Historical Sociolinguistics and Sociohistorical Linguistics

Home ] Up ]



 Possible Origins of Certain Nonstandard Verb Forms in the Dialect of Tristan Da Cunha

Bas van Elburg

Published: 21 December 2000 (HSL/SHL 1)

(print instructions) 

In an article on the speech of the people of St. Helena Hancock (1991:17) observes that the island, and the dialect of its inhabitants, are interesting from a number of  aspects: firstly because of the many similarities with island dialects elsewhere, and secondly because of its implications for the study of nautical English, and its relationship to creolized forms of that language. The reference to “island dialects elsewhere” in this quotation may very well include that of Tristan da Cunha, a South Atlantic island situated 1900 miles west of Cape Town. Besides phonological similarities (Hancock 1991:20), the dialects of St. Helena and Tristan da Cunha also have certain features in common as far as their verbal systems are concerned. For instance, the use of done as a preverbal perfective marker, which may be illustrated by the following examples taken from the dialects of St. Helena (SHDE) and Tristan da Cunha (TdCDE) respectively:

  1. He’s done gone up the road (Hancock 1991:22).

  2. Edwin’s done drawn his stone (Zettersten 1969:86).

Preverbal done in this function also appears to be a feature of Southern White American English (SWAE), where it occurs in the speech of the lower classes of society (Feagin 1991:162). According to Feagin, preverbal done is claimed to derive from the mesolect creole spoken by Blacks during the era of American slavery. Hancock (1976:283) comes to a similar conclusion when he argues that the form “has passed into Southern U.S. white speech through Afro-American dialects”. The following examples are taken from African American English (AAE) and SWAE respectively:

  1. … a li’l culled chu’ch what de masses an’ missuses in San ‘Tone done builted(Dillard 1972:47)

  2. The buses done quit runnin’ (Feagin 1991:167).

Basing himself on data from the 1950s, Dillard (1972:47) argues that black American English (which is now generally referred to as AAE) resembles certain West African English-based pidgins grammatically in that preverbal done expresses “Immediate Perfective Aspect” as opposed to been, which marks “Remote Perfective Aspect”, i.e. it functions as a past tense marker. Inorder to emphasize the pidgin or creole background of preverbal done, Dillard (1972:47) refers to parallel forms in creole languages with different superstrates, for instance French Creole fèk (from French faire), or Sranan Tongo kaba (from Portuguese acabar). In the creoles of Jamaica, Guyana, Sierra Leone (Krio) and Cameroon (WesKos),perfective done also occurs in clause-final position. The following examples from Guyanese creole and Jamaican creole respectively may illustrate this:

  1. Wen mi kuk done, me a hosl fid op me pikni (Bickerton 1974:40) ‘When I have finished cooking, I hurry and feed my child’.

  2.  Mi dis iit don (Feagin 1991:181) ‘I have just eaten’.

  3. According to Bickerton (1974:40) “there is a good deal of evidence to indicate that clause-final don represents an older layer of the creole system than preverbal don”.

  4.     Although preverbal perfective done also appears to have occurred in Middle English, Medieval Scots and early Modern English, it always had a preceding auxiliary have/has/had (Feagin 1991: 183). In TdCDE and SHDE there are also occurrences without a preceding auxiliary:

  5. I done went to the doctor (Zettersten 1969:85).

  6. We always done pass through that (Hancock 1991:21).

Feagin (1991:164) observes that done without a preceding auxiliary is a feature of AAE, whereas the form with a preceding auxiliary is rather common in SWAE. I have not found occurrences of preverbal perfective done in any British dialect.

    Another pidgin or creole feature which TdCDE has in common with AAE and SHDE is the frequent absence of inflected verb forms in past tense contexts. The following examples from TdCDE, AAE and SHDE respectively may illustrate this:

  1. So I rushed back and get those shoes … So I must rush back and get that, so I rush back for to got that (Zettersten 1969:135).

  2. When the day begin to crack, the whole plantation break out with all kinds of noise, and you could tell what was going on by the kind of noise you hear (Dillard 1972: 41).

