Appendix 1
The Origin of Surinam Creole

Acreole language must by definition have its origin in a pidgin language (Hall 1966:xii). Apidgin originates in a situation in which no communication is possible in any of the existing mother tongues. In these circumstances, people try to communicate in one of the available languages (usually the language of the power group), although they have no opportunity to acquire a good command of it. A reduced, simplified language arises that is used only in restricted situations: in trading, in the market, in the factory, on the plantation, or in the army. In all other situations - at home, in religious ceremonies, in court, in education - the mother tongue is still used. Thus a pidgin has a restricted use and may be regarded as a reduced language when compared to the model from which it arose.

The process of reduction can be studied if one knows both the model language and its reduced pidgin form. The trade language employed between Trio Indians and Djuka bushnegroes can be regarded as a pidginized form of the creole mother tongue of the Djuka. Not much is known about this pidgin. De Goeje (1908:204-19) gives a short word list and some expressions. These few examples show that the vocabulary of the model language has been reduced in a more or less systematic way. The Djuka interrogativesama (who?) broadened its range of application to include also ‘what?’, ‘which?’, ‘when?’, and ‘why?’. The Djuka demonstrativedisi (this) is also used as the third person singular pronoun and as the adverb ‘here.’ The unmarked member of a pair of relational terms is taken from Djuka (in the case of adjectives with the addition of an ‘essive’ Trio suffix =mee), and the marked member adds the Trio negationwa (‘no,’ ‘not’):

It seems as if the agents of the language change (the Trio in this case) reduced the vocabulary of Djuka at the expense of grammatical morphemes, which they took from their own language. This must be the basis of the concept of language mixing so often referred to in connection with pidgins.

A creole language develops from a pidgin as it becomes the native language of a group. If people speaking different mother tongues begin to intermarry and form a single community (as was the case, for example, when slaves of different origins were brought together on a plantation), and if a particular pidgin is the common language of this community, it will become the mother tongue of the children of mixed-language marriages and eventually of the whole community (though not of groups who do not intermarry with the others, such as the white settlers). As soon as this pidgin becomes the native language of a group, it must be used in many more situations than when it was only a contact language between people who had their own native languages, and we may thus assume that it undergoes a process of expansion to meet the new demands on it.

It is not possible to prove the reality of such a process in a direct way, because with the promotion of a pidgin to a creole status the pidgin ceases to exist. We can, however, find indirect evidence of it if, during the period of creolization, the speakers no longer had access to the original model language of the pidgin. This might be the cause of a remarkable distribution in the Zambuangeño lexicon of Spanish- and Philippine-based elements. Frake (1971) was struck by the fact that in this language a significantly greater number of unmarked lexical items had a Spanish origin and a significantly greater number of marked lexical items a Philippine origin. One must assume that, during the pidgin stage of Zambuangueño, the lexicon contained mainly unmarked lexical items and during the process of creolization, when the speakers no longer had access to the Spanish model, the marked items were added.

It is not, or not yet, possible to prove that the same development happened in Surinam Creole, although there is reason to believe that this language had a Portuguese pidgin origin and was creolized in contact with English.

Surinam Creole (also called Negro-English) is clearly an English-based Creole. In studying the origin of a large group of Creole verbs, the following results were obtained (Voorhoeve 1970):

The influence of English is greater than this table suggests. The verbs of English origin are often very basic verbs, e.g. ‘go,’ ‘come,’ ‘walk,’ ‘talk,’ ‘sleep,’ ‘want,’ etc., certainly more so than the verbs of

Dutch origin. Verbs of African origin may also be basic, e.g.nyam ‘to eat,’fom ‘to beat,’fufuru ‘to steal,’koso ‘to cough,’gonggosa ‘to gossip.’ The same holds true for verbs of Portuguese origin, e.g.sabi ‘to know,’kaba ‘to finish,’pina ‘to be poor,’pasa ‘to pass.’

