Taal en Tongval. Jaargang 48

[p. 35]

Anthony F. Buccini
New Netherland Dutch, Cape Dutch, Afrikaans

1 Introduction

Few topics in historical linguistics have been so passionately debated as the problem of the restructuring of Dutch in Southern Africa and the rise of Afrikaans as an independent language. The central issue has, of course, been what the primary driving force was behind this restructuring and over the past hundred years or so, the process has been ascribed to the effects of virtually all the conceivable factors by one scholar or another. While the topic remains to a degree a source of controversy, progress has clearly been made insofar as the overall number of competing theories has been considerably reduced, leaving essentially only two general views on the genesis of Afrikaans; in addition, it seems that now all interested scholars would agree that language contact was a factor in the genesis of the language.¹ The abiding source of controversy is the issue of the rôle of ‘pidginisation’ and ‘creolisation’ in the process and the two main competing views can, I believe, be characterised thus. On the one hand, there is the view, argued for by an increasing number of linguists in recent years, according to which the development of Dutch in Southern Africa was crucially influenced by pidginised or creolised varieties of the language which developed among the non-European inhabitants of the colony, namely the imported slaves and the indigenous Khoisan peoples.² On the other hand, there is the opposing view, as championed by Raidt (e.g., 1983), according to which the development of Afrikaans was neither especially abrupt nor in any meaningful way affected by the process of pidginisation or creolisation but rather was the result of a

1 A recent insightful review and critical discussion of the various scholarly views is in Roberge (1994: 18ff.). Scholarly work on the origins of Afrikaans is also treated at various points in Ponelis (1993).

2 Outside the literature specifically on Afrikaans and pidgin and creole studies, one notes the acceptance of forms of this view in recent works by scholars concerned with general theoretical models of the dynamics of language contact, namely, Van Coetsem (1988: 129-44) and Thomason and Kaufman (1988: 251-56).

[p. 36]

combination of natural, language-internal processes together with more modest contact-induced simplification or reduction arising from the linguistic absorption of both European and non-European alloglot groups.³

Progress in the study of such a problem as the origins of Afrikaans clearly must be grounded in careful and extensive examination of the relevant philological materials available from Southern Africa, as Raidt (1983: 45-6) has indicated in the course of her critique of other scholars' views on the topic. The analysis of the philological data is itself, however, of little use if not informed by relevant theoretical advances and it is in this regard that the proponents of the view that the development of Afrikaans was at most only modestly affected by language contact themselves become legitimate targets of criticism. Fortunately, some students of Afrikaans have in recent years brought together both empirical and theoretical advances and out of their work there is gradually emerging a more plausible and coherent, albeit more complicated, picture of how the language arose. The central result of this work will be a measured appreciation of the crucial rôle played by non-Europeans and the process of creolisation in the development of Afrikaans as a language structurally distinct from Dutch.

Despite the broadening of the perspective from which the development of Afrikaans is viewed, there remains an important body of data which has been almost completely neglected in the discussion, namely, the evidence for the Dutch spoken in the New Netherland colony in North America. In this paper, I will briefly discuss why the New Netherland Dutch data is relevant to the study of Afrikaans and indicate some specific issues in Afrikaans on which the American evidence sheds light.

3 For a concise and pointed statement, see Raidt (1983: 191), where attempts to discuss features of Afrikaans in connexion with pidgin and creole studies are characterised as ‘Mißverständnisse’. She further states: ‘Obwohl die Endergebnisse des afr. Sprachwandels oft mit. kreol. Kennzeichen übereinstimmen, wurden sie, wie aus dem Quellenstudium einwandfrei hervorgeht, nicht durch Pidgin-, sondern durch Interferenzfaktoren verursacht.’

2 New Netherland Dutch

Although Netherlandicists have been aware of the existence of New Netherland Dutch for some ninety years, the dialect has received remarkably little attention and, with but a few exceptions, those discussions which have appeared have generally been ill-informed and superficial.⁴ This neglect of New Netherland Dutch is surprising on several counts: first, the dialect and

4 Much of the literature on New Netherland Dutch is cited and discussed in Buccini (1995).

[p. 37]

its sociohistorical setting are themselves inherently interesting; second, the formation of the dialect sheds new light on sociolinguistic processes which were taking place in 17th century Holland; and third, given the abiding controversy surrounding Afrikaans, the study of New Netherland Dutch, the dialect of the only major colonial settlement of Dutch-speakers outside of Southern Africa, naturally suggests itself as a point of comparison and contrast with its African counterpart.