  3. They go round did they anything see (Hancock 1991:21).

It is true that the present tense form also occurs in past tense contexts in standard English (SE), but this is only the case in the historical (or dramatic) present. Moreover, in such contexts all the verbs are in the present tense. In this respect Dillard (1972:42), referring to example (10), suggests that “any facile assumption about historical present is broken by the occurrence of forms like could and was”. The verb forms begin, break and hear, Dillard continues, are “consistent with occurrence in the past insofar as the grammar of Black English is concerned. Such occurrence is non-redundantly marked in the language”. Non-redundancy is also a feature of creole languages. Tagliamonte and Poplack (1993:183) hypothesize that “a negative correlation between temporal disambiguation and morphological marking is a creole characteristic by virtue of the widely recognized predisposition of creoles to minimize redundancy in syntax and rely on context for interpretation”.

    Nonstandard verb forms also occur in present tense contexts in TdCDE. Such forms can be found in the present tense paradigms ofbe, auxiliary have, periphrastic do and of lexical verbs:

  1. … they’s similar to albatross (Zettersten 1969:34).

  2. Now has the ships come (Zettersten 1969: 84).

  3. Then my old great grandfather, see, he tell me, he don’t think she had enough cork in her ... again (Zettersten 1969:49).

  4. And we goes over there under sail … stay (Zettersten 1969:140).  

TdCDE shares this phenomenon with AAE:

  1. I has a uncle was one of the world’s heavyweight contenders (Dillard 1972:68).

  2. They treats me nice (Holm 1991:240).

Hughes and Trudgill (1979:16-17) observe that in certain British dialects the present tense paradigms of the above verbs is either completely regular as a result of the absence of the third singular –s suffix (e.g. in East Anglia), or show regularity of the opposite kind, i.e. with the –s suffix occurring with all persons of the verb (e.g. in parts of the north of England and in the southwest and south of Wales).     

    In order to account for the occurrence of certain verb forms in TdCDE with parallel forms in SHDE, AAE and SWAE, it is necessary to look at the historical and linguistic backgrounds of the present community of Tristan da Cunha. In 1817 the foundation for this community was laid by a Scotsman and two Englishmen. All three of them decided to stay and build a settlement on the island after the withdrawal of their regiment, which had been stationed on Tristan da Cunha for a couple of years, to the Cape. The Scotsman, William Glass, was in the company of his wife and child. According to some eye-witness reports Mrs Glass, née Leenders, was a Cape Creole who spoke a little Dutch, but preferred to use the English language (Earle 1966: 208; Beintema 1997:126). Mrs Glass may have spoken “Extreme South African English”. One of the features of this variety of English is the absence of the third person singular present tense marker (Lanham 1984:341), which is also characteristic of TdCDE (Zettersten 1969:83). During the next few decades after the foundation of the British settlement on Tristan da Cunha, more and more people with different nationalities came to live on the island, though a number of them left at some time or other. Most of the new settlers were the victims of shipwrecks. The use of the English language as a means of communication by the early settlers, and also their different linguistic backgrounds became important for the development of TdCDE. In 1827 five women from St. Helena, one of them in the company of four children, settled on Tristan da Cunha. According to Brander (1940:130) one of the women was black, while the others were mulattoes. Munch (1945:53) suggests that “a not inconsiderable part of the coloured blood introduced by the four women from St. Helena was Malayan blood”. Assuming that the language of the St. Helena women contained creole features, these may very well have influenced the speech of their children. Beside the importance of the mothers, as “matriarchs of the settlement” (Cassidy 1974:176), some of the women of Tristan da Cunha also acted as teachers in the absence of clergymen. The following report was given by the captain of HMS Racer, which touched at Tristan da Cunha in 1893:

 The only education for the children is imparted by Mrs Swain, a native of St Helena … I am told that they learn to read and write, but the presence of a minister or clergyman who would undertake this duty and act as head of the community would, I think, confer on it a great benefit (Evans 1994:256).