Surinam was under British rule for only sixteen years (1651-67). It is not absolutely necessary for the language of administration to serve as a model in the development of a pidgin or creole. The creole of Trinidad is French-based, although Trinidad has never been under French rule (Thomas 1869). The early history is of extreme importance. This type of language is used to try to bridge a communication gap between different components of the population. Once established, the language survives as long as it serves a useful purpose. It is therefore important to know as precisely as possible the growth of the population and the relative size of the different components during the formative period. The documents are often not very clear, and a certain amount of conjecture is necessary. Let me try to summarize the main facts (see also the population figures in the introduction, p. 3):

In 1661 there must have been about 1000 Europeans in the colony. In 1665 an epidemic eliminated many. Two hundred settlers left the colony that year, but 200 Portuguese Jewish planters from Cayenne, possibly with their own slaves, settled in Surinam. The colony was captured by the Dutch in 1667, recaptured by Barbados that same year, but handed back to the Dutch in 1668 in accordance with the terms of the peace treaty of 1667. Before Surinam was handed over, 67 of the most important English planters emigrated from the colony with 412 slaves. In 1671 a group of 517 people left the colony, followed in 1675 by a group of 250 whites with 980 slaves. In 1680 the last group of 102 Englishmen and slaves left, leaving only 39 Englishmen behind. The English planters were not allowed to take the slaves acquired under Dutch rule out of the colony, which meant that the ones who left were mainly the old slaves.

There are not many clues to the language of the slaves when they arrived from Africa. Slaves as far south as Angola were probably among the first groups (Daeleman 1973). The main contact language on the African coast in these early days was a Portuguese pidgin, although its influence was already declining in the seventeenth century. A distinction between old slaves (acquired before 1667) and new slaves (acquired after that date) seems important. After 1667 a rapidly increasing number of new slaves was brought in, and they soon outnumbered the old ones; this was already the case in 1671. The de-

parture of the experienced slaves with the best command of English after 1671 forces us to assume that the original Surinam pidgin had already become firmly established among the slaves by about 1675. If this had not been so, it would have been impossible for the old slaves to pass on an English heritage to the new ones.

This short formation period (1651-80) might give rise to the hypothesis that an English pidgin had already been acquired by the slaves in Africa. There is no historical support for the existence of an English pidgin in Africa at such an early date, however, whereas the existence of a Portuguese pidgin has been established beyond doubt. Moreover, the creole of Surinam shows clear cases of early Portuguese elements. It is therefore worthwhile to consider the importance of early Portuguese influence.

The Portuguese Component

The existence of Portuguese elements in Surinam Creole is a problem, especially because of the preponderance of Portuguese elements in the language of the Saramaccan bushnegroes, the descendants of runaways to the central part of Surinam. In that language the Portuguese component is almost as strong as the English one. We must therefore consider the rather early Portuguese influence, which has not yet been accounted for by prevailing theories.

For a long time people have been satisfied to attribute the Portuguese influence to the presence in Surinam of Portuguese Jewish planters. Herskovits, however, pointed out that typical Jewish terms such astrefu (food taboo) andkaseri (ritually clean, kosher) are found in the coastal Creole, the language of the descendants of slaves, but not in Saramaccan Creole, the language of the descendants of the runaways (Herskovits 1930/31). There thus appears to have been no Jewish influence on the language that shows the strongest Portuguese influence, so the problem of the origin of the Portuguese influence remains. Herskovits also pointed out that newly imported slaves probably had the strongest tendency to escape to the bush, where they gathered in runaway tribes. These runaways did not get the opportunity to adapt themselves to the prevailing slave language spoken on the plantations. The English items in their vocabulary can easily be explained by the influence of their stay on the plantations or by later contacts with coastal Creole, especially after emancipation. The Portuguese items can have been adopted only in Africa.

Portuguese pidgin was in fact the most common contact language on the African coast during the seventeenth century. The slaves must

have had contact with this language, even if it was only after they were transported to the African coast to be sold to European traders. They took this language with them to the New World. It might even have been the only means of communication with their fellow slaves and the crew on the ships. However, the same language was not used on the plantations in Surinam. Thus, after their arrival, the new slaves had to adopt the English-based pidgin, probably already in the process of becoming a creole.

