the use of Dutch in public life (Den Besten 1986:189).  Given the chronological context, I consider this to be an appropriate time to review the development of thought surrounding the slippery topic of the classification of Afrikaans, and venture a prognosis of possible new developments.  At the same time, I will provide a view of  more general developments in the field of creole studies against which the classification of Afrikaans as a case study should be seen.

The paper will take the following line of argumentation:  Section 1 will indicate that attitudes prevalent at the time of Hesseling’s publication made an objective investigation of linguistic and sociolinguistic factors determining the genesis of Afrikaans difficult, if not impossible.  Examples of the emotionality surrounding the issue will be quoted.  Sections 2 and 3 point to the increased awareness of linguistic data obtained through archival research, and discuss some of the results for the classification of purported creole languages, also elsewhere.  In Section 4, the fallacy of oversimplification by restricting research to the standard variety and assuming homogeneity of structure is touched on.  A brief survey of the heterogeneous nature of Afrikaans as a label for a number of varieties is given.  In attempting to arrive at the psycholinguistic locus of change, the process of the intergenerational transmission of language is revisited in Section 5, and the external factors influencing the direction of change are identified.  Subsequently, in Section 6, different processes of change linked to these external factors, which led to the diversification of dialects are discussed.  Finally, a prognosis of further structural change in Afrikaans is given, based on socio-political changes which affected the basis of standardization.  The conclusion is drawn that only did a shift in attitude influence insight into the nature of language change, but that attitudinal shifts are also likely to contribute to change itself.

La communication suivra le raisonnement suivant: La première section indique que les attitudes qui régnaient au temps de la publication de D.C. Hesseling il y a 100 ans (Het Afrikaans) rendirent difficile, sinon impossible, une enquête objective des facteurs linguistiques et sociolinguistiques qui ont déterminé la genèse de l’afrikaans.  Des exemples de l’émotivité autour de la question seront cités.  Les sections 2 et 3 font observer la conscience grossie de l’importance des données obtenues par la recherche dans les archives, et en discutent certains résultats pour la classification des langues créoles impliquées, ailleurs aussi.  Dans la section 4, on touche l’erreur de trop simplifier les faits en limitant la recherche à la variété courante (standard), et en supposant l’homogenéité.  Un exposé bref du charactère hétérogène de l’afrikaans comme nom pour un nombre de variétés est donné.  En essayant d’arriver au lieu psycholinguistique de changement, on revisite le processus de la transmission intergénérationnelle du langage dans la Section 5, et on identifie les facteurs externes qui influencent la direction de changement.  En suite, on discute dans la Section 6 des processus de changement différents qui sont liés à ces facteurs externes et qui ont mené à la diversification de dialectes.  Finalement, un pronostic de changements structurels en afrikaans est donné, qui est basé sur des changements socio-politiques qui ont affecté la base de la standardization.  On arrive à la conclusion que, non seulement un changement d’attitude a-t-il influencé les perspectives de la nature de changement de la langue, mais que des attitudes nouvelles peuvent mener elles-même au changement.

1. An emotional issue

We know that at the time of Hesseling’s publication, the Zeitgeist generally favored a purist attitude towards both linguistic and racial matters, and creolization was equated with miscegenation, both being regarded as deviations from the norm by linguists and laymen alike.  The Newgrammarian approach to linguistic description, based on the Stammbaumtheorie (or Family Tree Theory), relegated what were regarded as mixed languages to the position of ”black sheep of the family”, or illegitimate children, because the accepted line of descent was one parent per child  (a case of asexual procreation?).  Racialist attitudes characterized the investigation of the origin and history of languages in terms of this theory and in the prevailing spirit of the time, so that objectivity was often amiss, as I will indicate shortly.