New Netherland was located in the Middle Atlantic region of what is now the United States and comprised all of the territory of New Jersey, the eastern parts of New York, Pennsylvania and Delaware and the western part of Connecticut. A permanent Dutch presence was established there in the early 1620's but it was only in the 1640's that a reasonably sizable and fairly stable colonial population developed, largely concentrated in two settlement areas, one around New Amsterdam (modern New York City) and the other around Fort Orange (Albany). At the time of the English conquest of New Netherland in 1664, the colonial population is estimated to have been roughly 9-10,000. In the years following the conquest, immigration from the Netherlands quickly ceased and contacts between the Dutch-speaking community of North America and the Netherlands soon became extremely limited. With English rule and the rapid influx of English speakers to the region, much of the Dutch-speaking population became bilingual and in some areas a complete shift to English soon began: Already by the mid-18th century, Dutch was moribund in New York City and by the early 19th century it had fallen out of use in many other areas (e.g., western Long Island in New York, the Raritan Valley in New Jersey). In two areas, the language continued to be acquired by children into the second half of the 19th century, namely, in Bergen County in northeastern New Jersey and in the central part of eastern New York around the Mohawk and Hudson Rivers.

Attestations of New Netherland Dutch can be grouped in two general categories. The first group includes the indirect evidence to be gleaned from various types of documents (journals, letters, deeds, etc.) dating from the 17th century and, more importantly, the 18th century.⁵ The second group includes materials which were intentionally gathered from and by some of the last native speakers of the colonial dialect from Bergen County.⁶ Despite

5 Such materials from New York State are discussed in Gehring (1973). Similar evidence from New Jersey is briefly discussed in Buccini (1995); I am continuing to gather further early material.

6 Unfortunately, the material published by Van Loon (e.g., 1938) which purports to be from the last stage of the dialect's survival in New York State is unreliable and possibly fraudulent.

[p. 38]

the considerable limitations of both bodies of evidence, the two together give us a coherent and surprisingly detailed picture of many aspects of the dialect.

There are two widespread misconceptions concerning New Netherland Dutch which appear in casual discussions of the dialect in works by historians. The first of these is that the dialect became early on a ‘creole’ (in the vulgar sense) or hotch-potch of Dutch, English and other elements. The second is that the original dialectal basis of New Netherland Dutch was southern and even specifically Flemish and/or Brabants. This latter view has also been expressed by several linguists (e.g., Van Ginneken 1913: 288; Shetter 1958: 247ff.; Stolz 1987: 306). In Buccini (1995), I have argued at length against both of these views. Focussing on the data from the early 20th century for Bergen (County) Dutch, what is most striking about the dialect is its remarkable conservatism despite there probably having been few if any Dutch monolinguals in the community for several generations. This is not to say that English influence is not to be seen, but it is largely limited to the lexicon and more superficial aspects of the grammar and phonology; the core of the dialect remained to the bitter end not only recognisably Dutch in a general sense but recognisably the descendant of a specific kind of Dutch.

On the basis of the late Bergen Dutch evidence, with corroboration from older evidence, the dialectal affinities of New Netherland Dutch can be determined quite clearly. Against the claims of others who have considered the question, my own analysis finds that there is no single feature from any language domain of late Bergen Dutch nor of New Netherland Dutch generally which points unambiguously to a southern Dutch (Flemish, Brabants or even Zeeuws) origin (Buccini 1995: 228ff.). Indeed, those features which have been claimed to be of southern origin are all phonological in nature and, when viewed in proper historical perspective, can be shown to have had analogues in the dialects of Holland and/or Utrecht in the 17th century.⁷ On the other hand, New Netherland Dutch does contain numerous lexical and

7 The ‘southern’ features of New Netherland Dutch are: 1) a low back rounded reflex of long a; 2) a front reflex of g; 3) a bilabial reflex of w. While the first two of these are widespread in southern dialects, they are also well known Utrechts features. The third feature is also now closely associated with southern varieties of Dutch but there is good evidence that the bilabial realisation was still widely known in the northern Netherlands during the 17th century. It should also be noted that the back rounded realisation of long a has also been a feature of Amsterdams. Of course, it is possible that these features' presence in New Netherland Dutch represent dialect internal developments rather than directly inherited metropolitan dialectal features, but while there is no way to judge the phonetic realisation of g and w in early stages of the colonial dialect, there are good indications that long a was a low back rounded vowel already in the period around 1700.