If the Portuguese components in Saramaccan Creole are ascribed to the African period, then the few basic Portuguese items in Surinam Creole could possibly also have found their way into the language in essentially the same manner. This, in my opinion, would mean that one can no longer take it for granted that Surinam Creole originated only from an English-based pidgin. The oldest source was in any case an Afro-Portuguese pidgin, which might have developed into the English-based creole directly or via an intermediate English-based pidgin stage on the plantations. The whole process must have taken place in a remarkably short period of time (between 1651 and 1680). Thus it is possible that an English-based pidgin was not spoken by the newer slaves in Surinam but that they adapted their Portuguese pidgin to the language of their masters, and of the earlier slaves, and at the same time converted it to a more general use as their own mother tongue. This process by which a pidgin develops into a pidgin or creole with a different base may be termedrelexification. The historical and linguistic arguments in favor of the relexification theory have been summarized in Voorhoeve 1973. Most of the Portuguese elements in the pidgin of the slaves were, as a result of this process, replaced by English elements without any great changes in the grammatical structure of the language.

The relexification theory explains the striking similarities in the structure of creole languages all over the world, which are based on different models (Thompson 1961). Taylor (1960) has drawn up quite a convincing list of these similarities. One of the most striking examples may be found in the verbal system of creoles. In Surinam Creole all (or nearly all) verbal tenses are formed by means of three verbal particles placed between the subject and the main verb:ben (past tense),sa (future tense or hypothetical), and(d)e imperfect tense or durative). These three particles, alone or in combination with each other, account for eight different verbal tenses and/or aspects. These eight tenses can be tabulated as follows (Voorhoeve 1957(a)).

There is no detailed description of all the creole languages on the basis of which a comparison of the verbal systems is possible. However, the verbal system always seems to operate essentially on the basis of three particles with roughly the same values, although the particles themselves have different forms and thus must have quite different origins:

The similarities in the basic system are such that a common origin seems plausible. This common origin is not always reflected in the form of the particles but is hardly likely to be different from the trade language of the Portuguese sailors, widely used by different nations until the seventeenth century. This language was relexified in different directions, giving rise to differently based creole languages.

Evidence from Older Stages of Surinam Creole

Relexification of a more limited scope is a rather common phenomenon. In all languages, isolated lexical items are replaced by new ones with a different origin without affecting the grammatical structure of the language. This process is usually called borrowing, but in creole languages borrowing becomes such a massive phenomenon that it may completely change the affiliation of the language. It is possible to demonstrate the phenomenon in Surinam Creole, as we have access to old documents in the language. The oldest text, inBeschrijvinge van de

volksplantinge Zuriname (Description of the Colony of Surinam), was published by J.D. Herlein in 1718. It is a curious document. It must have been constructed or elicited by a European, presumably the author himself, in any case by someone not aware of the subtle ways in which Surinamese Creoles show respect in their linguistic behavior. For example, the use of second or third person singular pronouns to refer to people of higher social position or people with whom one is not on intimate terms must often be avoided in Surinam Creole. This text used these forms freely and would therefore probably be jarring to modern Creole ears. It gives the impression of a European speaking with total disregard for polite Creole usage. It is possible that the author elicited the text from a Creole speaker, but in that case he has put the text consciously or unconsciously in a European mouth. It is highly unlikely that a Creole would have used such forms in everyday language. The lack of good manners should rather be attributed to the European author.

