Navigation – Plan du site

Complexities of speech in Palestinian refugee camps

Fieldwork results and rethinking the theory
Complexités du discours dans les camps de réfugiés palestiniens. Résultats du travail de terrain et réévaluation de la théorie
Nancy Hawker
Traduction(s) :
Complexités du discours dans les camps de réfugiés palestiniens

Texte intégral

1This paper is a report on work in progress. I have recently finished my fieldwork in three refugee camps in the West Bank: Shuafat RC, which is in the Jerusalem Municipal Area, Dheisheh RC, which is in the Bethlehem Governorate, and Tulkarem RC, which is in the north of the West Bank near the Green Line. I have been recording the speech of respondents in these locations to find out whether contact with Hebrew speakers is effecting language variation and change in spoken Palestinian Arabic. Some of the results have challenged my initial assumptions, but I haven’t had the time to reconstruct the theory in light of the findings. Please bear with me at this sensitive stage of the study as I try to sort it out.

2I will first lay out some of the basic theories and methods of sociolinguistic research as I tried to apply them to speech in the camps. Then I will explain my choice of locations and outline what I found there. Finally I will give some examples of findings that demand a re-examination of the initial theories.

3Three basic principles have been established in sociolinguistic research. The first is accommodation, which is the idea that people tend to speak like the people they speak to often. This means that when speakers live in a close-knit community, which is called a “dense and multiplex network”, then the manner of speech of that community will impinge very strongly on the individuals in it. “Dense” means that everybody knows everybody else in the network, and “multiplex” means that they know each other in a variety of roles: so your neighbour might also be your brother and your colleague, for instance. In a more open network the speech of individuals can be influenced by diverse factors from outside the community.

4The second principle is that of gendered difference in the formation of networks. In some societies, women tend to live in more open networks while men form closed networks. This would make women more likely to introduce new, “outside” elements to the speech of the network.

5The third principle is the function of “apparent time” as “real time”. It posits that since an individual’s speech stabilises in early adulthood, the speech of older generations now is a snapshot into the past, into the manner of speech at the time when the speaker was a young adult. This means that if you listen to a 70-year-old, you are more or less hearing the speech of fifty years ago, when you listen to a 40-year-old, you are hearing the speech of twenty years ago, etc. Therefore the “real time” lapse of twenty years can be captured by using the “apparent time” technique of interviewing different generations. This has been tested empirically.

6With these sociolinguistic principles in mind, I set out to record women and men of three generations in Shuafat, Tulkarem and Dheisheh RCs. I defined the generations as “those who remember 1948”, “those who were born around 1967”, and “those who were born just before the First Intifada”. These events are spaced by twenty years, but they also reflect defining moments that changed Palestinian society in ways that bear upon the sociolinguistic study.

7People from the eldest generation in the refugee camps still speak the dialects of the villages where they came from in pre-1948 Palestine. In the Palestinian village economy, tasks on the farm and in the market-place were quite evenly distributed among men and women, though of course certain domestic and communal tasks were gender-specific. In the generation that was formed by this economy one would therefore expect the speech difference between the sexes to be minimal. Interaction with Mandate authorities and the new possibilities this offered – mostly to men – might have pushed some men towards linguistic innovation, but in general the rural sociolinguistic networks would not have been gender-differentiated to the same degree as in the urban setting.

8It was mainly the dispossessed villagers, whose only source of livelihood had been their land, who ended up in refugee camps. Urban refugees who had some capital (in terms of education, professional training or finances) to start a new life did so in the towns that could absorb them. The villagers, however, having no capital (except for the land they had lost), were left with only their labour to sell, and became a floating workforce in search of a wage, and this workforce was strictly gendered. The second generation was born into new communities, refugee camps, where several dialects mixed, and where new networks were formed with new standards of speech. This is the generation that experienced the peak of the mostly male Palestinian workers’ day-migration to jobs in Israel in the 1980s. Therefore I expected greater linguistic differences between men and women among the forty-year-olds than among the first generation, and also some manifestations of Hebrew that wouldn’t have appeared among the eldest speakers.

9The third generation, the “intifada youth”, would be highly affected by restrictions to movement that were introduced during the 1990s and since. Here, I expected a greater difference between the holders of Jerusalem IDs (blue IDs) in Shuafat RC and the holders of green IDs in Dheisheh and Tulkarem RCs. The blue ID holders would still have access to the Hebrew-speaking job market as opposed to their West Bank counterparts, so I expected them to use more Hebrew. Tulkarem workers would still have access to some Israeli jobs due to their proximity to settlement industrial zones, for which it is easier to procure a work permit. In terms of networks, I assumed that the men would have more open networks due to the gendered socio-economic patterns in the camps, and that therefore the men, both of the second and third generations, would be introducing innovative forms of speech into their communities, including perhaps some Hebrew.