[p. 39]

particularly morpholexical elements that point to a particular affinity with northern Dutch and again specifically Hollands and/or Utrechts dialects and no such southernisms.⁸ A number of details concerning the phonology also point strongly toward a particularly close link between New Netherland Dutch and dialects of the western and central part of the northern Netherlands.⁹

As stated in Buccini (1995: 242), taking all the evidence into consideration, I conclude that New Netherland Dutch is a descendant of 17th century Hollands which bears some central dialect features, the origins of which should be sought primarily in the central dialect area of Utrecht. Flemish appears to have played no rôle in the formation of the dialect and there appears to be no need to invoke any particular Brabants or Zeeuws influence; those features of New Netherland Dutch which have analogues in Brabants and Zeeuws also have analogues in Hollands and Utrechts. I argue further (1995: 253ff.) that there are good reasons to believe that New Netherland Dutch had particularly close links to the area around Amsterdam and the city of Utrecht and quite possibly represents a kind of ‘levelled’ or restructured lower to middle class urban dialect that developed primarily through the interaction of local natives and ‘internal immigrants’ from elsewhere in the northern Netherlands, and especially from the province of Utrecht: in short, the speech of New Amsterdam was in a very real sense an outgrowth of the popular forms of speech to be heard along the canals and alleyways of old Amsterdam.

In saying that New Netherland Dutch likely bore a particularly strong connexion to the speech of Amsterdam, I do not mean to claim that this relationship is necessarily a question of the simple importation of an Amsterdams dialect to the New World. Rather, I would claim that a complex sociolinguistic situation, in some general respects similar to the one which obtained in Amsterdam, was created on a small scale in the settlement at the southern tip of Manhattan. Viewed from a linguistic perspective, the demographic heterogeneity of New Netherland bears this claim out.

8 The following examples of morpholexical elements will suffice here (Bergen Dutch forms are given in normalised orthography): BDu. weze ‘zijn/to be’; wij benne ‘wij zijn/we are’; hœ, hœbbe, hœvve ‘hebben/to have’; kœnne ‘kunnen/to be able’; zel ‘zal/shall’; hulle ‘zij (pl.)/they’; diminutives in -ie, e.g., pakie ‘pakje/pack (dim.)’. For further examples, see Buccini (1995: 241-2).

9 A particularly striking Hollandism which was preserved in 20th century Bergen Dutch is the tripartite development of Gmc. *ai, as exemplified by the forms klein, meer, stien ‘klein, meer, steen/small, more, stone’ (Buccini 1995: 225-6).

[p. 40]

With regard to the situation in Amsterdam and the other urban centres of Holland, it is a well known fact that a significant part of the rapid growth of the western Dutch cities in the 17th century was due to immigration. While linguists have traditionally focussed their attention on the numbers and alleged linguistic impact of refugees from the southern Netherlands, historians have better appreciated the volume and impact of immigration from other areas within the north itself. In this connexion, it appears that internal northern migration was throughout the period especially important in Amsterdam and I have conjectured that the striking lack of growth of the city of Utrecht may indicate that the surplus population of Utrecht province was especially well represented in nearby Amsterdam (Buccini 1995: 255). In addition to speakers of southern Dutch dialects and various northern Dutch dialects, the cities of central and southern Holland also received economic and religious refugees from Germany and French-speaking lands. Given that the cities were not able to absorb all of these immigrants into the mainstream economy, it is not surprising that the immigrant groups were well represented among the colonial population of New Netherland; the West India Company presumably recruited its colonists in the cities (and especially Amsterdam) and had the most success among those on the economic or social fringe.¹⁰

The surviving evidence for the regional origins (and, indirectly then, for the linguistic background) of the New Netherland colonists are incomplete but enough evidence does survive to give us a rough idea of the population's makeup.¹¹ According to the most detailed recent study of the topic, Cohen (1981) found that of 904 non-English, European immigrants to New Netherland during the 17th century, only 46.6% were from Dutch-speaking territories. Of these colonists, there were in addition 26.8% from Low German- and Frisian-speaking areas, 10.5% from French-speaking territories, 9.1% from Scandinavia, and 4.6% from High German-speaking regions. If we also take into consideration the presence of the English-speakers left out of Cohen's study, it seems reasonable to conclude that the colonial population travelling to New Netherland was probably at most only about 50%

10 A thorough study of the existing evidence for the recruiting practices of both the WIC and the VOC is a pressing desideratum.

11 The following discussion of the origins of European colonists in New Netherland is based primarily on Cohen (1981) and Rink (1981, 1986), though also informed by wide reading on the history of New Netherland. For further discussion, see Buccini (1995: 242 ff.). Note that while Cohen's data base was necessarily rather limited, his findings probably give a good, representative cross-section of the colonial population.