The text of 1718 has been criticized by Jan Nepveu, governor of Surinam from 1768 to 1779, in a manuscript of 1765. Nepveu made a few direct criticisms, mostly introduced by: ‘Creoles would rather say ...’. But he also added some vocabulary items and expressions that contain indirect criticisms of the 1718 text. Where Nepveu used different words for the Herlein items, we may assume that the Nepveu items were the normal ones in 1765. We reproduce here the original text from 1718 in the first column. The second column contains the corrections that can be deduced from the Nepveu manuscript of 1765. Corrections are introduced by a colon. If only one work in the Herlein text is corrected, this word is repeated in italics to the left of the colon. The third column presents the same text in modern Creole, between square brackets if the text lacks the necessary refinement. In that case an alternative polite rendering is added in parentheses. The fourth column presents as literal a modern English translation as possible. Italics in the original text draw attention to possible corruptions and printing errors. All translations of the original Creole text are based on the Dutch translation that accompanied it. For a later reprinting of the 1718 text with a Dutch translation, see Schuchardt 1914. For a reprinting with a literal English translation, see Rens 1953.

A careful comparison of the texts reveals that for a considerable period of time alternative possibilities of Portuguese and English origin coexisted in Creole. The clearest case is the pairbun/P. ‘bom’ vs.belwel / E. ‘very well,’ of which only the Portuguese-derived item survives in modern Creole. The Portuguese-derived itemgrande did not survive as a regular adverb, ‘very,’ or adjective, ‘big,’ but it is still present in composite nouns asgrantangi (many thanks),granmasra (big master),granman (governor),granmama (grandmother), in the composite adverbgranwe (long ago, in 1765 still found asgrandi wey), and maybe also in the verbgrani (to be old). The itemgrandebon may even have survived in the wordgranbun (holy communion). In modern Creole the regular adverbheri / D. ‘heel’ for the meaning ‘very’ can also be rendered byfu tu / E. ‘for true.’ The 1765 text contains the English-derived itemkweti / E. ‘quite’ without indications as to its use. It is possible that this item replacedgrande and survived for some time. It is still used in modern Creole but only in negative sentences. The pairkaba / P. ‘acabar’ anda reddi / E. ‘already’ also shows that the Portuguese item survived in modern Creole, although the English-derived item is still found in the dictionary of Focke in 1855, viz.arede. It appears that Portuguese items that managed to survive until 1718 were so firmly rooted in the language that it was no longer possible for them to be replaced by the coexisting English-derived item.

Alternative possibilities of items of English and of Dutch origin can also be found in the early texts. See the pairshansom / E. ‘handsome’ vs.moy / D. ‘mooi,’windau / E. ‘window’ vs.fensre / D. ‘venster.’ In both cases the English-derived item was eventually lost. The opposite occurred with the pairswani / E. ‘want’ vs.wil (ifwil is really from Dutch),agterdina / D. ‘achter’ and E. ‘dinner’ vs.bakadina / E. ‘back dinner’ (ifagter is really from Dutch).

These alternative possibilities may reflect old dialectal differences. The unpublished dictionary by C.L. Schumann (1783) for instance contains the wordbrens / E. ‘brains,’ with the remark that this word is still used on the old English plantations but not in town (Voorhoeve Donicie 1963(b): 23). It is quite possible that Schumann or his Creole informant (whose oral comments he often cites in this dictionary) had in mind the plantations that were first cultivated by the English pioneers before 1680. The English influence may in that case have been strongest and longest on the old English plantations and may have diminished only gradually with the spreading influence of the town language. Focke 1855 does not contain the wordbrens. A

more detailed study of the Surinam Creole vocabulary in a historical perspective may reveal more about the development of the language.

If we accept the hypothesis that most slaves came from Africa during the seventeenth century with a basic knowledge of an Afro-Portuguese pidgin and in Surinam found a few slaves from Barbados with a more deeply rooted knowledge of an English-based creole or pidgin, we may assume that the English-based language quickly became dominant but that for some time there was a certain competition within it between Portuguese-derived and English-derived items with an identical meaning. The British planters, overseers, and indentured laborers reinforced the English, so that the English-derived items won the battle in most cases. The Portuguese-derived items that remained in the language after the English influence had ceased were so firmly rooted that they could not easily be replaced by subsequent developments. The English-derived items were in a much weaker position and many could thus be replaced by Dutch-derived ones.