10All these assumptions make sense, and fall into place, when one sees networks from a historical-materialist, or economy-driven, perspective. This, in fact, is why I chose the refugee camps as the locations of the study: here the externalisation of labour is the most extreme of any sector of the Palestinian economy. Since the floating workforce is attracted to capital, and because most capital is in Israel, and Israel is the source of Hebrew, it seemed logical to look for Hebrew in the refugee camps.

11There would be other, attitudinal, constraints on the spread of Hebrew in Palestinian speech. Palestinians in refugee camps encourage girls to study more than boys. While women are encouraged to work and bring home a second wage as long as it doesn’t interfere with childcare duties, it is customarily not acceptable for women to work in menial jobs even in the services sector. Women find it more acceptable to engage in office work, or as teachers and nurses, for which some education and training are necessary, while boys, who are expected to support the family, work in menial jobs and are often sent to earn a wage as soon as they are physically able. Educated and professional women’s jobs therefore expose them more to written Arabic, and this could be a source of linguistic innovation in the community. In this case, for these women, it would not be true to say that their sociolinguistic networks are more closed than the men’s, but rather that they are open in a different direction. This occurs often among siblings or among couples: the women work in administrative jobs and the men as workers. However, since my study is focused on the introduction of Hebrew into Palestinian speech, I did not follow this line of inquiry far.

12Most people at this point in the exposé would ask: “And what about the political constraints? Isn’t Hebrew the language of the enemy and, therefore, wouldn’t it be frowned upon to use it in a Palestinian conversation?” The answer is that the sociolinguistic network principle, especially if applied in a historical-materialist manner, does not lend itself to integrating abstract political ideas into the analysis. It is not the abstract categorisation of a language as “enemy language” that limits its use, but the fact that the military occupation and the segregation seem to prevent the formation of dense and multiplex sociolinguistic networks that would include both Hebrew and Arabic speakers.

13Speech is too spontaneous to be governed by abstract ideas. Life in general is too complex to be governed strictly by political principles: during the First Intifada, men who threw stones at Israeli army jeeps at night went to work in Israeli factories in the morning. Finally, the evidence shows that some Palestinians use some Hebrew in their speech regardless of their political stance, and the clearest example of that is the case of political prisoners.

14My recordings in the camps supported most of my assumptions in terms of the findings: I found that men use Hebrew more than women, that workers use technical terms in Hebrew, and that this was especially true of middle-aged men in Dheisheh RC and of younger men in Shuafat RC, where Hebrew was more common in general. In Tulkarem RC, the gender differentiation was moderated as the absence of men – a high number of whom had been killed or imprisoned during the Intifadas – forced women to take up menial jobs easily accessible just across the Green Line in agriculture and cleaning. Therefore, the notion that individual Hebrew items were introduced into speech in the camps by the medium of the Palestinian day-migrant workers in Israel, was supported. However, casual observations also uncovered some surprising uses of Hebrew that challenge the historical-materialist view of networks. Let me first say a few more words on the expected findings.

15In the workers’ lingo, be it in Shuafat or Dheisheh, certain technical terms are more often referred to in Hebrew. One is klūš (pronounced tlūš in the original Hebrew), meaning “payment slip”, which denotes a formally regulated job rather than a “cash-in-hand” arrangement. Another is manof, which means “construction crane”, and mazleg “forklift”. Some words have spread into general use, such as mazgan “air conditioning”, az’aka, meaning “security alarm” and ramzor, used interchangeably with its Arabic equivalent, išārāt, for “traffic lights”. Another technical term that evidently comes from contact with the Israeli army is maxšīr “walkie-talkie”, but used also in civilian contexts. Widespread are words relating to the military and bureaucratic constraints on Palestinian life: ubiquitous is masūm, “checkpoint”, which is the only word used by all the respondents in my sample, rarer is išūr, “permit”, and te’uda, “ID card”, which are used interchangeably with the Arabic equivalents.

16In Shuafat, due to the residents’ access to Israeli social services, some Hebrew words relating to them are widely used: kupat xolim, “health service” and hitxayvut, “contract requiring monthly social payments”. In addition to this, Shuafat and Tulkarem workers, men and women, will use some words that are not used in Dheisheh speech: among them are xōfesh “day off”, maskōret “wage”, binyan “construction” and nikayon “cleaning”. In contrast, workers in Dheisheh would use the Arabic equivalents ‘otle, rātib, buna and tanīf. Construction work and cleaning services are the most common Palestinian employment in Israel. Other words denote Israeli products popular in the Palestinian market: uga “doughy cake or pastry”, naknik “sausage” and šamēnet “soured cream”.   

