Romani in Britain: The Afterlife of a Language

Publication: Romani Studies
Author: Acton, Thomas A
Date published: June 1, 2012

Romani in Britain: The Afterlife of a Language. Yaron Matras. Edinburgh: Edinburgh University Press, 2010. 255 +xvpp. ISBN 978-0-7486-3904-5-9 (hbk.)

Reviewed by Thomas A. Acton

This is the most important study of EngUsh Romanes for a quarter of a century, a major empirical offering in the meagre corpus of Uterature on this Unguistic phenomenon, and a theoreticaUy tantaUsing discussion of the question which has vexed EngUsh Romani Studies for more than a century as to what EngUsh Romanes actuaUy is. Is it a language in its own right, a dialect of Romani massively overlain by EngUsh (cf. Hancock 1984), or a dialect of EngUsh influenced by Romani, or a kind of creole? Or if not actuaUy technicaUy a creole, some other kind of contact language like other 'para-Romani' dialects, such as southern-Spanish Calo, which Matras previously (2002: 243) presented as an unproblematic category, in the tradition of Bakker and Cortiade (1991)? Oris it not a language at aU, but a register or special lexicon that is used in the context of another language, English in this case?

Based on meticulous analysis of a corpus of transcribed interview recordings from around forty individuals describing themselves as English or (south) Welsh Gypsies, mostly caravan -dwelling, Matras incUnes to the special lexicon answer; p. xi baldly asserts 'Speaking "Romani" today involves in the British contexts essentiaUy speaking EngUsh with Romani word insertions.' But Matras also finds this answer unsatisfactory and insufficient in turn. The fluid variabüity, the sheer mutabiUty of the vocabulary and the sentence structure lead Matras to see what he is studying as something less than a language or dialect, but something more than a mere vocabulary or register, and he comes up with a thought-provoking, if slightly imprecise, characterisation: this is the 'after-life of a language'. What we see here defies technical Unguistic description in Matras's terms because its function is only secondarily the linguistic one of communication, and primarily the social one of ethnic identification. It is not a language, but the ghost of one. This is gloriously poetic, but eminently contestable: it can only be the beginning in a new phase of debate over the study of what I would prefer to caU the daughter languages of Romani.

The strength and vividness of the interview material, much of which is cited verbatim, is backed up by a clear contemporary social contextualisation of the matter discussed, and an interesting discussion of the impact of Romani pentecostalism, particularly of the Light and Life Mission, which valorises Romani, but brings Anglo-Romani speakers into contact with inflected Romani (especiaUy Vlach Romani) preachers. This means, of course, that many of Matras's own informants themselves see English Romani as merely a decayed form of the Romani language spoken by miUions, which they need to learn if they are to play a role in the international councils of Romani evangelicaUsm. A similar process could be observed forty- years ago amongst the first French Manouche pentecostals. In the 1950s and 1960s, especiaUy under the influence of Breton Manouche speakers of inflected Romani, they made substantial use of Manouche Romani in their worship. In the 1970s this aU but coUapsed as the dialect of the Paris Kalderash became informaUy recognised as the standard Romani of the movement.

Is there any reason to contest the denial that EngUsh Romani remains a language? Perhaps. Although the empirical work in this book is very soUd, it cannot be regarded, and it does not claim to be, a comprehensive study of varieties of English Romani. Its survey of the amateur coUections of EngUsh Romani vocabulary is sketchy, although these were used to produce a preUminary list of EngUsh words likely to eUcit Anglo-Romani equivalents, and these were then used to induce informants to recaU conversations where the words had been used. The lexicon in Appendix 1 very properly Umits itself to items actuaUy in the recorded interview material (though more from other sources is listed in the Manchester University Romani website Anglo-Romani word Ust.) Probably many readers could, if pushed, think of other weU-known EngUsh Romani words which did not happen to get used in this particular corpus of utterances. ActuaUy to try to produce an exhaustive lexicon of extant EngUsh Romani through a translation questionnaire would require considerable additional effort, and would not have helped this book's project of analysis of natural language use. The author recognises that use of natural language in an interview context is not the same as its use in the weddings, fairs and above aU funerals, which are generaUy believed to eUcit most EngUsh Romani utterances.

