Demography of early St. Thomas to study emergence of Virgin Islands Dutch Creole

Several sources point to the Zeelandic/Flemish lexifier of Virgin Islands Dutch Creole. Hesseling mentions that, Logeman, etcetera. In 2000 I published the article ‘Negerhollands, Negerzeeuws, Negervlaams?’ in which I showed that not only Zeelandic, but especially  West-Flemish is of importance as lexifier of VIDC. Since lexical items from these dialects do not appear in Dutch texts from the Danish Antilles at the end of the seventeenth and the beginning of the eighteenth century, I considered the appearance in VIDC a proof for emigrants from the south of Zeeland and the north of West Flanders.

Demography underlines this. In my article from 2013, of which I have just found out it is digitally available, I study the heritage of the colonists on the basis of a few censuses of St. Thomas, of which the one of 1691 is the most important.

Van Rossem 2013


Digitale Bibliografie van die Afrikaanse Taalkunde

Interested in Dutch related languages? The following link will lead you to the digital bibliography of linguistics of Afrikaans.


With regard to the study of similarities between Dutch related languages, F. Ponelis, 1988. ‘Afrikaans en Taalversteuring’, in: Tydskrif vir Geesteswetenskappe , 28 (2) : 119-149, Jun. is of interest. The full text is also available on the page Scanned publications.

Christmas: earliest VIDC translation

Among the earliest VIDC translations of the New Testament and the Gospel Harmony we have seven versions of Luke 2: 1-21, the Birth of Jesus Christ. The earliest is 321 (manuscript by Böhner, about 1773) and is followed by 322 (manuscipt Böhner, before 1780), 315 (Magens’ printed New Testament), 3231 (1785-1790, Auerbach’s manuscript version), 3232 (about 1795, incomplete manuscript), 318 (1802, Moravian printed New Testament) and 3110 (1833, printed Moravian Gospel Harmony). Most of these texts are available in the Clarin-NEHOL database.

A comparison of these texts presents an interesting insight in the way the translators thought to connect as good as possible to their audience of Creole speakers. The following text is the oldest text, 321, without glosses, however to my opinion quite close to the German source text and therefore quite easy to understand.



New publication Philipp Krämer: Combien de néerlandais?

Recently I received a new publication by Philipp Krämer (Freie Universität Berlin):

‘Combien de néerlandais? Histoire linguistique et histoire de la linguistique dans les Îles Vierges Danoises’,  in: Histoire Épistémology Langage 38/1 (2016), p. 103-120. It is digitally available at: Combien de Néerlandais?

Krämer’s English abstract and keywords are the following:


For centuries, the Dutch-based Creole language of the Danish Virgin Islands was documented not by the Dutch but mainly by German missionaries and Danish colonialists. This article sheds light on the role of the Dutch language in this complex colonial universe. Historical sources from the 18th and 19th century will show which sociolinguistic role Dutch played in the society of the islands and which (meta-)linguistic knowledge of Dutch the authors of these sources (C.G.A. Oldendorp, J.M. Magens, and E. Pontoppidan) had. Some reflections on the discursive and epistemological foundations of the sources and the significance they attribute to the Dutch language will conclude the article in order to show that the linguistic compexities of this archipelago are diferent from most other Creole-speaking areas.


Colonialism, Christian mission, Creole languages, cariole (“Negerhollands”), Dutch, universalism, racialism, Danish Virgin Islands”





Dating the VIDC Moravian grammar manuscript

Already in the eighteenth century two descriptions of Virgin Islands Dutch Creole were published: Magens’s Grammatica over det Creolske sprog, som bruges paa de trende Danske Eilande, St. Croix, St. Thomas og St. Jans i Amerika. Sammenskrevet og opsat af en paa St. Thomas indföd Mand (1770) and the description by Oldendorp in the ninth section of his Geschichte der Mission der evangelischen Brueder auf den caraibischen Inseln S. Thomas, S. Croix und S. Jan (1777).