17These individual items constitute loanwords that are used for ease of communication. Their use reflects a certain type of encounter, translated into language contact, similar to loanwords that appear in spoken Arabic from Turkish and other historically dominant languages such as English or French. In fact, these types of borrowings are common in all languages, since it is normal for languages in contact to borrow from each other: one need but think of French words in English. The question for linguists studying language change is whether the Hebrew borrowings in Palestinian Arabic will last into the speech of future generations. Judging by the evidence in the speech of the youngest generation recorded in my fieldwork, it is likely that some of these words will remain.  

18On the other hand, it seems that there are relatively few loanwords and that their spread is limited. The volume of these loanwords and their type – only nouns – are not quantitatively or qualitatively comparable to other cases of colonial language contact, such as the French in Algeria or the English in India, where the administration relied on a local elite educated in the colonists’ language, which took over governance after independence. In the cases of Dheisheh, Tulkarem and Shuafat RCs, it would seem that in the sociolinguistic networks that do include a Hebrew speaker, as the foreman on a construction site for instance, the link to that speaker is uniplex (as opposed to multiplex), i.e. the speaker is only known in that one role. Moreover, the network that involves that construction foreman in the example is open, i.e. only a few people in the community are in contact with him and act as intermediates between him and the dense community network. For these reasons, not much Hebrew can spread into Palestinian speech.

19Now I will turn to the evidence that neatly fits neither with network theory nor with the assumptions I had initially made. This is the introduction of entire Hebrew phrases in Palestinian speech for stylistic effect. While the loanwords’ meaning is literal and communication with them is understood prima facie, the use of these phrases depends on connotations of power, status and prestige – though as we will see, these connotations can also be reversed and made ironic. Such features of language use are commonly associated with the alternation of codes, be they different languages or formal and casual registers of the same language, when two codes have been in prolonged contact and speakers are fairly fluent in both, and the codes have established social connotations. This practice is known as codeswitching. It has been found in very diverse contexts, such as Los Angeles teenagers’ use of Spanish to demarcate who is “in” and who is “out” of certain groups, or the authority of German in the Hungarian-speaking part of Austria when disciplining a child or choosing a spouse.

20Again, my observations indicate that Hebrew-Arabic codeswitching is very limited in the refugee camps I studied. However, there are limits to what I can observe as a foreign researcher, because these instances only appear outside interview settings, and so the method of data collection could skew the evidence. I would classify the codeswitching I did have the opportunity to observe into three broad categories: ironic power speech, reported untranslated speech, and “cool slang”. The first two are to be heard in all three refugee camps, the last only in Shuafat RC, where the proximity to Jerusalem seems to affect the prestige value system.

21Let me give some examples of each category. What I call “ironic power speech” always involves an element of teasing, and of using Hebrew’s connotations of Israeli dominance in paradoxical situations. It would seem that the speakers who use it are known to be witty, and have sufficient social status, or at least some advantage over the addressee, to “get away” with using Hebrew in this way. In one example, a man shouts across the table to a younger man who is a member of the youth group the first man is organising, after a communal meal in Shuafat Refugee Camp: Efo ha-kafe, xabībi? “Where is the coffee, man?” Axarei ōxel tsarix kafe. “One must have coffee after a meal.” The younger man smiles perplexedly and asks in Arabic, “Should I bring some coffee?” The conversation continues politely after that, and the youth group leader gets his coffee. Using Hebrew in this case had the effect of sounding rude and bossy, but in a jokey way and so with no offense taken. In general, ironic power speech is a witticism that reinforces the speaker’s advantageous social status.

22Another example of this involves the word menahēl “boss”. Usually, it is used to simply mean that – the manager of the workplace – as an equivalent to the Arabic mu‘allim. One sharp-tongued prominent woman in Shuafat RC, however, used it to describe her brother-in-law who had had the impertinence of snubbing her sister (his wife) by marrying several wives on top of her. She was outraged that an educated man would do this to a wife “from a good family”, and summed up the baseness of his behaviour by commenting huwe il-menahēl li-kbīr “he’s the big dick” (literally “big boss”).