Language not being an exclusively linguistic matter, the stimuli to the discussion about the genesis of Afrikaans ranged from linguistic curiosity about the differences between Cape Dutch (as it was called initially) and Continental Dutch on the one hand to emotional views on the other about the symbolic value of the language as ”kitchen language”, ”white man’s language”, more recently ”language of the oppressor”, etc.   M.F. Valkhoff in his 1966 publication adds the following perspective to his view that Afrikaans resulted from partial creolization ”in the mouths of both the Coloureds and the Whites”:

For many in this country this opinion is still a grave heresy, and with the prevailing idea of White supremacy the supposition that Coloured people contributed to the shaping of Afrikaans may be looked upon as malicious by some of the White population and may be exploited as hostile propaganda by others. (1966: x)

Reacting to these words, a South African linguist, HJJM van der Merwe (1969:33) remarks in similarly emotional vein:

Apparently the drift of this is that fornication usually was accompanied by a dialogue in the language of the woman, for which everybody who indulged in these lusts therefore had to know Portuguese Creole, which would have strongly promoted the Portuguese lingua franca at the Cape.  (translation EFK)

2. Archival research

After the turn of the half-century, concurrently with Valkhoff and Van der Merwe, certain scholars, whom Den Besten (1986:191) calls ”the South African philological school” (inter alia Scholtz, Raidt & Pheiffer), started to research the archival resources so as to test the various hypotheses concerning the purported creole status of Afrikaans.  The reason for the new approach is phrased as follows by Scholtz (1963:274):

The study of the history of Afrikaans had for a long time been caught up in the grip of divergent genetic theories.  To a large extent these theories were aprioristic in origin and were kept afloat by arguments, pro and contra, on the basis of sociological arguments (translation EFK).

Den Besten (e.g. 1978, 1986, 1989, etc.), likewise undertook intensive archival research to investigate cases of syntactic interference in Afrikaans, e.g. the parallelism between the Khoekhoe (Hottentot) double negation and that which developed in Afrikaans, a hypothesis previously suggested by Nienaber.  This data-oriented research coincided with a more general search for objectivity in the field of creole studies elsewhere, which increasingly led to the identification of criteria by means of which creoles could be identified.  Authors such as Taylor, Bickerton and Markey compiled checklists of features to try and categorize languages which demonstrably evolved as a result of contact between typologically divergent languages (cf. Romaine 1988:47).

3. A first criteria-based classification

As a demonstration of the application of such a checklist, Markey (1982) compares Afrikaans with Negerhollands, the Dutch-based creole which was spoken in the Virgin Islands until the 1940s.  Whereas Negerhollands fulfills all the criteria, Afrikaans proved only to possess two features out of 12 typical of a true creole (namely lack of inflectional gender marking in all nouns, and lack of nominal case inflection), and are creole-like in two further areas.  He concludes that Afrikaans does not fit into the class of true creoles, and particularly not in terms of its properties of negation (Romaine 1988:59), and can rather be regarded as a transitional language located on a continuum between creole and non-creole.  Den Besten’s (1986:201 et seq.) argument regarding double negation, namely that a rather intricate rule became part and parcel of the Afrikaans grammar through the influence of creole speakers, casts more than a shadow of doubt on the validity of decontextualized features as indices of creolization.

Markey’s findings were to a large extent consonant with those of the South African ”philologists”, who focused their attention on linguistic features of Afrikaans that were absent in Modern Dutch, but could be found in Dutch dialects, particularly in those spoken in the 17th century, when the first Dutch colonists arrived in the Cape.  Although they recognize a variety of ”external” influences shaping the eventual form of Afrikaans, it was maintained that the language essentially remained within the confines of Continental Western Germanic and could not be regarded as a creole (or even a creoloid) language. However, an important observation was made by Gilbert and Makhudu (1984), who noted that Markey’s analysis was based on a normative grammar of Standard Afrikaans as spoken by white speakers, and did not take cognizance of the fact that there existed (and still exist) other varieties of Afrikaans with significant numbers of speakers, varieties which display a considerably higher percentage of creole features.  (According to Makhudu, the Afrikaans spoken by blacks displays 9 out of a set of 13 creole features.)  Not only Markey, but most other researchers to date have restricted their comparative investigation to Standard Afrikaans as the only ”legitimate” variety amenable to analysis.  If Afrikaans is then, as Romaine (1988:62) indicates, ”a loose label for a set of varieties ranging along a scale from highly European-like (i.e. similar to Dutch) to moderately creole-like (i.e. similar to Negerhollands)”, what are these varieties, and to what extent can they be regarded as a continuum, in casu a post-creole continuum?