In the first part of the beautiful edition of the manuscript of Oldendorp’s History (2000: 681-724, sections 112-117) we find the complete contemporary description of which Oldendorp (1777) was only a kind of summary.

In the Unitäts Archiv in Herrnhut (Germany) another grammar is preserved. This manuscript, Grammatik der Creol-Sprache in West-Indiën, ms. 214 according to Stein (1986b), is not dated, however it looks early nineteenth century. In some publications (Van Rossem & Van der Voort 1996: 288) it is dated as ‘shortly after 1802’. In 1903 D.C. Hesseling obtained a copy of this manuscript from A. Glitsch, the then archivist of the Unitäts Archiv in Herrnhut (Grammatik der Creol-Sprache in West-Indien 1903).

While working on the section on Creole word order of this Grammar it appeared to me that some examples must have been taken from biblical texts, of which most are dated. The Grammar should then of course be younger than the youngest example.

In the first example we focus on the use of function word dan ‘then’:

(1) Dan em a see na die ander: hoeveel joe ben skuldig.

then 3SG PST say to the other how much 2SG are owing

The Gospel Harmonies (before 1780) have daarna ‘next, then’:

(1a) Darnah em a spreek tot die ander: en joe, hoe veel joe ben skul=dig (na mi Heer)? (321: 75)

(1b) Daarna em a see na die ander: en joe, hoe veel joe ben skuldig? (322: 75)

The German source text (Lieberkühn 1820: 162) has Darnach sprach er zu dem anderen (…).  Magens’s translation of the New Testament (1781) has asteran ‘next, further’:

(1c) Asteran hem ha seg na die ander: en ju, ju veel ju skylt? (315: Luke 16: 7)

The youngest texts of the Moravian Brethren however, use dan:

(1d) Dan em a see na die ander: Maar joe, hoe veel joe ben skuldig? (318, 1802) Lucas 16: 7)

(1e) Dan em a see na die ander: En joe, hoe veel joe ben skuldig? (3110 (1833): section 75)

The example shows that the early sources were much more according to the source text  than the younger texts, which appear to be changed to connect to the audience of Creole speakers. Is that so? See the English translation of Lieberkühn (from 1771): Then said he to another, And how much owest thou? The change from darnah to dan may well be a change of German influence (from source text) or Dutch influence of superstrate into English influence of other source text or from new vernacular on the Danish Antilles. 

The next example shows something similar. Compare for instance the use of vor ‘in order to’ and the positioning of die ‘that’. The source text has: ‘Aber das Sitzen zu meiner Rechten und meiner Linken, stehet nicht bey mir, euch zu geben‘ (Lieberkühn 1820: 182).

(2) Maar vor set na mi rechter en na mi slinker Hand, die no staan bij mi, vor gie na jender

however to sit on 1SG right and on 1SG left hand, that NEG stand by 1SG, to give to 2PL

(2a) maar die sett na mi Rechter en na mi Slinker Hand, no staan bi mi, vor gie die na jender; (321: 83)

(2b) maar die Sitt na mi Rechter en na mi Slinker (Hand), no staan bi mi, vor gie die na jender; (322:83)

(2c) Maar vor set na mi rechter en na mi slinker Hand, die no staan by mi vor gie na jender, (318: Mark 10: 40)

(2d) maar vor set na Mi rechter en na mi slinker Hand, die no staan by mi, vor gie na jender, (3110: section 83)

A similar thing as in (1) happens here: the early text use die sett , the sitting’ for das Sitzen, while the younger texts use vor set ‘in order to sit’. Lieberkühn (1781) has: but to sit on my right hand and on my left hand is not mine to give; but it shall be given to them (…). Again this resembles the younger texts: the noun set of the early texts has changed into the verb set.

My last example is:

(3) Toen die Tien a hoor die, soo sender a koom ontoevreeden over Jacobus en Johannes.