    Soon after the arrival of the St. Helena women, Tristan da Cunha began to be frequented by American whalers for supplies. Several of the crew members of these ships decided to stay on the island, and married local women. Moreover, most of the young men of the second generation were employed in the whaling industry, which took them away from the island, but a number of them returned to Tristan da Cunha sooner or later (Munch 1945:28-29). In this respect it is interesting to note that there appears to be a link between nautical English and English-related creoles (Hancock 1991:27). A verb form, such as preverbal perfective done, not only occurs in certain English-based creoles, but is also a common feature in nautical English. Alan Crawford, the author of several books on Tristan da Cunha, informed me that about forty to fifty years ago, he had observedTristanians expressing the past tense as American seafarers used to do, with the form done followed by the verb. For instance, when Tristanians say I’se done eat, they mean “I’ve eaten”, according to Alan Crawford. Hancock (1976:283) also stresses the importance of the role played by seafarers when he refers to the “widespread occurrence of preverbal done in nautically derived English dialects (e.g. St. Helena, Tristan da Cunha, Trinidad etc.)”. It may be concluded that the occurrence of preverbal perfective done in TdCDE was the result ofinfluence from the speech of the women from St. Helena, and that later syntactic convergence took place with the same form occurring in the nautical English of a number of the male members of the island community.

    Preverbal done may also prove Hancock’s (1991:26-27) argument that there appears to be a link between English-related creoles and nautical English. This could be true as well of the frequent use of the present tense form of verbs in past tense contexts. Indeed, the absence of inflectional morphology appears to be a creole feature (Rickford 1986:272). At the same time it may be assumed that if the crew members of a ship have different nationalities, non-redundancy in the verbal system could be a means to promote ease of conversation in one language. Ships, which were wrecked off the coast of Tristan da Cunha, often had people from different parts of the world on board. For instance, among the crew members of the Emily schooner, which had run on to a reef near the shore of Tristan da Cunha in 1836, were an American, a Dane and a Dutchman. All three of them decided to stay on the island and married local women (Brander 1940:137).

    The isolated position of Tristan da Cunha may have caused verb forms, such as those discussed above, to become fossilized. Intelligence tests, which were conducted in the early 1960s1, show that the Tristan children had “an uncertain grasp of the past tense” (Keir 1965:152). It may be argued, however, that this does not have anything to do with the level of intelligence of the Tristan children. The use of uninflected forms in past tense contexts is rather part of the native grammar of the people of Tristan da Cunha.Indeed there are instances from interviews with Tristanians in the early 1960s2 showing that it was the uninflected verb form that came to the informant’s mind first:

  1. So when we come – came home, some people was just flopping out … beach (Zettersten 1969:40).

  2. Increasing contact with SE, improvements in the field of education3, overseas training courses4 and the modern media5 during the past four or five decades have affected the speech of the islanders, particularly that of the members of the younger generation of Tristanians. Interviews with Tristan informants show that this increasing contact with SE may lead to the use of hypercorrected forms. For instance, in a recent BBC television documentary about Tristan da Cunha,6 one of the informants says:

  3. I had an opportunity to went to England.

    Past tense forms for infinitives also occur in AAE, which makes Dillard (1972:50) suggest that such forms “have no tense significance for the speaker”. A hypercorrected verb form which is, to my knowledge, not attested in any English-based creole or British dialect is the use of the past tense of a verb with the –ing suffix:

  1. And so they start to tooking an end, they start to send reports … it (Zettersten 1969:138).

    In another television documentary,7 there is a clear example of the influence of the sociolinguistic variable age in the interviews with two islanders: Harold Green, a middle-aged Tristanian, uses dialect forms, whereas Elaine Repetto, a Tristan adolescent, speaks SE.Such interviews are relatively formal, and it may be assumed that once the islanders are in each other’s company, they will revert to their local dialect, which is relatively hard for outsiders to understand. Jim Flint, a former head teacher of the elementary school on Tristan da Cunha, observed that “the children simply spoke two languages – School English and Island English” (Evans 1994:274). Booy (1957:97-98) gives an example of style-shifting, referring to a situation in which he was in the company of a number of Tristanians:

At these social assemblies I was frequently aware of a curious feature of the islanders’ conversation: when they addressed us, or obviously intended us to be included in the talk, their speech – once we had learnt the accent – was perfectly intelligible; but when they exchanged remarks among themselves, not intended for our attention, they relapsed into a dialect that was incomprehensible … Perhaps we were not intended to hear the substance of these exchanges. 