And most of the English Romani Pentecostals who have attended Romanichal evangelical gatherings in the United States are convinced - rightly I beUeve from my own experience - that Romanichals in at least the southern part of the United States speak EngUsh Romanes more frequently and conservatively, with a greater density of Romani-derived lexicon than is usual in England.

One has to ask then whether in other circumstances one might see EngUsh Romani as a language in its own right. What would this mean? It would have to mean that not only the Romani part of the lexicon, but also the English part of the lexicon, as weU as the grammatical and syntactical structure, were particular to EngUsh Romani, and it would be essential, in order to produce utterances regarded as authentic by another native speaker, to get the EngUsh right, as weU as the Romani.

The late Prince Nathaniel Petulengro Lee, the fortune-teUer of the PortobeUo Market, had a large English Romani vocabulary, which he employed whenever there was someone there to understand it. There are more Romani terms in his section of Sandford (1975: 29-50) than in those of any other contributor. But I never met any native EngUsh Romani speaker, even if they could rokker less than him, who regarded Lee as a native speaker, even if they regarded him as fluent and amusing. (The broadcast 'blessing' with which he freed a footbaU ground from an aUeged ancient Gypsy curse was, in its entirety, "Jukkel's kori 'n matchiko's minge!") What was inauthentic about his utterances? It was not any lack of Romani lexicon, but two other things. First, he did not slip Romani words into his EngUsh; he crammed them in, wiUy-niUy as many as he could, in a way no TraveUer ever would. And second, and more importantly, he just did not talk EngUsh Uke a TraveUer.

Yet even in natural settings - Uke weddings fairs, and funerals - most TraveUers, except for perhaps the least educated, most isolated, most rural ones, do not always speak Uke a TraveUer. Kenrick and I suggested in 1984 (9-11) that TraveUers switch between an Anglo-Romani code and an English dialect familiar to them, within which they may quote Romani words.5 What we did not do, however, is produce any recorded utterances (apart from one song) where we could identify the turns at which people switched codes, but rather relied on our memories.

Certainly, there are EngUsh and EngUsh-dialect derived words in EngUsh Romani which have become archaic in mainstream English, for example, my favourite, bidduv aradikal, as weU as brazen, dunnocken and pismire. (Matras 2010: 177 derives pishun from Romani pishom, a flea. But it is possible for EngUsh and Romani lexical items to converge. Most Romanichals beUeve jumper, 'frog', is from the EngUsh verb to jump rather than Romani zamba.)

But is it the case, as Kenrick and I certainly believed in 1984, that the lexicon of EngUsh words that could be included in authentic-sounding English Romani was quite severely limited? To examine this hypothesis we would need to look at extended transcripts, identify the points at which code-switching occurs, and then analyse the EngUsh content of the parts of the transcript identified as English Romani. Some of the longer passages cited by Matras seem to me to be educated English quoting Romani words; others, usuaUy shorter, are the kind of thing that one might think only a TraveUer would say'. Except of course that if I am modest, assert clearly that I am a Gaujo, deny any knowledge of Romani by saying nobody jins them ol' lavs any more, but go on pukkering only kovels I've shunned waver foki penning, then I can get TraveUers who do not know me a bit confused: "Ihem words you're using, does your Mum and Dad know them?' I beUeve it is getting the EngUsh right (and also getting the code-switching right) that makes TraveUer suspect I might be a self-denying TraveUer.