Before providing a brief description of varieties in Afrikaans, a further corollary of Markey’s finding should be mentioned at this stage.  It had become abundantly clear that languages under the umbrella of ”études créoles” are not all equally creole in nature, and researchers have consequently created a multiplicity of typological labels to indicate varying degrees of ”creoleness”, such as creoloid, semi-creole, interlanguage, koine, etc., not to mention the proliferation of labels to describe the range of contact varieties between pidgins and creoles.   What is clear from this collection of labels, is that not only do the various languages under the creole umbrella differ in terms of features; they have also been subjected to different processes of language change.  I will return to this point shortly.

4. The heterogeneity of Afrikaans

Various authors (cf. Den Besten 1986:185) discern three distinct historical varieties in Afrikaans, namely Cape Afrikaans, Orange River Afrikaans and Eastern Frontier Afrikaans.  The latter became the basis of the variety spoken by white trekkers who settled in the northern parts of the country from 1830 onwards and which was eventually accepted as Standard Afrikaans because of the preponderance of political and economic power, which eventually became concentrated in the north (cf. Van Rensburg 1983).

Although the focus of researchers was restricted to the standard variety as the form representative of all other varieties, the heterogeneity of the language was not neutralized by standardization - on the contrary.  In addition to the survival of the historical dialects as vernaculars, at least one, namely Cape Afrikaans, was codified to make provision for the needs of (a) the Malay community, who continues to utilize a formal variety of Afrikaans for religious purposes in spite of a diglossic incorporation of English in formal contexts; and (b) the social and political liberation struggle on the Cape Flats, for which a phonetic spelling of vernacular forms was used in poems and dramatic productions.  I will distinguish henceforth between Formal Malay Afrikaans and Literary Cape Vernacular.

The differences between these two codified registers are both cultural and stylistic.  FMA is based on the syntax of Standard Afrikaans and uses a considerable lexis of Arabic and a few Malay words, as well as non-standardized Afrikaans lexical items, many of which are either semantically opaque or unusual to the user of Standard Afrikaans, or morphologically peculiar.  In the following example (from a cultural newsletter/magazine called Masjied Boorhaanol Islaam) the relevant non-standardized items are underlined:

Genoegsaam is die Qiessa van Nabie Loett oor die swakheid en onreg van

Sufficient is the story of (the prophet) Lot about the weakness and wrongfulness of

selfsgeslag as ‘n manier van omgang.  Dit is werklik ontstryding teen die basiese

homosexuality as a way of intercourse.  It is really contradiction against the basic

grondwet van mens soos die Hoë Alla dit beplan het.

design of man as the High Allah has planned it.

Literary Cape Afrikaans, on the other hand, covers a wide spectrum of topics, and attempts to reflect informal spoken discourse, particularly by adapting the orthography to the typically Cape pronunciation, and incorporates lexical, semantic and syntactic characteristics.  Here is an example of a poem by Peter Snyders (1982:3) reflecting a meta-sociolinguistic view of the variety:

Moetie rai gamattaal gebrykie;

Don’t use that ”gamat” (Muslim) language

dit issie mooi nie

It isn’t seemly

dit dieghreid die coloured mense - of hoe?

it degrades the colored people - or does it?

wat traai djy om ‘n coloured culture te create?

why do you try to create a coloured culture?

of dink djy is snaaks

or do you think i(t)’s funny

om soe te skryf?

to write in this way?

of hoe?

or what do you say?