This quotation suggests that the Tristanians use their own dialect as a marker of their identity. This phenomenon seems similar to that encountered by Labov (1963) when he studied the language of the people of Martha’s Vineyard, an island lying off the coast of Massachusetts. According to Labov, a number of informants on Martha’s Vineyard showed a high degree of centralization of the [ai] and [au] sounds in order to maintain their own identity against outsiders. Labov (1963:300) observed a marked contrast between the Vineyarders who intended to leave their island for careers on the mainland and those who did not. The latter showed strong centralization of the [ai] and [au] sounds, while the former show little, if any. The difference between the situation on Martha's Vineyard seems to be, however, that their reaction against outsiders appears to have been a largely unconscious one, while that on Tristan da Cunha may well be a rather more conscious process, in that the speakers deliberately close themselves off from contact with outsiders by switching to their native dialect, even in their presence.

    In the television documentaries about Tristan da Cunha referred to above, it is especially the members of the older generation of Tristanians who show much indifference towards people from the outside world. This attitude may have been the result of the lack of cooperation of the British government in the early 1960s when the majority of the islanders had made it clear that they wished to return to Tristan da Cunha once the situation was safe again on the island. Negative reports about the Tristanians in the press at the time (Munch 1971:223) may also have contributed to the islanders’ attitude towards outsiders. These two factors could have created an anti-visitor mentality among a number of Tristanians, manifesting itself by their use of dialect. At the same time, a recent development in the relationship between Britain and Tristan da Cunha could have a similar linguistic effect on certain Tristan adolescents as it has had on a number of members of the younger generation of Vineyarders. For many years emigration to Britain was hardly possible for Tristanians since the British authorities had stipulated in the early 1960s that those who wished to return to their island would no longer be looked upon as British citizens. The previous Administrator of Tristan da Cunha, however, informed me that the British government agreed to restore the islanders’ right to British citizenship as from 16th March 1999. This may lead to an increase in the use of SE forms by those who wish to make use of the agreement. Contrary to what a number of informants say in the television documentaries, referred to above, the Administrator does not expect Tristanians to emigrate to Great Britain. Further research may shed more light on this, and give an answer to the question to what extent the developments which have taken place on the island during the past few decades have influenced the linguistic attitude of the Tristanians. Nonstandard verb forms, such as those discussed above, may also call for further research into the relationship between TdCDE, SHDE, AAE and nautical English. 



Beintema, A. (1997), Het Waterhoentje van Tristan da Cunha, Amsterdam: Uitgeverij Atlas.

Bickerton, D. (1974), Dynamics of a Creole System, Cambridge: Cambridge University Press.

Booy, D.M. (1957), Rock of Exile, a Narrative of Tristan da Cunha, London: J.M. Dent & Sons Ltd.

Brander, J. (1940), Tristan da Cunha 1506-1902, London: Allen & Unwin.

Cassidy, F. (1974), “Review of Arne Zettersten’s The English of Tristan da Cunha”, Language 50: 175-77.

Dillard, J.L. (1972), Black English, New York: Random House, Inc.

Earle, A. (1966), Narrative of a Residence on the Island of Tristan D’Acunha in the South Atlantic Ocean (1 edn 1832), Oxford: Clarendon Press.

Evans , D. (1994), Schooling in the South Atlantic Islands 1661-1992, Oswestry: Anthony Nelson.

