Published in British Academy Review, No. 31 (Autumn 2017).
The print version of this article can be downloaded as a PDF file.
Dr Uri Horesh is a British Academy Postdoctoral Fellow, in the Department of Language and Linguistics, at the University of Essex.
When we learn languages at school, even our own native language, the rules of grammar that we are taught may seem rigid and permanent. However, while languages do of course have rules and grammars, the study of linguistics is based on the premise that these rules actually emerge from people’s everyday use of language, rather than being imposed by authors of textbooks and teachers. Consider the oft-recited ‘rule’ of English that forbids ending a sentence with a preposition, a rule many English speakers seldom adhere to (pun intended).
Since the 1960s, a subfield of linguistics, sociolinguistics, has been examining language with three extra insights in mind. The first is that many grammatical rules are variable rules. This is to say that not all members of a speech community speak the same all of the time, and even individual speakers say the same things differently in different situations and to different people. The second insight follows from the first, namely that language changes over time. Today’s speakers of English do not speak like their predecessors in the 16th century. And in most families, people of different generations exhibit different features in their language that distinguish, say, grandparents from their grandchildren. The third is that in addition to language-internal factors – some sounds and words are ‘easier’ to pronounce than others – there are social factors as well that shape the way our grammars are constructed. It is this third insight that I wish to illustrate. In my work, currently supported by a British Academy Postdoctoral Fellowship, I study the effects of social forces on the grammar of Palestinian Arabic. In particular, I have identified three such forces, which are likely to (and in some cases, are proven to) have an effect on variation amongst speakers of the language:
- contact between speakers of different languages
- religious affiliation
Among the speakers of Palestinian Arabic, nearly 2 million are citizens of Israel. Inevitably, the majority of them are bilingual, speaking Hebrew in addition to their native Arabic. I conducted a study in Jaffa, a highly mixed Arab-Jewish town, focusing on the phonology of Arabic – that is, the sound system and pronunciation of the language. My results reveal that several changes in these speakers’ phonology are significantly tied to the degree of contact these Arabic speakers have with Hebrew speakers. Of particular interest here are the Arabic consonants that linguists refer to as pharyngeals. These sounds do not occur in English, and are often referred to in non-specialist parlance as gutturals because they are produced in the nether areas of our vocal tracts.
School years form a crucial time period in people’s lives, during which they learn, among other things, how to interact verbally with their peers. For each of the speakers in my sample I chose to examine the language in which they were taught in primary and secondary schools. It was not surprising to discover that speakers who had been taught in Hebrew-speaking schools – if only for a portion of their educational career – showed a higher tendency to weaken their pronunciation of the pharyngeals. This can easily be explained as a result of the influence of their second language, Modern Hebrew, on their pronunciation of their native Palestinian Arabic. Although Hebrew originally had pharyngeal consonants similar to those found in Arabic, Modern Hebrew has mostly lost the pharyngeal pronunciation of consonants – with the exception of certain speakers of Middle Eastern and North African descent, for whom relics of pharyngeality still exist.
In my current study, I am extending the geographical range of localities from which I draw my speaker sample to include towns and villages further north, where Arabic-speaking Palestinians reside in de facto segregated communities. While these Arabic speakers do often interact with Hebrew speakers in other domains of life – such as work, commerce and higher education – they do not live amongst Hebrew speakers, and therefore contact between the two languages is significantly reduced for them compared to the Jaffa sample. In a preliminary study, in which I compared the Jaffa speakers to Palestinians living in East Jerusalem and Ramallah, both in the West Bank, where contact with Hebrew is even more limited than in segregated communities in Israel, significant differences emerged between the two groups. These differences were apparent both in their use of pharyngeal consonants, and in other domains of language, such as the use of Hebrew loanwords and syntactic constructions that are typical of Hebrew, but not of Arabic. What I expect to unearth in the near future is whether bilingual speakers belonging to different communities within Israel also differ in similar ways.
A good number of previous studies have linked religion with dialect variation in Arabic. In Iraq, for example, the linguist Haim Blanc noted over 50 years ago that Muslim, Christian and Jewish Baghdadis spoke different dialects, divided along religious lines. Professor Clive Holes FBA has found similar patterns with linguistic practices distinguishing between Shi‘i and Sunni speakers – both Muslim groups – in Bahrain.