Traai om ôs lieweste op te lig;

Rather try to uplift us;

ôs praat mossie soe nie ...?

we don’t talk like that, not so ...?

of hoe?

or do we?


In addition, vernacular varieties exist as a first language among black speakers (one of which will be the topic of a another paper at this conference), and also as a second-language pidginized form and as a component of the range of slang forms called Flytaal (also tsotsitaal or Isicamtu).  Time does not allow me to provide examples of each here.  However, it should be clear that the ”loose set of varieties” Suzanne Romaine refers to originated in widely differing contexts.

5. Transmission or not transmission?

And here I wish to briefly revisit the actual process of the intergenerational transmission of language, a stage in the diachronic development of language which could represent the crucial difference between creoles and non-creoles in the eyes of the linguistic beholder.  This point is highlighted by Thomason and Kaufman, who distinguish between ”genetic” and ”non-genetic” paths of development (DeGraf, e-mail message of 10/03/1999), the former arising via ”normal transmission” and the latter via ”imperfect transmission” as with abrupt creoles.

DeGraf (1999) strongly questions the assumption of ”perfect transmission”, and argues that grammars are always created anew from innate mental resources (e.g. the human language faculty, and mechanisms for the acquisition and processing of language) coupled with the ambient of environment-specific primary linguistic data available to the learner.  This view is not new.  Antoine Meillet describes it (1929: 74) as follows:

[C]haque enfant  doit  acquérir par  lui-même la  capacité  de comprendre  le parler des gens  de son groupe.  ...  La langue ne lui est pas livrée en bloc,  tout d’une pièce. ...  Pour chaque individu, le langage est  ainsi une recréation totale faite sous l’influence du  milieu qui  l’entoure. Il  ne  saurait y  avoir discontinuité plus absolue.

[Each child  must on his own acquire  the  capacity to understand the speech of people in his community. ...  Language is not given to him en bloc, all in one piece. ...  Thus, for each individual, language is a total  re-creation, carried out under the influence of the surrounding environment. There could not exist a more absolute discontinuity.  [Translation by M DeGraff]]

The grammar that eventually materializes, then, is the result of a combination of internal and external factors.  The internal factors include the inherent creativity of language users by means of which an intricate system evolves, based on limited and heterogeneous linguistic data.  The external factors include social factors such as (a) peer group influence, (b) varying fluencies of the model speakers (who provide the primary linguistic data) and (c) the diversity of the model speakers’ native tongues.

6. What happened to the varieties of Afrikaans?

From a diachronic perspective, it follows that the various identifiable historical dialects originated in varying social contexts.  Now let us briefly consider those contexts for each of the three varieties which have been accepted as such by researchers by and large.

After the first European settlement in 1652, there was a short period during which,  in addition to the mother tongue(s) of the settlers, officials of the Dutch East India Company, only a form of pidgin Dutch was spoken by the Khoekhoen, .  However, after 1658, when the first large groups of slaves were imported, three clearly discernible social groupings with linguistic correlates started to develop.  The Dutch, Germans, and later the French Huguenots, used a form of Dutch based on the dialect spoken in the province of Holland; the indigenous Khoekhoen used their mother tongue, Khoekhoe, and Hottentot-Dutch, which had probably become a stable pidgin; and the slaves from Asia, Madagascar and Mozambique spoke (in addition to their respective first languages) Pasar Malay, Creole Portuguese and their own form of Pidgin Dutch, which was based on that of the Khoekhoen (Den Besten 1997).

These three groupings formed the basis from which the three historical dialects of Afrikaans developed.  At the beginning of the 18th century, movement into the interior set the stage for a geographically based diversification of dialects.  While a large percentage (if not all) of the Khoekhoen living at the Cape fled northwards in the direction of the Orange River before a devastating small-pox epidemic in 1713, a gradual migration of white settlers took place in an easterly direction.  The various configurations of speakers of Dutch at the Cape, in the north-west around the Orange River and at the Eastern Frontier experienced different external sociolinguistic (or ecolinguistic) circumstances which correlated with different processes of change.