When the ten PST hear that, so 3PL PST become displeased about James and John

(3a) En as die tien a  hoor die, sender a kom ontoevreden over die twee Broeders, Jacobus en Johannes. (321:83)

(3b) En as die (ander) tien a hoor die, da sender a neem die goe Qualik van die twee Broer, Jacobus en Johannes (322: 83)

(3c) Toen die Tien a hoor die, soo sender a kom ontoevreden over Jacobus en Johannes. (318 Marcus 10: 41)

(3d) Toen die Tien a hoor die, soo sender a kom ontoevreden over die twee Broeders, Jacobus en Johannes. (3110: 83)

The German source text has Da das die Zehen höreten, wurden sie unwillig über die zween Brüder, Jacobum und Johannem (Lieberkühn 1820: 182) The English has And when the ten heard it, they began to be much displeased with James and John (Lieberkühn 1781). The three Dutch Creole Gospel Harmonies and the German source text mention the brothers, while the English source text and the New Testament of 1802 do not.

These examples show that, unless an earlier manuscriptal version of the 1802 New Testament existed, the examples were from the New Testament of 1802. Examples (1) and (2) may leave a small possibility that 3110 was used, however the difference between 318 and 3110 in (3) gives a decisive answer.

This young text was prefered above the older texts which were available, which may point to the importance to use a more English related language in this period in which English (Creole) became the new vernacular on the prejudice of Virgin Islands Dutch Creole.

Cefas van Rossem






Berbice and Skepi Dutch, and Virgin Islands Dutch Creole, a lexical comparison

In 1989 Ian Robertson published his article ‘Berbice and Skepi Dutch, a lexical comparison’ in Tijdschrift voor Nederlandse Taal- en Letterkunde. As far as I know it is the best source for the lexicon of Skepi Dutch, the Dutch related Creole of which we know so little.

In this article the lexicon of these two Guyanese Creole languages are also compared to that of Virgin Islands Dutch Creole. A Zeelandic lexifier has already been linked to all three languages, however Skepi Dutch and Virgin Islands Dutch Creole seem to be closer related than Berbice and Skepi Dutch. The past years especially Peter Bakker (Aarhus University, Denmark) studied the links between these languages.

The article is digitally available on the so-called Digitale Bibliotheek voor de Nederlandse Letteren. You can find the link it here: Robertson 1989

Berbice Dutch Creole in collection Meertens Institute

In the West Indies only three Dutch related Creole languages existed. In Guyana two languages existed. Of Skepi Dutch, which was spoken on the borders of the Essequibo river, we only know a few sentences and a Swadesh list full of words. The first known sentence of this language, from 1780, was found only a few years ago by Marijke van der Wal in one of the letters which were studied in her Letters as Loot project. Robertson (1989) shows a resemblance in vocabulary of Virgin Islands Dutch Creole and Skepi Dutch, which may point to a similar Dutch/Zeelandic superstrate.

Berbice Dutch is a unique Creole language from which the African influence can be traced to one language which is spoken in Nigeria. There is far more to tell about this language and interested ones should at least consult Silvia Kouwenberg’s dissertation and the work of Ian Robertson on the discovery of the language in the 1970s.

The Dutch Meertens Institute (which is one of the institutes of the Royal Dutch Academy of Sciences) is, among Dutch folklore and onomastics, specialized in Dutch dialectology in the broadest sense. In its last newsletter it presents their conservation of Silvia Kouwenberg’s material of Berbice Dutch, which is safe for the future because of the Data Seal of Approval.

More information, in Dutch, can be found here.



THREE publications Virgin Islands Dutch Creole in July!

Last month three publications about, or related to, Virgin Islands Dutch Creole, appeared, at least in my mailbox. Alphabetically the first one is:

Kouwenberg, Silvia. 2016. ‘Review Article (…) Nineteenth-Century Creolist Work and Its Reflections on Language and Community’. In: Historiographia Linguistica 43: 1/2, p. 206-222. >On the occasion of Krämer, Philipp (ed.) (2014).