Feagin, C. (1991), “Preverbal done in Southern States English” in P. Trudgill and J.K Chambers, eds, Dialects of English, Studies in Grammatical Variation, Harlow: Longman Group UK Limited.

Hancock, I. (1976), “Nautical sources of Krio vocabulary”, International Journal of the Sociology of Language 7: 49-59.

Hancock, I. (1991), “St. Helena English” in F. Byrne and T. Huebner, eds, Development and Structures of Creole Languages, Amsterdam and Philadelphia: John Benjamins Publishing Company, 17-28.  

Holm, J. (1991), “The Atlantic Creoles and the Language of the Ex-Slave Recordings” in G. Bailey, N. Maynor and P. Cukor-Avila, eds, The Emergence of Black English, Amsterdam and Philadelphia: John Benjamins Publishing Company, 231-48.

Hughes, A. and Trudgill, P. (1979), English Accents and Dialects, Baltimore: University Park Press.

Keir, G. (1965), “The Psychological Assessment of the Children from the Island of Tristan da Cunha” in C. Banks and P.L. Broadhurst, eds, Studies in Psychology, London: University of London Press Ltd, 129-72.

Labov, W. (1963), “Social Motivation of a Sound Change”, Word 19: 273-309.

Lanham, L.W. (1984), “English in South Africa” in R. Bailey and M. Görlach, eds. English as a World Language, Cambridge: Cambridge University Press, 324-51.

Munch, P.A. (1945), Sociology of Tristan da Cunha, Oslo: Det Norske Videnskaps-Akademi.

Munch, P.A. (1971), Crisis in Utopia, New York: Thomas Y. Crowell Company.

Rickford, J.R. (1986), “Social Contact and Linguistic Diffusion: Hiberno-English and New World Black English”, Language 62: 245-89.

Tagliamonte, S. and S. Poplack (1993), “The zero-marked verb: Testing the creole hypothesis”, Journal of Pidgin and Creole Languages 8: 171-205.

Zettersten, A. (1969), The English of Tristan da Cunha, Lund: Berlingska Boktrycheriet.


1 The intelligence tests were given to the Tristan children, who attended lessons at different English schools, in the early 1960s when the entire population of Tristanians was in Britain after a volcanic eruption on Tristan da Cunha. The ages of the Tristan children who were given the tests ranged from 5 to 15.

2 The interviews were conducted in the early 1960s by the BBC in cooperation with the University College, London. Zettersten (1969) used the interviews for his analysis of the English of Tristan da Cunha.

3 In 1950 compulsory education was introduced for all the Tristan children between the ages of five and fifteen (Evans 1994:281).

4 The Tristan Times, a monthly issued by the Administrator of the island, frequently reports on the progess which Tristan adolescents make during their training courses in Britain and on St. Helena.

5 The Tristanians have access to radio broadcasts, provided by the BBC World Service. The present Administrator on Tristan da Cunha, Brian Baldwin, recently informed me that there are no television broadcasts on the island. The Tristanians do have a possibility to borrow video tapes, which are imported from South Africa. Mr Baldwin also told me that the elementary school on the island has a number of computers, though it is only the Administrator’s computer which can be linked to the Internet. The Tristanians are allowed to make use of the computer in Mr Baldwin’s office in order to communicate with friends and relatives in the outside world through e-mail (another “satellite phone” for sending and receiving e-mails is expected to be installed in the near future) and get information from the Internet. However, since the island’s Internet provider is in the UK, and surfing the net is over £ 5.00 per minute, the Tristanians cannot have frequent access to this medium.

6 This television documentary, called The Forgotten Island, concerns a “Gruppe 5” production for BBC Bristol in association with WDR/Arte (1998).

7 This television documentary, called Tristan da Cunha: No place like home, is a Warner Sisters production for Granada Television Ltd in association with Wnet/New York (1989).


List of abbreviations:

AAE        = African American English

SE          = Standard English

SHDE       = St. Helena Dialect English

SWAE      = Southern White American English

TdCDE     = Tristan da Cunha Dialect English


Contact the author.