At the Cape of Good Hope, a socially stratified community was formed, consisting of Europeans, slaves and maybe some remaining Khoekhoen.  The Dutch spoken by the Europeans was subjected to (a) koineization as a result of which dialect differences (comprising Dutch and Low German dialects) were eliminated, (b) mother tongue interference in the speech of the non-Dutch and (c) lexical and structural borrowing, subconscious and otherwise, from the other languages with which Dutch was in contact.  The slaves spoke (depending on the stage of assimilation) either Pidgin or Creole Dutch, in addition to the contact languages Creole Portuguese and Pasar Malay and (as insofar as language shift had not yet occurred) their respective mother tongues.  The Dutch creole spoken by the Khoekhoen formed the basis of that used by the slaves.  At the Cape, then, a wide array of processes of language change took place simultaneously.

At the Eastern Frontier, the koineized form of Dutch spoken by the settlers and the creolized form of the Khoekhoen who accompanied them, coexisted for an extended time, with the social predominance of the settlers probably determining the norms for the form of Dutch which prevailed.  The lower intensity of language contact than at the Cape resulted in a more conservative variety  of Cape Dutch being spoken.

In the Orange River region, the Khoekhoen predominated, and inasmuch as language shift towards a form of Dutch Creole had not yet occurred, they spoke one or more Khoekhoe dialects.  Transferred characteristics of Khoekhoe in Creole Dutch as a result of interlanguage stabilization would either have persisted because there was no corrective by Cape Dutch speakers, or even reinforced because of maintained contact with Khoekhoe dialects still spoken in the region.

To conclude this argument, then:  A multiplicity of processes seem to have been operative in all the discernible varieties of Afrikaans, which include pidginization, language shift, creolization, koneization and borrowing.  The variety which seems to have undergone (or retained) the least change as a result of sustained contact with either creole speakers or speakers of other languages, is Eastern Frontier Afrikaans, which in time formed the basis of Standard Afrikaans, as we have observed above.

7. Accelerated language change because of socio-political change?

As a result of the socio-political changes which occurred in South Africa since 1994 and the concomitant democratization of the broad South African speech community, the previously privileged white speakers do not constitute the exclusive basis of standardized speech any more.  As more and more speakers of Cape Afrikaans take up new positions in areas where speech role modeling is important, particularly in the media, previously nonstandard variants are sanctioned and accepted.

To name one or two examples:  The Afrikaans verbal system has slowly been undergoing a process of regularization, in that remnants of the preterite in Standard Afrikaans are gradually being eliminated.  Most auxiliaries in Standard Afrikaans have retained a preterite form, e.g.

ek wil kom - ek wou kom
I want to come - I wanted to come

ek sal kom - ek sou kom
I will come - I would (have) come

ek kan kom - ek kon kom
I can come - I could come

However, in Cape Afrikaans, the preterite has disappeared, so that, on the basis of a single example already existing in Standard Afrikaans, namely

ek mag kom - ek mag gekom het
I may come - I may come have

the main verb is marked for the past tense, while the auxiliary is used as in the present:

ek wil kom - ek wil gekom het
I want to come - I want come have

ek sal kom - ek sal gekom het
I will come - I will come have

ek kan kom - ek kan gekom het
I can come - I can come have

A more noticeable change is the disappearance of the infinitive form of the only two main verbs still possessing such forms, nl. (have) and wees (be) in favor of the indicative form (of course, these verbs are also used as auxiliaries, as in most Germanic and Romance languages).  The first to have gone this way, seems to be , as in the following example, which recently appeared twice in a feature article of a renowned daily newspaper (e.g. Die Burger, 14.06.99):

Nadine Gordimer sê ‘skrywers moet groot ore het’ (hê)
Nadine Gordimer says ‘authors must large ears has’ (have)

In spoken language, the indicative form of wees (be), namely is, is also occurring increasingly among educated speakers of Cape Afrikaans, e.g.