While Jews (predominantly speakers of Hebrew) do not generally live in the four communities I have begun examining recently – Nazareth, Umm Al-Fahm, Kufur Yasif, and Mghar – we do find different constellations of religious groups in each location. Nazareth is a historically Christian-Palestinian city, though Muslims comprise a majority of its residents nowadays. Umm Al-Fahm has been, and remains, a Muslim-Palestinian community. Kufur Yasif has an interesting mix of a Christian majority, a significant Muslim minority, and a small yet culturally significant group of members of the Druze religious community. In Mghar, Druze form roughly half of the population, with the remaining half split almost evenly between Muslims and Christians.
In my ongoing study in these four Palestinian communities, I am now examining the linguistic divide along religious lines, which manifests itself in several domains of language, such as phonology and lexicon. Interestingly, Christians residing in Galilee, even in relatively small, rural communities, tend to sound ‘urban’ [note 1]. What constitutes this urban-seeming flavour of Christian Palestinian Arabic is described in detail in the next section. It has to do predominantly with the rearrangement of the sound inventory of these speakers’ dialects. Rural Christian Palestinians tend to have fewer phonemes – or distinctive sounds – than rural Muslim Palestinians. For example, in rural Muslim and Druze speech (which is linguistically more representative of older varieties of Arabic), the words thaani ‘second’ and tamir ‘date’ (the fruit) begin with different sounds. Christian speakers from rural areas will usually pronounce both words with an initial [t] sound: taani, tamir.
This divide along religious lines is somewhat puzzling. Its historical origin is unclear, and there is emerging evidence that it is not as clear-cut as previously believed. Data I recently collected in the Christian-majority village of Kufur Yasif show that veteran Muslim and Druze speakers in the village actually do share many phonological features with their Christian counterparts. It is mostly newcomers to the village, who happen to be Muslims, whose speech patterns differ from the old Kufur Yasif norm, which may have mistakenly been attributed to Christians only.
Arabic dialects often exhibit different patterns for urban versus rural dialects. A number of villages in Palestine have expanded and become regional urban centres in recent decades, and in my current project I am examining the effect of this process on the urbanisation of the dialects spoken there as well.
If we look again at the various field sites mentioned above, we see that they differ not only in their religious make-up, but also in their population sizes and municipal statuses (Table 1). Nazareth used to be the only Palestinian community since the establishment of Israel as an independent state in 1948 to have the status of a city. In 1985, Umm Al-Fahm became the first Palestinian village to be formally upgraded from the lesser status of ‘local council’ to that of a city (several others have since followed suit). Kufur Yasif and Mghar are still small and rural enough to have maintained the status of local council.
|Umm Al-Fahm||City (since 1985)||52,000||100%||0%||0%|
Since population size is the predominant criterion used by the Israeli ministry of the interior to determine municipal status, it is not unlikely that Mghar will soon be declared a city as well, despite its rural nature and both its physical and societal structures. By the same token, it may be argued that life in Umm Al-Fahm is more like that in a village, albeit a very large one. In fact, some of my interviewees have said just that, not only in Umm Al-Fahm, but even in Nazareth. Although Nazareth is known to have been an urban centre since at least the early 12th century (it was established as a Catholic diocese in 1108), several residents there have told me that it is more a conglomeration of villages than one cohesive city. But the truth of the matter is that each of the two cities has both urban and rural characteristics. Alongside independent houses densely lined up in steep, narrow allies, are several wide streets and thoroughfares within the city limits. Nazareth has long been a centre for commerce, banking and government services. As one enters Umm Al-Fahm from the main road, one immediately encounters multiple restaurants (frequented by Palestinians and Jewish Israelis alike), industrial zones, shopping centres and government offices.
Then we have the linguistic evidence. Traditionally, Nazarenes have spoken a dialect that is in line with other major Mediterranean cities of the Levant, such as Haifa and Jaffa in Palestine, Beirut in neighbouring Lebanon, and Damascus in Syria. This includes such features as merging interdental consonants (such as the ‘th’ in English ‘this’ and ‘three’) with their dental counterparts (d and t respectively); and the pronunciation of historical /q/, originally a somewhat deeper [k]-like sound, as a glottal stop (like the ‘t’ in many informal British pronunciations of ‘water’ – wa’eh). On the other hand, Umm Al-Fahm’s traditional dialect has the interdentals intact, /q/ is pronounced [k], and /k/ in turn is pronounced like ‘ch’ in English ‘child’.