[p. 41]

originally from Dutch-speaking territory but quite possibly the Dutch element constituted a bit less than 50%.¹²

Considering the likely dialectal background of the colonists from Dutch-speaking regions, we find the following.¹³ Of Cohen's 904 colonists, 355 were presumably native Dutch-speakers and of these, there were in turn approximately 180, that is, ca. 50%, from North and South Holland, with 90, that is, ca. 25% of the native Dutch, coming originally from Amsterdam. In contrast, the numbers of colonists from more southern dialect areas which have often been claimed to have been the foundation of New Netherland Dutch are relatively small: Zeeland 18 (ca. 5%); Flanders 6 (ca 1.7%); Brabant 24 (ca. 6.7%). Even adding to this the 18 colonists from Cohen's study who are from unidentified places in the Spanish Netherlands (including presumably Limburg), the total of this by no means linguistically homogeneous group of southerners is but 66 or ca. 18.7% of the native Dutch speakers. There were, however, more substantial representations from the areas to the east and southeast of Amsterdam: 't Gooi 13; western Utrecht 18; unidentified localities in Utrecht province 25; eastern Utrecht and western Gelderland 40. In addition, a further 44 colonists came from the area between the Lek and Waal rivers from Gorinchem to Tiel just to the south of Utrecht province. While together these groups obviously did not form a dialectally homogeneous group, many specific dialectal features were shared by larger or smaller subgroupings of them.

If we consider these demographic facts in light of the dialectological analysis presented in Buccini 1995, it is not surprising that no specifically southern features are found in New Netherland Dutch. The demographic facts lend, moreover, support to the claim that the colonial dialect was in essence a form of Hollands, with elements from the neighbouring central dialect of Utrecht. Nevertheless, it is certain that the native Dutch-speaking colonial population under Dutch rule was dialectally mixed; consequently, it

12 Concerning the English presence, I refer here not to the purely (or largely so) English settlements within New Netherland territory but rather to those Englishmen who gradually filtered into the Dutch settlements.
Note that, from a linguistic standpoint, such demographic analyses cannot take into account the degree to which colonists with origins ultimately outside Dutch-speaking territory may have acquired Dutch to some degree before departing for New Netherland. Note too that Low German-speakers (and possibly to be included herewith are some of the Scandinavians) were linguistically-speaking in much the same boat as Dutch dialect speakers from peripheral regions of the Low Countries.

13 See further Buccini (1995: 244ff.). It is assumed here that a given colonist spoke a variety of Dutch typical for his region of origin; obviously, this assumption is surely something of a distortion but, alas, an unavoidable one given the limitations of the data.

[p. 42]

is equally certain that there took place within the colony itself a process of dialect ‘levelling’ or restructuring, akin to (and also possibly affected by) the process of standardisation, a process which in New Netherland generally favoured features from central Holland and western Utrecht. A very similar process was surely also taking place in Amsterdam during the 17th century. Equally certain is the fact that to some degree at least, the colonial population included very considerable numbers of people for whom no form of Dutch was native. Yet, the Germans, French, Scandinavians and others who took up residence among the Dutch in New Netherland were not only linguistically absorbed but they were linguistically absorbed without affecting in any discernible way the structure of the colonial dialect. As a result of this linguistic absorption, New Netherland Dutch underwent no radical grammatical simplification, no phonological skewing, nor even any significant lexical influence: if we were to judge only on the basis of the 18th century texts and the late Bergen Dutch materials, we would never suspect that so many of the founders of the Dutch-speaking population were themselves not originally Dutch speakers.

3 New Netherland Dutch and Cape Dutch

If one considers the figures which have been proposed for the regional and national origins of the early European colonists to the Cape of Good Hope alongside those presented above for New Netherland, the similarity is, to say the least, striking. Concerning the constitution of the colonial population in Southern Africa, Ponelis (1993: 9) indicates:

The total of adult free men with known origin between 1657 and 1707 was 1059. The main population element in the sample were the Dutch immigrants (473/1059; 45%), who with the Cape born (a small fraction at 73/1059; 7%), formed a Dutch bloc of 52%. The German men numbered 357 (34%), and the French 90 (8%).

For New Netherland, the figures from Cohen's data are 46.6% Dutch, 32.8% German (Low, High and not localisable combined) and 10.5% French. These similar demographic figures were surely neither the result of blind chance nor planning on the part of the WIC and VOC; rather, they are a consequence of the social conditions which prevailed in Holland, where the colonial populations of both New Netherland and the Cape Colony were recruited. The agreement is further reinforced if we consider the figures for localisable Dutch progenitors at the Cape from 1657 to 1820; out of 600, 301 or 50% were from North and South Holland, with 170 or 28% from Amsterdam (Ponelis 1993: 122); these figures match almost perfectly those derived from Cohen's New Netherland data (of Dutch settlers, ca. 50% from North

[p. 43]

and South Holland and ca. 25% from Amsterdam). While the limitations of the data allow for the possibility that all of these figures are to some degree inaccurate, the remarkable agreement across the two colonies strongly suggests that the data sets are representative and reasonably reliable.