Dit sal baie moeilik is (wees)
It will very difficult is (be)

It therefore seems as if, just as a change in attitude towards creolization as a phenomenon led to a larger degree of objectivity in the description of languages characterized by the effects of contact, a change in attitude towards long-held norms because of the democratization of society can lead to rapid changes and the acceptance of erstwhile stigmatized forms which would be regarded as creole-like simplifications.

Although a thorough sociolinguistic investigation is essential to verify these observations beyond all reasonable doubt, it certainly seems as if one of the significant effects of attitudinal change towards creolization in Afrikaans is an acceleration in standard language change.


Degenhardt, Ruth et al (eds.),  1997, Afrolusitanistik - eine vergessene Disziplin in Deutschland?  Dokumentation des 2. Bremer Afro-Romania Kolloquiums vom 27.-29. Juni,  Bremen: Universität Bremen.

Den Besten, Hans, 1978: «Cases of possible syntactic interference in the development of Afrikaans», in Muysken, Pieter (ed.), Amsterdam Creole Studies II, Amsterdam, University of Amsterdam:5-56.

Den Besten, Hans, 1986, «Double negation and the genesis of Afrikaans», in Muysken, Pieter & Norval Smith (eds.), Substrata versus universals in creole languages.  Papers from the Amsterdam Creole Workshop, April 1985, Benjamins, Amsterdam: 185-230.

Den Besten, Hans, 1989, «From Khoekhoe foreignertalk via Hottentot Dutch to Afrikaans: the creation of a novel grammar», in Pütz, Martin & René Dirven (eds.), Wheels within wheels.  Papers of the Duisburg Symposium on Pidgin and Creole Languages, Frankfurt am Main, Peter Lang: 207-249.

Den Besten, Hans, 1997, «Kreolportugiesisch in Südafrika: Malaio- oder Indoportugiesisch?», in Degenhardt, Ruth (et al), Afrolusitanistik - eine vergessene Disziplin in Deutschland?  Dokumentation des 2. Bremer Afro-Romania Kolloquiums vom 27.-29. Juni, Bremen, Universität Bremen: 317-351.

Gilbert, G.  & D. Makhudu, 1984, «The creole continuum in Afrikaans: a non-Eurocentric view», unpublished manuscript, Department of Linguistics, Southern Ill. Univ at Carbondale.

Hesseling, D.C, 1899, Het Afrikaans.  Bijdrage tot de geschiedenis der Nederlandsche taal in Zuid-Afrika, Leiden, Brill.

Holm, John, 1988, Pidgins and creoles.  Volume I: Theory and structure, Cambridge, CUP.

Markey, T.L, 1982, «Afrikaans: creole or non-creole?», in Zeitschrift für Dialektologie und Linguistik IL:169-207.

Meillet, A, 1929, «Le développement des langues», in Continu et discontenu, 1119ff,  Paris, Bloud & Gay.

Pütz, Martin & René Dirven (eds.),  1989,  Wheels within wheels.  Papers of the Duisburg Symposium on Pidgin and Creole Languages,  Frankfurt am Main: Peter Lang.

Romaine, Suzanne, 1988, Pidgin & creole languages, London, Longman.

Scholtz, J du P., 1963, Taalhistoriese opstelle, Pretoria, JL van Schaik.

Snyders, P, 1982, ‘n Ordinary mens, Cape Town, Tafelberg.

Valkhoff, M.F., 1966,  Studies in Portuguese and Creole.  With special reference to South Africa,  Johannesburg,Witwatersrand University Press.

Van der Merwe, HJJM, 1969, «Die ontstaan van Afrikaans», in Van der Merwe H.J.J.M., Afrikaans: sy aard en ontwikkeling, Pretoria: JL van Schaik: 15-66.


Go Afrikaans