However, language, as we now know, changes with time. And this is governed, as we also know, by both linguistic and social constraints. New data from Umm Al-Fahm show that younger speakers (born from around 1975 onward) rarely, if ever, pronounce ‘ch’. So words like ke:f ‘how’ is not pronounced che:f by youngsters as it is by (some) older speakers. As with the Jaffa case described earlier, education plays a role here too. I interviewed an older Umm Al-Fahm man (born 1956) with only a partial primary education. He exhibits this and other traditional features of the dialect, much more than his younger, more educated counterparts. But even this man is shifting gradually towards the ‘new’ pronunciation, which resembles that of nearby urban dialects.
In the Northern village of Kufur Yasif, we see a different aspect of the urban/rural dichotomy. On the one hand, this relatively small village has traditionally had an ‘urban-sounding’ dialect, whose phonology is much more like that of Nazareth than that of Umm Al-Fahm. This is probably attributable to the preponderance of Christians in the village, who have been long known to differ in their speech from Muslim and Druze neighbours. On the other hand, Kufur Yasif retains some archaic lexical items, which used to be prevalent in the Levant, specifically in the Horan region of north-central Jordan and southern Syria, but have all but disappeared east and north of the border. One such word is shēle (also shēle bēle) ‘not at all’, which is ubiquitous in today’s Kufur Yasif dialect (in fact, it is present in virtually all of the interviews I have conducted in the village).
Society and language change
Sociolinguists argue both that societal forces affect the way we speak, and that variation in speech can be indicative of trends in social structure, attitude and behaviour.
In the narrow case described above, we can see both elements of this. On the one hand, we see that factors like religion, urbanisation and contact with speakers of other languages and dialects have dictated, or at least contributed to, the various manifestations of speech patterns in Palestinian Arabic. The flipside of this is that we may take the linguistic data at face value, and deduce from it what it means to pronounce a particular sound in a specific way. In other words, while, for instance, pronouncing a [ch] sound is associated with an older, less-educated, rural speaker (probably Muslim), the apparent reversal of this historic process, whereby the historically older [k] is pronounced, carries a meaning of its own, signalling to the hearer that the speaker is (probably) younger, more educated and more attuned to processes of urbanisation, despite hailing from the same village-cum-city as the speaker who pronounces [ch].
Consider a similar feature of British English. Words like ‘cut’ and ‘cup’ are pronounced in different ways in Britain. For some speakers, ‘cut’ rhymes with ‘put’. These are usually people who grew up roughly from Birmingham northward. For southern English speakers (and also, for example, Canadian and American speakers of English), ‘cut’ and ‘cup’ are pronounced with a different vowel than ‘put’ and ‘foot’. It is safe to assume that most Britons will identify a northern speaker by the sound of their speech. This has, in fact, much to do with this very phenomenon (though there are others, of course). But what happens when a northerner is made aware of this feature of their accent, for instance upon taking a job in London or Cambridge? In many cases, this northerner may try to sound ‘less northern’. Sometimes, speakers know what it is about their accent that makes them sound ‘other’, and adjust their speech accordingly. But often people just think they know what they ‘need’ to change, and in fact do an inadequate job of adjusting to their new environment.
And of course this isn’t specific to Arabic or English. Similar stories can be told about virtually any language, at least the languages that have been thoroughly studied in this kind of framework. There has been much discussion in the media about the ‘decay’ of this or that language, or of language in general. But what really happens is that language naturally changes over time – sometimes within very short periods of time. And this happens either due to language-internal forces, or due to social forces, such as wanting to sound more like a particular group of speakers, or wanting to sound less like one’s original speech community. Many times, however, it’s not a matter of either/or, but rather of intricate combinations of linguistic and extralinguistic factors.
The Postdoctoral Fellowship scheme is the British Academy’s flagship funding programme for early career researchers, and is one of the most prestigious and sought-after of its type anywhere in the world.
In October 2017, the Government announced that it will be allocating a further £10 million from its Global Talent Fund to the British Academy to support 40 more British Academy Postdoctoral Fellowships, taking the number we can support starting this year from 45 to 85. Alun Evans, the Academy’s Chief Executive, said: ‘This is tremendous news and shows the confidence that the Government has in the Academy to deliver excellent, cutting-edge research in the humanities and the social sciences.’
1. This was also noted by Blanc in another study of his, focusing on the dialects of northern Palestine.