As with New Netherland Dutch, a number of writers have claimed that Afrikaans has particularly strong affinities to southern Dutch dialects; fortunately, such claims with regard to a particular Flemish, Zeeuws or Brabants connexion have long been recognised to be groundless. In the most comprehensive treatment of the subject, Kloeke (1950) comes to the conclusion that Afrikaans is especially closely linked to the dialects of South Holland, a finding which is obviously somewhat at odds with my own conclusions concerning New Netherland Dutch and Afrikaans. Given the likelihood that the demographics of the two colonial populations were roughly the same, the discrepancy between Kloeke's and my own findings warrants concern. In this regard, I call attention to a more recent consideration of the dialectal affinities of Afrikaans, namely that of Ponelis (1993: 121ff.). In a short but measured and well documented discussion (including a critique of Kloeke's view), Ponelis indicates that the dialectal roots of Afrikaans are Hollands but also indicates that there is a particular tie to the Amsterdam dialect specifically (p. 122).¹⁴ Ponelis (1993: 128-9) also calls attention to the limitations of a ‘purely regional interpretation of the Dutch base of Afrikaans’ and the importance of consideration of the rôle of ‘an incipient urban Holland koine.’ In Buccini (1995: 256-7), I argue that a number of features of New Netherland Dutch, which contrast with features of either the Hollands or Utrechts dialects are to be explained precisely as results of interaction of the various dialect speakers, without necessarily involving direct influence from the cultivated variety of Dutch at the time:

A number of typical and widespread Hollands and Utrechts dialect features are, however, missing in the colonial dialect... For example, with regard to the consonantism, it seems that those dialectal features which were not generally shared by the dialects of North Holland, South Holland, and Utrecht were not retained in the colonial dialect. Thus, the reduced form (e- or Ø-) of the prefix ge- which was so widespread in the Hollands dialects lost out to the full form of Utrecht (and possibly also the ‘levelled’ dialects of the cities of central Holland). Another widespread consonantal feature of Hollands which New Netherland Dutch ignored is the velarized nasal, as in hongt ‘hond/dog’ (BD hōnt); again, this feature was not present in the

14 Note that the basic text of Buccini (1995) dates to 1993, before Ponelis (1993) was yet available to me; most of the material in that paper was presented in a series of conference papers during 1991-2. Thus, my conclusions concerning the dialectal affinities of New Netherland Dutch and Cape Dutch were arrived at independently. That the same conclusion was arrived at from very different directions lends, I believe, support to its validity.

[p. 44]

dialects of Utrecht. Similarly, the apocope of -t, which was certainly a characteristic of Utrechts and some of the South Hollands dialects, was disfavoured and the colonial dialect, like most dialects in North Holland, has maintained final -t. In light of these facts, it is not surprising that dialectal features with more limited geographical (and sociolectal?) distributions, such as the maintenance of sk- vs. the development to sx-, leave no traces in New Netherland Dutch.

It will of course be immediately noticed that, while New Netherland Dutch and Afrikaans agree in their rejection of the loss of the ge- prefix and the Hollands velarised nasal, they disagree with respect to the apocope of the final -t in clusters and the cluster sch- vs. sk-. To these issues we will return below.

Given the space restrictions for this article, it will not be possible to review the numerous correspondences between New Netherland Dutch and Cape Dutch/Afrikaans but from a perusal of the New Netherland data presented in my 1995 article, it should be readily apparent that the two colonial varieties are sprung from essentially common origins, at least with respect to European parentage; in light of the demographic evidence discussed above, such a conclusion is inescapable. Yet, New Netherland Dutch, even in its 20th century Bergen Dutch form, after two centuries of contact with English, remained indisputably a form of Dutch, while Cape Dutch evolved into a structurally distinct language. Insofar as scholars of Afrikaans have and still do place considerable importance on the sizable non-Dutch European population of the Cape Colony in the restructuring of Cape Dutch, the fact that the European population of New Netherland was equally heterogeneous and even had roughly the same mixture of non-Dutch speakers, should indicate that the presence of German- and French-based xenolectal varieties in the colony, in and of itself, cannot have been a crucial factor in the restructuring. The divergent paths of linguistic development in the two colonies must therefore be accounted for either by sociolinguistic dynamics within the European colonial community on the Cape radically different from those in America or by differing relations between Europeans and non-Europeans in the two colonies.