23The “reported untranslated speech” category is quite straightforward: it simply means that when relating an anecdote, a speaker quotes what was said in Hebrew with the expectation that the listeners will understand, without it being necessary to translate into Arabic. I witnessed an exchange between two young men recounting their experiences at checkpoints, where one related how a guard had asked him ma šlomxa? “How are you?” and then had punched him. However, other, less hostile reported speech that was evidently originally made in Hebrew, is quoted in Arabic. This type of Hebrew codeswitching is done for stylistic effect and to give the account verisimilitude.

24The last category, the “cool slang”, is where I have the least evidence and where more fieldwork is needed – preferably to be done by someone young and local who can get closer to the necessary data. This slang is the kind of thing I overhear in Jerusalem from Palestinian youth sitting at the back of buses, where I cannot record. They talk of fashion, mobile phones and where to watch girls. In four of the seven families where I did interviews in Shuafat RC I found younger members of the family, usually in their late teens, who seemed to be more interested than the others in mobile phone ringtones, flashy accessories and consumer-culture symbols in general. These were generally youths who chose their partners and married young (both men and women). They had not completed secondary education and worked hard in menial jobs in Israel. I was discouraged from interviewing them by other members of the family, with the comment, “they aren’t educated” or “they can’t speak seriously”. In the few conversations I managed to have with these youths, they found it difficult to avoid Hebrew expressions: when having the choice between Hebrew and Arabic equivalents from the list in the paragraphs above, Hebrew would be used.

25The most common marker of this type of speech is bsēder “okay”, which is used as a tag at the end of a sentence in the example: minuff hōn u minrū ‘a-l-mat‘am, bsēder? “we’ll park here and go to the restaurant, ok?” More rarely, it is said as an answer to ma nišma?, Hebrew for “what’s up?” Bsēder is a particularly easy individual item to adopt and spread as a marker of slang speech, because it is separated syntactically from the Arabic sentence by a comma, it is used in isolation and not in relation to other grammatical items in the sentence.

26I will give you one very puzzling example that I witnessed at a child’s birthday party in Shuafat RC. The grandmother of the child in question had refused to take part in the festivities on the grounds that it had been ruled sinful by some religious edict. Finally she was persuaded to join the fun by the young and trendy parents (who fit my crude description of consumerist youth outlined above). At the end of the party, the father asked the grandmother in Arabic, kīf kān? “How was it?” After a moment’s reflection she answered yōfi “fine” in informal Hebrew. He seemed as taken aback as I was to hear an elderly religious lady speak slang, but gathered his wits enough to respond tōfi, completing the slang expression yōfi tōfi, meaning “okey dokey”. My interpretation is that the woman’s choice of language implied that the party was fine in that trendy, modern, perhaps unislamic, way that the young couple aspired to. He confirmed that he understood that implication when he answered equally in Hebrew slang. This example challenges my initial assumptions most radically, because women from the first generation who have never worked in Israel should have absorbed the least Hebrew, especially since in this case the woman’s lifestyle seems to favour conservatism and religiosity, not shopping and hairstyles!

27The problem these examples pose for the network theory is that their meaningfulness relies very much on the political and social connotations of Hebrew that had been deemed “too abstract” earlier in this paper. True, it is not a simple “enemy versus friendly language” dichotomy, but it does bear upon identity politics and social models. You can find, as I did in my sample, two young men, brothers close to each other in age, both working in Israel, both having dropped out of school at sixteen, living in the same house and frequenting overlapping networks of acquaintances in the family and in their neighbourhood, but one being interested in politics and the other in shoes, and the latter using a lot more Hebrew. It means that the historical-materialist analysis of networks is not an exhaustive explanation for language variation.

28My suspicion is that the problem with sociolinguistic network analysis, especially in its historical-materialist application, is that it is too deterministic. There needs to be more space in the account for human agency, and for the role of discourse. Evidently, in the Palestinian speech spectrum, using Hebrew indicates not only the language of the workplace, of technology and of bureaucracy but also the discourse of power, which can be subverted with irony, and the language of consumerism. A speaker can make linguistic choices depending on what discourse he or she embraces. The speaker thus retains some agency despite having to work in Israel to earn a living, despite living in a refugee camp and despite living under military occupation. The challenge for my study is to bring the discursive aspects into the analysis.

Haut de page

Pour citer cet article

Référence électronique

Nancy Hawker, « Complexities of speech in Palestinian refugee camps », Bulletin du Centre de recherche français à Jérusalem [En ligne], 21 | 2010, mis en ligne le 01 mars 2011, Consulté le 25 mars 2017. URL :

Haut de page


Nancy Hawker

DPhil Candidate, University of Oxford

Haut de page

Droits d’auteur

© Bulletin du Centre de recherche français à Jérusalem

Haut de page