To get the evidence which might definitively support such a hypothesis would require a great deal of difficult work. One of the things that would make this more difficult is the practice, in certain contexts of creating coinages, often punning caiques on EngUsh which are almost Uke crossword-puzzle clues, which older speakers often compete with each other to invent, and deUver with a dead-pan face. Some of these become widespread enough to become part of everyday language, and are in the lexicon - for example, sastergrai (Matras 2010: 180) - though I was a Uttle disappointed to see that for coffee' Matras has only kalapani and not the witty chinkerdi (cough-ee). The point is that such neologisms can earn the speaker kudos if they are elegant and appropriate. For example, shun'n'putch, weU estabUshed in the United States as meaning 'telephone', always eUcits respect when I use it in such a way that its meaning is clear, though it was at least ten years before I heard anyone else use it in England. Karing the mush on the jukkel'n'kokkalo (a caique on cockney rhyming slang dog and bone), however, eUcits Uttle more than a groan. But it is the kind of Uberty one can take with the language among close friends. Perhaps one of the reasons why native speakers of EngUsh Romani who are themselves academics have not written more about the language is that by the time they become old men, they are no longer sure what parts of their EngUsh Romani repertoire they made up themselves. Groups of old acquaintances refine and elaborate their in-group lexicon progressively to exclude others, and this private pleasure would be endangered by pubUc discussion. Perhaps no book could ever quite catch the sheer fun of competitive rokkering. Perhaps, after aU, 'language' is not the right category.

So it certainly remains to be seen whether a codification of English Romani with both its Romani-derived and English-derived components, similar to that of Yiddish, is possible. The drive in the 1970s to create an Anglo-Romani literature, which culminated in the still amateurish and unfocused Acton and Kenrick (1984), left behind fragments of Bible translation, children's books, a few poems, and has largely withered on the vine.

Notwithstanding the speculations above, this book presents an elegant, punctilious and rigorous case study of the knowledge of English Romani of middlingly weU-off English and Welsh Romani businessmen (and their laments about its decline). Over the past ten years several of the interviewees have remarked to this reviewer their appreciation of the way this research has been carried out. The book very clearly raises the outstanding theoretical questions. It wiU be a benchmark for all future work on English Romani; whether that work happens is now largely a matter for the EngUsh Romani community itself, although it would greatly assist to encourage such work if a cheaper edition could be pubUshed.

5. Incidentally, although Matras asserts that the Acton-Kenrick vocabulary was collected with the aid of Manfri Wood, this is true only of very few of the examples in Kenrick's paper in that book The main lexicon of Acton and Kenrick (1984) was based on a collection by the evangelist Tom Wilson and his late father. Items in that lexicon which had not been heard personally by Acton and Kenrick are asterisked as 'archaic'. In fact it is demonstrable that the vocabulary in Wood (1973) was partly plagiarised from lists collected by Acton and Bartlett in 1971. Wood (1973: 122) gives becker as meaning 'fruit'. This is a mistaken transcription of a handwritten list by Bartlett which gave becker as a scrawled 'frog'. In fact, as Acton confirmed with members of both the Cooper and Turner families, becker means 'toad', while 'frog' is jumper. It is interesting to observe that Matras continues the tradition of Borrow, and Acton and Kenrick, in including dubious translations which may serve to trap the plagiarists of the future. I won't spoil things by citing them.


Acton, Thomas and Kenrick, Donald, eds. 1984. Romani rokkeripen todivvus. London: Romanestan Publications

Bakker, Peter, and Cortiade, Marcel, eds. 1991. In the margin of Romani: Gypsy languages in contact. Amsterdam: Institute for General Linguistics

Hancock, Ian Francis, 1984. The social and linguistic development of English Romani. In: Acton, Thomas and Kenrick, Donald, eds. Romani rokkeripen todivvus. London: Romanestan Publications. 89-122.

Matras, Yaron, 2002. Romani: A linguistic introduction. Cambridge: Cambridge University Press.

Sandford, Jeremy, ed. 1975. Gypsies (2nd edn.). London: Abacus/Sphere.

Wood, Manfri Frederick, 1973. In the life of a Romany Gypsy. London: Routledge and Kegan Paul.

Author affiliation:

Thomas Acton is Emeritus Professor of Romani Studies. Address: 22 Northend, Warley, Brentwood, Essex CM14 5LA. E-mail:

The use of this website is subject to the following Terms of Use