For possible differences between the European colonial societies, one can perhaps point to the relative importance of religion. By all accounts, the presence of the Reformed Church in the Cape Colony was not especially strong for much of the colony's early history, while among the Dutch community in North America, religious life came to be an extremely important issue, albeit a divisive one with ties to political, ethnic and linguistic tensions. Yet, religion among the American Dutch seems only to have become a central aspect of community life toward the end of the 17th century, after

[p. 45]

(and likely in some sense connected to) the English conquest; during most of the period when the New Netherland dialect was taking shape, roughly from 1640 to 1690, opportunities for organised worship and religious instruction were extremely limited. Similarly, organised education for the Dutch community in America only developed to an appreciable degree well after the English conquest. Consequently, while religious and educational opportunities in New Netherland may have been somewhat better than in the Cape Colony, any significant differences in these regards arose in the early 18th century, after the New Netherland Dutch dialect had stabilised and was already coming under pressure from English. A further possible difference which might be mentioned here involves the volume of maritime traffic arriving at the chief settlements of the two colonies, Cape Town and New Amsterdam; presumably, Cape Town saw considerably more activity than New Amsterdam, given its position on the trade route between the East Indies and Europe. Finally, and perhaps most importantly, there seems to have been a contrast between the degree to which the two colonies' European populations remained geographically concentrated. In New Netherland, the colonial population remained strongly concentrated in two areas, one around New Amsterdam and the other around Fort Orange; in both areas, colonists outside the two towns developed more or less compact agricultural settlements and ultimately new villages and towns.¹⁵ In Southern Africa, the colonial population was also largely concentrated in and around a few major settlements (Cape Town, Stellenbosch, Drakenstein) during the early years of the colony, but around 1710, a part of the colonial population began to expand well beyond the area of relatively compact settlement, taking on a pastoral and even nomadic lifestyle.¹⁶ No such diffusion of colonists over wide expanses of territory ever took place in New Netherland under Dutch rule nor in the following period under English rule; when expansion finally came, it came in the form of the establishment of new, fairly compact settlement areas centered on towns (e.g., New Brunswick and the Raritan Valley in New Jersey).

While there appear to have been some differences of social and cultural organisation among the Europeans which could have had an impact on linguistic developments, these pale in importance alongside the one major demographic difference between the colonies and likely only gain significance in conjunction with that demographic difference. This major demo-

15 For an overview of the history of New Netherland, see Buccini (1995: 213ff.), with further bibliographical references.

16 See, for example, Ponelis (1993: 37-8), with further references.

[p. 46]

graphic difference involved the presence of non-European groups in the colonial settlements and thus also the nature and frequency of contacts between Europeans and non-Europeans. In Southern Africa, a slave population of diverse origin was present in the core area of the colony almost from the start and, while the number of company slaves remained fairly small, the slave population belonging to the burghers grew rapidly in the beginning of the 18th century, soon equalling and then surpassing in size the burgher population itself (Armstrong 1979: 91). Along with the trekboers who expanded into the interior areas after 1710 went small numbers of slaves but, given that trekboer families lived to a considerable degree in isolation from one another but in regular contact with their own slaves, the frequency of interaction between trekboers and slaves must have been relatively high. Probably of greater linguistic importance was the involvement of Khoisan peoples in the developments on the colonial frontier during this period; with the disintegration of Khoi society under multiple pressures including European expansion and disease, together with an increasing need of the colonists for cheap labour, large numbers of Khoi came to be employed by the Europeans (Elphick 1979: 19ff.). For the trekboers, who could not afford many slaves, the Khoi were an available source of labour. This development affected not just the fringes of the colony but also the agricultural settlements of the core area (around Stellenbosch and Drakenstein) and this widespread inclusion and, in at least a certain limited sense, integration of large numbers of Khoi into certain spheres of colonial life cannot have been without linguistic repercussions.¹⁷ However limited our attestations of it may be and whatever individual scholars may judge its rôle to have been in the development of Afrikaans, a creolised variety of the colonial Dutch of the Cape Colony clearly did exist in the 18th century among (at least significant parts of) the slave and Khoi populations.

With regard to interactions between Europeans and non-Europeans in New Netherland, conditions in the American colony contrast sharply with those in Southern Africa. There was from early on an African slave population in New Netherland and ownership of slaves in New Netherland was, moreover, not limited to the elite of society but extended to a considerable number of modestly well-off colonists. Aside from the West India Company itself, however, there seem not to have been any individual major slaveholders and the overall percentage of slaves in the population, though signifi-

17 Indeed, Ponelis (1993: 39) claims that ‘the incorporation of the Khoi was the single most important demographic contribution to the Afrikaans speech community during the eighteenth century.’