In the first place the following note of Kouwenberg is remarkable:

“Contemporary reference is simply to die creol taal (the creole language) or variants thereof; the term Negerhollands (literally: Negro Dutch) was coined by a Dutch linguist in 1840, and has been in use among students of the language ever since (Van Rossem & Van der Voort 1996: vii), despite some attempts to rechristen it (something like) Virgin Islands Dutch Creole. Given that its speakers did not refer to the language as Negerhollands, and taking into account the modern connotations of the term, it seems to me that that term is overdue for retirement.”

In our institute in Nijmegen we try to use Virgin Islands Dutch Creole as consequently as possible. Our German colleagues Krämer and Stein prefer to use the term Carriolsch, since it was the first name which was given to this language in written texts. In Troels Roland’s article a separate section (Sprogets navn, p.181-183) is dedicated to this subject.

This review of Philipp Krämers book about 19th century Creolistics is does not only refer to Virgin Islands Dutch Creole material, of course. Page 215 and 216 are about Pontoppidan’s contribution to the study of Virgin Islands Dutch Creole.

Roland, Troels Peter. 2016. ‘”Ju ben een Creol waer-waer”’. In: Kulturstudier 1 (Juli), pp. 159-187.

Troels Roland’s article is in Danish, but accompanied by an abstract in English which makes me very curious about the content of the entire article. Not only because of the interesting period 1750-1850 in which most written texts appeared and in which Dutch Creole was replaced by English Creole, but mainly because of the Danish perspective and the use of (Danish) sources which I did not see in references related to Dutch Creole before. The article is digitally available at

Of the third July article I received a pdf from Peter Stein. In this French text, which is richly illustrated, Stein presents an insight into the early years of Virgin Islands Dutch Creole

Stein, Peter. 2016. ‘La documentation ancienne du créole disparu des Îles Vierges Danouises’. In: Les Cahiers créoles du patrimoine de la Caraïbe/Pawol maké asi mès é labitid an Péyi Karayib 6: Les langues créoles / Palé Kréyol!, [Guadeloupe]: CANOPÉ. pp. 37–39.

De Josselin de Jong Het Huidige Negerhollandsch available as scan

On the page Scanned publications you will find a scan of de Josselin de Jong’s most important publication about Virgin Islands Dutch Creole:

Josselin de Jong, J[an] P[etrus] B[enjamin] de. 1926. Het huidige Neger­hollandsch (teksten en woordenlijst). Verhandelingen der Koninklijke Academie van Wetenschappen te Amsterdam, Nieuwe Reeks, Deel 26, no. 1. 124 pp.

In 1922-1923 De Josselin de Jong visited both St. Thomas and St. John on an archaeological expedition to find out more about the native American people who lived on the islands before the European colonization. In his spare time he read Hesseling’s Het Negerhollands der Deense Antillen and interviewed several native speakers of Virgin Islands Dutch Creole.

His work about the language was published only three years after his departure. The book about his archaeological work in the Caribbean was published twenty years later, in 1947.

Review of ‘Language contact in the Danish West Indies: Giving Jack his jacket’

About this moment, only four years ago, Robin Sabino’s Language contact in the Danish West Indies: Giving Jack his jacket appeared. This work is without a doubt THE description of spoken Virgin Islands Dutch Creole and contains a bulk of information about all stages and aspects of this Dutch related Creole.

In the latest volume of Journal of Pidgin and Creole Language Peter Bakker (Aarhus University) published his review:

Bakker, Peter. 2016. Review of Language contact in the Danish West Indies: Giving Jack his Jacket. By Robin Sabino. Leiden: Brill 2012. Pp. 337 ISBN 978-90-04-22540-4 (Brill’s Studies in Language, Cognition and Culture, Volume 1) (…). In: Journal of Pidgin and Creole Languages 31: 1, p. 223-228.

Other reviews of Sabino (2012) are Migge (2013) and Van den Berg & Van Sluijs (2015).