[p. 47]

cant, never approached the levels attained already in the early 18th century in the Cape Colony: the estimated population of New Amsterdam at the time of the English conquest in 1664 was 1,500 European colonists with approximately 300 slaves and 75 freed slaves (Goodfriend 1992: 13). While farmers living outside of New Amsterdam often did own a few slaves, these farmers lived in the proximity of many other Dutch-speakers, as mentioned above, and the social conditions for interaction between colonists and slaves which developed among the trekboers did not prevail anywhere in New Netherland. Given the limited number of slaves and the concentration and cohesion of the European population, it is not surprising that the Africans exercised no obvious influence on the development of the local Dutch dialect.¹⁸

Relations between the Europeans of New Netherland and the indigenous Algonquian and Iroquoian groups with whom they were in contact also differed in important ways from those between the colonists and Khoi in Southern Africa. One particularly striking difference is the willingness of the Dutch in New Netherland to attempt to learn the local languages, which in the southern part of the colony resulted in the formation of a contact variety based on the indigenous language; this was a xenolectal form of the (Algonquian) Delaware language commonly called Pidgin Delaware (Buccini in press). While some Indians surely learned Dutch to some degree and there may even ultimately have developed stabilised xenolectal varieties (i.e., ‘Delaware Dutch’ or ‘Mohawk Dutch’), the relatively numerous attestations of Pidgin Delaware indicate that even some Indians who had regular contact with the Dutch had perhaps not acquired the colonial language to any useful degree. In any event, it is striking that, whereas from the Cape Colony there are many early citations of ‘broken Dutch’ from the mouths of non-Europeans, in New Netherland we more often find evidence of ‘broken Delaware’.

As in Southern Africa, the native peoples of New Netherland came under intense pressures, both directly and indirectly, as a result of the colonial presence. For the Iroquoians of New York State, however, the Dutch, as allies and trading partners, were also in a sense a benefit. Nevertheless, the Iroquoians maintained strongly both their social and political structures and their own cultural and linguistic identity; the Mohawk and the Dutch of Fort Orange/Albany and Schenectady were neighbours who interacted but their

18 Evidence for the linguistic behaviour of the Africans in New Netherland is extremely sparse. It should be noted, however, that the free-black population of New Amsterdam apparently assimilated to Dutch ways to a considerable degree but there are some indications that there existed a particular (creolised?) dialect of Dutch which they spoke within their own community (Prince 1910: 460, 468).

[p. 48]

interactions also remained limited. For the Algonquian peoples of the Lower Hudson region, the Dutch presence had more drastic and negative effects. Yet, when the new conditions effectively destroyed their ability to carry on their traditional ways, those who survived the epidemics and occasional warfare for the most part emigrated to the west or north. Of those who remained near major Dutch settlements toward the end of the 17th century, some did in effect become marginal members of colonial society, but there was no widespread incorporation of Indians as an underclass of labourers.

4 Conclusion

In the foregoing discussion, I hope to have to shown that in all reasonable likelihood the divergence of the developments of New Netherland Dutch and Cape Dutch is crucially linked to the differing social and linguistic relations between Europeans and non-Europeans in the two colonies. Insofar as the rôle of non-Dutch-speaking Europeans is concerned, I would not rule out the possibility that deviant features in their Dutch may have had a greater impact on the overall linguistic developments in Southern Africa than in North America, but I would argue that such a greater influence probably could only have developed in the context of, and so was in a sense dependent upon, the presence of the non-European groups who needed to acquire Dutch to some degree. From the philological evidence and the evidence of modern non-white varieties of Afrikaans, it is beyond doubt that creolised varieties of Dutch developed among the non-Europeans in Southern Africa. The existence of such varieties in the Cape Colony and their absence in New Netherland is the one major difference in the linguistic environments of these colonies and surely is the crucial factor in their divergent linguistic histories.

Viewed from this perspective, some specific features of Afrikaans, which have been explained simply as the reflexes of Dutch dialectal forms that gained general currency in the Cape Colony, need to be reconsidered. Here I can only touch on a few of these.

First, I would like to turn back to two of the dialectal features mentioned above in section 3; the initial cluster sk- and the apocope of -t. These features were indisputably present in the speech of some of the Dutch settlers in both of the colonies and therefore it cannot be ruled out that their success in Afrikaans is owed to the selection and generalisation of these within the native Dutch or European population. Yet, if Afrikaans grew out of an urban Hollands koine or even just involved a parallel process of koine-formation, it seems odd that these features, rejected in standard Dutch, rejected in the

[p. 49]

metropolitan urban koine, and rejected in New Netherland Dutch, took root on the Cape. Sober consideration must lead to the conclusion that in all probability the occurrence of these features in Afrikaans is not the result of an uncharacteristic selection of fairly highly marked dialectal features among the Dutch themselves but rather at most a selection of the most easily acquired variants by non-Dutch speakers; quite possibly, they represent imposed phonological features stemming from xenolectal varieties. Other such possible dialectisms which appear in Afrikaans but not in New Netherland Dutch deserve to be reconsidered carefully.

Similarly, the grammatical features which distinguish Afrikaans from Dutch deserve also to be considered in light of the New Netherland Dutch evidence. For example, New Netherland Dutch, at least in its later stages, (and then probably in connexion with the contact with English) gradually lost some morphological distinctions, such as nominal gender and the use of zijn as a perfect auxiliary. On the other hand, such striking features of Afrikaans as the wholesale elimination of the strong/weak verb distinction and the loss of pronominal and adverbial clitics have no parallel in the American dialect. Given that these drastic reductions were absent from New Netherland Dutch and also apparently from the acrolectal colonial variety of Cape Town, as represented in Duminy's journal of 1797 (Roberge 1994: 60ff.), it would be gravely misguided to attribute such reductions to the combined effects of language internal tendencies and composite ‘interference factors’. First, the loss of clitics can quite obviously not be considered a language internal tendency for Dutch. Second, the loss of the irregularities is something of a language internal tendency, but the thoroughness and relative speed with which strong verb forms were eliminated in Afrikaans cannot be accounted for by simple invocation of the bland and poorly defined notion of interference. These reductions can, however, be explained in terms of strongly imperfect language acquisition, that is, the process that lies at the core of the notions of pidginisation and creolisation.

In this paper I have tried to indicate in a preliminary way some areas in which a comparison of New Netherland Dutch and Cape Dutch/Afrikaans can yield a useful perspective on the extremely complicated linguistic relations and developments in the Cape Colony. Further detailed investigation of these issues should form a component of the ongoing discussion of precisely which developments were likely tied to the interaction of more conservative forms of Cape Dutch with strongly xenolectal and creolised varieties.

[p. 50]

References

Armstrong, James C.
	1979	‘The slaves, 1652-1795.’ In Richard Elphick and Herman Giliomee (eds.), The shaping of South African society, 1652-1820, 75-115. Cape Town: Longman.
Buccini, Anthony F.
	1995	‘The dialectal origins of New Netherland Dutch.’ In Thomas F. Shannon and Johan P. Snapper (eds.), The Berkeley Conference on Dutch Linguistics 1993, 211-63. Lanham, MD: University Press of America.
		In press. ‘Swannekens ende wilden: linguistic attitudes and communication strategies among the Dutch and Indians in New Netherland.’ In Anneke Prins and Bettina Brandt (eds.), Publications of the American Association for Netherlandic Studies, 11.
Coetsem, Frans van
	1988	Loan phonology and the two transfer types in language contact. Dordrecht: Foris.
Cohen, David Steven
	1981	‘How Dutch were the Dutch of New Netherland.’ New York History 62: 43-60.
Elphick, Richard
	1979	‘The Khoisan to c. 1770.’ In Richard Elphick and Herman Giliomee (eds.), The shaping of South African society, 1652-1820, 3-40. Cape Town: Longman.
Gehring, Charles T.
	1973	The Dutch language in Colonial New York: An investigation of a language in decline and relationship to social change. Ph.D. dissertation, Indiana University. Ann Arbor: University Microfilms.
Ginneken, J. van
	1913	Handboek der Nederlandsche taal. Part I. De sociologische structuur der Nederlandsche taal. Nijmegen: Malmberg.
Goodfriend, Joyce D.
	1992	Before the melting pot. Society and culture in Colonial New York City, 1664-1730. Princeton: Princeton University.
Kloeke, G.G.
	1950	Herkomst en groei van het Afrikaans. Leiden: Universitaire Pers.
Loon, L.G. van
	1938	Crumbs from an old Dutch closet. The Dutch dialect of old New York. The Hague: Nijhoff.
Ponelis, Fritz
	1993	The development of Afrikaans. (Duisburger Arbeiten zur Sprach- und Kulturwissenschaft 18.) Frankfurt am Main: Peter Lang.
Prince, J. Dyneley
	1910	‘The Jersey Dutch dialect.’ Dialect Notes 3: 459-484.
Raidt, Edith H.
	1983	Einführung in Geschichte und Struktur des Afrikaans. Darmstadt: Wissenschaftliche Buchgesellschaft.

[p. 51]

Rink, Oliver A.
	1981	‘The people of New Netherland: Notes on non-English immigration to New York in the seventeenth century.’ New York History 62: 5-42.
	1986	Holland on the Hudson. An economic and social history of Dutch New York. Ithaca: Cornell University Press.
Roberge, Paul T.
	1994	The formation of Afrikaans. (Stellenbosch Papers in Linguistics Plus 23.) Stellenbosch: Dept. of General Linguistics, Univ. of Stellenbosch.
Shetter, William Z.
	1958	‘A final word on Jersey Dutch.’ American Speech 33: 243-251.
Stolz, Thomas
	1987	‘Kreolistik und Germanistik: Niederländisch-basierte Sprachformen in Übersee.’ LBe 110: 283-318.
Thomason, Sarah Grey, & Terrence Kaufman
	1988	Language contact, creolization, and genetic linguistics. Berkeley: University of California Press.